Home

MTS Operators' Manual - Deep Blue

1. PKW PLO PMW PQ PQW PRLOW PUW Q QLL SAT SPAREJT TAPEQ TAPEQL TFB TEP TMB TMP UAM UBJ 485 Disk paging weight Paging load low threshold Drum paging weight Mean page queue length last interval CPU queue weight BJPMOD paging rate low threshold CPU Utilization weight Mean CPU queue length last interval Queue length limit used by BJPMOD System saturated flag 0 gt off 1 gt on Desired minimum number of spare job taples Permits operator to use tape queueing 0 off 1 on Tape queue length Too few big jobs threshold Too few privileged jobs not currently used Too many big job threshold Too many privileged jobs threshold Unattended mode flag 0 gt off 1 gt on Upper big job limit real pages SYSTEMSTATUS 486 TAPES TAPE HANDLING Users wishing to submit tapes must first run tapesubmit from a terminal and then place identifying sticky labels on the reel One of the labels contains spaces for two pieces of information 1 the external tape ID 1 10 characters and 2 the receipt code which is assigned by tapesubmit This label looks something like the following fof Receipt If the tape is to be initialized a green sticker with an I in the center must be placed on the reel If the user wishes to keep the tape from being written on two NO RING stickers should be placed on the reel one on the front a
2. ABNORMAL SITUATIONS SUMMARY 312 Failure in the instruction processing part of the machine Printed when the processing damage bit in the machine check interrupt code is on and both key error and storage data failure are off kMCHK VM REFRESH RETRY Uncorrectable storage error on a page for which we have good copy paged out The sytem will attempt to recover the page from the paging device MCHK INTERVAL TIMER ERROR Interval timer location 50 failure MCHK ENVIRONMENTAL WARNING Power air conditioning or other similar failure Printed when the warning bit in the machine check interrupt code is on MCHK PROGRAM INTERRUPT IN MACHINE CHECK HANDLER Program interrupt while processing a machine check interupt I O ERROR PATH pppp CSW cccccece cccecece LCL 11111111 Interface control check channel control check or channel data check on path pppp The CSW CHANNEL STATUS WORD and LCL LIMITED CHANNEL LOGOUT formats are described on the system 370 reference card SOFTWARE FAILURE A SUPERDUMP just occurred NO LOG AREA AVATLABLE The machine check or channel error being processed will not be logged since too many logs are already pending Recovery from the error will not be effected HRE IPLHF The system is down ABNORMAL SITUATIONS SUMMARY 313 1 Punch Errors PCHn Machine ERROR CARD PUNCH STATUS UUIIFFSS Program UU Program flag values are in
3. 123 OK Task Termination Message On Off ooooooooooooooooooo 125 ONLINE Make Device CU Etc available oooooooooooooo 126 OPERATOR Start 3277 As Operator Console oooo oooooooooooo o 128 PDP Initiate Paging Drum Processor ooooooooo ooo ooo ooo eee eee 130 PDROP Remove Paging Devices Form PDP 2c eee ee eee eel 131 PGET Add Paging Devices TO PDP Libia ds dd 132 PN lt s hoadsiwPN UGS BUuFTer ec A A ae S 134 Pili toads PLL UCS Butter tia e a Nea a Eon dele ea aa e a Sige ROR Bag aed L35 ON yec Loads ON UCS BUPROL sous ices e Eee aire sabe a She a e Gaels r a woe 136 SEv a LST 12 Words in Device Tabi sessed seed Saeed oR Sc ete eee ao eased wed 138 SHUTDOWN cShUtSs Down Terminales i wes mia aaae aa a ad 139 12 SIGNONM Signon Message TO User ooooooooooooooooooooooooooooo 141 SOFTCHK Controls Recording Of Recoverable Machine Checks A Paws E A eee RS A A a Fd 142 STARTUP Cancels Effect Of SHUTDOWN ooooooooooooooooooooo 143 STAT Record Syst Stats On Tape o o oooooooooooooooooooooo 144 STATSW Dynamic Change To STAT Parameters 1 145 STOP FOrerbly Terminate A TASK paana e aaa a ead eee eed L46 TABLMOD Dis Or Mod Info In Shared File Table 1 147 Tape Commands OPeE At A AA WANS A A e A 149 TASKS sa Check Status Of A TASK cuna a id 150 TMT
4. SYSTEM SHUTDOWN POWER DOWN PROCEDURES 64 00002 00002 07861 SYSTEM SHUTDOWN AND START UP PROCEDURES 23 50 59 OPERATOR T 7859 23 51 00 OPERATOR Specified job s not found 23 50 48 TABLMOD Cancelled 23 51 56 07862 BLAST 1492 01492 23 51 56 MTS Stopped BLST 98 024 14 626 23 52 03 07863 SHUTDOWN ALL 07863 23 52 04 SHUTDOWN The NETP MM job reports that 1 ports were affected 6 After running SDM use the command T M to see if all MTS jobs have terminated you can ignore any jobs that don t have a valid ccid or that have the word BUSY Take job dumps of jobs that won t STOP and log them using DMP or check with a systems programmer If SDM cannot be run see the descrition of the SHUTDOWN job program NEVER SKIP THIS STEP If you are shutting UM MTS down i e if MTS is running on the 3090 go to the ES3090 300J console for UM and press the STOP key Wait until all of the CPUs on the UM side of the machine have stopped before proceeding If you are shutting UB MTS down i e if MTS is running under control of VM go to the VM operators console and issue the command FORCE UB MTS VM should respond COMMAND COMPLETE At this point MTS is down and you could reload Wait for the CPU status to show STOP or WAIT STOP If after 15 seconds or so the machine still hasn t stopped make a note of the fact for IBM and go ahead At this point
5. TABLE OF OPTIONS TO REQUEST TO OMIT BY OPTION OPTION DEFAULT TL NTL NTL TL VALUE VALUE 1 32767 EL NEL NEL EL VALUE VALUE 1 32767 EP NEP EP CP NCP NCP PP NPP NPP PP LEVEL LEVEL 0 3 PR NPR NPR FE NFE FE MI NMI NMI RE NRE NRE EXT OLTS 420 PLOTTING SYSTEM NOTE There is a ring binder below the operator s console which contains two techical manuals to be referred to for pen cleaning assembling etc One of the manuals is for the CalComp model 1051 The other two section manual is for the model 936 The second section of the 936 manual Pen and Paper Users Guide shows as examples pens that we don t use For information on deferred plots read the material in the section entitled CCD For information on plot tape initialization descrition of CCL in this manual the For information on the length of the impending plot session and the number of user tapes involved CCT see To print indentifying information about a plot tape actually from it including its name e g TUE 1000 when it was last written and the minimum time to reuse see PLB in this manual Note To get a copy of the current plot receipt sheet sys plotreceipt to a terminal or printer Plot Times copy Plot sessions are started automatically by the clock watcher program These sessions are started at the following times Plots with spec
6. 559 IBM ES3090 300J Water Temp Flow 17Mar92 Note The water flow can influence the temperature of the IBM water even if customer water is at ideal temperature Usually home EE IBM water 20 9 22 5 Celsius Customer Water 4 11 Celsius Flow Rate 70 80 Litres per minute if the water temp is way off the system will call IBM water temp is bad call IBM service after checking customer chiller water temp If customer water temp call the A C people is bad IBM ES3090 300J WATER TEMP FLOW 560 Xerox Page Printer Notes Powering On and Software Loading Procedures 1 Press the LOCAL button on the system controller panel 2 Press the system POWER ON button on the system controller panel 3 If the user interface screen is blank press the ON OFF toggle switch at the back of the user interface console to the ON position The POWER indicator lights on the system controller and system diagnostics appear on the screen for approximately 30 seconds A and a blinking yellow cursor appear on the display If the system is configured so that the user interface is not directly powered by the 9790 only a blinking yellow cursor appears 4 Press the CAPS LOCK key so that the red light turns on and enter B for BOOT on the keyboard to load the operating software The warm up cycle is complete only after operating software loads When loading is complete the display shows
7. PLOTTING DEFERRED PLOTS Run PLT in the usual way Note that the deferred tape is added PLOTTING SYSTEM 430 to so that you may not want to plot all of the plots The WHICH TAPE message from PLT will say that the deferred tape is being plotted Example MTS PLOTS FROM DEFERRED TAPE TAPE WRITTEN 13 46 15 03 11 78 MTS 1 BLAK 4 2 BLAK 3 3 BLAK 6 MTS ok TO STOP DEFERRING USER PLOTS This should be done to allow a users plots to be processed in the normal way after they have been deferred Start the MTS job in the file CCD You should be prompted for a ccid Enter all ccids for which you wish plots to no longer be deferred one per line Terminate the list with an end of file CNCL Example 02799 21 10 21 mts ccd 02799 21 10 24 MTS ENTER SIGNON ID 02799 21 10 24 MTS mts 02799 21 10 29 MTS PLOTS FOR MTS WILL BE PROCESSED NORMALLY 02799 21 10 29 MTS sys 02799 21 10 30 MTS PLOTS FOR SYS WILL BE PROCCESSED NORMALLY 02799 21 10 30 MTS CANCELLED This is of course exactly the program which defers plots It just changes the status of any id you give it as 02805 21 10 54 02805 MTS ccd 02805 21 12 45 MTS ENTER SIGNON ID 02805 21 12 46 MTS sys 02805 21 12 56 MTS PLOTS FOR SYS WILL BE DEFERRED 02805 21 12 56 sys 02805 21 13 00 MTS PLOTS FOR SYS WILL BE PROCESSED NORMALLY 02805 21 13 00 MTS CANCEL
8. d pce at mconpce PCE AT 246948 1052 PCESAVEA 01000000 00000000 REGISTERS 13 gt 12 00246948 601E5950 90109144 02016214 002379F8 00000000 00238788 00000000 00000000 00000003 00000000 00000000 00000000 00000000 001C7018 001E5724 PCEEPSAV 00000000 PCENEXT 0024699C PCEEWF WORK PCEID 1052 The field PCEEWF shows that the PCE is waiting for WORK This shouldn t normally be the case The register 15 value above indicates the restart address for the PCE and can be used to check the HASP listing to see where the wait really is xcalc 1c9144 hasp 00002144 This turns out to be near the beginning of the routine MBCPUGET which gets the next logical record in this case the console command from a CPU type remote The PCE is stopped at the instruction WAIT WORK waiting for the command to appear in its buffer However this is never supposed to happen i e the record should already be there since it suspends the one and only remote console PCE Check the buffer from the field DCTBUFAD above res prot off it s protected so do this first dis buffer at lecd40 long TPbuffer at 1ECD40 TOBFLAG1 42 IOBFLAG2 00 IOBSENSO 00 IOBSENS1 00 IOBECBCC 00 IOBECBPT 000000 IOBFLAG3 00 IOBCSW 00000000 000000 TIOBSIOCC 00 IOBSTART 24C974 IOBDCBPT 00000000 IOBREPM 00 IOBRESTR 24C974 TPBMXREC 00 TPMCONCH 000000 TPBLCCC 11 TPBLCCAD 1ECEEA TPBRECNT C5 TPBFDATA 1ECDC5 BUFECBCC 00 BUFACHN 000000 BUFTYPE 80 BUFDCTE 24FB1
9. JOBS PROGRAMS 117 JOBS To start a task that 1 gathers statistics about the current system load and does load levelling on the system using these gathered statistics or 2 display the statistics being gathered JOBS XXXX If the xxxx parameter is not specified the JOBS program will read commands from the console Otherwise the xxxx is an initial command JOBS LOA prints out the system load statistics on the operator s console every 20 seconds use GOOSE or STOP jobs to get rid of See the section on System Status below and MTS Vol I for a complete list of JOBS commands This job is not normally used from the 3270 operator s console the SSRTN SYSTEM STATUS commands are used instead JOB PROGRAMS JOBS LOA 118 MOD NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES NOTES JOB PROGRAMS MOD JOB PROGRAMS MOD To modify a location in real memory MOD xxxxxx yyyyyy ZZZZZZ This command is used to modify a real memory location or a string of contiguous real memory locations The first parameter i e xxxxxx t tyyyyyy is evaluated as a real memory address It may be a hexadecimal constant or a hexadecimal expression i e hexadecimal constants separated by plus signs The real memory address so evaluated will be the location at which modification begins The zzzzzz parameter is a variable length hexadecimal string that denotes
10. SPACE AVAILABLE to summarize the total amount of space and the total amount of free space in the system STOP END DONE to end the run An end of file may also be used A single command may be specified in the PAR field of the RUN command For those commands that accept a volume name a list of volume names may also be given eg LIST MIS001 MTS002 EXAMPLES POLR MTS009 ADD MTS099 NODMGR MTS008 OPERATOR FILES DSK 210 NAME FUNCTION USAGE REMARKS OPERATOR FILES DSN Starts the user HASP status display at NUBS This display is a CONRAC monitor driven from the video output of an ANN ARBOR TERMINAL CONTROLLER which is hardwired to an SCP Is normally started automatically by MNS Requires only mts dsn Takes no parameters If the user screen seems not to be functioning e g the cursor isn t moving and the time displayed on the screen is behind the following actions may remedy the situation Check and see if there is a job on CB7D If there is none start one by typing mts dsn If there is a job present check with a systems person OPERATOR FILES DSN NAME FUNCTION OPERATOR FILES 211 DSO This program prints the total number of disk pages the number of free disk pages and the percent of free disk space available You also may initiate DSK or run sys dsk using a LIST OPERATOR FILES DSO 212 DWB NAME FUNCTION USAGE N
11. The online offline switch is in the offline position Switch it to online and rerun HUH lt name gt does not respond Check and reload if necessary Either lt name gt is powered off ora serious error has occurred If the power is off it should be turned on and HUH rerun If the power is on then the PCP has probably crashed in such a way that the High Core monitor is not running To see if this is the case select lt name gt at the Super Console terminals and type HELLO If there is no response the machine has crashed and must be restarted from the front panel of the PCP If it says lt name gt monitor then high core is running and xDL should be run If it says name gt PCP MINOS version xxxx or PCP MCS then the operating system is running but the channel interface is not operating properly If the latter happens doa UNITS D lt name gt OP S on MTS and try to cantact a UMnet UMnet Firefighter If none are available restart from the front panel and run xDL Another somewhat unlikely possibility when this message appears is that the control unit the path from the control unit to the channel the channel itself or some combination is logically offline The UNITS command UNITS D xxOP where xx is CA CB CC or MN should provide the information necessary to see if this is the case If it is the ONLINE job can be used to place OPERATOR FILES 253 things online Message lt
12. Type starting input tape number syst 2 meaning mts001 tape 2 or in the case of DEADFILES 1 for the first original tape and 2 for the second original tape etc The following questions are asked only if a digit other than 1 was typed in the above command ARE WE A STARTING WITH A NEW OUTPUT TAPE SYST OR B CONTINUING IN THE MIDDLE OF AN OUTPUT TAPE SYST ENTER A OR B FROM ABOVE SECTION OPER OPERATOR FILES FSC 216 OPERATOR FILES Type starting output tape number syst uqu oper Where is input tape number 2 for mts001 ring out syst T9IOX oper Where is output tape number 1 for mts001 ring in syst T9OX oper FSC is normally terminated with an end of file Copying of weekly filesave tapes which produces a list is usually done in two parts producing two consecutinve listings This six week filesave copy produces only one continuous list See Filesave Schedule and Tape Storage Comments If an error occurs on the output tape the operator starts the job over and the tapes will space out to where the error occurred for a retry If an error occurs on an input tape record FSC will mention this and copy the error as is No operator action is necessary FSC will copy labeled input tapes to non labeled output tapes If the output tape is labeled incorrectly the NOCHECK option T9OX NOCHECK option must be used if one actually does want to
13. For just draw the x coordinate for each line to be drawn is requested this is the distance in inches entered as a floating point number always include the decimal point from the left side of the sign Examples of each format are on the following pages Let the systems programmer in charge of the plot routines currently Gail Lift know what you want implemented under other or if you need more characters in the alphabet e g if it doesn t currently have a semicolon or if you don t like the way it works or COMPUTING CENTER HOURS 392 The following example produces 4 signs They are shown in order at 1 4 size on the next page HR W035 HOLIDAY 1000 EXECUTION BEGINS WHICH FORMAT 4 LINE 3 LINE REPEAT OTHER 4 SET UP THE PLOTTER ENTER TITLE LINE uc off OK A Title CENTER LINE UP OR JUST DRAW LINES c TEXT FOR LINE 1 first line TEXT FOR LINE 2 second line TEXT FOR LINE 3 third line TEXT FOR LINE 4 fourth line WHICH FORMAT 4 LINE 3 LINE REPEAT OTHER 4 SET UP THE PLOTTER ENTER TITLE LINE Title CENTER LINEUP OR JUST DRAW LINES 1 LINE TO BE CENTERED third line centered WHICH LINE IS THAT 3 1 TEXT FOR LINE 1 first line TEXT FOR LINE 2 second line TEXT FOR LINE 4 fourth line WHICH FORMAT 4 LINE 3 LINE REPEAT OTHER SET UP THE PLOTTER WHICH FORMAT 4 LINE 3 LINE REPEAT OTHER
14. P16 Starts 16 PAG jobs mts p16 267 OPERATOR FILES P16 268 RCP NAME FUNCTION FEATURES LIMITATIONS USAGE ERROR MESSAGES EXAMPLE OPERATOR FILES RCP OPERATOR FILES RCP To produce MTS S8 cards Produces receipt and deck or just deck S8 cards Start and end number set at run time Punch must be drained 1 Make sure you have a free punch E g use a free punch or drain pchx 2 Put blank S8 cards in hopper and ready punch At console type MTS RCP 4 Program will ask for a DO YOU WANT RECEIPT CARD YES DECK CARDS NO b THE PUNCH TO USE i e pchx c A SIX DIGIT STARTING NUMBER i e 110001 da A SIX DIGIT ENDING NUMBER i e 110500 Ww If you ask for 500 cards you will get 1 500 deck cards in ascending order 2 500 each of receipt and deck cards in ascending order PUNCH IS BUSY NOT OPERATIONAL OR NOT A PUNCH RESPONSE WAS NOT YES OR NO NUMBER JUST ENTERED CONTAINS ILLEGAL CHARACTERS OR IS NOT SIX DIGITS LONG FIRST NUMBER IS LARGER THAN SECOND drain pchx mts rcp MTS MTS RECEIPT CARD PROGRAM MTS ENTER THE NAME OF THE PUNCH MTS gt pchx MTS ENTER YES IF RECEIPT CARDS WANTED NO IF ONLY DECK CARDS WANTED MTS no MTS ENTER SIX DIGIT STARTING RECEIPT NUMBER MTS 300000 MTS ENTER SIX DIGIT ENDING RECEIPT NUMBER MTS 300010 OPERATOR FILES 269 MTS ALL RECEIPT CARDS HAV
15. When UB MTS runs under VM XA it is very important that no tape drives be attached to UB MTS except for the ones that UB MTS is using at the moment If a tape drive is attached to UB MTS but in use from the UM MTS system it is possible that VM XA could cause the tape to be rewound unexpectedly this would obviously cause problems for the UM MTS user of the tape To avoid problems of this sort it is necessary to issue an explicit ATTACH or DETACH command at the VM XA operator console to make a tape drive available or unavailable to UB MTS These commands are issued automatically by UB MTS for tapes that are normally mounted however the commands must be issued by the operator under other circumstances For example it will be necessary to issue explicit ATTACH and DETACH commands in taking a UB MTS superdump VM XA refers to tape drives by number rather than by name as does MTS The following table lists the MTS name and the VM XA device number for each tape drive the device number must be specified on the ATTACH and DETACH commands as shown below Name Number Name Number Name Number Name Number T900 0680 T909 0689 T922 1682 T947 1087 T901 0681 T910 068A T923 1683 T948 1088 T902 0682 T911 068B T940 1080 T949 1089 T903 0683 T912 068C T941 1081 T950 108A T904 0684 T913 068D T942 1082 T951 108B T905 0685 T914 068E T943 1083 T952 108C T906 0686 T915 068F T944 1084 T953 108D T907 0697 T920 1680 T945 1085 T954 108E T908 0698 T921 1681
16. power open the side panel and follow the instructions Next type t11 ptr2 echo at the operators console or using the system ststus oper command This not only loads the UNIVERSAL CHARACTER SET UCS buffer but also sets the carriage tape and echos the contents of the UCS buffer on the printer so the paper must be alligned at the top of the page before the T11 job is used If a power surge occurs that stops the 3403 printer but doesn t affect MTS drain the 3403 as soon as possible Reload the UCS buffer using the T11 job Next start the printer A power off and on may be necessary here If the 2501 card reader fails to come up after a power interruption open the back panel turn the LOCAL REMOTE switch to LOCAL and power the reader on Tf the LOCAL REMOTE switch was in REMOTE set the switch back to that position Tf the 2520 punch loses power open the back panel and place the LOCAL REMOTE switch in LOCAL use the POWER ON switch and put the punch back in REMOTE INTEL 3805 3825 If power has been lost to the INTEL 3805 or 3825 and they are in REMOTE status with the ENABLE switch on a warm start will be done automatically Tf POWER FLUCTUATIONS amp FAILURES RECOVERY 90 SYSTEM SHUTDOWN AND START UP PROCEDURES either of these boxes was in LOCAL mode then they must be manually powered up anda manual warm start should be executed see the section entitled Virtual Memory earl
17. 22 431 PLB Displays Identifying Information About A Plot Tape 432 PLP3 6 Andi PUTS T a5 veh eo ate ta wae A A ALA A EA a 433 Message TO USER ti A Ga A a Shades ia 435 Mutt Pen SPOS dad AAA A od ra 435 REE P TOE TAG dr A A tada 437 ATESNTELOM THESCTEUD ES A NA A AA A AR 437 Tape LI O Trouble issiro iso o eS EAS 6 ebb EA ESE ORE SS ia 438 PES anecs dae A A ee aoe he oS es oS Sw ee aS 438 OQut Of Paper SwRtch Mode DIG ss 438 NomAQueued PLOES 4 s Sa eee Ii 439 Testing PFOGMAM e esie ase eRe ae ae Bee ee OMS DRA ee ORS ct 440 Specifi Test Descriptions ceser ea ii 441 PEN Seaan is oe ores eevee hehe Ns NS DEE a A a a a aa acne 442 SDN ee ect A uA ide 442 DOTS tas ta he eS cee A la ta Mae le ee eee caja 442 Ta NENG A O A A we sheen ace te ee oP 442 GRIDS eke Ste Sik Sorta tt hte cee RA Sir Sl Ss RMS cael BE Sete Ei cs ae Ml AA Sie esl e 442 CHARS A a A mete hey e A cd mesons AI ADA A 443 LOUIS a A Rice te A wee eR A aah a A we a SA eee Rew deed eens 443 STARE A A ss laa 443 NOW ES AEA a a es aut cet kth Ded 444 BISA CK Se oie steno eta ot tol e s ena Mees ou Seite arias bias E ee id 444 TAS Hess acs A ec id a A ad Slate des 444 ERTS Seca se Menten aires se ates am A a a erehane ote ene 5 ane ha isis 445 SAMTES LS a A aN ae aie Nas 445 PM2 An STC Error Analizing Program oooooooooooooooooooo 446 Holiday Rates dativo a 451 SEAN A A SS cae cle Rt RE LAR A Ad 454 Security Clerk ui oA Eee wee esd a Eo
18. FUNCTION To place a shutdown message in PCP s attached to the UB system This procedure is used to enter a one line UMnet UMnet message which will be printed for any user attemting to access which ever system is down It will not disconnect any users currently connected to the UB or UM systems One message line is read from the console terminated by a null line or an end of file and placed in the PCP s The shutdown facility is then activated Any previous message is replaced by the new message USAGE mts sdb REMARKS This procedure should normally be run after SHUTDOWN ALL has been entered at the console and you have checked to see that all PCP jobs have stopped Use t t mnet to verify this SDB will skip any processors that are logically offline but may hang if any is Online but not running SDB shutdown message should normally be OPERATOR FILES SDB OPERATOR FILES 285 run after SHUTDOWN ALL or SHUTDOWN PART has been entered and all UMnet MERIT jobs have stopped Use t t mnet to check Note SDB may be run on the UM console if necessary OPERATOR FILES SDB 286 SDM OPERATOR FILES NOTE SDM should be run after the SHUTDOWN ALL COMPLETE appears SDM can be run on the UB system if the UM system is already down Also see SDB NAME SDM NOTE If SDM cannot be run for some reason go to a network terminal and at which host type HG signon ccid
19. ONLINE FBOx PGET FBOx 13 Pass the information recorded in steps 2 and 3 on to INTEL VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 75 The PDP program is restartable at any time if the same devices are used or only when no MTS jobs running if fewer devices are used Restarting PDP 1 Do t n pdp to find the task number of pdp where x is the current unit number of a fixed head file Example t n pdp 2 Stop Task of PDP if still active 3 Enter pdp or whatever devices desired and you should be back in business Example 00003 22 14 01 pdp error 300000 01 40000000FFO00 0000004C 07 00003 22 14 09 pdp ff00 not ready 00107 22 24 47 t d fhf0 00107 22 24 50 00004 PDP 027000 DORMANT D341 FFOO ff08 00108 22 25 57 stop 4 00004 22 26 02 PDP stopped KILL 35 905 71 582 00109 22 26 08 PDP 00109 22 26 39 PDP PAG101 on D230 NOTE If a PDP device NOT READY message is received while MTS is running check the lights on the devices and their control units If there are no error lights on and the units are READY Restart PDP Error Typeout Formats on operators console 1 CHANNEL LOGOUT I O INBOARD ERROR PATH xxxx 2 Direct Access Error dddd mtsnnn error c ffffff xx ssssssssssss aaaaaaaa 00 DEV VOLUME ERROR PROGRAM SENSE SENSE SEEK LAST OP NAME MAY BE CODE FLAGS FLAG DATA ADDRESS CODE ABSENT SEE ATT LIST VIRTUAL MEMORY PDP 76 SYSTEM SHUTDOWN AND START
20. Once the message Do you want to run the current system is displayed on the console proceed as follows 1 The message Do you want to run the current system is displayed 2 Normally you should just press ENTER 3 Several lines of text will be displayed There will be a short pause while MTS checks all of the I O devices If any devices particularly the UMnet PCPs are hung this pause may take up to 5 10 minutes 4 Eventually the screen will clear and MTS will print the message Local time is 5 If the time and date are correct press ENTER and proceed with step 6 a If the time or date is incorrect and you are loading UB MTS i e you are running under VM then the only way to reset the time is to shut VM down and reload it See the sections Shutting VM Down and Loading VM for information on how to do this b If the time or date is incorrect and you are loading UM MTS i e you are not running under VM then you can reset the clock To do this enter the command SET ZONE zzz DATE mm dd yy TIME hh mm ss and press ENTER zzz should be the time zone EST or EDT at UM mm dd and yy are the month date and year hh mm and ss are the hour minute and second c The system will print the message Enable TOD clock d Go to the console for the UM side of the ES3090 300J and press the alt key and while holding it down the TOD key VSE SYSTEM SHUT
21. RDVTYP X 80 on gt incorrect length is an error bits 0 3 X 40 on gt 24 byte sense X 20 on gt check for I O Ops that take too long X 10 on gt on block multiplexer channels RDVTYP 1 gt direct access bits 4 7 2 gt 2250 display 3 gt magnetic tape 4 gt bisynch line 5 gt printer 6 gt 3066 display 7 gt channel to channel adapter 0 gt other RDVBTS X 80 on gt sense in progress X 40 on gt I 0 queue busy X 20 on gt DE not received X 10 on gt proceed X 08 on gt walt taken with UE or UC X 04 on gt program error on inop device X 02 on gt unit check x 01 on gt unit exception RDVBTS2 X 80 on gt spurious device end received X 40 on gt device has subchannel X 20 on gt wait for CE and not DE X 10 on gt start I O pending X 08 on gt unused X 04 on gt select on same path X 02 on gt use RDVRSTRT as CAW x 01 on gt wait for any interrupt RDVBTS3 X 80 on gt OLTEP Attn glitch X 40 on gt GTUNIT queued X 20 on gt control unit end since SIOF X 10 on gt one sec device end pending X 08 on gt OLTEP operation X 04 on gt HALTIO completion pending X 02 on gt unused SYSTEMSTATUS 474 X 01 on gt unused CSW and SNS are the channel status word and device sense data 1 byte length 1 byte SNS flag SNS data ATNX SNSX PCIX and ENDX represent the Attention auto sense PCI and Endexit area addresses respectively The remaining lines are th
22. TASKS descriptor The TASKS command prints one line of output for each job which fits the descriptor category Each output line contains from left to right The job number The job name The job table address The number of virtual memory pages in use by the job if relocatable such as MTS The job status The job table parameters if any The I O devices attached to the job if any For MTS the job table parameters are the four character signon ID blank if the job has no user signed on the four character project number IDLE if no user is attached and BUSY if a user is attached but not signed on the receipt number if it is a batch job and the file name if it is a file job The descriptors specify the type of jobs to be displayed The valid descriptors are XXXX job number xxxx B all batch jobs F all jobs full M all MTS jobs N xxxx all non MTS jobs or job name xxxx D xxxx device name xxxx T XXXX device type xxxx U XXXX user signon ID xxxx P XXXX project number xxxx If a job number is specified the first nonnumeric character terminates the number the remainder of the input line is ignored If the first nonblank character in the descriptor is B F or M then the remainder of the input line is ignored e g it is permissible to enter MTS instead of M If the first nonblank character is D T U or P then the argument for the descriptor begins with the next no
23. TERMINAL ANSWERBACK CODE PAGE PRINTER SHEETS PAGE PRINTER LINES PAGE PRINTER PAGES PAGE PRINTER IMAGES PLOT TIME CHARGED INCHES OF PAPER USED LINEAR TAPE MOUNTS FLOPPY DISK DISKETTE MOUNTS PAPER TAPE READER MOUNTS PAPER TAPE PUNCH MOUNTS MAG TAPE DRIVE TIME FLOPPY DISK DISKETTE DRIVE TIME PAPER TAPE READER TIME PAPER TAPE PUNCH TIME PAPER TAPE PUNCHED Using the Statistics Utilities SCAN provides basic utility routines to search copy and print statistics records Subroutines may be written which act as command processors and which may call the utility routines SCAN 458 Before a command processor can be used it must be loaded via the LOAD command This establishes the name of the command handled by this command processor and the entry point to be called to do the processing That entry point is called with the following parameters LINES 2F line numbers STRING nC character string from command line terminated by X 4000 TV nA transfer vector STAT routine to call to get statistics records CMD routine to call to get cmdstat records LSTAT routine to list statistics records LCMD routine to list type 1 cmdstat records LES PAN routine to do SPAN NULL LENUMB routine to CVB a number LEBREA routine to do BREAK STATFD fdub for temporary file containing statistics records See SCANTR in the file SCANDS for a transfer
24. The program will request a sign format 3 line 4 line repeat or other other isn t implemented yet A one character response 3 4 or R will suffice An end of file terminates the program Next the program types SET UP THE PLOTTER the plotter is assummed to be available Respond with a carriage return when the plotter is ready for use The position at which you leave the pen becomes the lower left hand corner of the sign unlike the postprocessor If the format is repeat the previous sign will be redrawn if there is a previous sign Then the program prints ENTER TITLE LINE Enter the title exactly as it is to appear on the sign to get lower case on a terminal through the data concentrator use CONT A CONT A UC off The title will be centered Next the program requests the format of the hours lines centered center each line or just draw a given line linr the other lines up with it or just draw specify an x coordinate for each line The first character of each type C UL J will suffice for a response For centered lines the program will request the text for each line in order from the top of the sign Enter each line exactly as you want it to appear For lineup format the program first requests the text of the line which is to be centered It then asks which line this is 1 2 3 or 4 numbered from the top Then the text of the other lines is requested just as with centered
25. then filenames and end of file c To restore entire volume type ALL Program will ask for volume names to restore NOTE As used in restore a FILENAME will have one of the following formats a Userl User2 will cause all files of UserID1 to be restored under UserID2 The copies of the files under UserID1 will be left unchanged If UserID2 is omitted all files of UserID1 will be restored under that UserID Both Userl and User2 must be 4 character UserIDs b UserID1 UserID2 Filename will cause the file given by UserID1Filename to be restored as UserID2Filename c Filename where Filename is the full disk type name ie UserID concatenated with UserFilename or System File Name If this format entry is given only that particular file will be restored Up to 250 file restore requests can be processed at one time EXAMPLE 4 RST will OPERATOR FILES 281 d Useridlfilename userid2newfilename simil TAPE name and Repl lar to ask le is restored llowed by the file restored b WHERE IS THE RESTORE ly with the physical device Tape to be used will be indicated on the form 5 When the fil The CSID fol COMPLETED will be written on the operators console the former when it begins restoring the file the latter when it finishes 228 44 47 28 28 28 28 29 29 12y AO 35
26. 14 SYSTEM SHUTDOWN AND START UP PROCEDURES the screen Type DISC and press ENTER The message DISCONNECT AT time should appear Ina minute or so the VM XA logo should appear again At this point the message Do you want to run the current system should appear on the UB MTS operator console See the section on Loading MTS to proceed Switching to a backup UB MTS operator console To start the backup UB MTS operator console you must signal an external interrupt To do this 1 2 10 11 Go to a terminal connnected to VM XA At the LOGON prompt type UB MTS and press ENTER to logon to the UB MTS virtual machine console for the UB MTS virtual machine The terminal should respond with the message Enter Password Type the password and press ENTER If you don t know the password ask your shift supervisor The terminal should respond with the message Reconnected at time Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of the screen Type EXTERNAL and press ENTER Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of the screen Type DISC and press ENTER Hold down the ALT key and press the CLEAR key The VM logo should appear Backspace over the word LOGON type DIAL UB MTS 1F6F and press ENTER UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 3
27. 2 The device has NO ERRORS that is there are no current error indications for the device Note that although the general status line appears twice once for each path the device status line appears only once The device status line always appears beneath the description of the path that has been most recently used for an I O operation EXAMPLE 3 Units D ptrx JOB PROGRAMS UNITS CHO gt CU 9 gt DEV ptrx OFF ADDR 082 This example shows the notation employed in the path description to describe OFFLINE status The characters OFF appear after any channel control unit or device that is logically OFFLINE A channel control unit device may be logically OFFLINE because someone has made it so using the OFFLINE job because it was initialized as logically OFFLINE when the supervisor tables were assembled or because it was marked logically OFFLINE because it was found to be physically offline when the system was IPLed This sort of OFFLINE notation together with the path OFFLINE notation described in EXAMPLE 1 make the total of four different ways that a device can be OFFLINE All four of these ways in simultaneous effect would result in a path description like this OLD STYLE USING MTA UNITS O JOBS PROGRAMS 163 CH 0 OFF CU 9 OFF gt DEV ptrx OFF ADDR 082 To make this device operational someone would have to ONLINE not only the device itself but a
28. OPERATOR FILES 181 CCD The program causes the deferment of long or difficult plots so that they may be plotted at a more convenient time mts ccd You will be promted for a ccid Enter all the ccids for which plots are to be deferred one per line Terminate the list with an end of file CNCL The deferred plots were queued by the user in the normal way but placed on a separate plot tape To defer a users plot a program is run to tell CCP that all plots for a particular Signon id are to be deferred After this is done CCP will ignore that users plots when doing the scheduled plot runs the deferred plots will be left in the queue The deferred plots can be put on a tape at the operator s convenience by running CCP using a special plot tape The system ensures that only deferred plots are placed on this tape When it is no longer necessary to defer the user s plots a program is run to tell CCP to process the plots normally mts ccd ENTER SIGNON ID sys PLOTS FOR SYS WILL BE DEFERRED mts LOTS FOR MTS WILL BE DEFERRED fendfile CANCELLED This program writes information into SYS CCPPRIO telling CCP which ccids to defer plots for All plots for the ccid will be deferred until CCD is instructed otherwise see below Running Regular Plots Run CCP in the normal way Deferred plots OPERATOR FILES CCD 182 OPERATOR FILES will be left in the
29. PB II must be powered down before PB I NOTE Before attempting to power up a PB I make sure that the individual Battery Disconnect breakers located on each battery cabinet and the main Battery Disconnect breaker for each system located on wall boxes are open as well as the Main Rectifier Charger breakers located on the lower front of each PB TI Then make sure that on the PB Is the Main Input breaker the Main Output breaker and the Output breaker s are open Make sure that the Main Input Switches located on the blue cabinet on the south wall as well as the System 0 breaker in the Generater Room are closed Powerung up PB 1 1 Close Main Input breaker 2 Clear out erroneous alarms ALARM TEST AUDIO RESET 3 Check incoming power by pressing Input Voltage A B B C C A Results should be 480 volts plus or minus 10 4 Close Main Circuit Breaker this doesn t do anything as far as the system is concerned 5 Get output power press and hold transfer initiate METERING SECTION 554 while rotating the transfer switch to Bypass BYP Then release both switches The load on Bypass alarm will be on Clear this with alarm test and audio reset 6 Check output volatage by pressing Output Voltage A N Musso BAN and Puse A N one at a time while observing the digital panel meter Values should be 227vac 2v 7 Get rid of error conditions u
30. TASKS nnnnn tasks number TASKS B batch jobs TASKS P project number TASKS N jobname non mts This job is not normally used from the 3270 operator s console the T command being used instead With the M parameter the status of all active MTS tasks is given With the U userid parameter the status of those tasks on which user userid is signed on will be printed With the D device i d parameter the task that currently owns device device i d will have its status printed With the T device type parameter the status of all those jobs owning one or more devices of device type device type will be printed With the nnnnn parameter the status of task nonnn will be printed With the N parameter alone the status of all non MTS tasks will be printed e g HASP PDP HASPLINGs etc If the N parameter is followed by a jobname then the status of all tasks executing with that job program is printed note that for jobnames having more than four characters the name must be split into two parameters the latter of which must contain the last four characters of the job name JOBS PROGRAMS 151 With the P prjid parameter the status of those tasks on which a user with project i d prjid is active will be printed With the B parameter the status of all currently executing batch tasks will be printed With no parameter the status of all tasks will be printed this includes all of the
31. These lines are used primarily for communicating with jobs in the system although several other forms of output will also appear here Basically message lines and message response lines will be inserted at various points in the message area as described below and lines will be removed from the top of the area to make room for more output The handling of each output form is described in the following list 1 Message lines from jobs the format will be job number 5 characters job name 1 to 8 characters message text 65 to 72 characters maximum unless SWRAP is ON in which longer messages will be displayed on two lines Example 00017 MTS REMOVE TAPE FROM T901 These lines will always be added at the bottom of the message area 2 Message response lines When a job requests input via SVC READ the message area will be scanned for the last message line from the same job This line will be intensified anda blank line will be inserted immediately after it in which the response should be entered These two lines will not be removed from the screen until the response has been entered If no message line for the same job is found on the screen or if the line found is completely blank or null the message in the normal message line format will be added first This special message will not be copied to the printer 3 Output from System Status commands e g T These lines will be displayed with
32. occur if system crashes or if install DISKCOPY 358 12 Send a message to Steve Donnelly about the disk just copied He will change the label cards in the plastic pouches on the physical disk boxes on the third floor to reflect the new labels 13 Put a notice in the Systems Changes Forum conference FILEINFO The program in the file ACC FILEINFO can be used to determine how much disk space would be retrieved if the DEADFILES program were run on a given date The program ACC FILEINFO often takes about 15 minutes to run to completion To use the program from the operators console signon ACC run acc fileinfo The FILEINFO program supports the following keyword parameters STAFF causes a list of file staff usage to be produced DATE mm dd yy specifies the last expiration date to be used for determining if files are dead DEAD causes all active files to be grouped together and dead files to be listed according to first letter of project number instead of vice versa DISKS FILEINFO 359 DEADFILES Schedule and Miscellaneous Notes The DEADFILES program is usually run on the last Thursday of each month There are a few exceptional months During these months the DEADFILES program is usually run on different days depending on current fiscal considerations On any month the DEADFILES program may be run at an unusual time if necessary e g when the system is running lo
33. press ENTER This returns the panel display to FCA nnn NOTE Following a power failure the 3805 may not automatically power up because it is not connected to the IBM ES3090 300J POWER DISTRIBUTION UNIT PDU If the FAULT light on the same side as the EPO knob is on a F E must be called INTEL 3825 Reseting amp Warm Start The INTEL 3825 operates almost the same as the 3805 Instead of a service panel at the rear a drawer in the front contains an ASP Alphanumeric Service Processor The procedure is given in detail in chapter 10 of the 3825 Utilities and Diagnostic manual or Chapter 5 of the FAST 3825 Operators manual The 3825 manual mentions SLOW START which is the same as WARM START for the 3805 Both terms mean the COLD START info is still valid in non volatile memory The 3805 has battery backed up CMOS for these tables while the 3825 has E PROM VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 73 Since one of the 3825 has no MOTOR ALTERNATORS to smooth over power spikes it is more likely to crash following power failures etc Reseting the 3825 If you should need to reset the INTEL 3825 e g after a power failure this is what you should do 1 Press the white alarm reset button on the 3825 s front panel this Just turns off the alarm 2 Write down the time and date and any number displayed on the status indicator on the front of the 3825 3 Pull open the draw
34. 2 means it was from the RM layer of things 1 means the DSP layer 023 are messages from auxiliary processor tasks and in the 2023 category 2xx messages are BITNET related When running SYS RMOPR the default verbosity level is 5 so the headers do not come out normally Enter VERBOSITY 10 to see them Other not so obvious things are linked to the verbosity level For example you may have a job that is waiting to print and you want to see how many pages it is With the default verbosity doing DISPLAY JOB RM024572 will yield only DWB RM024572 24572 is awaiting print Pl after 2 jobs but if you enter VERBOSITY 10 and then DISPLAY JOB RM024572 you get CIJ 2012002 DWB RM024572 24572 is awaiting print Pl after 2 jobs Entered at 20 55 32 Sun Sep 25 88 TYPE PRINT DELIVERY ARGUS2 HOST UM PRIORITY MINIMUM PROUTE CNTR TWOSIDED NO IMAGES 7 PAGES 7 SHEETS 7 PROJ WDWB PROC PR98 SOURCE AF14 USER DWB USERNAME Don Boettner there appears to be no way to just display one attribute of a job you either get none or all 4 A note on job names and job numbers The job name is a user selectable eight character name qualified by the userid The job number is a six digit integer which is generated by the RM and always unique The job name even when qualified by the userid does not have to be unique When the user does not specify a job name the RM generates one consisting of the letters RM
35. 32 SYSTEM SHUTDOWN AND START UP PROCEDURES VM refers to tape drives by number rather than by name as does MTS The following table lists the MTS name and the VM device number for each tape drive the device number must be specified on the ATTACH and DETACH commands as shown below Name Number Name Number Name Number T900 0680 T907 0687 T914 068E T901 0681 T908 0688 T915 068F T902 0682 T909 0689 T920 1680 T903 0683 T910 068A T921 1681 T904 0684 T911 068B T922 1682 T905 0685 T912 068C T923 1683 T906 0686 T913 068D Determining whether a tape drive is attached to UB MTS if you are unsure about whether a tape drive is attached to UB MTS or not issue the following command at the VM operator s console QUERY 0680 068F 1680 168F Some possible responses are nnnn indicates the device number TAPE nnnn OFFLINE TAPE nnnn FREE TAPE nnnn ATTACHED TO UB MTS nnnn R W The last message TAPE nnnn ATTACHED indicates that a drive is attached to UB MTS The ATTACH command To attach a tape drive to UB MTS issue the following command at the VM operator s console where nnnn is the device number for the tape drive ATTACH nnnn TO UB MTS nnnn The DETACH command To detach a tape drive from UB MTS issue the following command at the VM operator s console where nnnn is the device number for the tape drive DETACH nnnn FROM UB MTS UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 33 Loading
36. CSW given Example FIO DS06 0 0C000000 In the example above the underlined portion of the CSW is of most concern to the one issuing the FIO commands Some of the codes in this byte are listed below 02 control unit end 08 channel end 04 device end 2C control unit end channel end device end FIO should be used very judiciously probably under the supervision of a systems programmer JOB job This command given without a job number generates a table showing the user id the master source device name the job number number of virtual pages in use and CPU time used for each active MTS job as follows ecce dddd jjjj vvvv tt UEM L L CPU time in whole seconds L page wait L virtual pages L vertical bar if disk attached L job number L blank neutral privileged non privileged L device name or file L user id A summary line is then generated as follows hh ss nB nT nVP l L HASP jobs held L HASP punch queue L HASP print queue L HASP execute queue l L HASP spool pack in use L real pages L virtual pages L active terminal jobs nRP n n XEC n PTR n PCH n HLD SYSTEMSTATUS 476 l L active Ltime If a job number i job is given L j L job number batch jobs s given L ob name l L BTS2 id L task st
37. If PAR LAST is specified the latest entry is deleted the entry with the highest sequence ID If SEQ n is specified the entry with sequence ID n is deleted Tf the TAPE rack and VOL label parameters are specified all entries from the specified tape are deleted FILE filename This command specifies the file filename as the directory file GET type PFIELD m SFIELD m SEQ n PAR LAST This command gets an entry from the tape directory The parameter type specifies the type of tape entry to be accessed The tape directory program searches the directory for the first entry which matches the search values indicated by the PFIELD SFIELD SEQ or LAST parameters If LAST is specified the entry with the highest sequence ID is returned If two search values are given the second is searched upon only after the first is satisfied In this case only the PFIELD and SFIELD parameters may be used The order in which they are specified is the order in which they are used for the search The PFIELD and SFIELD parameters specify the pri mary and secondary search field values to be used for the search The format of nmi must TAPE DIRECTORY PROJECT 514 Command Explanation Command Explanation Command Explanation correspond to the format specified in the master entry for type For Julian time formats m may be given in either integer format giving the actual number of m
38. It is desirable to keep a minimum number of pages available on each pack for expansion purposes currently 500 pages When this message appears ALL packs are below the minimum The appropriate person should be notified to add another public volume DESTINATION OPER SOURCE FILE SYSTEM 482 17 CREATE OPEN etc MISCELLANEOUS MESSAGES 410 gt gt gt JOB nnnn WAITING ON FULL SHARED FILE TABLE User Explanation The system keeps track a pun of all currently active files ina table in shared memory When this message occurs either the table has overflowed legitimately and hence should be increased in size or some program has gone wild and activated a large number of files In the latter case the job should be stopped A systems programmer should be notified to determine which condition occurred DESTINATION OPER SOURCE FILE SYSTEM 482 027 ICLOCKx ICUNLKx gt gt gt MOUNT MESSAGES TO THE OPERATOR User Explanation Messages to the operator for MOUNT requests and reports of mount errors are produced in this section DESTINATION OPER TEXTLOC MOUNT 101 ST12 MOUNTCMD MNT01270 SOURCE MOUNT 101 ST12 MOUNTCMD MNT01270 gt gt gt MTS DEBUG JOBDUMP FOR SYSTEMS User Explanation The operator is notified that the MTS debug job dump is occurring System Explanation The SETEXIT entry into MTS from the supervisor was taken and the EXITDUMP flag was on with appropriate ex
39. LISTSTAT Type Output Codes and Meanings SCAN JOB ID PROJ TIME DATE TIME ZONE EL TIME CPU TIME VMI CPU VMI WAIT LAST SIGNON TIME ZONE LINE RATE COST FILES C DSK LINES PAGES CDS P CDS R SUPERVISOR JOB NUMBER USER 1D PROJECT ACCOUNT NUMBER TIME USER SIGNED ON DATE USER SIGNED ON ELAPSED TIME SECONDS CPU TIME USED SECONDS STORAGE INTEGRAL VMI WITH CPU PG SEC STORAGE INTEGRAL VMI WITHOUT CPU PG SEC TIME AND DATE OF LAST SIGNON LINE NUMBER IN STATISTICS FILE RATE NUMBER TOTAL COST FILE CHARGE CURRENT DISK SPACE NUMBER OF LINES PRINTED LINE PRINTERS NUMBER OF PAGES PRINTED LINE PRINTERS NUMBER OF CARDS PUNCHED NUMBER OF CARDS READ it is NOTE To enter SET contain note that STATISTICS will be locked while the STA command is executing during periods of heavy load STATSCAN DEFAULT causes the FD CRE DES MERIT DSK OPS PAGE INS ANSBACK PPRSHT PPR LNS PPRPGS PPRIM PLT TM PLT PAP MNT FMN RMN PMN T DRV F DRV R DRV P DRV PTPP DSK USED 457 NUMBER OF DISK FILES NUMBER OF FILES CREATED NUMBER OF FILES DESTROYED MERIT TIME SECONDS CUMULATIVE DISK SPACE SINCE LAST SIGNOFF PG MIN DISK OPERATIONS NUMBER OF VM PAGE INS PAGED IN BY PDP
40. MTS CMD 8 If CMD shows as LOCK 9 ICTAB it means that it is waiting to use the incore file table The command DIS LOCK 9 will show you which job currently has the incore file table lock set and the J n command will in turn show the status of that job If that job seems to be using CPU time wait several minutes and check again If the job isn t using any CPU time take a job dump and stop the job 9 If CMD is doing DISK I O then it is waiting for the disk manager Check to make sure that the disk manager is running using the command T N DMGR If the DMGR isn t running get help from a systems programmer or if no systems programmer is available try restarting the DMGR If it is running use the OPERATOR FILES CMD OPERATOR FILES 197 DMGRSTAT job to check its status 10 In all cases check each job several times using the T n or J n commands before STOPping them If the Situation changes start the checking over from the top If CMD goes into CMDSTAT WAIT the problem has probably cleared itself up If you need to stop a job always try to STOP a job first and only if that fails after waiting several minutes should you consider BLASTing the job If you are working from an MTS terminal rather than the operator s console the Systemstatus CLS can be used to issue the Two commands the LOCKSTATUS command can be used in place of TABLMOD and MTA UNITS O in place of the UNITS job If you do work fro
41. MTS is down and you could IPL If you need to power the ES3090 300J off go on to step 9 SYSTEM SHUTDOWN POWER DOWN PROCEDURES SYSTEM SHUTDOWN AND START UP PROCEDURES 65 EMERGENCY POWER OFF MEASURES To be re written SYSTEM STARTUP EMERGENCY POWER OFF 66 SYSTEM SHUTDOWN AND START UP PROCEDURES VIRTUAL MEMORY The INTEL 3805 and 3825 are used by MTS as DIRECT ACCESS STORAGE devices for paging After IPL only the PDP Paging Drum Processor uses the 3805s and 3825s and any problems releated to these devices will therefore affect the PDP NOTE See the sections on PGET and PDROP for adding or removing virtual paging devices At present we have four virtual paging devices These include two 3805s and two 3825s One of the 3825s is in a 3805 cabinet with two motor generators Additional information can be obtained from manuals put out by INTEL These manuals are located in book shelves near the INTEL equipment NOTE Each Intel box has its own manual These manuals are on a shelf in the first floor machine room Each box has a sticker with a serial number which should correspond to a serial number on one of the manuals At present the numbers are 105 v01319 117 and 299 INTEL 3805 OPERATOR ACTIONS The INTEL 3805 The Intel 3805 can either emulate an IBM 2305 or can be used in NATIVE mode It runs in NATIVE mode at this installation This means that it is addressed differently and that is is not
42. Meaning Message Meaning Message Meaning OPERATOR FILES 251 Unable to lock MNET AUTH The file MNET AUTH is locked for MODIFY or DESTROY Rerun MUP in a few minutes and if the file is still locked try to determine what job has it locked using the MTS LOCKSTATUS command or the TABLMOD job and BROADCST a message to this job MNET AUTH is not permitted read others MNET AUTH is not correctly permitted let a member of the Communications Group know MNET AUTH is incorrectly formatted One or both of the first two lines in MNET AUTH is missing running MNS should fix it up Next MUP checks the state of the UMnet PCP and one of the following messages will be printed Message Meaning Message Meaning Message Meaning lt name gt seems to be running normally Everything looks good lt name gt is busy Try again later This message will appear if the xxOP device is busy when HUH is run Try HUH again after a short wait lt name gt is logically offline Either lt name gt was not up when MTS was IPLed or it has been offlined using the OFFLINE job In either case to online lt name gt sign on to the CCID OPERATOR FILES MUP 252 Message Meaning Message Meaning OPERATOR FILES MUP OPERATOR FILES STRT and SOURCE STRT ONLINExx where xx is AA AB AD AE AN or CC Once this is done rerun HUH lt name gt is physically offline
43. Presumably CMDSTATc was marked F by CMDPIKUP So either CMDPIKUP was confused someone marked the file F in CMDDIR by hand or the file really wasn t marked F and CMDTAPE was confused The only thing to do is to tell the person who maintains this stuff about it LINE 1111 IN CMDFDIR DUPLICATED WRITING CMDSTATc The data in CMDSTATc had exactly the same time range as some file already on the tape This is unlikely to occur without a mistake by a person fixing something Check whether the two tape files are really duplicates and if so delete one Delete the one which is not described in 1111 NEW FILE FROM time OVERLAPS FILE IN CMDFDIR LINE 1111 This is similar to the problem above except that the two files are not duplicates This message is generated only if the overlap is more that a minute shorter overlaps sometimes occur because of the way the buffers are filled and emptied If this occurs just once ignore it but if it occurs several times over a couple of weeks tell the person who maintains this code about TE NO ROOM IN CMDFDIR AFTER 1111 CALL SYSTEMS PROGRAMMER This message goes to the console Spread out the lines around 1111 in CMDFDIR it is safe to renumber the file then enter a null line at the console CMDTAPE will make a new attempt to write out the directory line This problem is unlikely to occur the probable cause is that files were not written to t
44. REPLOTTING After the plot tape has been dumped into a file plots can be replotted without remounting the tape as long as these plots were included in the latest tape to file operation After the operator has typed run plt Operator can type plot all replots everything in the file plot 014 plots block 14 only plot 007 009 plots range of blocks 7 9 plot s 8 replots all blocks under s 8 given ATTENTION INTERRUPTS NOTE After PLT has been interrupted and is waiting fora response on the operators console do not enter a command until the pen carriage has stopped with the plotter in automatic Plotting can be interrupted by GOOSING the supervisor number of the plot job This will result in the printing of the comment PLOTTING PLOT XXX NOW WHAT DO YOU WANT The following are options which can be used at this time pri n n being the time in seconds that the plot routines will wait after last usage of the pen before priming it default 2 min bac start the current plot over bac n start at plot N continue to previously specified end con continue stop at end of current plot con n continue stop at end of plot N mts call MTS NOTE PLT is still loaded at this point and can be REStarted PLOTTING SYSTEM 438 sta go back to what is to be plotted End of file go back to whatever you were doing at time of interrupt ret n Retraces N number of records NOTE If the c
45. appropriately scheduled times USAGE mts ccp System replies with a receipt number in the following manner PRINT ASSIGNED RECEIPT NUMBER 610418 Where is a plot tape system asks Operator mounts appropriate check sticker for day and time tape on a 9 track unit and responds with T90X X being the number of the unit to be used System responds with USER W175 W175 FILE CCPFILE 22902992 CCP is done approximately 26 min plotting Blue used NOTE If colors other than black are to be used this will be mentioned in above line A tape has now been prepared for plotting NOTE CCP is usually started automatically by CLK OPERATOR FILES CCP CCr NAME FUNCTION USAGE OPERATOR FILES 187 CCT To let the operators know how much time the plotting session would take and how many user tapes would be used if the plots were dumped on tape and run at the time CCT is run mts cct Response on operator s console might be 01721 PLOTTING TO DO 27 MIN REQUIRING NO USER TAPES OPERATOR FILES CCT 188 CHK NAME FUNCTION USAGE REMARKS OPERATOR FILES CHK Displays list of OFFLINE equipment on OPER mts chk CHK is atumatically started fifteen minutes after the beginning of each shift CHK should always be run following a reload OPERATOR FILES CHK OPERATOR FILES 189 CLB NAME CLB FUNCTION This program retrieves the label from a tap
46. as you might therefore have guessed it takes a good deal longer NOTE If one of the streams fails call Steve Burling and don t restart the individual stream but instead stop the individual streams and restart SVW otherwise the wrong tapes will be called for Use the program FSC to make copies of the filesave tapes See the write up in this manual NOTE When the comment IT IS OK TO LET THE LINES OUT NOW appears do the following Type STARTUP Tf there are no clerks or DPAs on the first floor offline ptrx ptrx and ptrx as well as pchx OPERATOR FILES SVW 298 OPERATOR FILES e Start RMS RMP LAS and MNS release ex Do not shut the system down for any reason before the IT S OK TO TAKE THE SYSTEM DOWN NOW message is printed on the operators console If the system crashes after the lines have been let out and before the above message appears contact one of the FIREFIGHTERS It is not necessary to restart the weekly filesave If there is trouble concerning the running of SVW call Steve Burling If SVW is started on the wrong date the system will ask if you really meant to run it This is to guard against SVW being run accidentally If SVW suspects that it is being run at the wrong time it will ask if it is ok to proceed Responding NO to the question will terminate SVW with no other bad effects If one wants to run a special large weekly type filesave S
47. example it has commands to print out the job table and various sections of real or virtual memory In addition PRINTDUMP is smart enough to print a nice formatted listing for MTS jobs The best documentation for MTS PRINTDUMP is its own EXPLAIN command A garden variety listing of an MTS job dump is produced as follows Mount dump17 9tp dump volume dump17 ring in run mts printdump 1 dump p 100 vm mts Endfile Note Don t forget the page limit Depending on how much VM is being used PRINTDUMP can produce zillions of pages of output 100 pages is usually enough room to dump all of the system segment and most of user memory Note Always mount dump tapes with ring in This makes things easier for the operators Note This long hand method of typing in the MOUNT and RUN command is rarely used It is easier to run program DUMP DUMPER which takes care of this stuff automatically We ve set up a ccid DUMP which is used to handle most jobdump processing It has a sigfile which only allows it to print dumps in batch mode and which uses CKID in interactive mode so we no longer have to worry about security with the password the password is DUMP DUMP DUMPER also keeps a log in files DUMP LOG and FIRE FIGHTERS GUIDE 372 DUMP TAPES Hopefully this log will sometime replace the dump log notebook the operators now keep but not yet 5 OPERATOR FILE DMP This file is used by
48. information appearing under the column heading STATUS This information appears as a series of letters which should be interpreted as follows D gt means node has been deleted X gt means node has an exit set S gt means exit is synchronous Q gt means node has a non empty queue of messages receipts and notifications B gt means node has been blasted The optional modifiers are NOHEADER gt to suppress the printing of the summary of the information from the network control block NONODES gt to suppress the printing of the individual node descriptors Prints list of legal commands TT IT 114 JOB PROGRAMS LIST SENSE DMGRCTL HELP NOTE This program may also be run as an MTS program i e run mts it IT JOBDUMP NAME FUNCTION PROTOTYPE USAGE PARAMETER JOBS PROGRAMS 115 JOBDUMP To force a dump on tape of all of real core and of all tables and virtual memory associated with a single task JOBDUMP nnnnn A jobdump is of use only to systems programmers and should be taken only for a good reason e g if an MTS task SNARKs if a task hangs up for no apparent reason or if requested to do so by a systems programmer All processing stops until all of real core is written out onto tape Before the JOBDUMP request is given the operator should mount a dump tape on a nine track drive and make sure the ring is in After the command is given the message DUMP SP
49. mts avs AVS is normally started by MNS OPERATOR FILES AVS 176 AWS NAME FUNCTION USAGE OPERATOR FILES AWS starts up ports AW10 AW6F mts aws AWS is normally started by MNS OPERATOR FILES AWS AXS NAME FUNCTION USAGE OPERATOR FILES 177 AXS starts up ports AX10 AX6F mts axs AXS is normally started by MNS OPERATOR FILES AXS 178 AYS NAME FUNCTION USAGE OPERATOR FILES AYS starts up ports AY10 AY6F mts ays AYS is normally started by MNS OPERATOR FILES AYS NAME FUNCTION USAGE OPERATOR FILES 179 AZS starts up ports AZ10 AZ6F mts azs AZS is normally started by MNS OPERATOR FILES AZQ 180 NAME FUNCTION USAGE KkKKKKK KKKKKK KKKKKK OPERATOR FILES BBS OPERATOR FILES BBS To start up the babysit program which watches for an i d This is used when it is suspected that someone is fraudulently using an i d and notification is wanted if that i d signs on mts bbs MTS ENTER USER ID S ONE PER LINE MTS 0ab2 MTS An end of file condition was entered on the last line above to terminate the list When someone signs on with the i d s being watched for the following is printed on the operator s console MTS USER OAB2 SIGNED ON To get rid of a watchdog task use the STOP job CCD NAME FUNCTION USAGE EXAMPLE
50. online ptrx online ptrx rdrx To online control units online cu 2 online cu ptrx To online a channel online ch b Only one channel may be onlined in a single command To online paths between a channel and control units online ch b 2 online ch b 2 3 online ch b ptrx rdrx JOB PROGRAMS ONLINE 128 OPERATOR NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES JOB PROGRAMS OPERATOR To start up a 3270 as the operator s console OPERATOR console device printer device The OPERATOR job program is the task which controls the 3270 operator s console and echoes console messages to the console printer This job is normally started automatically after the system is IPLed See the sections DEVICE COMMANDS and 3066 or 1052 SUPPORT AS BACKUP for details on the operation of the OPERATOR job and for instructions on switching between a backup console and a 3270 console If the OPERATOR task stops or is stopped it can be restarted by entering the OPERATOR job request If the parameters are omitted it will be restarted using the default 3270 display and printer 1 console device specifies a 3270 display to be used as the console If the parameter is omitted the usual display will be assumed 2 printer device specifies a 3284 1052 3211 or 1403 to be used as a console printer If it is left out the usual printer is assumed To start the OPERATOR job without a print
51. request a job dump If that happens cancel the dump clear everything out and try again Then to start things up again a Issue the RM command START WAYNESTT It sometimes happens that things get really hosed up and you need to get rid of the whole RM and restart it For this case use the predefined MTS files to drain and restart it so if we add new steps to the startup we don t have to find every last copy of this note and modify it What you should do is a Use operators console or SY O and issue MTS RMD D for Drain b It should print a message on the operators console that the RM was drained If nothing happens do a T on the MTS RMD job and see what it is doing If it is sitting in Monitor wait status the RM intertask network is probably also hosed up so proceed to 5 below c Use the systemstatus command TASKS RM or T T RM to make sure all tasks have gone Be patient d Make sure the messagesystem task connected with all of this has gone Either look for a job that has POST NETLOG BIT locked should be id POST and RM M showing as front of msource name when SY displays it actual name is RM MSG or else use SYS TASKS and enter INTERTASK RMGRNET to find it If you find it stop it e Issue MTS RMS to start up the system f and then issue MTS RMP to start up the processors Tf you try running SYS RMOPR and it immediately produces three lines of error messages from
52. t11 ptrx 81pi g11 ptrx 81lpil For more information on Print Buffer loading look for appropriate heading in the Contents The next line will appear on the operators console on the VSE SYSTEM SHUTDOWN AND START UP PROCEDURES 61 second floor OPERATOR OPERATOR OPERATOR OPERATOR MTS MTS STARTED AT hh mm ss ON dd mm yy MODEL xxx SERIAL xxxxxx USING DISPLAY DS02 amp PRINTER PTRx 3284 STARTED AT hh mm ss ON dd mm yy USING DISPLAY DS32 amp PRINTER PTRx 3284 ORELOAD AT hh mm ss on dd mm yy SYSTEM UL123 WRITTEN AT hh mm ss on TUE JUL 12 ENTER INITIALS AND REASON FOR RELOADING Type your initials a space and the reason for loading PDP MTS MTS MTS COM COM COM COM COM COM COM COM COM MTS PAGOO1 ON Dnnn CONTENTS OF IPL FILE IPL O0O UL153 DENTRY 45680 PSECT 100008 VIRTUAL A0000 1432A8 WRITTEN BY ID W163 AT hh mm ss FRI JUL 15 83 18 10 JUL 14 CHANGES TO STACK DESCRIPTOR FORMAT REQUIRED CHANGES TO STACK POPPING IN CMDSHELL AND SYGNOF CHANGES TO RESET THINGS PROPERLY IF THE USER HITS ATTENTION AT THE ENTER PASSWORD PROMPT CHANGES TO TSA TO SNARK IF IT HAS A RATE NUMBER OF 0 HAPPENS ON NDSK SYSTEMS CHANGES RELATING TO THE STACK MAINTAINED BY MACPUSH MACPOP THESE MAKE THE KUDGLE EVE KLUDGIER BUT WILL GO AWAY EVENTUALL FILESAVE DIRECTORY UPDATE COMPLETED THE ACCOUNTING FILE IS AVAILABLE POST IPL SYSTEM LOADE
53. t908 T908 NOT READY VOL CHG186 OWNER TECUMSEH DENS 1600 ENTER TAPE NAME eof if no additional tapes NOTE The present policy permits a staff member to give out tape label information under the following conditions The tape receipt is brought to the Computing Center by the person requesting label information The user on the phone is able to give the name phone number and tape id and be reached at the number given on the rack sticker After running FLB we call the person back only at the phone number on the sticker with the label information Tf a user other than the owner of the tape is requesting label information and cannot meet the requirments mentioned e g can t be reached at the phone number on the rack sticker he or she must have the owner call us to request the information The owner of course must be reachable at the phone number on the sticker OPERATOR FILES FLB OPERATOR FILES 215 FSC Name FSC Function FSC is a program designed to make copies of original filesave tapes Usage mts fsc Operator and System interaction IS PRINTED LISTING OF FILES DESIRED YES NO SYST ves or no oper If yes response is given system will ask if no tape copy is to be made but only a listing IS COPYING OF TAPES DESIRED YES NO ves or no oper TYPE STREAM NAME MTSOON FILES OR ONLINE online for SVW oper OR files when DEADFILES is being run oper
54. the Security office phone 31131 and or the Fire and Police departments phone 9 911 should be called immediately If any damage has been done to the Computing Center building or to the equipment inside of it one of the following people should be called Mr L G Mitchell Asst Director of Operations Dr A R Emery Deputy Director 9 761 6960 SECURITY CLERK 461 Doors leading to the outside of the building should be checked on each tour to make sure that they are locked These doors are locked by the use of an alan wrench which is on the key ring The Security Clerk is expected to answer the phone that is the trunk lines 764 2121 22 and 23 The phone in the second floor machine room can be made to ring by turning the switch on the switching box on This box is located immediately to the left of the two black recording phones by the key ring SECURITY CLERK 462 Emergency Power Off Measures Emergency Power Off Measures June 1988 Power Off Buttons Emergency pull type At present there are seven pull type Emergency Power Off EPO buttons located in the Computing Center building 1075 Beal Two of these buttons are located in the NSF computer room and will only power the NSF computer equipment off The other five will turn all computing equipment off that is the ES3090 300J equipment and its peripherals as well as the NSF computer and its peripherals The two EPO buttons for the
55. 0o00 print release print To print the list on a page printer run pagepr scards 000 or for onesided pages OPERATOR FILES MSS OPERATOR FILES 249 run pagepr scards 000 par onesided OPERATOR FILES MSS 250 MUP NAME FUNCTION USAGE The operator file the file MNET AUTH following messages Message Meaning Message Meaning Message Meaning OPERATOR FILES MUP OPERATOR FILES MUP To determine the current state of the UMnet Computer Network mts mup MUP runs the file MNOP MUP MUP first checks and if problems are found one of the is printed Outbound MTS access to UMnet is restricted from HH MM to HH MM MTS users are not able to MOUNT outbound network connections during the period specified This would normally be the case when either hardware or software testing on the UMnet PCP is necessary Running MNS will remove this restriction Outbound MTS access to UMnet will be restricted from HH MM to HH MM MTS users will not be able to MOUNT outbound network connections after the time specified This would normally be the case when hardware or software testing on the UMnet PCP is scheduled Running MNS will remove this restriction Error while accessing MNET AUTH A call to GETFD failed or a return code greater than 4 was returned by READ while accessing MNET AUTH If this message appears show it toa member of the UMnet staff Message
56. 223 HASP Status Display 210 HASP Transmitters Receivers 384 HELP ERROR EXTENDING CATALOG 407 HELP SNARK IN MTS 408 568 Hermes UMnet 400 HGS 218 HIMs 387 HLB 220 HLG 221 HLM 222 Holiday Rates How To Set 451 Holiday Sign Program 390 Host Interface Machines HIMs 387 HOURS Comp HPS 223 HUH 224 IBM ES3090 300J Water Temp Flow 559 TG SHUFFLE 227 TG Usage Summary 227 ILLEGAL INPUT 408 ILLEGAL REPLY 408 IMPROPER DEVICE NAME 409 IMPROPER DEVICE TYPE 409 Incomplete Logout 311 INCORRECT OR MISSING RACK ETC Cntr 389 409 INIT 84 112 217 228 Initemd 451 INIT UC3330LOG 212 Input Output Configuration Data Set 395 Instruction Processing Damage 311 INTEL 3805 66 INTEL 3805 Or 3825 Low Spares 362 INTEL 3805 3825 66 INTEL 3825 reseting 72 Intertask Communications Test 229 Interval Timer Error 312 Introduction To MTS 17 INVLD 147 IOCDS Installation 395 I O Devices Controlling UB VM XA 38 IPLBOOT 78 IPLINIT 78 IPLREAD 78 IPLREADER amp Friends 77 IPLREADER ITSELF 80 IPL Sequence Results 84 IT 113 IT IS TIME TO ADD ANOTHER ETC 409 IT MTS 114 TTX 229 JOBDUMP 115 308 See Also Superdumps Jobdumps Printing 365 JOB N WAITING ON ETC 410 Job Program 21 Job Programs 91 JOBS 18 117 Job Scheduling Program Restart 192 JOBS LOA 117 Keyboard OPER 33
57. 3 SET UP THE PLOTTER COMPUTING CENTER HOURS 393 ENTER TITLE LINE title line CENTER LINEUP OR JUST DRAW LINES J X FOR LINE 1 2 TEXT FOR LINE First line at 2 X FOR LINE 2 Bah TEXT FOR LINE second line at 3 X FOR LINE 3 As TEXT FOR LINE third line at 4 WHICH FORMAT 4 LINE 3 LINE REPEAT OTHER EXECUTION TERMINATED NOTE On the next page are examples of the signs which were generated by the program Continued on next page COMPUTING CENTER HOURS 394 A Title first line second line third line fourth line first line second line third line centered fourth line Title first line second line third line centered fourth line Title Line first line at 2 second line at 3 third line at 4 COMPUTING CENTER HOURS 395 How to Install a New IOCDS Operations Version K Bosley C Lever IOCDS Input Output Configuration Data Set To install a new IOCDS perform the following steps 1 Shut down UM MTS or VM and its guests and press the STOP key on the ES3090 300J console 2 Type F IOCDSM and press ENTER to select the IOCDSM IOCDS management frame on the 3090 3 The ES3090 3000 saves 6 IOCDSs for each side of the machine A0 A5 on the A UM side BO B5 on the B UB side only 4 happen to be defined for the UB side currently Observe that each is tagged with a date of the
58. 300 repeats of this sequence the tape is rewound and tested at 1600 BPI and then again at 6250 BPI The test will be skipped with a comment for any density not available on the drive being tested For 7 track tapes the test is similar except that each repeat in the sequence of 300 is done at a different one of the 15 7 track modes so that all densities both parities the translator and the data converter are all tested Because of this only one 300 repeat sequence is done for 7 track tapes mts tpd The program will ask for a tape unit name and parameters The unit name may be any 7 or 9 track tape drive The optional parameters may be entered in any order after the unit name each item separated by at least one blank OPERATOR FILES TPD OPERATOR FILES 301 EXAMPLE The program will ask for a tape mts tpd ENTER TAPE UNIT NAME AND PARAMETERS t900 1600 The above example would cause drive T900 to be tested at 1600 BPI only The parameters which may be used to modify somewhat the default behavior of TPD follow 800 For 9 track tapes the tape drive is tested only at 800 BPI For 7 track tapes this program is ignored 1600 For 9 track tapes the tape drive is tested only at 1600 BPI For 7 track tapes this parameter is ignored 6250 For 9 track tapes the tape drive is tested only at 6250 BPI For 7 track tapes this parameter is ignored ALL This parameter causes each of the tests to
59. 300J and VM XA on the other half UB MTS and DOS VSE runs under control of VM XA VM XA must be loaded and running before UB MTS can be loaded UB MTS must be shut down before VM XA can be shut down At the present there are three things that operators will need to do to run VM XA 1 Monitor the VM XA operator console 2 Load VM XA 3 Shut VM XA down Monitoring the VM XA operator console Messages are written to the VM XA operator console one line at a time When the screen is full the word HOLDING is printed at the lower right hand corner and no more information is displayed until the operator holds down the ALT key and presses PA2 It is important to check the VM XA operator console every few minutes to make sure that no backlog of messages builds up Some may be important In general important error messages on the VM XA operator console will be highlighted In any case the operator should try to keep an eye on the VM XA operator console and look for messages that may indicate a potential problem The VM XA operator console is an area in which we do not have much experience all we can do is try to build up some experience as we go This writeup will be updated to describe common VM XA problems as soon as we have run VM XA for long enough to discover what kind of problems commonly occur There are two binders on top of the 3370 disks in the first floor machine room which contain three VM XA manuals which contain
60. A RS Se A Ea he Beers 490 Task A Running Invocation Of A Job Program ooooooooooo ooo 21 UB MIS Under VM osar LA SE ew 29 UM MTS Running Of On 3090 Natlve sicer essa ea sa ts 26 UNIX Filesave Saves Files On UNIX System o oooooooooooooooo 557 Virtual Memory Paging DEVICES boscosa a Nees 66 MSE 30 9 Oh A A AUD BSS Bile ca ade Rubles ee Bids Rv bu Reed ters 50 What The TPL Sequence Did kerestesi eredue oe Soe he aS SOS Oe elie bs Win lS ae hee Bees OS 84 xerox 9 90 Photo Receptor Bele ends ds aaa rado 562 INTRODUCTION TO MTS THE MICHIGAN TERMINAL SYSTEM 17 OVERVIEW OF THE SYSTEM MTS is a system developed by the University of Michigan to take advantage of dynamic relocation hardware Basically MTS allows a large number of users to make use of the computer as if they each had their own computer It does this by using Virtual Memory and Multi Tasking Techniques The virtual memory concept is simply a method of simulating a computer with an extremely large memory This is done by dividing both REAL and USER memory into pages of 4096 bytes each USER pages are loaded if possible into real memory however since users may request more pages of memory than is or ever will be available in the real computer a large number of user pages must be kept on Direct Access storage devices primarily Fixed Head Files or their Solid State replacements but a disk may be also used under heavy loading Of
61. A2 for the POWER ON RESET This should be done on the UM side only since LPAR doesn t exist on the B side of the 3090 To power up the IBM ES3090 300J 1 Go to the Processor Controller and press the green Power On button This action will power up the CPUs the MGs and the Processor 2 Go to the 3880 Disk Controller and press the Subsystem Power switch This should power on the 3380 disks If some of the disks do not power up open the panel doors to the set of drives and turn the Front Drive and Rear Drive switches off and back on If you have to use the Front and Rear Drive switches to power up the disks then you ll have to go back to the Disk Controller and one at a time turn the Storage Direct Power switches SD1 and SD2 off and back on ES3090 300J POWER UP DOWN 26 SYSTEM SHUTDOWN AND START UP PROCEDURES Running UM MTS on the ES3090 300J There are two ways to run MTS on the 3090 native and under control of VM Currently at UM we are running UM MTS native and running UB MTS under control of VM XA The following discussion applies only to running UM MTS native on the 3090 see the sections Running MTS under VM for information on running UB MTS under VM In order to run MTS native on the 3090 you need to know how to do five things 1 Load MTS 2 Enable the time of day clock set control 3 Switch to a backup console 4 Force a superdump 5 Stop the CPU after a shutdown Loading UM M
62. CATION St ii dee ed sheng Stace wo Side eels tala 307 CONSOLE ERROR EONDETTONS rag h a A eh SRI Se A Bete ROS 308 SNARK GS so Se siete woe ten eS doa eis A A AA onan AA SENSE en LR eon Bd 308 HARDWARE ERRORS 100 9d alc hs bee le el ad aig De ie ge hs alee ew Mel eek Se ye Rs ele Oe 310 EPIA A ete ter A A A A SESS awe Fale eee bbe 310 Machine Check Handling Messages o ooooooooooooooo o eee teens 311 PUNCH ERRORS A E eter ahs oe eh ch tenses 313 PRINTER ERRORS ecg Shes Sk Suthers Gevalia ete Shel Selby a id 313 CARD READER ERRORS ssc 24 seeks ar Sst ote he Ne aa ate aha a Gatien a a ie Tecan Gv whe be ar Ea 314 DISK URATES Cheeks it a ah eves ame ee 316 Disk And Fixed Head File Unit Checks Printing 317 HEREDA SVG od sacs eee testa ge braun a id Si NE Gud catela G 318 Memorex Tine HUNG AR uo A aire had wh eels iat areal Bs 322 Memorex T270 ii AA A We A a eve Si 324 Miscellaneous Operations Information 22 2 eee eee eee 325 Carfiage Control Tape Eor MTS dana en sie ve GaSb Gost bleed NGS e ene 325 Eogigal Carriage Control ira daa oe gy oles 325 Machine Carriage C MtEOL unicas a a deal ge ES 325 Carriage Control Tape Bor HASP ts ti ls 326 EMBDSTA A A A A A aise A AS A toners 32 7 ConsoLe Operator 3270 ii ds 335 THe Sereen A taa ios 335 The POLOS ii A AA A Seg bo o 337 The Keyboarder gaia A AA A AI SAA A ce anaes 338 DEVICE Commands Hi a BANS SLES lo Bee wo AA we IS ae
63. Cc C COMPUTE WIDTH OF A TAPE SLOT Cc SLOTW STRIPL NSLOTS c C NUMBER OF LABEL STRIPS TO BE PREPARED c NSTRIP 46 Cc C FUDGE FACTOR FOR LETTER HEIGHT Cc FACTOR 0 669 Cc C COMPUTE LETTER HEIGHT Cc LETRH 0 4 STRIPH FACTOR Cc C NUMBER OF CHARACTERS IN A SLOT LABEL Cc NCHAR 4 Cc C LENGTH OF LETTERING FOR A SLOT LABEL Cc LETRL PSYMLN LETRH NCHAR C INITIALIZE VARIOUS THINGS FOR PLOTSYS ROUTINES c sc 1 0 THETA 0 0 CALL PLTXMX 48 0 CALL PLTOFS 0 0 1 0 0 0 1 0 1 0 1 0 Cc C PREPARE THE GRID FOR ALL LABELS Cc CALL PGRID 1 0 1 0 SLOTW STRIPH NSLOTS NSTRIP Cc C SET FIRST NUMBER Cc K 0000 Cc C EXECUTE NESTED LOOPS TO WRITE LABELS IN THE GRID BOXES Cc DO 200 I 1 NSTRIP YLOC NSTRIP 1 STRIPH STRIPH 2 0 LETRH 2 0 DO 100 J 1 NSLOTS XLOC J 1 SLOTW SLOTW 2 0 LETRL 2 0 Cc c POSITION FOR LETTERING IS XLOC YLOC AND THE LETTER CODE TAPE RACK STRIP LABELS 506 Is C OR N OR WHATEVER YOU WANT TO MAKE IT C CALL PNUMBR XLOC YLOC LETRH K THETA C ZI4 SC E C INCREMENT THE SLOT NUMBER E 100 K K 1 200 CONTINUE CALL PLTEND END TAPE RACK STRIP LABELS 507 TAPE DIRECTORY PROJECT The tape directory facility is designed to manage the tapes that are used for various functions within the system such as system statistics command statistics billing etc Currently these functions are all handled
64. Computing Center will be locked and NUBS will be closed To display unattended mode UAM O OFF 1 ON d uam To enter unattended mode m uam 1 To leave Unattended mode m uam 0 NOTE When entering unattended mode on either system do the following mod uam 1 mod termprio 1 mod btchprio 1 release minimum When the systems are taken out of unattended mode issue mod uam 0 mod termprio 0 mod btchprio 0 Users will be able to use MTS via remote stations or terminals but will not be able to use parts of the system which require operator intervention such as mounting magnetic or paper tapes When the operations staff is ready to leave on Saturday and Sunday nights there are a few things to be done as unattended UNATTENDED MODE 519 mode is entered The operator broadcasts a message telling of the impending unattended state of the computer At this point the operator should put in a SIGNONM message The message should be THE SYSTEM IS IN UNATTENDED MODE USERS COPY UAM Other things to be done by the operator at this time are drain the printers and drain pchx The printer doors should be left open and the printer gates should be opened stopping the print train or chain motors The operator should enter unattended mode by typing MOD UAM 1 on the operators console This entry will prevent users at terminals from mounting tapes and will rerun batch jobs which attempt to mo
65. DATE OR ALL R C1234 02 05 76 03 10 76 operator U WSG 03 20 76 U W035 R P10 RACK DATE TIME USER PROJ JOB TYPE UNIT MODE STAT FLAGS C1234 02 23 76 13 45 32 K1FO 9TP T903 OK LP DUMP 03 01 76 21 32 76 W035 9TP T902 OK RING LP DUMP 03 03 76 17 31 39 W035 9TP T900 OK RING LP DUMP 03 08 76 18 15 47 W035 9TP T903 OK RING LP DUMP 03 17 76 17 23 51 W035 WOPR 01519 9TP T900 160 OK RING LP DUMP 03 18 76 19 38 32 W035 WOPR 01870 9TP T901 160 OK RING LP STAT 03 01 76 07 53 41 W035 9TP T901 OK RING LP OPERATOR FILES MSS 242 OPERATOR FILES Continuation of Examples 0011 08 30 73 18 30 06 W035 S011 09 17 73 21351332 S11 01 13 75 21 56 54 W035 s188A 04 13 72 22 01 02 W035 s188B 04 13 72 22 06 58 W035 S417 02 23 73 19 58 33 W035 YYY P100 P101 P102 103 08 18 75 16 08 18 75 16 09 02 75 23 01 10 76 21 2 00 20 2 02 22 2 19 02 2 01 20 P EXECUTION TERMINATED NOTE 02 12 76 18 14 10 W035 SDDZ SDDZ K2LE SBLM 9TP 9TP 9TP 7TP 7TP 9TP 9TP T909 T908 T900 T700 T700 T900 T900 PTPR PTPR PTPR PTPR PTPR PTPR PTPR PTPR OK NL OK NL OK NL OK NL RING OK NL RING OK NL RING OK NL RING OK OK OK OK Up to six s can be used in the rack number section letter being ignored beginning with C12 RACK Rack Number
66. F if request from file job B if request from old style batch job TAPES TAPE HANDLING rrrr XXX 489 is number of 9 track tapes desired is a the rack name alphanumeric or b POOL for a pool tape is 9 for 9 track tape 7 for 7 track tape is OUT for ring out on tape IN for ring in on tape paper label must match what is written on a paper label affixed to the tape Valid responses are OK rrrr Txxx comment To indicate that tape with Rack Number rrrr is on drive Txxx e g Ok c3296 t902 When mounting a pool tape the response must indicate which pool tape was mounted for example ok pool91 t907 NA rrrr comment To say NOT AVAILABLE to tape request the comment must appear and will be printed on user s output to indicate why operator did not mount tape Do not use if all tape drives are in use use BY for terminal requests and RER for batch requests e g na 9876 no such tape ABORT rrrr comment To say NA to this tape request and skip remaining tape mounts of a multi tape mount request comment must appear and will be printed on user s output to indicate why tape request was aborted The alternative to AB is an NA for EACH of the tapes of a multi mount e g ab 9876 no such tape LE rrrr comment Label Error to be used to signal a label error to the user This response should be given only after the TAPE LABEL INCORRECT message to the operator
67. ILLREP MOUNTCMD MNT02802 SOURCE MOUNT 101 ST101 MOUNTCMD MNTO1505 MISCELLANEOUS MESSAGES 409 gt gt gt IMPROPER DEVICE NAME User Explanation This is the system message given to the operator when the operator replies to a mount request and specifies an invalid drive name e g T990 instead of T909 for a 9 track tape drive name DESTINATION OPER TEXTLOC MOUNT 101 BADDEVN MOUNTCMD MNT02805 SOURCE MOUNT 101 ST106 MOUNTCMD MNT01648 gt gt gt IMPROPER DEVICE TYPE User Explanation This is the system message given to the operator when the operator mounts a tape and specifies the wrong type of drive e g 7TP instead of 9TP or a drive which doesn t support the mode specified by the user via the MODE parameter e g a 6250 bpi drive when MODE 800 is specified DESTINATION OPER TEXTLOC MOUNT 101 BADDEVT MOUNTCMD MNT02807 SOURCE MOUNT 101 ST33 MOUNTCMD MNT01678 SOURCE MOUNT 101 ST114 MOUNTCMD MNT01781 gt gt gt INCORRECT OR MISSING RACK NUMBER User Explanation This is the system message given when the operator replies with an incorrect rack number where it is required in a reply to the MOUNT routines e g C1324 T900 instead of C1324 T900 to mount tape C1324 on drive T900 DESTINATION OPER TEXTLOC MOUNT 101 BADRNUM MOUNTCMD MNT02804 SOURCE MOUNT 101 ST104 MOUNTCMD MNT01594 gt gt gt IT S TIME TO ADD ANOTHER PUBLIC VOLUME METHINKS User Explanation
68. If none are available restart from the front panel and run xDL Another somewhat unlikely possibility when this message appears is that the control unit the path from the control unit to the channel the channel itself or some combination is logically offline The UNITS command UNITS D xxOP where xx is AA AB AD AE AN or CC should provide the information necessary to see if this is the case If it is the ONLINE job can be used to place things online lt name gt Hicore system is running Start xDL lt name gt is running but must be dumped and reloaded using ADL BDL CDL or NDL Once this is done rerun HUH OPERATOR FILES HUH 226 Message Meaning OPERATOR FILES Unexpected status from lt name gt show this output to This message should never appear but if it does it isn t clear just what the problem is or how to fix it Re running HUH might result in a better message the second time In any case please show the output to a Communications Group member HUH may also be run from an MTS terminal with the following command run sys huh Output is written on SPRINT and or SERCOM OPERATOR FILES HUH IG SHUFFLE NAME FUNCTION USAGE REMARKS OPERATOR FILES 227 1G SHUFFLE This program sources the file GRAF 1IG SHUFFLE and causes subsequent summarization of IG usage Normally started by the CLOCKWATCHER program If it were to be st
69. In addition there are four modifiers that are selectable by adding a parameter following the command Legal modifiers are B Busy Select only those devices that are allocated to a job W Waiting Select only those paths for which there is an I O operation I O queue active or START I O pending A Active Select the most recently active path s S Status Print detailed device status if possible O Offline Select those paths which are logically offline in some way V Verify Verify that the tables are internally correct CLR CLRIO Does both a FAKE I O anda CLEAR 1 0 Use only with extreme caution B W and A imply 8 If no modifier is given detailed information is not printed devtype is a device OUTPUT from UNITS The output from a UNITS request has several parts PART 1 The path specification This part shows how a given device is connected to the system through which channel and control unit and the logical ONLINE OFFLINE ness of each part of JOB PROGRAMS UNITS OLD STYLE 160 JOB PROGRAMS the path PART 2 The address This part specifies the address associated with the path given in PART 1 PART 3 The job number This part appears if a particular task owns the device in question and gives the task number of the job PART 4 The general status information This part appears if the S B W or A options follow the units command It gives selected gen
70. LASER PRINTING SYSTEM VERSION XXXXXX DISK 1 D XXXXXX ENTER DATE MM DD YY 5 Enter the date with the numerals such as 02 14 18 You may use a space a colon or a slash between the month day and year Once you have entered the date by pressing the enter key the display should show it to you in a written version e g 1 23 86 becomes January 23 1986 6 The display will request that the time be entered ENTER TIME HH MM SS Enter the time usint the 24 hour system i e 2 30 P M is 14 30 Both the date and time will now be displayed 7 Display asks 9790 POWERING ON LOADING PROCEDURES 561 ARE THE DATE AND TIME CORRECT AS DISPLAYED Y N Enter Y if correct If incorrect enter N and reenter the date and time correctly 8 Finally the screen asks if you want the time status displayed in the upper right corner of the display LOAD AND ENABLE TIME OF DAY STATUS DISPLAY Y N In most cases you should answer yes Y If you respond yes to this question the system displays the time and system status in the upper right hand corner of the screen If at any time you wish to disable this feature enter TIM E OFF To re enable the feature enter TIM E ON Tf you select no at this time you cannot add the time status display without re booting the system 9 Allow 4 to 7 minutes for the printer to completely warm up 10 When the printer is ready a multicolor screen appears on the U
71. Mount C3297 119 leek 786 on T902 Ring Out 00097 09 39 00 MTS SQB1 119 leek 786 Accepted on T902 To let the user know that the tape was mounted OK If the Rack ring status Device Type or Volume Label is incorrect then an error message is generated If the user just wants a tape to use temporarily and not save he requests a pool tape mount pool on 9tp p and the request on the operator s console has POOL for a rack number There are a set of special pool tapes to use These tapes are labeled magnetically as well as visually POOL91 POOL92 always mount these with the ring in For detailed information on the possible responses and error messages from MOUNT see below RESPONSES TO MOUNT If more than one tape is being mounted by the mount request a preliminary message is issued to the operator listing all the rack numbers the device types densities and whether or not the rings are to be IN or OUT With this information the tapes may be gotten out ahead of time For example MOUNT C3276 C0701 The operator is aware a tape mount is desired when the following message appears on the operator s console ccid q tt n 9TP rrrr ON tTP RING xxx paper label ccid is the signon id of the user requesting the tape mount a is Q if the request is from the mount queue a is n if this is the first instance of the request tt is HB if request from hasp batch job T if request from terminal
72. OF PLOCE SESS NOM oak sete Be es Sl Site A A A OS GARI oe 431 ECMD sPPODLEMS Aina a a e SEE da gated A IA IA it ds eve 195 EMDSTAT Problems iii A AA A coe Sh aes E dak A 195 Consoles Malin amp BackuD iii A a A eaters 22 COMEMES a id muse lao e 3 Deferred PIOS A A A sees erie Sees E 428 Deterred Plots PLOCCING Li id ada Weed sb vie Mle 429 IBM ES3090 300J Water Temp Flow o o o ooo ooo 559 NO A O A E A a a a irae te ARs 565 FNHEPOAUCELON To MIS gt A A A AS A A ii Da 17 Job Program Job Programs For Each System Function 21 TOD Programs A A A A A AA AS EGR Eee Be 91 BLAST Forcibly Removes Job If STOP Will Not 92 BROADCST Sends Operator Messages To Users ooooooooooooooo o 94 DIS Display Contents Of Real Memory oooooooooo ooo oooooooo 97 DMGRSTAT 4 DESK MATT PULA ELON A A A A tc 98 FLUSH Clear Requests From HASP For I O OPS o ooooooooooo 1 108 GOOSE Att Interrupts Batch amp Oper Cons 1 110 INI Thes tnitiat s System AEPD A hee le eee a 112 TT Names And Lists NetWOrkS oaod ese sd ds ad ss a 113 JOBDUMP lt Forces DUMp OM Tape axe sedd we a a ele ween ad 115 JOBS Gather Display Load Statistics oooooooooooooooooooo o 117 MOD Modify Location In Real Memory o o oooooooooooooooooooooo 118 MT Ses Starte An MTS Task A A A a ica 19 OFFLINE Make Device CU Etc unavailable
73. PROCESSOR command shorthand for DISPLAY PROCESSOR In addition JOB shorthand for DISPLAY JOB will give information about a specific job And DISPLAY DEF can be used to find out current definitions of things A QUEUE If you type QUEUE ALL or Q ALL or just Q it will print one line for each of the 25 currently queues giving the queue name how many jobs are in that queue and what the criteria are for assigning jobs to that queue Jobs are assigned to queues by going down the list of queues in the order printed and putting a job in the first queue for which all the criteria are met If you just want to look at one queue you can ask by name Q WAYNESTT will just print the entry for queue WAYNESTT Outgoing BITNET traffic will be found in queues the same name as the link WAYNESTT UMICHUB Incoming BITNET traffic ends up in the queue MESSAGE if it was mail or in the queue IMPORT if RESOURCE MANAGER 537 it was a file The normal good case is for all the outgoing queues to have no entries and for the MESSAGE queue to also have no entries The IMPORT queue will normally be quite large over 100 as the jobs that represent incoming file transfers stay there until the user copies IMPORT into their own file Most other queues are currently zero or very small except for NETPURGE and MSGPURGE which are usually very large because they are defined with a 48 hour retention period The comm
74. Phone Pager Numbers Note Every pager has an ordinary telephone number After calling a pager s number from a touch tone phone you can enter numbers using the phone keypad that are then transmitted to the Digital Display pager and shown on its screen By entering the phone number where you can be reached the person with the pager will know where to return your call To reach someone 1 Dial the pager number from a touch tone phone 2 Listen for the ring and a tone 3 At the end of the tone enter up to 23 numbers to be displayed on the pager your phone number 4 If you enter a wrong number press three times then re enter all numbers to be displayed 5 Press the symbol and hang up The following currently have pagers assigned to them Baker Jane 996 6822 Deneau Carol 996 6174 Donnelly Steve 668 3692 Run Sys Fire for list of Firefighters currently on duty Fire Fighter 930 8086 Systems Manager 930 8976 See next line Only F F allowed to beep manager Don t call manager unles 1 Firefighter has been called but can t fix problem or 2 Firefighter hasn t responded within 20 minutes Gluski Kari 930 8976 Gold Steve 668 3954 Hankins John 996 6759 Jaworski Dennis 996 6282 Voice Kissling Gene 996 6823 Knox Jim 996 6504 Lone Ken 668 5699 Voice Martin Dave 668 5083 Masten Deb 761 0881 Mitchell Lee 996 6806 Myers Jack 761 0920 Olsen Steve 930
75. REQUIRING NO USER TAPES PLOT TAPES IDENTIFYING INFORMATION PLB NAME FUNCTION USAGE EXAMPLE PLOTTING SYSTEM PLB This program prints identifying information about plot tapes i e the name e g TUE 1000 when it was last written and the minimum time to re use To run from the Operators console enter mts plb The program will print ENTER TAPE NAME Respond with the name of a drive on which a plot tape is mounted e g T911 The program will print several lines of tape information and then prompt for another tape name Enter an end of file CNCL to terminate the program mts plb ENTER TAPE NAME t911 TAPE MON 2000 LAST USED 11 06 78 22 00 11 433 MIN REUSE TIME 6 DAYS 0 HRS OMIN NEW PLOTS REPLACE OLD ENTER TAPE NAME When running the program from somewhere other than at the Operators console run sys pident When prompted for a tape name enter either the name of the drive e g t901 or a pseudo device name e g tape The program reads tape names from GUSER and prints its output on SERCOM ERROR MESSAGES NOT A TAPE The tape name entered was not the name of a drive or proper pseudo device name or the drive was busy NOT A PLOT TAPE either the volume name was not plot or the name of the first data set was not IDENT FORMAT ERROR ON PLOT TAPE The tape was a plot tape but some error was found in the identifying
76. START UP PROCEDURES the screen 8 Type ATTACH nnnn nnnn where nnnn is the device number of the tape drive where the dump tape is located See the section Attaching and Detaching Tape Drives for a table of the tape drives and device numbers For example to take a dump using T901 the command would be ATTACH 0681 0681 Note that the device number appears twice in the ATTACH command 9 Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of the screen 10 Type SYSTEM RESTART and press ENTER 11 The UB MTS operator s console should request a UB MTS superdump 12 Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of the screen 13 Type DISC and press ENTER The message DISCONNECT AT time should appear In a minute or so the VM XA logo should appear again Stopping the CPU UB MTS VM XA You must do this as the last step in shutting down the UB MTS system 1 If you are planning to reload UB MTS right away just proceed to the instructions on Running UB MTS under VM XA Loading MTS 2 If you are planning to shut VM XA down or don t plan to run UB MTS for a while perform the following steps 3 Go to the VM XA operator console 4 Type FORCE UB MTS and press ENTER UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 37 VM XA UB MTS Attaching and Detaching Tape Drives
77. Sans Sat a ease we ee 460 Signon Message MessagesysteM ooooooooooo ooo o o ooooooo 463 SENSES VALUES E A A ge AA A Ae es 466 SUPESLGQUMP S venera PRES WB A A A REA A AA AAA 467 System Relwability Firefighters isis a bee ee hk ees 468 System Status Recorded Message cee eee ee ee ee eens 469 Systemstatus tor Operators Mint e e r a a elg wen acetates or a 471 TADEO dl dad 486 tape Hand ing e riedia A A A A a ee hie dt 486 TADS BELOS ia sis Si hela Su 8 ately Ween ge a a OB a deller Wy gene Rtede Ol Aa Gi 5 atelev heels 493 Tape Error COSES Lai oie sees Bae eae Wk ee ie as ee Ba ane eke Se Se OS 494 Tape Mount SUSDEMSWON a it iaa 499 Tape Diiskette MATTING eros io daa a we be wad ae 500 Diskettes t Matti A ld is 500 Tapes MAD A A O A AS e 500 Tape Racks Electric Problems 25 3 6 sis ad Se os ri a a e aia 502 Tape RaCk Strip aDeSe av a 503 Tape Directory PrOPSCE oe ss 507 Unattended Mode UAM e A A iis A a A es 518 WAM DA gai eee ee ange EA ber ae Cada AA DARE a td 519 UNYN Terminals And SCOPS BEG mais ai a ss 520 Vendor Informati n in E A ap Eco has eee OE 527 NUBS Terminal Stations TAS 6 8 seresa co A iris 531 Maintenance Agreements seredan ae Sib ese ere Gs AA a A ge 533 Resource MANAGES is AS A 9 ea Sas he a Bees 535 san Dreg S per EOMPpuUESLE ss sla se ere die ta 543 special Formes Jobs artis Searle ale oer ele Wee Wedel be a eee ele Pee 547 UPS Uninterruptable Power Supply ooo
78. Sikes ac ales Bh a N A aua en L TAPE COMMANDS S OPpErATCOT mii ci A AS A EE AA L TASKO ua ii a di dal an L 168 FABS E A A wien eas a a Saar gue do le L L70 L71 L72 L73 L74 L75 L76 L77 L78 L79 180 181 184 186 187 188 69 O States aS aoe 189 A a e A E O T A ee ree E A tigi Pe cag ores ee E RA ee ne ree ee a a ere ey es er ea oe Dee er A ne Crear oe CPW ee ee ee er gate E Be pee Rear ter etme ete O gers a O DA a pe ado O a Le a o oe ae ESE AICA q DER EM ne Ae Ea apie Cy a Po A A A EENAA NA MSA SO a Pn T T O S NS N ate ee ee ee ee a ere She q EC AS DAL DAA AA A MAPA Ea RRL AE SO oan S ee ee ee ee ees ae A o O am E A E E ee E Ge o epee nar i EE OO N re ee eye ees ne A A MEDI oe aL ae hi WRU A Daa aaa A eye a ee ey eee ee S a O A tore mer manana ert ep a a a ee ee aa 554 A tee ea eae ee esa Aas A A I MLE EE RIBS Ore ae ey te age ne sea tere aegis ae peer ee te eee ee oe A ALL SAARI ERE ua AS GL a ae INIT CCU es acme ERA ADS ee Pere ee A a ea r A ee eee er oe are e ROO al ada ren E nat ae ee ee ee ee ng eee ete er nee ace as A OO o oa S i E E eta erg Bae a Lens hit Ack eerie Lik OR PRE unk LCG ae Bee et trig ty RS sap Ro ee te en ee ra re is eA UD Did BRIE Cy cs A ae ee En o od poi a O E A A ae ae ae Ge eer oir T ee emg ee aye ee tare fee A ne Re gia Sak a E e eee ee ee tere ey Oca as pes iene ae i Aa O RSPB SCRIP TE ap a ari A As A in ean ie D
79. Sraa TES E MPSS a a A ta hs ios one es AA ta hee ee ws cl 153 TN Lo6ad TN VES BUESA Git Sees teases ns Gest a eae or Steaua eRe sare e aaa 154 Taide Loads TLL UESTBUPESE A A ee re 155 UNITS Display I 70 Confirguration id tt od L56 UNITS Old Style Display I O Config CHs Cus Etc 1 157 Loading UB MES La A A A 32 Memorex LI A a a Gude ea ele aS a a eS 324 Micsellaneous Operations Information Host Interface Machines HIMs Sica Rar E taut et mae A A o rt ee eget 387 Tranemtitters Recelvers HASP see S VE AA oe eee eee 384 Miscellaneous Operations Information ACC FILEINFO Dead File Space Retrieval TOE OMA TOD a Ae 358 Acct Listing File Charge Update ooo oooooooooooooooooo 368 Carriage COntrodk TAPES is A otis Son dk ae dae a 325 326 CCDs 5 DO terreduD IES ii aero isle ade E a tenga toe oo elt weenie BN 420 CCI PLO Tape INTETALUZGELOR od axe eiers 4 ites acre sad de 420 CCP Gathers User Plot Files For Tape o o ooooooooooooooooo o 421 CCT Length Of Plot Session No User Tapes Involved 420 Channel D OGOQUES lt PDE A AS A DE e A CAE Ts TD EMBSTAT A en SPd se bx A ala 327 computing Center Hours tote ee A ee ELS EPS eee 389 Console Backup Support MTS ES3090 3007 ooooooooooooooooooo 346 Console Operator S270 ox eee Shs A a AA AA TA Se 335 DAC ETEOr COS a A ebb a A a a 361 DASDT TR A A E A Sainte el ae ey whl RA AA ES A 354 DEADFILES Schedule amp Procedure oooooocoooo
80. T946 1086 T955 108F UB MTS UNDER VM 38 SYSTEM SHUTDOWN AND START UP PROCEDURES Determining whether a tape drive is attached to UB MTS if you are unsure about whether a tape drive is attached to UB MTS or not issue the following command at the VM XA operator console QUERY 0680 068F 1680 168F Some possible responses are nnnn indicates the device number TAPE nnnn OFFLINE TAPE nnnn FREE TAPE nnnn ATTACHED TO UB MTS nnnn R W The last message TAPE nnnn ATTACHED indicates that a drive is attached to UB MTS See the section Controlling Devices for a description of the meaning of each of these messages The ATTACH command To attach a tape drive to UB MTS issue the following command at the VM XA operator console where nnnn is the device number for the tape drive ATTACH nnnn TO UB MTS nnnn Note that the device number appears twice in the ATTACH command The DETACH command To detach a tape drive from UB MTS issue the following command at the VM XA operator console where nnnn is the device number for the tape drive DETACH nnnn FROM UB MTS Controlling I O Devices UB VM XA When UB MTS runs under VM XA VM XA has ultimate control over all of the I O devices attached to UB MTS In some cases it is possible for a device to be unavailable to UB MTS because VM XA is unable to communicate with it In such a case the problem must be resolved at the VM XA level from the VM XA operator con
81. Type Command stop It will take some time to mount and scan the tape Please be patient Use ATTN goose to abort Operator mounts tape s requested by RES w000 VIDEO1 10 48 07 07 10 77 SUNDAY COMPLETED Type command dis w000 d L DG saved TUE 10 30 84 03 17 01 DIRECTORY000 saved SUN 10 28 84 09 26 27 There are 5 older versions of this file L DISASTER saved SUN 10 28 84 10 56 36 There are 5 older versions of this file L DISIG saved 10 28 84 10 14 22 There are 7 older versions of htis file Type command dis w000 d s histroy 5 1 DISASTOR saved SUN 10 28 84 10 56 36 2 1 saved SUN 10 21 84 10 06 44 3 saved SUN 10 14 84 10 12 13 4 saved SUN 10 07 84 10 21 37 5 saved MON 10 01 84 01 41 28 There is 1 older version of this file 1 DISIG saved SUN 10 28 84 10 14 22 y saved SUN 10 21 84 09 20 31 saved SUN 10 14 84 09 30 50 2 saved FRI 10 12 84 03 03 43 ti saved WED 10 10 84 02 40 21 There are 3 older versions of this file OF WN Type command stop OPERATOR FILES RES 274 NAME FUNCTION USAGE OPERATOR FILES RIP Produce a ripple pattern on a printer with a PN train or chain mounted Drain the printer Enter mts rip ptrn where ptrn is the name of the printer To stop the rippling use the GOOSE or STOP jobs on the task OPERATOR FILES RIP OPERATOR FILES 275 RHB NAME RHB FUNCTION To Hold amp Release HASP R
82. UB MTS The loading of UB MTS should only be attempted after loading VM XA or after shutting UB MTS down To load UB MTS 1 Go to the VM XA operators console 2 If you are not sure whether the UB MTS virtual machine is logged on type QUERY UB MTS and press ENTER If the UB MTS virtual machine is not logged on VM XA will respond with the message IB MTS NOT LOGGED ON If the UB MTS virtual machine is not logged on type AUTOLOG UB MTS and press ENTER Then skip to stop 13 If the UB MTS virtual machine is logged on proceed with step 3 3 If the UB MTS virtual machine is already logged on do the following or from VM XA operators console type force ub mts then autolog ub mts 4 Go to a terminal connected to VM XA 5 At the LOGON prompt type UB MTS and press ENTER to logon to the UB MTS virtual machine console for the UB MTS virtual machine 6 The terminal should respond with the message Enter Password 7 Type the password and press ENTER If you don t know the password ask your shift supervisor 8 The terminal should respond with the message Reconnected at time 9 Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of the screen 10 Type IPL 52C and press ENTER 11 Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of UB MTS UNDER VM 34 12 Sr
83. UB filesave is done at 1 00am on Saturday Daily filesave is started automatically Tuesday through Friday at lam A partial filesave daily is also done at 8 30 pm on Saturday and Sunday Original full filesave tapes are kept in the machine room and the two latest copies copies are stored in the safe Six week copy tapes are kept at Bendix after they have been in the safe for a few days awaiting transportation New tapes should be used when making six week copies from the originals Lists UM and UB The file save listsings that need to be printed and kept at the Computing Center are The original listing from each weekly filesave which should have Info printed on the first sheet and FILESAVE SCHEDULE AND TAPE STORAGE 334 The copy listing from the six week filesave FILESAVE SCHEDULE AND TAPE STORAGE 335 USING THE 3270 OPERATOR S CONSOLE THE SCREEN The Operator job operates using a 3270 3278 or 3066 display The display screen contains 24 to 43 80 character lines it is divided into three areas as follows 1 Lines 1 and 2 System status area 2 Lines 3 22 3 32 3 41 Message area 3 Lines 23 24 33 34 42 43 Request area The use of these three areas is further described below SYSTEM STATUS AREA These two lines will contain current system status information as follows line 1 Output from the system status routine SSRTN line 2 A list of keywords alerting the op
84. UP PROCEDURES DEV NAME IS UMMPS NAME OF UNIT DRM1 D237 ETC VOLUME IS AVAILABLE ONLY FOR MTS PACKS PROGRAM FLAGS ARE AS FOLLOWS IN HEX ADD BITS IF NECESSARY 800000 2321 400000 2314 200000 2301 100000 RETURN ON FATAL ERROR 080000 CCW LISTS CONSISTS OF A NON TRIVIAL SEEK 040000 RETURN ON TRACK OVERFLOW 020000 RETURN ON END OF CYLINDER 010000 RETURN ON NO RECORD FOUND 008000 OPERATION GOT UNIT CHECK ON SIO 004000 IMMEDIATE OPERATION 002000 NOT FIRST ENTRY 001000 BUS OUT CHECK OCCURRED 000800 INTERNAL I O IN PROGRESS 000400 READING RECORD 0 000200 RECOVERING FROM TRACK CONDITION CHECK 000100 REDOING A STAND ALONE SEEK 000080 EQUIPMENT CHECK HAS OCCURRED 000040 DOING A SEEK TO A DIFFERENT ADDRESS 2321 000020 LOOKING FOR A HOME ADDRESS ERROR ON 2321 SENSE FLAG VALUES 00 NO SENSE REQUESTED PROGRAM ERROR 01 GOOD SENSE DATA 02 UNIT CHECK ON SENSE COMMAND 03 DEVICE ALL FOULED UP NO SENSE POSSIBLE 04 CHAINING CHECK 08 INTERFACE CONTROL CHECK 10 CHANNEL CONTROL CHECK 20 CHANNEL DATA CHECK 40 PROTECTION CHECK 80 CHANNEL PROGRAM CHECK OTHER PROGRAM ERROR SENSE DATA SEE SRL FEMI ETC SEEK ADDRESS LAST ADDRESS USED IN A SEEK COMMAND FORMAT IS CCCCHHHH OP CODE OPERATION CODE IN LAST CCW BEFORE THE UNIT CHECK VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 77 DOCUMENTATION F
85. a few messages are printed on the console VSE1 will display the message 0I80D LOCK FILE ON 537 DUPLICATE NAME ON VOLUME DOS LOCK FILE 0D07D ENTER RESPONSE Type DELETE at this point and press the ENTER key 8 Shortly after the LOCK FILE message you MAY see a message like 0137D LABEL AREA ON 422 OVERLAP ON UNEXPIRED FILE DOS LABEL FILE FF3712723090 AREA1 If this message is displayed VSE will stop loading 1 e nothing else will be printed on the VSE console after this message until you reply Type DELETE and press the ENTER key This message if it appears at all will be printed in the first minute or so after LOCK FILE message If you don t see it in the first minute or so it s safe to assume that it won t appear 9 Regardless of whether or not the LABEL AREA message appears after a few minutes you should see the message F1 001 10121 VSE POWER INITIATION COMPLETED FOR SYSID 1 At this point you can go to the VSE2 console and start loading VSE2 VSE2 is loaded the same way as VSE1 except that you logon to VSE2 instead of VSE1 and you should never get a LOCK FILE message while you are loading VSE2 You MAY get a LABEL AREA message as with VSE1 within the first few minutes of loading If you do geta LABEL AREA message you reply by typing DELETE and pressing the ENTER key 10 After the POWER INITIATION COMPLETED messages there VSE 52 SY
86. a minus sign prefix instead of the job number job name prefix Each new line will be inserted immediately after the last output line for this command or after the command itself if this is the first line thus output from a System Status command will not have other messages interspersed OPERATOR CONSOLE 3270 337 4 Completed requests from the request area All requests will be echoed by adding them at the bottom of the message area Successful new job requests will be prefixed by the 5 character job number all others will be displayed as is 5 Error messages from invalid requests e g JOB NAME NOT IN JOB LIST Tf a request is in error it will generally be left in the request area and will not be added to the message area The error comment will be added to the message area instead 6 Dump and machine check error recovery messages When a dump request or machine check occurs the entire screen will be cleared and the messages from these routines will appear at the top of the screen When the dump is completed or the machine check recovered from the original contents of the screen will be replaced with these messages added at the bottom of the message area An exception to this form of operation will occur if a JOB DUMP is requested as the result of a serious error within the console management program itself In this case the system console will be transfered to the backup console support see belo
87. and only after the operator has re verified that the correct tape has actually been mounted If the wrong tape was inadvertently mounted the correct one should be mounted and the OK response should be given again e g le c1104 TAPES TAPE HANDLING 490 IM rrrr comment IM user gets response p name NO UNIT AVAILABLE WHICH SUPPORTS COMBINATION OF MODES REQUIRED Example IM C4097 The IM response can be used if a user tries to mount a tape currently recorded at 800 BPI on a 6250 tape drive The IM response can also be used at installations which have no 6250 tape drives if someone tries to mount a tape using 6250 mode II rrrr comment II user gets response p name INCORRECT RACK NUMBER OR TAPE ID Example IR C4097 BY rrrr comment BUSY to be used to tell user that no drives are currently available This response will not work for HASP jobs use RER for them REPEAT The mount request is to be printed again MSG rrrr comment To print a message to the issuer of the mount request The read is then repeated for an OK etc EXAMPLE msg c5823 please call operator RERUN rrrr This will RERUN the job works only for HASP jobs and should be used whenever the operator desires to run the job at a later time normally because enough free drives are not currently available The HASP command RELEASE JOB n is used later to release the job The options enclosed in
88. applicable This is useful for determining what RES thinks it s going to do if there is some uncertainty Restores all files queued for restoration Terminates command input restores all files queued for restoration and then terminates execution of the program lt end of file gt or CNCL key on Operators console Same as STOP command lt zero length line gt CANCEL Ignored Removes all files queued for restoration by any previous RESTORE commands lt attention gt goose If STOP has not been issued yet an attention interrupt terminates the current command and requests another If STOP has been issued while the program is in the process of restoring files no more files are restored a list of unrestored files is printed and the program terminates The following is a sample session using RES Operator input shown in lower case mts res Type Command dis w035 video 1 w035 video saved tue 08 09 77 04 55 44 on tape 1308 ver 2 3 4 saved sun 07 31 77 10 43 31 on tape 1208 ver saved sun 07 24 77 09 07 18 on tape 1105 ver saved sun 07 17 77 08 40 42 on tape 1008 ver N N WW OPERATOR FILES RES OPERATOR FILES 273 5 saved sun 07 10 77 10 48 07 on tape 1505 ver 1 6 i saved sun 07 03 77 10 30 24 on tape 1408 ver 2 Type Command res w035 video 07 10 77 newname w000 video1 1 W035 VIDEO SAVED SUN 07 10 77 10 48 07 ON TAPE 1505 VER 1 W035 VIDEO has been queued for restoration
89. as 07 01 83 Ccids that expired on or before the date given will have DISKS DEADFILES 360 10 11 12 their files user file catalogs Restore Message and Userdirectory data removed from the system If no date is given yesterday s date will be assumed by default The last thing printed by this step will be the number of files to be saved Make a note of this number for comparison with the number printed by the next step Use the Deadfiles program to save the files that are to be purged from the system on either 9 track 6250 bpi magnetic reel tape or on 18 track IBM 3480 cartridges df_save When asked enter the name of the tape drive to be used Compare the count from step 3 to the counts printer in step 4 If the totoal number of files doesn t agree with the count of files from step 3 STOP Use the filesave copy job to make a copy of the magnetic tape that was created in step 4 by entering mts fsc from the operators console The stream name to use is FILES be sure to include the trailing perioed in the stream name When FSC asks if a listing is desired reply YES Compare the count of the number of files saved in step 4 with the count printed by FSC in step 6 If they don t not agree STOP If there have been any problems in steps 1 through 7 DO NOT go on with the remaining steps instead report the problems to a senior member of the systems group Now destroy the files that belong to
90. be possible to pass operator OPERATOR CONSOLE 3270 345 console input to these jobs without transfering to the backup support again If it is not possible to restart the OPERATOR job for some reason system operation may continue using the backup support but any requests for input that were made to the 3270 operator s console cannot be satisfied The only alternatives to running from just the backup support are contacting a systems programmer or IPLing Problems with the 3270 operator s console support should be called to the attention of the programmer responsible for the OPERATOR job currently Paul Pickelmann at UM OPERATOR CONSOLE 3270 346 CONSOLE BACKUP SUPPORT CHANGES FOR MTS ES3090 300J When a task requests a reply to a message mts hangs a read on the operator s console and waits for a reply for that task from the operator This has been changed in mts ES3090 300J so that these requests are queued by the supervisor which after having queued a request puts out a message on the operator s console giving all the tasks numbers of the tasks that have requests pending EXAMPLE PENDING 10 15 12 The support for the 3066 display console is such that it looks as if a read is always hanging on the console i e The operator doesn t have to press a REQUEST button before typing a line Thus the 3066 console is active or alive at all times Input is always entered on the line 2nd from the bottom of the screen The
91. be repeated continuously until the end of tape marker is reached instead of just 300 tries OPERATOR FILES TPD 302 NAME FUNCTION USAGE EXAMPLE OPERATOR FILES TPS A file to run UNSP TAPEDRIVES This program informs the operator of the number of available tape units If a tape drive is OFFLINE or in use the program will consider it unusable by the operator The program can distinguish between 800 1600 and 1600 6250 drives mts tps At This moment there are AT MOST 1 7 Track and 2 9 Track 1 6250 1600 BPI 1 800 1600 BPI drives free OPERATOR FILES TPS OPERATOR FILES 303 TSH Obsolete NAME TSH FUNCTION Adds tapes to or deletes tapes from UNSP TAPESHARE USAGE mts tsh EXAMPLE mts tsh addr lt rackno gt lt owner CCID gt EOF to terminate If an error was made while entering the tape e g the wrong CCID was used it can be removed and re entered delr lt rackno gt lt owner CCID gt then re add it using ADDR REMARKS When a tape has been entered in TAPESHARE via addr or addrack the owners may permit the tape any way they like by running UNSP TAPESHARE NOTE To find out if a tape is in TAPESHARE and which CCID owns it type t lt rackno gt Example t s1091 OPERATOR FILES TSH 304 OPERATOR FILES NAME TSN FUNCTION The weekly full file save slave stream SVW will start 6 to 8 of these Not normally used direc
92. can run but Minimum priority jobs can t Don t change the line numbers in this file The line Beginning with Day should be line 0 and the next line 0 x Day Begin End Low Def Min SUNDAY 00 00 06 59 ON ON ON SUNDAY 07 00 17 59 ON ON OFF SUNDAY 18 00 23 59 ON ON ON MONDAY 00 00 06 59 ON ON ON FRIDAY 18 00 21 59 ON ON OFF FRIDAY 22 00 23 59 ON ON ON WEEKDAY 00 00 01 59 ON ON OFF WEEKDAY 02 00 06 59 ON ON ON WEEKDAY 07 00 17 59 OFF OFF OFF WEEKDAY 18 00 21 59 ON OFF OFF WEEKDAY 22 00 23 59 ON ON OFF SATURDAY 00 00 06 59 ON ON ON SATURDAY 07 00 17 59 ON ON OFF SATURDAY 18 00 23 59 ON ON ON 2 Change the file of initial reload commands to include setting the special holiday rates This is done so that if the machine crashes during the holiday and is reloaded the rates will still be set to minimum Signon INIT Duplicate INITCMD INITCMD lt Edit INITCMD The lines below to be inserted into file initcmd should be placed after the line probably line 40 which contains mod csdw2 12 insert 40 or whichever line is appropriate SPECIAL RATES HOLIDAYS 452 insert 40 inserts a line or lines after line 40 when in edit mode The three lines which follow should be inserted MOD btchprio 1 MOD termprio 1 O fRelease minimum Press Enter Note Btchprio and Termprio values O indicates that rates are based on the time of day 1 indicates minimum rates 2 indicates deferred ra
93. change the label It is not necessary to strip copy tapes before using them since FSC will back space the output tape if it cannot complete the writing of a record on it and will ask for a new output tape NOTE The original filesave tapes should be kept at the Computing Center for use as restore tapes The copy tapes do not always agree exactly with the listings OPERATOR FILES FSC FSM NAME FUNCTION Usage NOTE OPERATOR FILES FSM Merge File Save mts fsm 217 FSM is normally started by SAV SVW or at IPL INIT OPERATOR FILES FSM 218 HGS NAME FUNCTION USAGE este CPU TIME A 3 DMGR 3 293 K9SK 0 Etc Etc Etc Etc Etc Etc OPERATOR FILES HGS OPERATOR FILES HGS Program to run MTA HOGS MTA HOGS will look at all jobs running on the ES3090 300J for about thirty seconds to determine which of these jobs are using the most CPU time VIRTUAL MEMORY Disk I O DRUM READS REAL pages and which jobs need the most REAL pages Ten rows of job information will appear under each of the six headings CPU DSK 1 0 VM DR and RPN This job information is the Supervisor the CCID or job type i e PDP and an amount pertinent to the heading Under CPU the first column contains ten supervisor numbers the second user ID s or job type and the third contains amounts of CPU TIME used in the last thirty seconds by each job Under VM first column equals supervisor n
94. control units offline cu 2 offline cu ptrx To offline a channel offline ch b To offline paths between a channel and control units offline ch b 2 offline ch b 2 3 offline ch b ptrx rdrx JOB PROGRAMS OFFLINE NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES JOBS PROGRAMS 125 OK To set a switch so as to print or not to print the task termination message on the console when the termination code is O K OK ON OFF Normally when a job terminates with code O K a termination message to that effect is not printed on the console This job sets a bit so that this message can either be bypassed not printed or be printed Note that if OK was ON then when any MTS task started by HASP terminates you will get the O K message which will tax the console just a little bit more ON or OFF ok on ok off JOB PROGRAMS OK 126 JOB PROGRAMS ONLINE JOB PROGRAMS ONLINE NAME ONLINE FUNCTION To make a device control unit channel or the path between a channel and a control unit available This is the inverse of the OFFLINE job PROTOTYPES 1 online devname devname 2 online cu cu cu 3 online cu devname devname 4 online ch chan adr 5 online ch chan adr cu cut 6 online ch chan adr devname devname USAGE If device names are given 1 then the devices are marked as online and available for use If the string CU followed by by contro
95. counselor for assistance NO ACCESS TO file PLOT ABORTED The file containing the plot was not permitted to the postprocessor Either the plot file was not permitted properly at the time it was queued but CCQUEUE was not able to determine that in which case CCQUEUE printed the message PERMIT STATUS OF file UNKNOWN before assigning the receipt number or the permit status of the file was changed after it was queued In either case the file must be properly permitted READ PKEY CCQUEUE and requeued See the section Queueing in the CCQUEUE writeup in MTS volume 2 ERROR IN READING file PLOT ABORTED Some error occured while the postprocessor was reading the file containing the plot See Gail Lift Computing PLOTTING SYSTEM 428 Center 763 6051 for assistance The next two errors occur when the user changes the contents of the file containing the plot after running CCQUEUE but before the postprocessor is run CCQUEUE merely places the plot file name on a queue the postprocessor actually copies the plot file to a system tape for plotting See the sections Queueing and Postprocessing in the CCQUEUE writeup in MTS Volume 2 Sometimes the postprocessor is run late or the plot file is not copied at the next plot time after it was queued because the file was locked or special pens were required The DPA at the output window can check the time that the postprocessor was actually run
96. course the CPUs can only work with instructions and data contained in real memory thus a method must be devised to make User memory pages available to the CPU when the CPU requires them Pages are made to move back and forth between real memory and the paging device under control of the Paging Drum Processor PDP job program The supervisor communicates with the PDP telling it what pages should be read from the paging device and what pages are eligible to be moved out to be moved out to the paging device Multi Tasking is a technique that allows the computer to be used by many users at the same time Since the CPUs can only perform one instruction at a time in a single processor system it may seem at first glance that there is little to be gained from Multi Tasking This of course is not true the reason being that the CPU spends much of it s time doing nothing except waiting for I O operations to be completed Multi Tasking allows this CPU wait time to be used by other users of the system In addition more than one I O operation can be going on at once For example while the CPU is busy doing a compute function many I O devices can be going at full speed Unfortunately something cannot be gained for nothing and both Paging and Multi Tasking require resources of the computer time and memory to manage or supervise these functions Normally the gains far outweigh the losses and throughput is increased But under extr
97. data are capable of deblocking these tapes INPUT TAPES Blocked or unblocked tapes which may be unlabeled or labeled STAT STATM or STATC Labeled tapes with other volume names are not acceptable OUTPUT TAPES Blocked by this program into blocks not USAGE longer than 4352 bytes Tape format is U 4352 Tapes may be unlabeled or labeled STATC mts cmb Strip about 40 feet from the front of the new output tape The program will ask for the first output tape must be mounted with ring in Reply with the tape unit name on which it is mounted for example T1901 The program will then ask for successive input tapes Input tapes will be requested in the following manner ENTER INPUT TAPE NAME AND VOLUME NAME OR DONE To end the combining process enter DONE in place of the tape unit name NOTE Generally this program is used on the first or second day of the month When an output tape is asked for a new tape remember to strip about 40 feet from the beginning of the new tape labeled txxx 6250 statc umcmpctr should be used When input tapes are asked for use the output tape from the last CMB run if it isn t full then use the OPERATOR FILES CMB 194 OPERATOR FILES monthly statistics tape from the preceding month When an output tape has been filled a new output tape should be mounted Beginning and ending dates will be printed on the operators console These dates should be written on the stick
98. drives will be needed Wait until enough drives are available If there aren t enough drives available stop mounting tapes on UM and UB until you have enough drives Mount the tapes on the drives and ready the drives Keep track of which tape has been placed on which drive 6 Go to the VM XA operator s console For each tape mounted issue an attach command ATTACH rdev TO VSE1 vdev where rdev is the number of the real tape drive on which the tape was mounted see the section Running UB MTS under VM XA Attaching and Detaching Tape Drives for the numbers corresponding to each real tape drive and vdev is the number of the tape drive requested by the VSE1 mount request It is important to make sure that each tape is ATTACHed to the same vdev as indicated in the mount request on the VSE console 7 Go to the VSE1 operator s console and type 0 and press ENTER That s a zero not an oh VSE SYSTEM SHUTDOWN AND START UP PROCEDURES 57 8 The VSE1 batch job should resume operation and use the tapes 9 When VSE1 rewinds and unloads a tape go to the VM XA operator s console and issue a detach command to remove the tape drive from VSE DETACH rdev FROM VSE1 where rdev is the number of the real tape drive VSE 58 SYSTEM SHUTDOWN AND START UP PROCEDURES Loading MTS To start the loading process you should follow the instructions given in the section Loading MTS on the 3090 or Loading MTS under VM
99. dumps J Hansen HASP dumps etc Dumps of MTS jobs should be scanned to determine as specifically as possible where the problem was e g CLS DSR file system etc then routed to the appropriate person If you don t know how to read a dump listing contact someone who does that s the best way to learn Whenever there is a dump the operators will use DMP to produce a batch listing of it They should also log it in the dump log book Sometimes DUMP DUMPER ot MTS PRINTDUMP will be used to print a dump immediately in crisis situations The file DUMP TAPES contains a line for each current unreleased dump The batch job initiated by DMP writes a line to this file each time it prints a dump The receipt number can be used to pick up the listing When the dump is routed you should edit in the name of the routee over the receipt number then when the tape is released delete the line from the file If there is ever confusion about the contents of a dump tape the operator job FLB described in the operator s manual can be used to quickly print out the dump header information The program MTS PRINTDUMP will do the same thing The dump tapes in the machine room are kept in two rows The released tapes are on top When a new dump request comes up the operators will use one of these tapes for the dump The un released tapes are on the bottom row Whenever the operators write a new dump tape they put it in the u
100. error messages may or may not be printed depending on whether IPLBOOT can get through to the console After trying to print the error messages IPLBOOT goes into wait state with the PSW right half containing error status as follows 00000001 gt unexpected external interrupt 00000002 gt unexpected SVC interrupt 00000003 gt unexpected program interrupt 00000004 gt machine check interrupt 00000005 gt unexpected I O interrupt 00000006 gt RC gt 0 from UMLOAD while loading IPLREAD 00000007 gt I O error occurred reading disk 00000008 gt input record gt 256 bytes should not happen 00000009 gt input record format error should not happen 0000000A gt ran out of memory to load IPLREAD a small machine Any error message printed by IPLBOOT contains the name IPLBOOT in the text of the message Typing rpsw to ACS on a 3090 at any time while IPLBOOT is executing or after it has gone into wait state because of some error causes the entire IPLBOOT process to be repeated IPLBOOT attempts to write error messages onto the console at address 3A8 by default If the halfword at real memory location X FEO is nonzero it will use that address instead xX FEO is set to zero initially but not after a PSW RESTART SYSTEM START UP PROCEDURES IPLREADER 80 SYSTEM SHUTDOWN AND START UP PROCEDURES IPLREAD the Iplreader Itself This component is loaded by IPLBOOT into the top end of real memor
101. file DOES NOT EXIST PLOT ABORTED The file containing the plot was destroyed between the time the plot was queued and the time the postprocessor was run the scheduled plot time PLOT file HAS CHANGED PLOT NOT PLOTTED The file containing the plot was changed between the time it was queued CCQUEUE was run and the time the postprocessor was run the scheduled plot time The usual cause of this is that the user put a new plot in the same file The plot time and charges for a plot not plotted for any of the above reasons can be rebated In most cases the charges are automatically rebated but fill out a rebate form anyway just to make sure In some cases a rebate for additional charges may be justified Fill out a rebate form and or talk to Gail Lift Computing Center 763 6051 Rebate forms are available at the output window DEFERRED PLOTS CCD When one user has some particularly long or obnoxious plots it may be easiest to handle them separately from the normal plot runs Such plots queued by the user in the normal way but placed on a separate plot tape are called deferred plots To defer a users plot s a program CCD described below is run to tell CCP that all plots for a particular signon id are to be deferred After this is done CCP will ignore that user s plots when doing the scheduled plot runs the deferred plots will be left in the queue The deferred plots can be put on a tape
102. file selection utilities Location The object module for SCAN is in the file W030 SCANX Philosophy This program has two classes of users those who use it almost daily e g the Business Office and those who make occasional use of it e g the programming and operations staff The syntax and verbosity are designed mostly for the heavy users who must live with it the most For the benifit of occasional users however it will print the syntax of a command once it figures out which one you re trying to use Also since the program expects to be searching through large volumes of data some things are sacrificed to run time efficiency for instance user supplied search programs use a non standard calling sequence Description Generally to request a search the user first specifies the type stat cmds then a list of the search criteria SCAN carries out the search and prints the results The results can be copied or printed in the whole or in part on any desired FDNAME The command descriptions follow STAT fdname NOLIST This command is used to request a search for statistics records After the search has been completed the statistics records found will be listed in LISTSTAT format on fdname fdname defaults to SINK If NOLIST is specified this listing is not produced SCAN will prompt for a list of search requests one per line Each request has the form CCID fromtime TO totime CCID i
103. followed by the job number In the above message the DWB RM024572 is the job name the number in parentheses following it is the job number Tf this had been an RMBATCH job instead of the RMPRINT print job that it was then you might have seen it as DWB RM024572 24572 awaiting execution at start but when it started execution a new job say 24573 would be generated for the print output This new job would however keep the same job name as the original input job and so when you saw it awaiting print it would say DWB RM024572 24573 awaiting print RESOURCE MANAGER 540 What to do if something is wrong 1 2 The basic idea is to get rid of the processor for the link that isn t moving and then start a new one up This often requires stopping both ends of the link making sure that both ends have stopped and then starting each of them up again For the link from UM to UB you can do both easily so I usually sign onto both sides stop things as described below on both sides and then start things up again on both sides The links are named according to the other end so on UM the link is UMICHUB and on UB the link is UMICHUM For the other links from UM I always try first stopping and starting the UM end If that doesn t work then I stop the UM end and then contact the NOC or else have operations contact the NOC to have them contact the people at the other end eg at Wa
104. for IPL writer SLOW Pauses between tracks This option should always be used when the system is busy to keep DASDI form hogging the system An end of file terminates the run EXAMP DASDI LES signon FILE set macros on macrolib cmdmaclib dasdi D830 MTSOO1 VX 1 IPL 355 D838 MTS002 VX 2 If asked for confirmation type OK if the Volume label is correct fendfile Anything else will cause the preceding input line to be ignored SLOW should be specified on the DASDI command if DASDI is used during the day A 50 MSEC TWAIT will be done between tracks NOTE The IPL parameter above means that the pack being dasdied is one to be IPLed from DISK COPY PURPOSE To copy a disk pack to another disk Last updated Apr 06 1992 by Tom Valerio Ahead of time you should 1 Find an unused volume to copy to by checking the file W225 DISKS The file contains the list of production volumes for various systems UM and VM followed by the list of idle disks followed by a list of disks and associated notes for disks that we got rid of long ago 2 Attach the MTS and FILE macro libraries gt macrolib mts cmdmaclib gt macrolib file cmdmaclib 3 Use VAMREC to insure the disk you ve chosen is not a production disk That is make sure the internal label on the disk is of the form XXXHHH A brand new disk may have a label which VAMREC cannot even read gt VAMREC Volume D200 Volume XXX200 Devi
105. form yy ddd where yy is the year and ddd is the day from 1 366 For example January 1 1988 is 88 001 Select the oldest IOCDS for the particular side A or B with which you are working 4 Release the write protect status of the IOCDS you ve selected by typing A3 Ix where x is the number of the line on the menu designating the chosen IOCDS the number should range between 1 and 12 5 Type F ACCESS and press ENTER to select the ACCESS frame 6 Type Al and press ENTER to select the access level 7 Type IOCP CPx where x is the number of a CPU On the UM side x should be 1 on the UB side myn should be 4 8 The ES3090 3003 alarm will ring and the message No Program Console Assigned will be displayed on the console 9 Press the ASSGN CONS key The ES3090 300J will display the Console Assignment frame 10 Select option A3 program mode and press ENTER then press the END key The ES3090 300J3 will display the IOCP general menu 11 Select option 5 Build IOCDS from Cards and press ENTER The screen will shift to the appropriate menu IOCDS INSTALLATION 396 12 13 14 15 16 17 18 Ready the IOCDS tape 82406 on any drive we ll use gt t904 here Tape drive addresses begin at 0680 for the gt T900 series drives and at 1680 for the gt T920 series for example gt t910 address 068A The tape will have been written with the new IOCDS you are abou
106. from last Friday s date 05 06 83 the last time PM2 was run to Monday s date the current PM2 run pm2daly print pm may0983 runs the program PM2 447 Note the gt on the SET command It is a flag indicating that this is a command for the macro processor and is required on this command Invoking the two command macros makes six command macros available for the duration of the current signon These macros may be used exactly as if they were MTS commands The prototype of each of them is given below followed by a description of the parameters they take Note that many of the parameters are for use only when debugging PM2 and are included here for completeness only Only the parameters given on the first line of each prototype are needed for normal use of PM2 To create or update a PM2 history file PM2HMAN CFIGIN config SAVEIN save PAR params PRINT printfdname VSSCMDS cmdfdname VSS vssobj PM2LIB pm20b DEBUG HISTFILE pm2history To dump a PM2 history file PM2HDMP PAR params PRINT printfdname VSSCMDS cmdfdname VSS vssobj PM2LIB pm20bj DEBUG HISTFILE pm2history To update the tape volume error history PM2TVDL PAR params PRINT printfdname TVOLCMDS volfdname VSSCMDS cmdfdname VSS vssobj PM2LIB pm20bj DEBUG HISTFILE pm2history TAPEHIST tapehistory To run the daily PM2 run PM2DALY PAR params PRINT printfdname TVOLCMDS volfdname VSSCMDS cmdfdname VSS vssobj PM2LIB pm20bj DEBUG HISTFILE pm2history TAPEH
107. hlg Normally started automatically by HSP OPERATOR FILES HLG 222 HLM NAME FUNCTION USAGE REMARKS The user OPERATOR FILES HLM Forces all access to the UM host through the Host Locator Also can be used to restrict access to the host MTS HLM This program is used for testing by systems people will be promted as follow For Host UM Enter YES to force Host Locator access or NO to release system to users After the operator enters the appropriate response a confirmation message will be given OPERATOR FILES HLM NAME FUNCTION USAGE OPER HPS ATOR FILES 223 To display the status of HASP This program is usually used when it is suspected that HASP is in some sort of trouble No input is required initiated mts hps after the program has been OPERATOR FILES HPS 224 OPERATOR FILES To determine the current state of all PCPs HUH NAME HUH FUNCTION USAGE mts HUH The operator file HUH runs the file SYS HUH Concentrator or UMnet PCP in turn and one of the following Data messages will be printed Message Meaning Message Meaning Message Meaning Message Meaning OPERATOR FILES HUH HUH checks each lt name gt seems to be running normally Everything looks good lt name gt is busy Try again later This message will appear if the xx0OP device is busy when HUH is run Try HUH aga
108. if required by some unusual situation The status of LOW and DEFER are printed whenever a status command is issued BACKLOG lists jobs in LOW DEFERRED or MINIMUM priority separately QUE or DISPLAY shows EXEC WL for jobs waiting in LOW priority and EXEC WD for jobs in the DEFERRED priority category Jobs in LOW priority are less expensive to run than those with a normal priority and jobs in DEFERRED priority are less expensive to run than LOW priority jobs MINIMUM priority jobs are least expensive LOW DEFERRED and MINIMUM jobs are not re computed after execution When the commands HOLD LOW and or HOLD DEFER are given by HASP priority 0 jobs in the active or waiting stage of print or punch are not affected Esssentailly normal jobs aren t put ahead of low jobs etc RM PRIORITIES CLASSES SCHEDULE 375 but rather the priority of the low job is lowered by a few points And similarly for other priorities Every two hours every print job that has been in the queue since the prvious two hour period has its priority raised by one up to a limit of 15 for normal 13 for low 12 for deferred and 11 for minimum So During the night the priorities will drift up but the jobs will remain in the same order since jobs are printed in order of arrival within a given priority class RM PRIORITIES CLASSES SCHEDULE 376 NOTE Jobs that are 1500 pages are automatically held and
109. in the units This black box with attached wire is usually kept on top of the structure that holds the operators console When this box has been plugged into the tape racks it can be operated by its white button There are two circuit keys near the operators console which are used on all of the electric racks The yellow one when plugged into the override slot overides the safety features A reason for using this overide key could be because the floor is not actuating its switches to let the racks move back and forth The other key is colored red on one side and green on the other This key when inserted in the slot below the overide slot can be used to turn the power on or off depending on which side of the key is facing up If the key is inserted with the green side up the racks will be powered on However the key must be left in the slot for several seconds until a beep is heard and the rack lights come on When this happens the key must be removed from the slot andthe safety reset button pressed The racks should now be ready for use If the small rectangular green light below the words in use is blinking this means that one of the sweeps is jammed or caught on something The sweeps are metal safety plates at the bottom of each rack almost touching the floor They are about 6 inches high and extend the full length of the rack TAPE RACKS ELECTRIC PROBLEMS 503 Tape Rack Strip Labels Below is a
110. information in the first file on the tape The tape should be re initialized before being written by CCP PLT36 and PLT51 The second step of running regular plots which follows can t be run from a library file as can the first one since it is necessary that the job be interruptable 1 Signon using id PLOT run sys plt51 or plt36 EXECUTION BEGINS WHAT IS TO BE PLOTTED PLOTTING SYSTEM 434 or 2 signon using id SYS run plt51 or plt36 EXECUTION BEGINS WHAT IS TO BE PLOTTED When using PLT51 after plots have been dumped onto tape but just before beginning plotting you will be prompted for a plotter Responses are PLT2 or PLT3 or 2 or 3 An endfile gets you back to the main prompt If the requested plotter is busy or unavailable for some other reason a message will be printed and the plot program will prompt again OPTIONS a A single block example 003 plots block 3 only b An s 8 number example 001476 plots all blocks associated with s 8 number 001476 c ALL plots everything d A range of blocks example 004 015 plots block 4 through block 15 e A range of blocks example 003 end plots block 3 to eot f an EOF via CNCL button or fgendfile will terminate execution After option is given the system will ask for a tape This is the tape which was produced during CCP The system will then read the tape dismount it and list the plots foun
111. is the name of the file containing control information for CNVTEREP LOGREC historyfile Historyfile is the name of the file containing output from CNVTEREP If this parameter is given CNVTEREP will not be run DEBUG If this parameter is given VSS and hence PM2 will be run under control of SDS HISTFILE pm2history Pm2history is the root of the name of the stack of history files for PM2 It is suffixed with N for the new history file created by this run by 0O for the first backup history file and by VO for the second backup history file and by nothing for the current history file TAPEHIST tapehistory Tapehistory is the root of the name of the stack of tape history files It is suffixed in the same way as pm2history described above Examples PM2 450 In all these examples the two set commands need to be given only once per signon They will be shown in the first example only to emphasize this To make the daily PM2 run and print the output use the following commands from user id SYS Another example of starting up PM2 follows SET MACROS ON gt SET VAR MACLIB 1 ETC PM2 MAC PM2DALY PRINT PRINT Alternatively to get the output 2 up on the page printer use PM2DALY RUN PAGEPR SCARDS PRINT PAR TWOUP To give an example of how parameters may be passed to PM2 suppose you want to retain entries on the tape volume history for 100 days and ignore printer errors on the
112. job during the previous interval virtual pages is the number of pages of virtual memory currently in use by the job real pages is the number of virtual pages owned by the job currently in real memory real pages needed is the number of pages currently estimated to constitute the job s working set paging activity is the average number of page in operations drum reads per second for the job over the previous interval disk activity is the number of file I O operations performed during the previous interval SYSTEMSTATUS 479 PT percent processor time is the percentage of total system processor time used by the job during the previous interval SVP percent virtual pages is the percentage of the total virtual memory being used by the job SRP percent real pages is the percentage of the total real memory in use that is being used by the job SPA percent paging activity is the percentage of the total paging activity being done by the job during the previous interval STATUS the status of the job e g PAGE WAIT PI percent processor idle is the same as above This command continues to print information every 20 seconds until interrupted by an attention interrupt PAl on a 3270 or until the JOBS job is STOPped or GOOSEd Examples LOAD LOAD 123 OPER text This command given without a parameter causes each line that appears on the UMMPS operator s console to be printed until
113. log book provided Reloading should finish within 5 minutes If it takes longer or doesn t seem to work call the operator Don t reload it again UNYN TERMINALS SCPS ETC 526 UNYN station map as of 18 15 25 Mar 1983 W051 UNYNMAP Keypunches 104 51 112 50 Decwriters 111 4 s113 6 s117 10 5206 18 8207 19 8213 25 s212 24 5214 26 8300 32 5302 34 s110 3 s114 7 s118 11 8205 17 5204 16 8210 22 8211 23 8215 27 218 31 303 35 Nomenclature lt SCP gt lt port gt lt cable gt Just lt cable gt 108 53 8116 52 S101 54 100 55 glass partition S s109 2 1 X105 5 5115 8 103 9 5203 15 s119 12 202 14 s201 13 5208 20 8209 21 s216 28 S217 29 219 30 S301533 5304 36 Consultants s102 45 46 47 48 s106 42 s107 40 41 43 Express terminals 49 316 37 8317 38 8318 39 Graphics terminal Operator s displ 5200 56 User display 319 57 so S101 1 is SCP S1 port 01 cable 1 means there is an unused cable installed An SCP name beginning with X means there is no terminal connected UNYN TERMINALS SCPS ETC S ay Zoanr 527 VENDOR INFORMATION Updated 09 28 10 Company Position Name Phone No Amdahl Manager Ron Pate 1 313 358 4440 9 1 Home Phone 9 1 313 887 8046 Sales Pam Russell 9 1 313 358 4440 9 1 9 1 AMDEC 1 800 638 1764 Service Alert Center 1 800
114. map of the terminal locations here at UNYN Look at the map and find out which SCP the dead terminal is connected to This is easily done since the code for each terminal gives the name of the SCP For example the consultants terminal is marked as S102 45 which means that it is on SCP 1 connected to port 02 of that SCP and attached by cable number 45 All you care about is the SCP name in this case Sl Find another terminal on the same SCP and test it using the directions given above in step 2 If the terminal works then the SCP is OK and you should try the first terminal again to see if it is still dead If it is an Ontel you should try pressing the PROG key while you have the SHIFT and CTRL keys held down This will reset the Ontel completely Press ATTN and see if you get any response If not power the terminal off and report it as being out of order If the second terminal doesn t work either then the SCP is probably dead You should then look very carefully at the lights on the front of the SCP Make sure you look at the correct SCP The lights are difficult to see because they are dim You should cup your hands above and below the lights to make sure you know which ones are on If either of the LOAD lights is on then the SCP program is being reloaded from MTS If these lights don t go off within three minutes call the MTS UNYN TERMINALS SCPS ETC 523 operator on the red phone and expla
115. must be preceded by an S 8 card and MTS user Jobs may be stacked in this example 4 In this invocation of MTS the task will gracefully disappear when the END OF FILE is sensed on msource Note that there is no syspch unit assigned so all defaulted calls to SPUNCH in the users programs will cause an error in the user s program As in 3 each SIGNON card must be preceded by an 5 8 card 5 This invocation shows how the operator starts an Operator file The file FYL contains the SIGNON and SIGNOFF cards and other commands also Note that since msource is a file msink will default to the MTS pseudo device DUMMY the infinite wastebasket JOB PROGRAMS MTS 122 JOB PROGRAMS MTS JOB PROGRAMS 6 In this invocation of MTS we have the same as in example 5 except that any output going to msink will be printed on the operator s console i e OPER Note also the comma delimited by blanks separating the file name from OPER The same use of the stand alone comma occurs in examples 2 3 and 4 7 And in this example we see the invocation of the operator file FILE2 Note that the name of this file is greater than four characters in length It is split up into four character strings starting from the left and padded in the last four character string if necessary with commas JOBS PROGRAMS 123 OFFLINE NAME OFFLINE FUNCTION To make a
116. network connections At times HIMs experince interface control checks IFCCs If these occur with sufficient rapidity MTS will fail and will have to be reloaded Under normal circumstances one or two HIM lines usually HIMI1 will be high lighted at the top of the console screen However when a him is in trouble usually IFCCS there will be a long line of individual HIM lines highlighted Sometimes error comments regarding IFCCs appear on the console with enough frequency to interfere with normal operation There are several ways to handle HIM problems If the screen is filling up with error comments signon w035 at a terminal sou strt offlinehimx offlines all HIMx connections mts him xstp stops HIMx Go to IBM System Console and choose the Channel Configuration screen chpid xx off offlines chpid channel to which HIM is attached chpid xx on onlines chpid Go into shop and reload the HIM using the INIT switch sou strt onlinehimx onlines HIMx connections If a line of HIMs is high lighted at the top of the screen or there is some other reason that a HIM must be stopped and started mts him xstp stops HIMx t d hx checks to see if there are remaining connections The remaining connections can be STOPped or BLASTed mts him x starts HIMx or another method HOST INTERFACE MACHINES HIMS 388 mts him xstp Go to IBM System Console and choose Ch
117. o ea Ll 89 CLK Starts Automatic Job Scheduling Program 1 190 C N Cleans Up MOHK ARO UC INTOs ii a e ee ea oe 191 CLR Restarts Job Scheduling Program oooooooooooooooooooooo 192 REMB 4 2Omb Ine Statistics Tapes A A A eed 193 CMD lt Command SEAETVSEUES s ato A A A A aid 195 CPW Change Password To Unknown oo ooooooooooooooooooooooo 198 CTD Cheek Disks ALOE PM as a a aia 200 DDG 4 LOGS Disk Errors ii A A AAA A AA Ea 203 DMP Initiates BATCH Job To Print Dumps o o oooooo o 204 XDRN lt c Drains HASP Remotes urna da Shee Ree es 205 DESDE RUNS DAS OS da 206 FDSK Add bist Delete DISKS a as a as 207 DSN Start User HASP Display NUBS so ce so cisnes e 210 DSO AMOUNT Free File SPACE ridad A ade Boe 211 DWB CMDSTAT Unit Checks To Tape oooooooooooo nooo ooo 212 FIX Update CCID Accounting Record oooooooooooooo ooo ooo ooo 213 ELB Check Tape Labels isi ae aea AA A a a a Bh A 214 FSC Make Copies Of File Save Tapes ooooooooooooooooooooo 215 ESM gra Pa We S ave MOE A A A E AS BENS Ses 217 HGS 10 Jobs With Most CPU DSK I O VM DRM RDS REAL PAGES REAG PAGES NEEDED yess ataie poe hae aa a ANS pe Se ein ee ace ad aus Ma ete eee 218 HLB Force Access To UB Host Through Host Locator 220 HlG6 6 2 HAS OG COMMECELON il A A ee Pe a SS 221 HLM Force Access To UM Host Through Host Locator 222 FHP
118. of F4 004 is the reply ID Occasionally instead of displaying F4 004 or whatever in a new message VSE will simply highlight SYSTEM SHUTDOWN AND START UP PROCEDURES 53 the F4 004 if it already appears somewhere on the screen So if you don t see a message like the one in the example look for a highlighted message Once you know what the reply ID is type the following command substituting the reply ID in place of the 4 4 CMSG MSG MIRLYN GOING DOWN IN x MINUTES ROUTE ALL SEND and press ENTER After waiting x minutes type the following substituting the reply ID in place of the 4 4 CEMT P SHUT I and press ENTER Sending the message is desirable but not absolutely necessary 5 Type Z NET QUICK and press ENTER 6 The shutdown will take a minute or two to complete While the shutdown of VSE2 is in progress you can begin the shutdown of VSE1 Eventually VSE2 will print something like F3 018 EBO2I VTAM IS NOW INACTIVE F3 003 EOJ VTAMSTART DATE TIME F1 001 10341 F3 WAITING FOR WORK 7 If 5 minutes or so have gone by and you still don t see the messages indicating the end of VTAM you have to cancel it To do this type CANCEL F3 and then hit ENTER VSE will respond with a message something like AR 015 1155D CANCEL VTAM REPLY YES OR NO In this example the 015 is the reply ID Answer the prompt by typing 15 YES substituting the actual reply ID fo
119. ones Next come the one pen plots beginning with the shortest Just before pens are requested a tape I D message will appear e g PLOTS FROM MON 1000 TAPE TAPE WRITTEN 10 12 47 12 19 77 The number after the slash refers to the pen size 4 narrow 6 medium default and 8 wide 1 BLAK 6 3 RED 4 2 BLUE 8 ok If ok is given PLT will begin plotting that plot If no is given PLT will go on to next plot 1 BLAK system asks as single pen plots begin ok PEN POSITION CHANGING Occasionally one may need to change the position of the pens for example if one of the pen holders isn t working properly In order to do this one must wait for the plotting system to print out 1l black 2 blue 3 red 4 green At this point for example you could type 3 2 4 1 which would expect the black pen in position 3 the blue in position 2 the red in position 4 and the green in position 1 If you typed 4 3 2 1 the plotting system would expect green in position 1 red in postion 2 blue in position 3 and black in position 4 What counts is what number you put in which position For instance when we said in the above example 3 2 4 1 the first postion tells where the black pen goes in this case in position 3 The second position tells where the blue pen goes in this case position 2 When all of the plots are done the system will comment PLOTTING COMPLETE PLOTTING SYSTEM 437 EXECUTION TERMINATED sig
120. only one BACK TAB Positions the cursor to the beginning of the field it is in or to the beginning of the previous field if it is already at the beginning Tab and back tab are the recommended means of moving the cursor from one field to another For example if the cursor is at the first message response line and you want to enter something in the request area back tab may be used RETURN Something like a carriage return This key has little use with the system console Cursor positioning These keys up down left and right arrow move the cursor one character position in the indicated direction Note that the two left arrow keys have exactly the same function DELete Deletes the character at the cursor position and shifts the remaining characters in the field left one character position INSert MODE Puts the display in insert edit mode which is indicated by a flag on the screen In insert mode characters are inserted into the field at the cursor position and the remaining characters in the field are shifted to the right to make room Insert mode is turned off with the reset key DUP Causes the character to be displayed on the screen the character X 1C to be entered into the field and the cursor to be advanced to the next field This character is used at the operators console to cause an automatic skip of the cursor past a read where a has been placed FIELD MARK Causes the ch
121. printed immediately before a user signs on For batch users it will appear on the first page of his output For terminal users it will appear after the brief output that specifies what line he is on and what MTS task is enabled for that line After the SIGNONM command has been given the SIGNONM job program will read a line from the operators console The operator may then type in a signon message Pressing the cncl button will cause the read to be repeated and a null line will delete any active signon message As with the BROADCST command the operator should keep in mind that the carriage width of some terminal devices is normally limited to seventy characters If the operator gives a SIGNONM message longer than this the message will be truncated at the current carriage width none signonm SIGNONM system will be up until 20 00 today There can only be one SIGNONM message at a given time and the message must be re entered following a system reload The MTS Message command provides an alternative method of setting signon messages that avoids these restrictions See Signon Message using Message System later in this manual JOB PROGRAMS SIGNONM 142 SOFTCHK NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLE JOB PROGRAMS SOFTCHK To control the recording of soft recoverable machine checks SOFTCHK OFF SOFTCHK ON n SOFTCHK SOFTCHK OFF disables
122. program upon the parameters given upon the job request and upon the actions of the system operator THE JOB PROGRAM amp TASK 22 The Main and Backup Consoles The main console is a 3270 terminal which is controlled by the OPERATOR job i e by an invocation of the OPERATOR job program The main console also called the Operator s Console is the primary terminal used to control the action of the system Most of the commands issued from the operator s console are job requests a complete description of the operator s console is given later in this manual Occasionally the operator s console may become unusable either because of hardware problems or programming errors in the system In such circumstances a second terminal the backup console may be started the system can then be controlled from the backup console until the main operator s console can be made to work again Sometimes the system may automatically start the backup console if it determines that the main console is unusable if not an operator can start the backup console by issuing a signal to the machine Normally the signal consists of an external interrupt that is use the ALT IRPT keys on the IBM console UM operators should consult the section Running MTS on the ES3090 300J CONSOLES MAIN amp BACKUP 23 Forcing a Superdump Occasionally the system may detect a serious error in its operation in this case a message is print
123. program is used to print some or all of the machine check I O outboard errors etc which are recorded on system disk files SYS EREP contains a driver program that runs this EREP under VSS It uses the following logical I O units SCARDS CMDSTAT type 4 10 25 records Default INIT UC3330LO0G L 40 INIT MCFILE however we are presently using the following INIT MCFILE INIT UC3330L0G LAST 500 INIT MCFILE UB INIT UC3330L Note The use of such a large number L 500 is for the purpose of getting a history that goes back far enough SPRINT EREP error report printed output Default PRINT SERCOM EREP messages and prompting messages Default MSINK GUSER Responses to prompts Default MSOURCE SPUNCH Error records in MVS history tape format Default ACCIN 0 Source of additional VSS commands to be executed immediately before calling EREP Default No additional commands are read If SCARDS is assigned to DUMMY and SPUNCH is assigned EXPLICITLY to the location of MVS history tape data then the CNVTEREP step will be skipped and the history tape data will be read SPUNCH must not be allowed to default in this case even if the default is correct since the file will be emptied as part EREP SYS 319 of the defaulting process EREP uses a machine carriage control on its printed output so remember to use MCC if you copy a file containing EREP output to PRINT SYS E
124. ring the alarm for the UB side of the ES3090 300J from one to three times Each time it does so go to the ES3090 300J console for the UB side of the ES3090 300J and press the ENTER key The ringing of the alarm in this case is a normal part of VM XA operation and doesn t indicate a problem 13 At this point VM XA is up and running and you can load UB MTS and DOS VSE To do this see the sections Loading UB MTS Under VM XA and Loading DOS VSE Under VM XA If you entered force in step 8 above type the following command and press enter autolog ub mts UB MTS UNDER VM 46 SYSTEM SHUTDOWN AND START UP PROCEDURES Note Do not enable anything VM will automatically enable the correct ports VM XA Error Messages Every message that is printed on the VM XA operator console is identified by a 7 character code The form of a message code is HCPxxxnnns where HCP indicates a message from the VM XA control program xxx is a three character name of the part of VM XA that sent the message nnn is the message number gu is one of the following severity codes A Immediate Physical Action p Decision ERM Error Mero Information An immediate physical action message indicates something that must be done for example readying a tape drive A decision message is a prompt or question to which you must enter some answer for example a message to which you must respond YES or NO An error messa
125. service state LSS actual lines is the total number of non batch MTS users currently signed on virtual pages is the total number of pages of virtual memory currently in use by all jobs real pages is the number of virtual pages occupying real pages for all jobs drum pages available is the number of pages available for use on the primary paging device s paging activity is the average number of page in operations drum reads per second over the previous interval disk activity is the average number of disk I O operations per second over the previous interval channel activity is the average number of channel I O operations per second over the previous interval percent processor idle is the average CPU idle wait state time over the previous interval queue is the current number of jobs on the system CPU queue tape queue is the current number of jobson the tape queue The values for DPA PA DA CA and PI may be followed by a plus sign indicating that the condition represented by the number is excessive i e that the system is overloaded in this aspect The presence of any plus sign causes the asterisk summary mark to be shown after the BP value If a job number jobs is specified then a sampling of information about the specified job is displayed every 20 seconds as follows delta time is the same as above delta processor time is the number of seconds of CPU time used by the
126. should be noted here that when running EXAMPLES JOBS PROGRAMS 121 batch with the BATCH parameter as opposed to letting HASP supervise the batch function 1 the unit record I O units used are directly allocated to the MTS task and therefore run slowly and inefficiently and 2 there is just one task number for all the batch jobs executed This is opposed to HASP where 1 HASP owns the unit record equipment and makes good use of it and 2 each user job is executed with a different MTS task number 1 mts la71 2 mts rdrx ptrx pchx 3 mts batch more rdrx ptrx pchx 4 mts batch rdrx ptrx 5 mts fyl 6 mts fyl oper 7 mts fil e2 The following explanations are in order 1 This starts a task on device LA71 probably a terminal with msource and msink assigned to that device syspch is not available 2 This starts a task using device RDRx as msource device PTRx as msink and device pchx as syspch Note that this is not a batch task BATCH was not given as a parameter Thus the input on msource should not begin with an S 8 card Also only one MTS job will run i e when the first SIGNOFF card is read the task will disappear quietly no matter if there are more card images on msource 3 This is an invocation of a batch task that will not stop until the operator STOPs it or it dies ungracefully The devices are the same as in example 2 However in this invocation each SIGNON card
127. someone RESOURCE MANAGER 542 the subtasking monitor and then sits in Monitor wait limbo the communications path to the RM is also hosed up The Monitor wait condition is the system equivalent of the have a happy day face with What me worry printed under it when you re dealing with programs that use the subtasking monitor a Run SYS TASKS enter the command KILL RM b then still under SYS TASKS enter the command TASKS RM to make sure they re all gone Wait till they are c then still under SYS TASKS enter the command INTERTASK RMGRNET If it finds any tasks KILL them There will probably be one task the RM MSG task d Now things should be cleaned up and you can restart the RM in the normal way with the operator files MTS RMS MTS RMP e Even though the RM has started again a given link may not be running again because both ends need to be drained and then both ends started So check to see if any of the links have nonzero queues and if so watch to see if the counts go down 6 If the processor BITMSG keeps reporting that it is Still initializing then there is a leftover copy with POST NETLOG BIT locked still around Use the systemstatus command to find all MTS jobs with the id POST SY T U POST You should find two tasks showing an msource name of RM M The newer task one with the larger task number should report FILE WAIT If this is the case stop the
128. something that can t be interrupted it will print a message saying so and continue A second interrupt will be processed by MTS When PRINTDUMP can be interrupted it will DUMPS PRINTING prompt for an input line from GUSER This inpu RESTART to continue the command interrupted or command to terminate it Example in batch MOUNT DUMP17 9TP DUMP VOLUME DUMP RING IN RUN MTS PRINTDUMP 0 or 1 DUMP TRACE VM CORE 7F10 A768 OPERATOR FILE DMP This file is used by the operators to log generate standard format listings of all MTS 367 t lines can be any PRINTDUMP DUMP 17 all dumps and job dumps A BATCH job is created to print out the dump at deferred priority MTS PRINTDUMP takes a long long t ime For MTS jobs a full dump is taken For other types more abreviated dump is taken enough it is you a start at figuring out what went of jobs a hoped to give wrong File DUMP DUMP CMDS contains the actual MTS PRINTDUMP commands that are used for each job type The receipt number of the BATCH job created by DMP should always be recorded in the dump log notebook A log of all dumps is kept in file DUMP LOG file DUMP TAPES has an up to date list un released tapes In addition of the current DUMPS PRINTING 368 ACCOUNTING MONTHLY LISTING AND FILE CHARGE UPDATE NOTE PRINT must be available before the following procedures sig sou sig s
129. sou mnop sdm If the HG machine isn t available net wu sig w035 sou mnop sdm FUNCTION USAGE REMARKS OPERATOR FILES SDM To place a shutdown message in the UMnet UMnret PCPs One message line is read from the console This procedure is used to enter a one line UMnet UMnet message which will be printed for any user attemting to access which ever system is down It will not disconnect any users currently connected to the UB or UM systems terminated by a null line or end of file and placed in the PCP s mts sdm This procedure should normally be run after SHUTDOWN ALL has been entered at the console and you have checked to see that all MERIT UMnet jobs have stopped Use t t mnet to verify this SDM will skip any processors that are logically offline but may hang if any is online but not running OPERATOR FILES 287 REMARKS SDM shutdown message should normally be run after SHUTDOWN ALL or SHUTDOWN PART has been entered and all UMnet MERIT jobs have stopped Use and t t mnet to check OPERATOR FILES SDM 288 OPERATOR FILES SFT FUNCTION Turns on SOFTCHK which enables reporting of soft machine checks USAGE Started automatically by AUTOSTART OPERATOR FILES SFT OPERATOR FILES 289 STA Note STA may require a jobdump if it fails use mts srk Notify Charles Engle if this happens NAME STA FUNCTION To save the billing information STATISTICS on magne
130. specific device control unit channel or the path between a channel and a control unit logically offline PROTOTYPES 1 OFFLINE devname devname 2 OFFLINE CU cut cut 3 OFFLINE CU devname devname 4 OFFLINE CH chan adr 5 OFFLINE CH chan adr cu cu 6 OFFLINE CH chan adr devname devname USAGE If device names are given 1 then the devices are marked as offline and unavailable for further use If a device is allocated to some job at the time it is taken logically offline it will remain available to that job until the job releases it If the string CU followed by by control unit numbers 2 is given then the control units are marked logically offline If the string CU followed by device names is given 3 then the control units attached to the devices are marked logically offline Form 3 of the OFFLINE command may only be used if the devices are attached to a single control unit Control units numbers are not in any way related to the address ofa device but the UNITS command may be used to determine the control unit number for a given unit If the string CH followed by a channel address is given 4 then the channel is marked logically offline If the string CH followed by a channel address and control unit numbers is given 5 then the paths between the channel and the control units are marked logically offline If the string CH followed by a channel addr
131. t d sa07 xxx Job not found RM PRIORITIES CLASSES SCHEDULE Remote Console Read Stuck 381 Obsolete HASP remotes command from a remote This situation is not supposed occasions recently for remote BSSR The dump analyzer program D A can be used to this situation for HASP stuck following is as uses a single PCE and DCT for all console messages to and from all If the remote console PCE gets stuck in a read waiting then HASP will all remote messages will queue up waiting to for a quickly run out of WTO buffers as for the read to complete happen but it has on several from The detect and recover on a remote console write an annotated HASP D A session showing how to detect a stuck remote console read and a way to recover without reloading Hr hasp d a 14 12 13 why No free WTO buffers d globals Globals at 1C7018 JOBOPTR 001CB8B0 JQFREE 001CD640 O01CFEFO 00000000 001CF2E0 001CD8D0 001CED20 S8NBR 650111 XEQTOTL NUMRJE 43 POOLECB FF 001D0510 JQENT 001CD250 JQPV 00000000 00000000 00000000 001D0050 001CCEDO 001CCC20 001CBCBO 001CC180 001D04D0 001CD250 21237 DACKPT 00 NUMPRT 57 000000 POOLECF FF HASPXWT 00 POOLCHG 00 POOLUBF FF NUMDA 2 CKPTRAK 10000003 CYLALOC 269 VCYLTOTL 2044 PRCHKPT 00245114 QUEFREE 00000000 BUFPOOL 00225860 TPBPOOL 001ECD40 LOWFLAG 00 MINXCLAS Normal INITDONE FF QCKPTEWF 00 HOLDPD
132. the UB machine On UB the same first four tasks will appear and there should be one LINK task with SA00 attached This ig UMICHUM and is the other end of the UM to UB BITNET link Currently the links and their SDA lines are on UM WAYNESTT SA07 link to Wayne State VM machine RESOURCE MANAGER 536 our link to the rest of BITNET UMICHUB SA17 link to UB machine MERIT SA18 link to NSFNET VM machine UMDSCVM SA19 will be link to DSC not yet operational on UB UMICHUM SA00 link to UM machine To find out which SA line is attached to which link look at the file SFIL DEFINITIONS This file is read by the RM at startup time and used as commands to modify its symbol table the source for the symbol table itself is in RMGR SYMTABLE gt SD but I would not advise studying this if you value your sanity Later definitions and redefinitions can be done at any time by feeding these commands to the SYS RMOPR program Each link has three processors defined for it as for example define device sa07 waynestl type link define device sa07 wsugout type sysou define device sa07 wsu in par stream any type sysin You can also find out this information by displaying the definitions using the SYS RMOPR program See next section 2 Using the SYS RMOPR program To enter commands to the RM run the program in SYS RMOPR There are two commands used mostly to look at the RM s status the QUEUE command shorthand for DISPLAY QUEUE and the
133. the directory program the directory file is specified either in the PAR field of the RUN command or by the FILE command When calling the directory subroutines directly the directory file is specified as the first parameter in each subroutine call The directory is a line file organized in the form of master entries and file entries For each function type within the directory there is a master entry that describes the function type and its characteristics i e the mode of the tape and the formats used for the primary and secondary search fields For each tape file catalogued within the directory there is a file TAPE DIRECTORY PROJECT 508 entry that describes the characteristics of the tape file i e the name of the tape the location on the tape and the ranges of the primary and secondary search fields In addition there is a master entry for the pool of available tapes type POOL and file tape entries that describe the current status of each tape within the directory i e whether the tape is in use and by whom The format of the directory file is given below Each entry will contain the following information Type code bytes 1 4 The type code identifies the function type of the tape Some of the possible type codes are POOL available tape tape currently unused DSTA daily statistics tape MSTA monthly statistics tape MRTB MERIT billing tape HNDB Hand billing tape CMDS Command statisti
134. the operation was successful The tapes used for the daily and monthly statistics are maintained in the directory file SYS DIRECTORY The STA program automatically maintains this directory and selects the correct tape The exact range of dates over which the statistics were accumulated are automatically recorded in OPERATOR FILES STA 290 OPERATOR FILES the directory by the program therefore it is no longer necessary to write this information on the tape cannister System Crashes and Recovery Operations during STA If the system should crash during the saving of the statistics file recovery operations will automatically be initiated when the system is reloaded Each time that the system is reloaded the STA program is automatically run to determine whether the system crashed during the statistics file save This is indicated by the statistics work file SYS STATISTICS still being full as described above the first thing that the program does is to copy STATISTICS into the work file and the last thing done is to empty the work file If the work file is still full at system reload the previous file save did not run to completion The reocovery operations consist of re saving the work file on both the daily and monthly save tapes When this occurs the message LAST SAVE OF STATISTICS UNSUCCESSFUL REPAIR OPERATIONS COMMENCING is printed on the console The operator may also manually check for a system
135. the operators to log all dumps and generate standard format listings of all MTS job dumps A BATCH job is created to print out the dump at deferred priority MTS PRINTDUMP takes a long long time For MTS jobs a full dump is taken For other types of jobs a more abreviated dump is taken enough it is hoped to give you a start at figuring out what went wrong File DUMP DUMP CMDS contains the actual MTS PRINTDUMP commands that are used for each job type The receipt number of the BATCH job created by DMP should always be recorded in the dump log notebook A log of all dumps is kept in file DUMP LOG In addition file DUMP TAPES has an up to date list of the current un released tapes 6 DUMP DUMPER PROGRAM This program is invoked by DMP but can also be used interactively to print dumps while you wait a To produce and log a dump listing via a BATCH job Run dump dumper par batch it will prompt for the tape number and the reason for taking the dump it will log the dump in file DUMP LOG and in file DUMP TAPES it will print the BATCH job receipt number b To log a dump then generate the listing on line Run dump dumper the tape number and reason are prompted for and logged as with par batch a log entry is made in file DUMP LOG but not DUMP TAPES the requested dump tape is mounted MTS PRINTDUMP is started up via RUN which unloads DUMP DU
136. vector dsect The command line read by SCAN is cmdname line 1 line 2 string line 1 and line 2 are a pair of line numbers They are passes in the parameter lines in internal line number form with linef1 first If either value is omitted in the command line a zero is used String ig an optional arbitrary character string on the command line which is passed in the parameter STRING to be parsed by the command processor SCAN does not guarantee that this string will exist after a call to any utility routine other than LESPAN LEBREA or LENUMB STAT searches the monthly statistics tapes for records described in an itemlist given by its caller It is a vector of items of identical length Each item has the format Jccid stsw stfrom stto whatever 0 4 8 10 18 hex ccid is a signon id If the high order bit of stsw is on a record without that signon id in STAUI cannot satisfy this item If the high order bit is off STAT does not check the signon id stfrom and stto specify a range of times in SCAN 459 Julian microseconds Only records with signon times within that range can be accepted whatever is ignored by STAT it may be any number of fullwords in length or omitted but every item in the itemlist must have the same length STAT sorts the itemlist in place by stfrom It scans statistics tapes beginnig with formtime of the oldest item and continuing until 12 hours aft
137. which aleady contains the program looking for C tapes run load 1 5 0 6 00 last 1 This runs the cleanup job for the N tapes and places the ouput records at the end of 00 Now the same procedure must be done on the UB system net ub sig w035 run sys msscn scards sys mountstat sprint 0 b OPERATOR FILES MSS 248 OPERATOR FILES r CHHHHH r n endfile run ftn scards w225 w32 tpst 57 run load 5 0 b 6 00 b copy w225 w32 tpst 57 w32 edit w32 aD Ele NN stop run ftn scards w32 spunch load 1 run load 1 5 0 b 6 00 b last 1 Now to combine the UB file 00 b with the UM oo file the UB file must be copied to the end of the file on the UM system copy rmt oo b lcl oo last 1 This procedure should take a minute or so To get back to the UM system mts At this point the C and N tapes from both systems should be in the file oo The next thing to do is to sort the tape list The following example sorts the tapes by date run sort par sort ch a 15 2 ch a 9 6 1 00 O 000 Finally to get a double spaced listing all of the blanks in column 1 must be changed to zeros zero being the carriage control character for skipping a line fedit 000 col 1 1 this confines editing to column 1 a a nv opl f 0 nv means no verification and opl means once per line stop To print the list on the line printer cont print printer line copy
138. words in the device table for a specified device SE device i d This command is used when the operator wants to examine the status of some device e g to see if it is waiting for a device end The format of the printed information can be found in an Assembler COPY section called COPY LLMPSEQU A device i d se t905 There is no SE job as such since it is a supervisor command This command can t be used on the 3270 operator s console Use the DEV command instead The DEV command can be found in the Systemstatus section of this manual SHUTDOWN NAME FUNCTION PROTOTYPE USAGE Note After necessary not EXAMPLES JOBS PROGRAMS 139 SHUTDOWN To SHUTDOWN the terminal portion of the system with the option that all MTS terminal tasks are to terminate immediately SHUTDOWN ALL PART If no parameters are given then all currently executing non batch MTS tasks will be allowed to terminate of their own accord When an MTS terminal user signs off the associated line will be disabled Terminal lines which are IDLE will be disabled immediately so that no user can sign on on any of those lines If the ALL parameter is given all non batch MTS tasks are terminated immediately as if the STOP command had been given for each task Terminal tasks are stopped first followed by file and OPER tasks If the PART parameter is given all MTS terminal tasks are terminated but any OPER and fil
139. you can do 1 and 2 above Pressing PSW RESTART is like pressing a break key and has the effect of putting this program back into command mode The console device address will be selected from X FEO and if that is zero which it is set to initially 3A8 will be used If something goes awry a message may or may not be printed In any event we go into wait state and the PSW right half is set to one of the following 00000100 gt external interrupt 00000200 gt unexpected SVC interrupt 00000300 gt unexpected program interrupt 00000400 gt machine check 00000500 gt unexpected I O interrupt 00000600 gt could not find console 00000700 gt could not find console look at roc Note the third byte byte 2 of the PSW right half is the error status area for IPLREAD and the fourth byte is the error status area for IPLBOOT SYSTEM START UP PROCEDURES IPLREADER 84 SYSTEM SHUTDOWN AND START UP PROCEDURES WHAT THE IPL SEQUENCE DID 1 IPL loads main storage directly from disk or from cards or tape 2 System configures itself i e it finds out what hardware is available initializes some tables etc and asks for the time and date and verifies it 3 Starts an INIT task which echoes time and date locates the system catalog on the disk packs and starts the OPERATOR and DMGR jobs 4 Starts an MTS task in which user INIT is signed on and runs the programs commands
140. 0 1270 1270 1403 2305 3066 3090 3211 3277 3284 3286 3288 3330 3330 3350 3805 3805 3805 3805 3825 3825 4305 8 Lines Per Inch 315 Line Hung 322 Memorex Enable Disable 324 128 342 Error Codes 361 128 346 Printers 60 128 128 342 342 342 354 Error Codes 361 Error Codes 361 EGC 66 INTEL 66 Or 3825 Low Spares 362 3825 Errors 361 66 reseting 72 Error Codes 361 Printing 60 8 LPI 60 575
141. 01 NLOG WOPN SLV ADO1 00002 23 45 50 OPERATOR 00658 MTS OOOABE80 57 I O on AEOO NLOG WOPN SLV AEOO 00002 23 45 50 OPERATOR 01492 MTS 000AC140 37 I O on DSOE BUSY DSOE 00002 23 45 50 OPERATOR 00666 MTS 000AC560 57 I O on AE01 NLOG WOPN SLV AEO1 00002 23 45 50 OPERATOR 00667 MTS 000AC6CO 57 I O on AF02 NLOG WOPN SLV AF02 00002 23 45 50 OPERATOR 00672 MTS 000AC820 57 I O on AN01 N 0 LOG WOPN SLV ANO1 0002 23 45 50 OPERATOR 00673 MTS 000AC980 57 I O on AE02 NLOG WOPN SLV AF02 00002 23 45 50 OPERATOR 00674 MTS OOOACAEO 57 I O on AE03 NLOG WOPN SLV AF03 00002 23 45 50 OPERATOR 00675 MTS 000ACC40 57 I O on AEO4 NLOG WOPN SLV AEO4 00002 23 45 54 OPERATOR 02824 MTS 000AE240 141 I O on AN0A POST WMTS NETM ANOA 00002 23 45 55 OPERATOR 01408 MTS 000AE920 130 I O Wait HOST WMTS FTPS AU42 23 48 32 07858 STOP 1408 00002 23 48 37 OPERATOR T 1408 00002 23 48 38 OPERATOR Specified job s not found 23 48 48 07859 SHUTDOWN ALL 07859 23 48 48 SHUTDOWN The NETP MM job reports that 1 ports were affected 00002 23 50 19 OPERATOR T 7859 00002 23 50 20 OPERATOR 07859 SHUTDOWN 000773A0 2 Time Wait 23 50 47 07861 TABLMOD 07859 23 50 48 SHUTDOWN One or more batch or terminal tasks hasn t stopped SHUTDOWN is NOT complete 07859 23 50 48 SHUTDOWN Reissue the SHUTDOWN ALL after the tasks have terminated or been STOPped or RERUN
142. 179 258 UMnet Hermes 400 UMnet MSU 400 UMnet Setdate 401 UMnet Set Time amp Date 239 UMnet Startup 238 UMnet Status 250 UMnet WSU 400 574 El JM Shutdown Message On PCPs 286 nattended Mode UAM 518 ninterruptable Power Supply 549 nit Check Messages 195 212 NITS 156 Output 159 NITS O1d Style 157 NITS O MTA 157 NITS O MTA NIX Filesave 557 ISP TAPESHARE 303 UNU 306 YN SCPs Terminals Etc YN Terminals SCPs Etc YN Terminals starting 306 PS 549 ser CMD Records 454 ser I D History 454 ser STAT Records 454 ser Tapes CCP 422 SER X FILE y FILE Z 414 1 750 543 Vector Processor Console 42 Vendor Information 527 VM Refresh Retry 312 VM Running Of UB MTS Under 29 VM XA Error Messages 46 VM XA Loading 42 VM XA Running On 3090 41 VM XA Shutting Down 47 VM XA UB Attaching Detaching Tape Drives 37 VM XA UB MTS Backup Console 34 VM XA UB Stopping The CPU 36 VM XA UB Superdump 35 gaga UB System 163 520 520 Gqqqqqqq GqgGrtQGqqaqg 3233 lt D D lt m Checking On VM VOLUME FOUND ETC 414 VOLUME NAME INCORRECT 414 Vol 11 443 VSE Loading 50 VSE Mounting Tapes 54 VSE On The 3090 50 VSE Shutting Down 52 Wait Counts System 265 WAITD 147 WAITM 147 WAITO 147 WAITR 147 Weekly File Save TSN 304 W030 SCANX 454 W035 Tapes 503 Xerox 9790 Page Printer 560 Xerox 9790 Photo Receptor Belt R amp R 562 1052 128 342 127
143. 1F17AF3 F640F2F3 P CNTR 11 36 23 230EF8 0000F4 40404040 00000000 00000000 00000000 E 230F18 000114 00000000 00000789 i note PCEEWF 1052 above remote number not in message d globals Globals at 1A9018 JOBOPTR 001AD898 JOFREE 001B0408 JQENT 001AF238 JOPV 001B0A78 001AF418 001B0C28 00000000 00000000 00000000 001B0828 O01AEEB8 001ADEB8 00000000 001B0898 O01ADBB8 001AEC88 0O01AFFB8 001ADA08 001AF238 S8NBR 624093 XEQTOTL 7671 DACKPT 00 NUMPRT 57 NUMRJE 43 POOLECB FF000000 POOLECF FF HASPXWT 00 POOLCHG 00 POOLJBF FF NUMDA 2 CKPTRAK 10000003 CYLALOC 287 VCYLTOTL 2044 PRCHKPT 00226EFC QUEFREE 0021A504 BUFPOOL 001EC4C8 TPBPOOL 001CC3E8 LOWFLAG 00 MINXCLAS Normal INITDONE FF QCKPTEWF 00 HOLDPDS 00 HOLDEX 00 DRAIN 00 WAYT 00 ERRDUMP FF HASPERR 00000000 ERRSAVE 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 IDLELMT 3600 LASTTIME 973C7F3D CURNTCLK 973C7F9E 7A56D400 WCONBYTE 96 WQUEBUSY 00000000 WSPOOLO 00000000 00000000 00000000 00000000 00000000 calc 96 80 from WCONBYTE above 22 compute remote number in decimal d rtab 22 display remote name RTAB at 1AC944 RTABID BSSR RTABRNUM 22 RTABFLGS INUSE PERM SGON UC RTABPAGE 38 RTABTNPG 0 RTABCARD 0 RTABPRRT BSSR RTABPURT BSSR RTABLIST 00 RTA
144. 2 T901 C0024A SAMPLE DATA 1978 file 28 block 7734 T901 invalid or insufficient sense data This error declares an error on the drive named T901 The tape mounted on the drive was tape C0024A line two and was a labeled tape as indicated by the label SAMPLE on line two The error occurred during a read operation see channel command at end of line one 02 and the dataset being read was called DATA 1978 This dataset was located at file 28 of the tape and the error occured at block 7734 in that dataset The error was caused by invalid sense data see line three and most likely signifies a problem with that drive s hardware TAPE MOUNTS SUSPENSION OF When operations staff members aren t scheduled to be in the second floor machine room the suspension of tape mounts as well as other operational aspects is necessary There are at least two ways to prohibit the mounting of tapes UAM When the system is placed in UNATTENDED mode by altering UAM users will receive a message instructing them to copy UAM and informing them that among other things tape mounting is prohibited To turn UAM on m uam 1 To turn UAM off m uam 0 NTM TAPE MOUNT SUSPENSION NTM UAM 500 At other times even during regular hours when the machine room is staffed it may be necessary to prohibit tape mounts for a period of time Altering NTM will produce the desired results here To turn NTM on m ntm 1 T
145. 270 screen will be OPERATOR CONSOLE 3270 338 cleared and the message devn INTERVENTION REQUIRED will be displayed until the printer becomes ready If it is not possible to ready the printer the console management program must be stopped and restarted using another printer or no printer see below If a fatal I O error occurs on the printer the message devn FATAL I O ERROR xxxx will be displayed on the 3270 screen and the equivalent of a PTR NONE device command will be executed see below xxxx represents two bytes of SENSE information If the printer becomes hung the PTR devn or PTR NONE device commands should be entered before the message area becomes completely full If the message area does become full before printers may be switched it will be necessary to stop and restart the console management program using the backup console support see below THE CURSOR The cursor will normally be placed at the beginning of the first message response line on the screen or in the request area if there are no message responses pending The operator may move it to any position on the screen and the system will reposition it only after completion of a request or a message response Input may be typed only when the cursor is in a message response line or in the request area If it is anywhere else typing will be ignored NOTE The DUP key may be used to skip a read on the console when the operator wish
146. 28 36 36 00419 15 00419 15 00419 15 00419 15 RESTORED 00419 15 00419 15 00419 15 TAPE 00419 15 00419 15 00419 15 00419 15 TAPE 00419 15 etc etc 34 53 57 30 35 38 55 59 31 mts MTS MTS MTS MTS MTS MTS MTS MTS MTS MTS MTS rst WHERE IS THE PRINTER oper WHAT FILES ARE TO BE 0ab2 k2t4 va encl WHERE IS THE RESTORE t901 K2T4ALFAR8 COMPLETED WHERE IS THE RESTORE t908 OPERATOR FILES RMS 282 RSV NAME FUNCTION USAGE OPERATOR FILES RSV Program to issue reserve commands to a group of disk drives when a single control unit and lts drives are to be used by both halves of a split configuration with each half having exclusive use of certain drives This program is only suitable for systems with more than one processor Right after IPL use the OFFLINE job to offline the drives that should not be used on that half of the system then enter mts rsv and when it asks for input enter the volume names of the disks on that controller it should be using one per line Use cncl to terminate OPERATOR FILES RSV SAV MAME FUNCTION USAGE Note The program OPERATOR FILES 283 SAV To save files which have been changed since the last file save mts sav The program SAV will ask for a particular tape e g FS0023 and when the tape is mounted procede to save files
147. 30BB20 0030CB20 00706178 5030D0E0 001A16A8 FPRS 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 Dev DS40 00 KILL job Stops the specified job This performs the same function as the STOP job program This is useful when there are no job tables available LOAD JOB If job is omitted various information about the total system operation is displayed at 20 second intervals For 3270 console operation this same information is constantly displayed at the top of the screen The information included is as follows DT delta time is the actual time between successive intervals of output nominally 20 seconds EXQ execution queue is the number of batch jobs waiting to execute PTR print queue at the currently released priorities is the number of jobs waiting to print PCH punch queue is the number of jobs waiting to punch BP batch preferred is the desired number of batch jobs the system would prefer to execute under the current system load SYSTEMSTATUS 478 An asterisk following this number indicates that the system is currently overloaded actual batch is the number of batch jobs currently executing batch per hour is the approximate batch throughput rate in jobs per hour measured over the last several minutes VP RP DPA PA DA DT DPT VP RP RPN PA DA limited lines is the number of terminal users currently operating in limited
148. 367 Unit Mgr Mike Bosak 9 995 5600 Sales Kent Foster 9 995 5608 Service 9 1 800 482 0164 Emergency Power Engineering UPS Also PDU at UNYN N Y Mgr George Hamis 9 1 800 438 7373 Sales Rathsburg Associates Tom Stuef 9 1 313 489 1500 Service John Drogheo 9 1 800 438 7373 Information 9 1 800 438 7373 FAX 9 1 201 265 1811 Graham Magnetics See Carlisle Memory Products Hewlett Packard Plotter NUBS Contact Nick Amato 1 800 824 9954 Hitachi Data Systems HDS Intergrated Service Control Ctr 1 800 446 0744 Sales C A Potts 9 1 313 540 2002 Systems Eng T Maslowski 9 1 313 540 2002 Service Jerry Aten 9 1 800 446 0744 or 9 1 619 458 2046 4 30pm 8am give phone no name problem and location to recording Site Reference Number 21300 25 IBM Manager Bob Hoeft 9 1 313 434 4850 Sales Pat McGrew 9 1 313 393 6100 or A A Office 995 5822 FAX 9 995 5838 A A Office IBM 9 995 5817 A A Office Sharon Ellison 995 5829 A A Office Pat McGrew 995 5822 A A Office Holly Knighton 995 5813 Acct Market Rep Marge Farrand 9 1 313 393 6230 or 9 434 4810 Service 24 Hour 9 1 800 426 7378 Service PCs 9 1 800 428 2569 Customer ref no for ES3090 300J 9021 720 5843392 VENDOR INFORMATION 529 Systems Mike Miotto 9 1 313 393 6100 or 9 1 313 393 6168 or Ann Arbor Office 9 995 5825 or 313 665 0816 Parts 9 1 313 262 3010 Supplies 9 1 800 426 2468 Distribution Center 9 1 317 658 2900 Physical Planni
149. 39 Plotting Out of Paper Switch Mod 936 438 Plotting Pens 438 Plotting Plotter Testing 440 Plotting Processor Error Messages 427 Plotting Re plotting 437 Plotting SYS PLOT36 SYS PLOT51 439 Plotting System 420 Plotting System Problems 427 Plotting Tape I O Trouble 438 PLOT36 439 PLOT51 439 PLO 264 423 PLT36 And PLT51 433 PM2 446 P P M2 STC Er Anal 446 Y 134 ECHO 134 POLY 441 443 POR 25 Position Changes Pen Plotter 436 POSITIONING FAILED 411 Post IPL System 60 Power Down Procedure 62 Power Failures 549 Power Fluctuations amp Failures Recovery 85 Air Conditioning 85 Amdahl Disks 86 Card Punch 89 Card Reader 89 Data Concentrators 88 Disk Storage 86 Emergency Off Knobs 86 Mag Tapes 86 Memorex 1270 88 Pages Printer s 89 Paging Devices 89 Paper Tape Reader 88 Power Panels 86 Terminals 88 Powering On Loading Procedures 9790 560 POWER OFF EMERGENCY 65 462 Power On Reset 25 Power Up Down ES3090 300J 24 Primary Communications Processor 258 Print Buffer PN 134 Print Buffer P11 135 Print Buffer QN 136 Print Buffer TN 154 Print Buffer T11 155 PRINTER ERRORS 313 Printer OPER 337 Printing Dumps 204 Printing 8 Lines Per Inch 60 Priorities Classes amp Categories 376 Priorities Also Copy Rates 374 Priorities Setting For Holidays 451 Priority LOW DEFERRED amp MINIMUM 374 Program Function Keys OPER 341 Program Interrupt In Machine Check Handler 312 PSW RESTART
150. 4 02 01 Sense Sense Overrun has occurred Got unit check on dummy read after run in Must do dummy read no feed after run in Bus out check occurred Command was initially rejected Command is read no feed Command is feed and stack no read flag as above should be 01 data 315 MEMOREX 1270 LINE I O ERRORS The following error messages are printed on the HASP log I O ERROR ON LINEm cc ssss iirr ttee where ce Channel command code being executed ssss Channel Status Code ii Unit Sense Information rr BSC Terminal Response Let Internal Sequence and Command Code ee Expected Response MEMOREX 1270 LINE I O ERRORS 316 DISK UNIT CHECKS If the program which dumps error records from VM into the files falls behind and it is not possible to log an error a message is written on the operator s console This message is DXXX ERROR C where c is the error code from the above list All other information about the error is lost in this case To print Disk Unit Checks See the programs EREP PM2 or MTA PUC mentioned elsewhere in this manual DISK UNIT CHECKS 317 DISK SOLID STATE DISK AND FIXED HEAD FILE UNIT CHECKS Another similar program for printing disk and fixed head file errors is also available This program can print output in the same well format as mta printuc3330 or can summarize the printout as signon W run mta puc sprint sin
151. 4 57 7 123 7 66 8 132 8 cut con t next column CARRIAGE CONTROL HASP 327 CMDSTAT This writeup describes the CMDSTAT buffer to file CMDPIKUP and file to tape CMDTAPE programs The first part covers what an operator needs to know for normal operations The second part describes the programs and explains the possible error conditions and recovery actions Everything covered in the first part is also covered in the second OPERATIONAL INSTRUCTIONS The CMDSTAT file to tape program DWB has been changed in an attempt to make it more reliable Hopefully when things start to go wrong it will ask either the operator or the firefighter for help rather than just letting things get worse as in the past If it is only suspicious about something it will senda MESSage to the firefighter using the names in SEG2 MONITORS If it thinks something is really wrong it will ask the operator to get some help before going any further There is also a change to procedures the program will now ask the operator to create new M tapes rather than waiting for some forgetful programmer to remember to do it New Tapes When DWB needs a new tape it will print on the console DWB MOUNT NEW CMDSTAT TAPE RING IN WITH RACK Mnnn and wait for an operator response Get a new blank tape put on the appropriate sticky labels for Mnnn hang the tape and enter the drive name to the program DWB will label the tape and rel
152. 41 570 OPER Keyboard 338 OPER Printer 337 OPER Program Function Keys 341 OPER Screen 335 Output Delivery 349 OUT STACK ERROR 411 Overview Of The System 18 OVR Startup 231 PAG 261 PAGE CONTROL BLOCKS 130 Page Printer Output Delivery 349 Page Printer Xerox 9790 560 Pager No s 347 Pages 17 Paging Devices 66 Paging Drum Processor 17 130 PAG Multiple Jobs 267 Paper Tape Delivery 349 Password Change 198 PB I Alarm Section 551 PB II Alarm Section 552 PB II Meter Panel 553 PB II Powering Down 555 PB II Power Up 554 PB I Powering Down 555 PB I Power Up 553 PCP AN 258 PCP AN Reload NLO 259 PCP Msg Re UB Shutdown 284 PCP Msg Re UM Shutdown 286 PDL Startup 231 PDP 66 130 PDP Errors 361 PDROP 131 PEN 442 Pen Position Changes Plotter 436 Permitting Tapes Unsp tapeshare 303 PGET 66 74 132 Phone Message System Status 469 Phone Pager No s 347 Pident 262 433 PLB 262 420 421 432 PLOTQUIK 264 423 Plot Receipt 420 Plot Schedule 421 Plots Deferred 181 420 Plot Session Length 420 Plots Message To User 435 Plots QUICKPLOTS 264 423 Plot Tape Description 421 Plot Tape Format Info 262 Plot Tape Identifying Information 420 Plot Tape Initialization 184 420 430 Plotter 186 Plotter Pen Position Changes 436 PLOTTERTEST 441 Plot Times 421 Plotting Attention Interrupts 437 Plotting Multi Pen Plots 435 Plotting Non Queued Plots 4
153. 48 It can on rare occasions be caused by reading a tape which was written by an improperly adjusted tape drive Load point reached but drive not backwards This signifies a tape drive problem The drive is reporting inconsistent status about itself Data check This error is usually caused by a flakey tape not necessarily a flakey tape drive If a drive is out of adjustment it may declare errors like this on tapes which can successfully be read from or written on another drive The error is reported by the drive when the data it reads from or writes on the tape is incorrect This is usually due to worn or damaged tape Data security erase check This message it isn t actually TAPE ERROR CODES 498 A sample an error is given whenever the tape drive has been told to erase an entire tape using the data security erase command The reason it causes this message to be generated is because this is a drastic action to take It does not signify an error Td burst check This error is signalled by the drive when 1t is unable to determine the density of a tape from the ID burst written on the front of all tapes If an attempt is made to read a tape which hasn t been initialized this error will occur because no ID burst appears at the front of a brand new tape If this occurs the tape must be written before it can be read Intervention required drive not ready This happens when the tape drive is not in the ready
154. 5 12 The backup console should start on the terminal Special notes 1 If the UB MTS backup console doesn t appear right away on the screen you may need to hold down the ALT and press CLEAR 2 Under certain circumstances the VM XA logo may reappear after step 12 In that case go back to step 10 and try again 3 When you re done using the backup console i e the regular UB MTS operator s console is back in service switch the TEST NORMAL switch to the right of the screen from NORMAL to TEST and back to NORMAL Ina moment the VM XA logo should appear 4 If you need to reload UB MTS while the backup console is running use the TEST NORMAL switch to get back to the VM logo and then proceed normally Forcing a UB MTS Superdump To initiate a UB MTS superdump you must signal a restart interrupt To do this 1 Go to a terminal connected to VM XA 2 At the LOGON prompt type UB MTS and press ENTER to logon to the UB MTS virtual machine console for the UB MTS virtual machine 3 The terminal should respond with the message Enter Password 4 Type the password and press ENTER If you don t know the password ask your shift supervisor 5 The terminal should respond with the message Reconnected at time 6 Mount a dump tape on a tape drive 7 Hold down the ALT key and press the PAl key The word CP READ should appear in the lower right corner of UB MTS UNDER VM 36 SYSTEM SHUTDOWN AND
155. 538 8486 F E Mark Klebba Autologic Phototypesetter Service Craig Pemberton 9 312 297 0560 See also Litho Service for Developer CalComp Sales Daria Gundle 9 1 462 2200 Supplies Carolyn Darryberry 9 1 800 225 2667 Tech Support 0 1 800 225 2667 Service Mgr Tim Kator 9 1 489 0060 Service 9 1 800 451 7568 Ser No s P1t1 2603 P1t2 2580 P1t3 3778 Customer Service Number 50385 Carlisle Memory Products Formerly Graham Magnetics Sales Steve Howell 9 1 313 963 2380 Tape Cleaner Service 9 1 800 433 0916 Serial No 3319 Casper Systems Corp See Spacesaver Canon See ICS International Calculator Sales INc Control Data Corporation Manager Joe Michon 9 1 313 552 6768 Sales John Penders 9 1 313 552 6861 Service 9 1 313 552 6785 ComputerVision Handles Intel Service 9 1 800 800 7746 SysId 17511 INTEL Manager John Clark 9 1 313 851 8905 Sales Larry White 9 1 800 468 3548 Service Bob Hunt 9 1 800 468 3548 Call service No and follow instructions Dan Oas C E Office 9 1 313 851 8905 S Ns 3805 FBOO 117 3825 FBO1 105 3825 FBO3 1319 3805 FB02 299 Site Code 01337AA Data South Printers See Terminal Services VENDOR INFORMATION 528 Data 100 Site Code 671723001 Sales Bob Turner 9 1 313 358 5065 Service or Service Mgr 9 1 313 968 3860 Digital Equipment Corp DEC 11 750s West Unit ID No 11750 86068326K S N BT16833 East Unit ID No 11750 82F58688H S N BT05
156. 8 KILL 146 477 Labelled Magnetic Tapes ANSI 232 234 sabelled Tapes ASCII 232 234 Labels Tape Finding 214 Labels Tape Rack Strips 503 LAR 230 LAS 190 231 LBL 232 233 L 21 210 LIM 235 LINE 441 445 LINES 441 442 LISTSTAT 454 455 LISTSTAT Output Meanings 456 LMT 236 Loading MTS 39 58 Loading UB MTS 32 Loading VM XA 42 Loading VSE 50 LOCKD 147 LOCKM 147 LOCKR 147 Logouts CPU amp Channel 318 LOW Priority 374 LRL 237 LSS 398 LSTAT 147 Machine Carriage Control 325 MACHINE CHECK HANDLER 310 Machine Check Handling Messages 311 Machine Check In M C Handler 311 Machine Check In Wait State 311 Magnetic Tapes ANSI Labelled 232 234 Mailing Diskettes 500 Mailing Mag Tapes amp Floppy Disks 350 Mailing Tapes 500 Mail Librarian 500 Maintenance Agreements 533 Manuals Technical CalComp Mod 936 1051 420 Memorex 1270 230 315 Memorex 1270 Draining 205 Memorex 1270 Enable Disable Etc 324 Memorex 1270 Line Hung 322 Memorex 1270 Line I O Errors 315 Memorex 1270 Startup 231 Memory 0 17 Real 17 User 17 Virtual 17 Memory Virtual 66 Merge Filesave 217 MERIT 258 MERIT UMnet 171 Message File 240 essage Purge Job 202 Messages Also See BROADCST 94 essages Miscellaneous 404 Message System 240 Message System Data Structure Checking Of 201 Messagesystem Signonmessage 463 Message To User Plots 435 Messenger 349 etering Sectio
157. 8 402C1B96 FPRs 00000000 00000000 00000000 00000000 00000000 00000000 TRANSMITTERS RECEIVERS HASP 385 00000000 00000000 Dev AF7E 70 5 If the task is dormant this will show after HASPLING rather than I O on A or it is using CPU time as shown here it s ok If it ig not dormant and not using CPU time then issue the MTS command run sys units To get the status enter d devicename s where devicename would be AF7E in the above example The output will look like AF7E Type MNET DevNum OCFE ON LPM 80 Sch 0324 Job 05969 device initialized no errors I O queue active task waiting I O operation awaiting channel end SCHIB mode flags sch ON lim chk OFF measur blk upd dev timing devnum ok int sbcls 3 MBI 0324 LPM 80 PNOM 00 LPUM 80 PIM 80 POM FF PAM 80 PATH 0C SCSW OF SCHIB key 0 Fmt 1 CCW Prefetch allowed Function start Activity sch actv dev actv Dev status unit check Sch status PCI program check DEVLST Config flags No timeout DEVTBL ATTN ON SNS ON SCSW OF DEVTBL key 0 The status should be i o operation awaiting channel end for this solution to work If the status is not stated as above then the problem is elsewhere and the action stated here will not fix it To be sure that you have the right problem issue the following MTS command run mta disschib oa Enter the device name at the prompt and the output will look like Path control word 0000B6D8 189B0CF
158. 8 BUFEWF 00000000 LCBMCB 00 LCBACK 00 LCBRCB 0000 BUFPCEAD 0024C908 TOBCCW1 27 000000 CC SLI 1 TOBCCW2 01 1CA06C CD SLI 4 TOBCCW3 01 1ECD76 CC SLI 2 TOBCCW4 02 1ECDCO SLI 400 TOBCCW5 01 1CA06C CD SLI 4 TOBCCW6 01 1ECDCO CC SLI 361 TOBCCW7 01 1CA07E CC SLI 2 TOBCCW8 02 1ECDCO SLI 400 TPBUFCHN 1002888F 1ECDCO 1002888F CFFOC179 5BE2C9C7 D5D6D540 OAS SIGNON 1ECDDO 40404040 404040D9 C5D4D6E3 C5F2F240 REMOTE22 1ECDEO C2E2E2D9 40404040 40404040 40404040 BSSR 1ECF40 40404040 40404040 40404040 40404040 The field TPBFDATA is the pointer to the next logical record which in this case points to the string FOC1795B However it should be RM PRIORITIES CLASSES SCHEDULE 383 pointing to a record control byte with value 92 the same as MDCTRCB in the DCT since HASP saw that before it initiated the console read dis lecdc5 1ECDC5 FOC1795B One way to recover from this is to alter the data in the buffer pointed to by TPBFDATA to be something HASP expects However before doing this it s best to STOP the remote line as outlined in the previous section The line DCT for the remote can be found from the field DCTDCB in the remote console PCE or BUFDCTE in the buffer For example d dct at 24fb18 DCT at 24FB18 SA07 DCTSTAT INUSE DCTPCE 2468F4 DCTBUFAD 001ECDCO DCTPSTAT LEASE ETX DCTDCB 000000 DCTBUFCT 1 MDCTATTN 0 DCTDEVTP LINE DCTPCODE FUL
159. 8 Beene 341 Abnormal System Condens esti li 343 ConsolesBaCkup aese dw td Se Se BSG eee A tica 343 Console Backup Support ES3090 3007 oooooooooooooo ooo 346 Phone Pager V NOS A a dace aw a a 347 1B Yl Fea te a et Ca ee ee eee eee ee 349 DTS Storage dos sees tre te dad a eee 351 Disk Manager DMGR tai A az 351 DIME a LI A Rd A aS te 351 PASDE is AA A A A A RA iaa 354 DESK CORY ea AS A A A ees a A AA ta iaa 355 PILEINEO aAG Gic ETLEIDEO a A ind 358 DEADELTLES A A a SR ata eee a 359 Direct Access Storages Hs a a bis 361 Unire Check Error COdES iia ds dd 361 DESK EPTOTS La E A bie we a esate or Sates 361 Faxed Head Eile Mrrorse it ace tears aed ates fers tue Ses Se aud ay ote id 361 Solid State Disk Error rS vaa ts WS ees a we ee ES 361 DAG Error COLES a we ieee a tangy we Sle eae Taio we on gba a BS 361 DUMP Sis PRIDE A A bal Stag es 365 Account Listing And File Charge Update Monthly 368 Pare Rag ters GUIAS Ni A o Boece dub eee 369 HOW PELOCLEEY ese ee AA A th St el en SA ae SE Bande el 374 DEFERRED Pr Ority gt 2 cies aki et ete a lee sake a Rotaract aches at tos 374 MINIMUM PETOCLEY cic Sete e Bae bee AS Be gia bee elev E 374 Print POLOLLEDES Aa aea A ad e ae aus 374 EXecution Priorities ici A eae HAS Se NS BS Fea See 374 Printe Categories sia lees ee es Sl A Shia Sean eRe 374 Priority Class 6 2 ness oe ee cod aw Bate Dard A eee Bac EE 374 HA
160. 8074 Palms Andy 747 5145 Public Facilities Techs 761 0967 761 0988 930 8088 761 0979 PHONE PAGER NUMBERS 348 Rowe Fred 668 5480 668 3429 668 3443 Sullivan Jim 996 2802 PHONE PAGER NUMBERS 349 DELIVERY Printer output Plots Diskettes Tapes etc The file DELIVERY which is changed from time to time contains information about delivery Diskettes Magnetic tapes output etc A messenger delivers page printer output to NUBS and BSAD leaving the Center at 8 am 11 am 1 pm and 3 30 pm Monday through Friday Magnetic and paper tapes and floppy disks and plots are delivered to NUBS but not BSAD according to this same schedule Magnetic tapes and plots are delivered to UNYN Because of budgetary and staff restrictions deliveries are limited to the above schedule The schedule is tentative and will change depending on the needs of the user community and availablilty of Computing Center staff and funds UM PAGE PRINTER DELIVERY To have all page printer output delivered to NUBS or BSAD issue the following MTS command before using PRINT for the first time in a terminal session SET DELIVERY NUBS To have a given PRINT job delivered to NUBS or BSAD use the following MTS command before the job is released CONTROL PRINT PRINTER PAGE DELIVERY NUBS To have a batch job printed on the page printer at the CNTR and delivered to NUBS or BSAD use a signon command similar to the fo
161. 83 PUC 317 PUC MTA 361 PUNCH ERRORS 313 PWC 265 P11 60 135 P16 267 QN 136 Queueing Tape Mounts 490 Queueing tapes 491 QUICKPLOTS 264 423 Racks Tape Electric Problems 502 RAMROD 77 Rate Classes Holding Releasing 275 RATE INIT Rates 374 Rate Schedule Copy Rates 374 Rates For Holidays How To Set 451 RA81 543 RCP 268 Receipts 268 Receivers Transmitters HASP 384 387 Recorded System Status Message 469 Recovered Machine Check 311 Recovered Storage Protect Failure 311 Recovery Power Fluctuations Failures 85 HHRE IPLHH Machine Error ERROR 412 571 Message 312 Reliability System 468 Reload Last IPL Information 237 REMOUNT TAPE WITH RING How 412 RES 270 Reserve Disks 282 Resource Manager 19 62 535 Responses Tape Mounts 488 RESTORE 270 Restore Files 270 Retrieving Tapes amp Diskettes By Mail 500 RHB 275 RIP 274 Ripple Print 274 299 RIP 274 TNR 299 RMC 276 RMD 277 RMP 278 RMS 279 RST 280 RSV 282 RUAC 543 Runaway Tapes 493 San Diego Super Computer Consortium 543 SANSEIF 1 443 Satatistics Tapes Combined 193 SAV 217 283 See Filesave For Schedule And Tape Storage SCAN 454 SCAN LISTSTAT Output Meanings 456 SCANX W030 454 Schedule Differential Rate Periods See Rates 374 Scheduled Shutdowns 389 SCPs UNYN 520 Screen OPER 335 SDB 284 SDM 140 286 SE 138 Security Clerk 460 SEG2 MONITORS 327 Sense Values 466 SETDATE MNET 403 Setdate UMn
162. A for information on how to do this 5 If the time and date are correct press ENTER and proceed with step 6 If the time or date is incorrect and you are loading UM MTS i e you are NOT running under VM XA then you can reset the clock To do this enter the command SET ZONE zzz DATE mm dd yy TIME hh mm ss and press ENTER zzz should be the time zone EST or EDT in Michigan mm dd and yy are the month date and year hh mm and ss are the hour minute and second 5c UM MTS will print the message Enable TOD clock 5d Go to the ES3090 300J console for the UM side of the ES3090 300J and press the alt key and while holding it down the TOD key 5e UM MTS should display the new time and date if it s OK press ENTER if not go back to step 5b 5f UM MTS may tell you to enable the TOD clock a second time if so repeat step 5d 6 At this point the MTS operator console and printer should start Note Looking at the UM ES3090 300J console in the lower left section of the screen see the Command word Just below this is displayed either UM or MVS Be sure that you are IPLing the correct system To change the systems UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 41 setlp um for the UM MTS system setlp mvs for the MVS system Running VM XA on the 3090 VM XA is an IBM operating system which runs on the 3090 At Michigan we run UM MTS on one half of the ES3090
163. A SYSTEMS PROGRAMMER Call someone Don t enter any response to the program just let it be until the programmer can look at it If fdname is CMDDIR CMD will probably become ill within the next hour or so unless someone fixes the problem If DWB prints the message DWB ERROR JOBDUMP AND OR CALL SYSTEMS PGMR it is confused in some way The best thing to do is call the firefighter if you can t get some help jobdump it and then STOP it DESCRIPTION AND FIREFIGHTERS INSTRUCTIONS Overview CMDTAPE writes files of CMDSTAT data to tape These files are filled with data from the CMDSTAT buffers by CMDPIKUP When CMDPIKUP fills a file it makes note of that fact in SYS CMDDIR and STRTJBs DWB CMDTAPE copies the file to tape and then empties the file so that CMDPIKUP can use it again Files SYS CMDSTAT1 and CMDSTAT2 are normally used for this purpose If CMDPIKUP fills both of them and needs more space it will create a temporary file named CMDSTATc where c is some character After CMDTAPE copies a temporary file to CMDSTAT 329 tape it will destroy the file In addition to writing the CMDSTAT records on tape both CMDPIKUP and CMDTAPE write records of certain types to other files When CMDPIKUP is hung up the problem is often that one of these files is locked and CMDPIKUP is waiting for it CMDTAPE will wait a while for any file it writes to then give up send the firefighter a message and skip tha
164. BLASTed Tf a task is BLASTed any termination exits that the task may have enabled are ignored If the task to be BLASTed is a terminal task the operator should wait at least three minutes after he has STOPped the task to allow the task to go through the full signoff procedure Before BLASTing the task the operator should know the task number of the task to be BLASTed the i d of the user if it is an MTS task and the device i d of the terminal if it is a terminal task If it ig a HASP batch task the operator should also get the HASP job number associated with this task He can obtain this information by initiating a TASKS job with the job number of the task to be BLASTed as a parameter The operator should use this information to restart the line later and to run FIX on that user i d see the section on OPERATOR FILES for a description of FIX The task will be terminated with a code of BLST WARNING Use of the BLAST job on an MTS task that is not actually hung up may cause damage to the file system that will keep the system down for several hours DO NOT USE BLAST except when absolutely required mnnnn a task number Make sure you BLAST the task you desire and not the task of another user who will become furious or a HASPLING task blast 42 JOBS PROGRAMS 93 NOTE The SYSTEM STATUS command blast nnnnn performs the same function and is especially useful if all job tables are in use JOB
165. BM 3180 24 Hrs 7 Days IBM 3864 24 Hrs 7 Days IBM 3380 24 Hrs 7 Days IBM 3370 24 Hrs 7 Days IBM 3880 24 Hrs 7 Days IBM 3180 24 Hrs 7 Days IBM Note For some IBM equipment see SORBUS below IBM 3274s 24 Hrs 7 Days IBM 3278s 24 Hrs 7 Days IBM 3287 24 Hrs 7 Days IBM 2520 See amp contact Sorbus MAINTENANCE AGREEMENTS 534 Lee Data 7am 4pm 5 Days M F Memorex Tape Drives 24 Hrs 7 Days Memorex Printers 6am 7pm 5 Days M F Memorex 1270 7am 6pm 5 Days M F ONTELS All locations Serviced by staff Spacesaver Electric Tape Racks Casper Systems Time amp Materials TAB Time and Materials TEKTRONIX Terminals Serviced by staff Telcom Engineering Ontels amp VISUAL 550s VISUAL 550 Time amp Materials XEROX 24 Hrs 5 Days 3 Only call for service during the times mentioned above There is no contract so on a time and materials basis Corp and days service is 4Tf problems occur with the CalComp plotter outside of the contract hours 8am 5pm don t call CalComp but rather leave a conspicuous note for the day shift so that they may call them right at 8am 5 On Call basis After hours much more expensive 6 On Call basis After hours much more expensive MAINTENANCE AGREEMENTS 535 Resource Manager Contributed by D Boettner General Comments 1 The RM is a great deal less fragile than HASP And it rebuilds its data structures from scratch using checkp
166. BMCNT 0 RTABPSWD RTABSPPG 0 RTABACPT 0 RM PRIORITIES CLASSES SCHEDULE 380 d det rm22 rdl find the line DCT in DCTDCB DCT at 232BDC RM22 RD1 DCTSTAT HOLD DCTPCE 000000 DCTBUFAD 00000000 DCTPSTAT SINON DCTDCB 2318F8 DCTBUFCT 0 DCTNO 22 DCTDEVTP REMOTE_READER DCTPCODE CON MRF CPU TRSP DCTPRINT 16 DCTPUNCH 16 DCTPRINC 0 DCTPRLIM 0F DCTFORMS 00 DCTCHARS Uc Pn DCTPRLEN 132 DCTPLINE P20 TRSP BSC DCTEWF 00000000 DCTFLAGS 0 DCTCHAIN 232C18 DCTDEVN RM22 RD1 DCTWAYT 00000000 DCTSJNO 0 DCTLOCK 00 DCTPRNTR Line DCTTOTPG 0 DCTTOTJB 0 DCTRESCN 00 MDCTRCB 93 MDCTDCT 232C18 d det at 2318f8 address shown for DCTDCB DCT at 2318F8 SA07 DCTSTAT INUSE DCTPCE 2286D4 DCTBUFAD 001C8C38 DCTPSTAT LEASE ETX POST DCTDCB 000000 DCTBUFCT 0 DCTATTN TIMER DCTDEVTP LINE DCTPCODE FULL BLK TRSP BSC MDCTCODE 001AC06C MDCTRSEQ 06 MDCTTSEQ 6C MDCTERCT 08 DCTPLINE TRSP BSC MDCTOPCT 00 MDCTOBUF 000000 DCTFLAGS 0 DCTCHAIN 231940 DCTDEVN SAO7 DCTWAYT FF000000 DCTSJNO 92 DCTLOCK 00 DCTPRNTR Line DCTTOTPG 0 DCTTOTJUB 0 DCTRESCN 00 MDCTRCB 00 MDCTDCT 232BDC MDCTFCS 8FCF0000 MDCTPSWD MDCTTIME 973C7E9E offending remote on SA07 DCTPSTAT shows it s a leased line sy t d sa07 find out task number 00092 HASPLING 04C860 Hsp FF231920 SA07 SA07 sy o stop 92 stop the haspling sy mod 2318f8 1c 80 set DCTFLAGS to X 80 sy
167. C LETRH K THETA C ZI4 SC the C can be changed to an N or a K or whichever character is needed If changed to an N the first slot on the strip would be NOOO After the necessary alterations if any to the program have been made the program must be compiled using Fortran To do this run ftn scards tapes If the program runs successfully EXECUTION BEGINS NO ERRORS IN MAIN EXECUTION TERMINATED should be printed The object program automatically is placed in a file named LOAD The next thing to do is to put the program in a form to be plotted sig w035 TAPE RACK STRIP LABELS 504 run load plotsys 9 plot Ready the plotter and use a 8 pen Now since the 1051 plotters are attached through the network one must be mounted To do this mou mnet plot d hXxCcb7 where xx is 00 for plt2 and 01 for plt3 Run sys plot51 run sys plot51 Execution begins WHERE FROM plot WHERE TO plot Bin SCALE FACTOR 1 0 usually PRIMING TIME SEC press enter HEADER CCIDDATETIMERECEIPT press enter NUBS no PENS press enter Below is a copy of the source program to be altered if necessary and compiled The name of the file is W035 TAPES REAL LETRH LETRL Cc C HEIGHT OF STRIP Cc STRIPH 10 0 16 0 C C LENGTH OF STRIP TAPE RACK STRIP LABELS 505 Cc STRIPL 23 1875 Cc C NUMBER OF TAPE SLOTS IN A RACK UNIT Cc NSLOTS 25
168. CAL Next to the DC Master Console and type HELLO go at each one of the MERIT PCP or DC postions If a response is printed usually something like High Core Monitor then you can go to the operators console and type assuming no that MTS is operating mts elo nlo etc depending on the machine type Tf responce appears then follow the insructions parts 1 amp 2 under the glass in front of Concentrators CB amp CC The instructions for loading the MERIT machine are attached to its front After these POWER FLUCTUATIONS amp FAILURES RECOVERY e En g h Tine SYSTEM SHUTDOWN AND START UP PROCEDURES 89 procedures go to the operators console and type mts nlo elo etc XEROX 9700 If possible it is better to let the DPAs take care of restarting the XEROX 9700 printer if a power failure occurs There are some instruction books under the wrapping table The first thing that should be attempted is a ROLLOVER More than two ROLLOVERS in a row will erase the disk pack on which the printout resides If the ROLLOVER doesn t work then the 9700 must be BOOTed The button to do this is in a lower panel on the Control Unit near the Disk Pack If a power surge occurs and 9700 power is still present press STOP and do a ROLLOVER Tf the ROLLOVER doesn t work do a BOOT MEMOREX 3403 If the MEMOREX 3403 line printer loses IBM 2501 IBM 2520
169. D OK When this comment appears start HASP but don t release execution This latest comment appears with the successful loading of shared virtual memory CMD THE CMDSTAT PICKUP PROGRAM HAS BEEN STARTED This comment may not always appear at this point but will be somewhere near here It appears after the cleanup procedure for CMDSTAT is finished and CMD begins This is to let you know that when you let out the lines things may stop until CMD gets started MTS 31 RECORDS Don t let lines out or do an mts rms or mts rmd until NAL finishes 00018 07 17 38 MTS NAL Complete There are seven of these VSE 62 SYSTEM SHUTDOWN AND START UP PROCEDURES rm mts rms Starts Resource Manager mts rmp Start Resource Manager Processors mts las mts mns The accounting file must be available the POST IPL SYSTEM must have been loaded and the number of records printed on the console before execution can be released release ex SHUTTING DOWN MTS 1 You should attempt to perform each of the steps outlined here but under some circumstances that won t always be possible Do the best you can but always do step 7 unless there is absolutely no alternative 2 From 30 minutes to several hours before the shutdown put in a signon message warning users of the shutdown Send broadcasts 60 30 10 5 and 2 minutes before the shutdown to repeat the warning Update the recorded s
170. DOWN AND START UP PROCEDURES 59 e The system should display the new time and date if it s OK press ENTER if not go back to step 5b f The system may tell you to enable the TOD clock a second time if so repeat step 5d At this point the operator s console should start 6 MTS will print the new date and time as it did in step 8 If everything is all right press enter Otherwise type no and go back to step 9 7 At this point the OPERATOR job should start up and you can proceed with system startup HR MIN AM PM 0 12 0 59 AM PM MO DAY YR Optional 1 12 1 31 last two digits of year Time Zone Either a positive or negative number giving the number of minutes relative to GMT or one of the following codes EST Eastern Standard Time EDT Eastern Daylight Time CST Central Standard Time CDT Central Daylight Time MST Mountain Standard Time MDT Mountain Daylight Time PST Pacific Standard Time BST British Summer Time GMT Greenwich Mean Time ADT Atlantic Daylight Time AST Atlantic Standard Time EXAMPLE IPL Enter MTSIPL or MTSIPL xxx where xxx is the plug address of MTSOO1 DO YOU WISH TO RUN THE CURRENT SYSTEM YES OR NO yes UMMPS 370 mm dd yy LOCAL TIME HH MM SS mm dd yy GMT HH MM SS MM DD YY OK VSE 60 SYSTEM SHUTDOWN AND START UP PROCEDURES The operator may then type ok yes or a blank line to accept the time as given No to reject
171. DS00 DS02 DS09 amp DSOB IBM 3274 1B DS20 thru DS2E IBM 3274 41D DS40 thru DS5F IBM 3274 41D DS80 thru DS8F Other MGR CLK MWT TPR ORC HIM MIH DSO Other CLK TPR This is the same as if the operator had typed in the job requests mts 1a00 mts la01 USAGE mts las OPERATOR FILES LAS 232 OPERATOR FILES Note LBH is a program to label tapes that are not to be processed by LBL the labelling program for cataloged tapes Users may use this program by running Label The the FUNCTION To label a tape so it can be used by MTS Any completely blank tape or any tape with an unreadable first record MUST have something written on it by this program before it is attempted to be used in MTS USAGE mts lbh The program will prompt for information Enter tape name density volume name owner t90x 1600 jake icpsr operator response above action wold label the tape on t90x at 1600 bpi using label jake and the optional owner id ICPSR t90x 6250 This action would write a few 6250 bpi file marks on the tape but would not label it The operator response for labelling an ANSI tape would be t90x 6250 mack engin lbltype ansi This last example shows t90x tapeunit 6250 density mack label engin ownerid and lbltype ansi causes the tape to be written in ANSI OPERATOR FILES LBH OPERATOR FILES 233 LBL NAME LBL FUNCTION To label a tape so it
172. Display address using TEST ADDRESS with TEST switch of Es A Roller 1 transmit B e 2 r 6 p eceive repare 7 enable F disable O dead or thinking Rol Buf Rol If ler fers ler CS CD or DSR lights are out line is not connected Examples Set roller to A To disable a line Set TEST ADDR switches to correct address 0101 SA05 Press CMD button Set TEST switch on Set switches in COMMAND REGISTER F Disable Flip SINGLE CYCLE switch Set TEST switch off CMD off To enable a line Set roller to A MEMOREX 1270 Set address switches at TEST ADDRESS Press CMD button Set TEST switch on Set switches in COMMAND REGISTER 7 enable Flip SINGLE CYCLE switch Set TEST switch off CMD off 325 CARRIAGE CONTROL TAPE FOR MTS The 1403 printers use 11 inch paper printing 6 lines per inch normally Thus there are 66 lines per page Printing is usually done in units of a logical page A logical page is defined as the 60 middle lines of the full 66 lines thus the first and last three physical lines on the page are not part of the logical page In either machine carriage control or logical carriage control i e USASI or ASA carriage control a request to skip to the top of the next page will be interpreted to be a request to skip to the fourth physical line on the next page this is where Channel 1 is defined to be LOGICAL CARRIAGE CON
173. Ds to be on Logic On RA81 disk LEDs SAN DIEGO SUPER COMPUTER 544 RUN STOP READY A Tube Console for printer DecWrite IV System Log Some Commands to be used on the DecWriter are Note The terms enclosed in brackets are the default that is if you just press ENTER that is what will take effect Username SYSTEM Password Use the following command to tell all online users of system shudown if not using the automatic shutdown sequence reply all system shutdown in five minutes Reply received from user SYSTEM at _XXA1 10 52 17 system shutdown in five minutes Press ENTER 3 terminals have been notified The following is the command to begin the automatic shutdown sequence sys system shutdown SHUTDOWN Perform an Orderly System Shutdown How many minutes until final shutdown 0 10 The answer to the following question will go out to online users Reason for shutdown Standalone Preventive Maintenance Example Do you want to spin down the disk volumes NO Press ENTER Do you want to invoke the site specific shutdown procedure YES Press ENTER The next question should be yes if the system is hung and the system needs to be back ASAP Should an automatic system reboot be performed NO y or Press ENTER if for PM When will the system be rebooted shortly via automatic reboot Shutdown options enter as a comma separated list REBOOT_CHECK Che
174. E 80008080 0324FF80 OCFFFFFF 0 Subchannel status word 00C040C0 00B26AE0 02A00080 TRANSMITTERS RECEIVERS HASP 386 If unlike this example the second half of the first word of the sub channel status word 40C0 in this case is hex 0 procede with the following Note This step will not be necessary once UNITS is fixed to print the information out correctly Also clr can fix a number of problems so it may work even if the problem is not exactly the one described here 8 Issue the MTS command run sys units Then enter d devicename clr This should free the stuck HASP receiver If the receivers appear to be free but thing are still stuck call the Firefighters since the problem may be with the UB transmitters rather than with the receivers A simpler version of the above may be tried See below t d ab7e t d ad7e t d ae7e t d af7e t d an7e The above tasking will get you the task numbers of the devices of concern Check the hex number under super time from the J tasknumber see above and if there s no change then do a units d a 7e clr One can also use status all on the UB screen wait a couple of minutes do it again and see if the jobs are changing TRANSMITTERS RECEIVERS HASP 387 Host Interface Machines HIMs HIMs 1 2 and 3 are located in the Electronics shop located at the east end of the second floor CNTR The HIMs are one means for users to connect to MTS using
175. E BEEN PUNCHED OPERATOR FILES RCP 270 OPERATOR FILES RES NAME RES Purpose To display the restore status i e the dates and times the file was saved and or to actually restore the file s Note Users can display and restore files using the RESTORE program However if a file is to be restored from one ID to another RES must be run by an Operator Usage The program is invoked by issuing mts res When execution begins RES expects the operator to enter a command that indicates which of several services is desired For example the DISPLAY command may be used to determine which versions of a file are available for restoration The RESTORE command may be used to indicate the name of a file to be restored Several successive RESTORE commands may be entered the file names are queued and then the files are restored after the STOP command is given The queue may be emptied with the CANCEL command After the stop command is entered the program automatically requests that an operator mount the proper tape s and searches the tapes for the proper files If a tape drive is unavailable or if the desired tape is in use a message will indicate which files were not restored Since tape mounting and searching are somewhat time consuming the resore operation may take several minutes An attention interrupt goose aborts the running of the program Comments RES does not check to determine whether files queued
176. E SMe ta as A AAA ASADA we ake AS 400 14 Unattended MOE a is whale IR AAA whee a ee GS Sr ete a Se ees 518 Unit Check Error Codes midana sa aan Sie a ae 361 UNYN Terminals SCPS REG e veda A a e 520 UPS Uninterruptable Power Supply o ooooooooooo ooo oo ooo ooo 549 Vendor Intormation dson A A a A A chee Baer 527 MESA ORINA o A A AAA A A aio 39 NLO Reload PCP AN No Dump o oooooooooooooooooooonoon nooo 259 Operator Full a did 168 FABS Starts ABT ABS AA e oa 169 ADS 5 SEarts AD LO ADIE IA duna 170 ARES Starts ARTO ABTS sinew A A 6 00 Onset eae asa 171 APS a Starts ARLO 5 ARI Li A A wed 72 FANS Starts ANIOS CANT et hal ak Ales oo AU A a A ds coat 73 FAUS 0 65 Starte AUTO AUGE G ooa kea e ee cvs AAA Oe ol 74 AVS Shar tis AVIO JAMON a ad 75 AWS lt a bar tS AWLOY AWG BY i agana a a ean ices ehh arse ete laa E Bien acer ehe ese ere 176 FARSA o tarte ARO AE a o R atau L77 AYS 2 oSbarts AYTO AYOP daria ed een BS a a a ad 178 ADS Starts AZLA ALOE Ci sive a id 179 BBG Monitors EDs Sig on off voada E ia TN eh a 180 CODA TO Defer PIOUS ia A A ad 181 CCG edna ZeES Phot Tapes murio ad dd a 184 CEP 2 Dump PLOES VO TAHE st A A Be a a 186 CCT Deng Of Plot SEBA a ta wel 187 CHK Checks For Offline EQquUIpment seseris eese nd ss a ad 188 CLB Retrieves Label From Tape And Inserts It In Tape Catalog RS ate Rent ate A A AS Sd he cots Sea ee
177. E1l mount process in detail 1 VSE1 will display a mount request on the VSE1 operator s console in a format similar to BG 000 BG 000 PLEASE MOUNT 1234 NOTIS DUMP A RING IN on 680 BG 000 PAUSE AND 5678 LIBRARY DATA RING 0UT on 681 BG 000 The last line or two will be highlighted The number at the end of the line on each mount request is called the virtual device number or vdev You will need to remember which tape is to be mounted on which vdev 2 Go and get the tapes from the racks and check the labels just as if you were mounting MTS tapes If the rack numbers and labels are OK skip to step 4 Otherwise proceed with step 3 3 If there is any error in the rack number external VSE 56 SYSTEM SHUTDOWN AND START UP PROCEDURES label or anything else in the mount request kill the batch job by performing the following Type 0 CANCEL and press ENTER That s a zero not an oh VSE1 should type O CANCELBG 000 11201 JOB name CANCELLED DUE TO OPERATOR INTERVENTION BG 000 1100D READY FOR COMMUNICATIONS BG 000 Type 0 and press ENTER That s a zero not an oh VSE1 will print a few lines and kill the batch job Tape mount processing for this job is complete at this point just wait for the next job 4 If the mount request makes sense and the rack numbers and external labels match continue processing the mount request 5 First determine how many tape
178. ECIFY TAPE DRIVE will be printed on the operators console The console will be in read status awaiting a reply If you wish to actually CANCEL the dump i e you really didn t want a dump type a c followed by an ENTER Otherwise type the name of the nine track tape drive on which you have mounted the dump tape e g t903 followed by an ENTER If you make a mistake in responding cancel the line and retype the response There are certain instances in which a task will internally request a job dump The operator should comply with all unsolicited requests If he has enough dump tapes Some internally requested job dumps will be followed by a message on the console explaining the reason for the job dump these messages will usually be typed after the job dump has been taken A dump tape is a 9 track labeled tape with volume label of DUMP mnnnn The task number of the task to be dumped JOB PROGRAMS JOBDUMP 116 JOB PROGRAMS EXAMPLE jobdump 453 DUMP SPECIFY TAPE DRIVE t903 After taking a dump log it in the dump note book together with the the other information requested in the book NOTE When the dump has finished completely DMP should be initiated by the operator DMP has to do tapes See the section on DMP appearing with printing dump later in this manual JOB PROGRAMS JOBDUMP NAME FUNCTION PROTOTYPE USAGE
179. G OR IT IS HUNG is printed on the console it means that an MTS job has been unable to log a STATISTICS record by calling CMDSTAT The first time the message is printed for a given job it means that the call to CMDSTAT has failed 60 times one Call a second The message will be repeated every two minutes every 120 calls until a call succeeds While the job is doing this the T command will usually show the job in TIME WAIT When this is going on the job has the file STATISTICS locked for modification so any other jobs that try to signoff will end up in a FILE WAIT for STATISTICS including CMD if it is trying to sign off When you see this message you should do the following 1 Check to see if CMD is running The simplest way to do this is to say T U SYS on the console and look for CMD 2 If CMD isn t running you should restart it MTS CMD and then look back at the console to see if you can figure out why it stopped 3 If CMD is running check its status several times using the T nnn command If it shows CMDSTAT WAIT the problem has probably cleared itself up 4 If CMD is in a FILE WAIT use TABLMOD to see which file OPERATOR FILES CMD 196 OPERATOR FILES CMD is waiting for If CMD is waiting for STATISTICS you should start a new CMD job MTS CMD 5 If CMD is waiting for a file other than STATISTICS use the output from TABLMOD to figure out which job has the fi
180. H TEST The available tests are currently CHARS draws characters LINES draws lines in various directions GRID draws a grid DOTS draws dots SIN draws a sin curve POLY draws polygons and circles BLACKEN draw lines close together DASH draw dashes LINE draw without picking up the pen JOTE print something in 2 inch letters STAR draw lines intersecting in a single point SYMTEST draw characters in decreasing sizes The details of each plotter test are described below For each test the program will plot for a bit then plot the time and then repeat that until you hit the attention After an attention the plotter will plot a bit more ie it won t stop immediately then the program will ask WHICH TEST again SPECIFIC TEST DESCRIPTIONS For each test there are several variable parameters Values are suggested for each to get you started feel free to try other values as well Whenever the program requests two or more numbers on the same line enter them separated by commas Most of the test ask which pen number to use PLOTTING SYSTEM 442 PEN Respond with a number between 1 and 4 inclusive SIN plots a sine curve along the drum Program will ask PERIOD AMPLITUDE INCREMENT IN Enter 3 floating point numbers separated by commas The increment is the smoothness of the curve a smaller number is a smoother curve For starters try 2 1 01 Finally the pro
181. I 90 display with printing options You may select these options with the mounse or use the keyboard to enter commands e Refer to the Xerox UI 90 Operator Guide for instructions on using the UI 90 screens to operate the 9790 11 The time and or date can be changed without re booting the system The system must be off line and idle with no background tasks running and the logon level must be at a class two or higher to make a change enter SET TIME MM DD YY HH MM SS You may enter the time or date or both Follow the same procedures explained in steps 5 and 6 If the time of day status display option was selected when the software was loaded the time change appears in the upper right corner of the display SETTIME is available whether or not the time of day status display is enabled The system records of the date change and the command line displays the time change 9790 POWERING ON LOADING PROCEDURES 562 Removing PR belt 1 10 Put on plastic gloves Open covers Cam down developer module using Xerox tool Swing out laser shield using screw driver Remove edge exposure lamp at left Remove detrack cortron at top of belt Remove laser lamp at bottom left Raise and latch C cortron Remove drive and support shaft nuts Screw in short extension shaft to the drive roll at upper right Screw in the long extention shaft to the drive belt roll at lower left Slide photo receptor
182. IST tapehistory CMVTEREP cnvterepobj CNVTCTL ctlfdname LOGREC historyfile To pop the stack of history files PM2POP HISTORY TAPEVOL HISTFILE pm2history TAPEHIST tapehistory To push the stack of history files PM2PUSH HISTORY TAPEVOL HISTFILE pm2history TAPEHIST tapehistory The parameters required for normal use of PM2 are as follows PAR params PM2 448 params is any set of parameters from chapter 5 of the STC PM2 manual that is appropriate for the PM2 program being run The defaults are as given in the STC manual except that PM2DALY defaults P N if and only if not explicit PAR param parameter is given Also PM2STAT defaults to PAR P N TYPE OT but this is not given much use since PM2STAT requires a DATE or DTTM parameter which must be given on the PM2STAT command If more than one parameter is given they must be separated by commas and enclosed in quotes PRINT printfdname The printed output from PM2 will be directed to printfdname If this is a file the output will be automatically appended to the end of the file you need not specify L 1 on the fdname The default is PRINT PRINT TVOLCMDS volfdname Volfdname is an fdname from which will be read tape volume control functions as described starting on page 5 16 of the PM2 manual Normally volfdname will be a file The default is TVOLCMDS DUMMY CFIGIN config Config must be an fdname from which will be read configurati
183. IVE TASK WAITING I O OP AWAITING CHANNEL DEV END END DEVICE BUSY SIO PENDING JOBS PROGRAMS JOB PROGRAMS UNITS 165 No other ops to control unit until sense following unit check is done CC 2 to SIO or SIOF Status from a CSW Retry every so often even if we don t get a CU end following a CU busy lt n gt I O operations outstanding for the channel lt n gt I O operations outstanding for the CU Path was in use There is a device table for this device No UNIT CHK UNIT EXCEPTION or INOP CHANERR CC 3 to SIO or SIOF or channel error in CSW Status from CSW Status from CSW A SIOF has been done for the top entry on the active I O queue Task has done an SVC WAIT or an SVC ENDEXIT Subchannel still in use Subchannel free but I O op not complete Status from CSW I O for Top entry on active queue is waiting super to do a SIOF OLD STYLE 166 SENSING JOB PROGRAMS ON SPECIFIC PATH USING RESTART CAW ERROR EXIT ACTIVE SPUR DEV END WAIT FOR CE NOT DE LAST I O OP INITIATED BY OLTEP OLTEP ATTN 1 OLTEP ATTN 2 GTUNIT QUEUED CU END RECEIVED FAKE DEVICE END PENDING DEVICE STATUS CHANNEL ATTENTION STATUS MODIFIER CONTROL UNIT END BUSY Task request software command chaining sensing following UC DLG Software command chaining Super doing sense op following unit check Top entry o
184. JOB PROGRAMS GOOSE 112 NAME FUNCTION PROTOTYPE USAGE PARAMETERS JOB PROGRAMS INIT JOB PROGRAMS INIT To initiate the system at IPL time INIT The INIT job is automatically started when the system is IPLed It is this job that prints out the message TIME AND DATE HAVE BEEN SET TO This job also starts an MTS task with parameters INI T If you have gone through a normal IPL there should be no reason for the operator to start the INIT job If he does he might run the chance of destroying all scratch files in use that might be the least of his worries however None NAME FUNCTION Function JOBS PROGRAMS 113 LT To obtain a list of all of the intertask Networks in existence and to sense a particular Network The IT program does not process parameters but rather reads commands directly from the console Currently there are only three commands LIST SENSE and HELP LIST to obtain a list of the names of all networks in existence The list command does not take parameters SENSE net name lt modifiers gt Function HELP Example to sense a particular network The network name must be specified A summary of the information in the network control block and one line of information per node is printed The information in the summary is self explanatory The information in the node descriptors is largely sef explanatory except perhaps for the
185. KM BH DP M GUE Finally a summary line is printed giving the number of jobs printing waiting to print punching and waiting to punch This command continues to print information until interrupted by an attention interrupt PAl on a 3270 or until the JOBS job is STOPped or GOOSEd SYSTEMSTATUS Example SC CNTR Staff Only Status SOS 481 Note A Signonm message should be installed when the system is in SOS mode Don t turn SOS on unless execution is held MOD SOS 1 begin SOS MOD SOD 0 end SOS While SOS is on users are prohibited from signing on TAPEQUEUE descriptors This command permits the operator to exercise control over the queue into which tape mount requests have been placed See the section TAPE HANDLING in this manual for instructions on queueing process TAPEQUEUE prints the size of the queue TAPEQUEUE LIST list the current queue TAPEQUEUE POP 6250 800 releases the top reel tape job in the queue TAPequeue POP 38000 releases first cartridge request from queue TAPEQUEUE RELEASE releases all jobs in the queue TAPEQUEUE RELEASE JOB releases job number TAPEQUEUE ABORT aborts all of the jobs in the queue TAPEQUEUE ABORT JOB aborts job number To turn tapequeueing on mod tapeq 1 To turn tapequeueing off SYSTEMSTATUS the 482 mod tapeq 0 To display the status of TAPEQ dis tapeg
186. L BLK TRSP BSC MDCTCODE 001CA06C MDCTRSEQ 01 MDCTTSEQ 00 MDCTERCT 08 DCTPLINE TRSP BSC MDCTOPCT 00 MDCTOBUF 000000 DCTFLAGS 0 DCTCHAIN 24FB60 DCTDEVN SA07 DCTWAYT 00000000 DCTSJNO 0 DCTLOCK 00 DCTPRNTR Line DCTTOTPG 0 DCTTOTJB 0 DCTRESCN 00 MDCTRCB 00 MDCTDCT 250EEC DCTFCS 8FCF0000 MDCTPSWD MDCTTIME 985F4427 The offending remote is on SA07 BSSR again so stop its HASPLING and set the STOP bit in the line DCT sys t d sa07 00027 HASPLING 04E950 Hsp FF24FB40 SA07 SA07 sys O stop 27 sys t d sa07 xxx Job not found sys mod 24fb18 1c 80 turn on stop bit in DCT for SA07 Finally change the data in the remote console buffer to give an end of file to the MBCPUGET routine the magic sequence is 92800000 sys mod lecdc5 92800000 address form TPBFDATA dis lecdc5 1ECDC5 92800000 Fire up the remote console PCE by clearing the PCEEWF field offset X 50 in the PCE sys mod 246948 50 0000 The PCE will get an EOF notice that the line DCT was stopped get rid of it and go onto the next remote console write The WTO queue should clear up a few minutes later RM PRIORITIES CLASSES SCHEDULE 384 Procedure for Freeing UM HASP Transmitters If HASP seems to be stuck attempting to transmit UB jobs to the U The following procedure may clear the impaired UM receivers Note If you follow the procedures below 1 run hasp d a 2 Enter d dct type mnetreader all The s
187. LAR RIBBON COPIES 416 OLTS Aug 6 1974 VERSION There is a version of OLTS available in MTS which is the same as in release 21 7 of OS The procedures required to run it are somewhat different than before and are given below To run OLTS signon under any staff userid and enter source mta oltepcmd This will load a special version of FAKEOS and start OLTEP When this is completed OLTEP will print IFD102I OLTS RUNNING and further communication will be as described in IBM Manual C28 6650 6 on pages 41 47 The operation of OLTEP in MTS is as described in this manual except for the following 1 Pages 1 40 describe how to install and invoke OLTEP in OSs and are irrelevant They are replaced by the paragraph above 2 The procedure required to stop running a test is to press the attention key on your terminal instead of the procedure described on page 47 3 All communication with OLTEP is done via your terminal not the Operators console 4 It is not necessary to type r 00 xxx since xxx is treated exactly the same 5 All devices to be tested must be ONLINEd in order for OLTS to work The Channel s Control Unit s and paths to the devices may be OFFLINEd if desired to keep normal system I o from them If anyone has more test sections they wish to have added they should give them to M T Alexander The printed output from OLTEP referred to as the DIAGMSG DD in the manual will be p
188. LED INITIALIZATION OF PLOT TAPES CCL This program initializes plot tapes for use by CCP PLOTTING SYSTEM 431 Run the program by entering mts ccl at the operators console The program will print ENTER DRIVE DENSITY NAME Type in the name of the drive e g T903 the density 1600 or 6250 and the name of the plot tape e g TUE 1400 separated by blanks The program will label the tape with volume name PLOT and write a file used by CCP to identify the plot tape The same information may then be entered for another plot tape Use an end of file to terminate the program The name of the plot tape must appear in exactly the right format e g MON 1000 but not MON 1000 The weekday names used are MON TUE WED THU FRI SAT and the times are OPTS 1000 1400 1800 2200 0200 OPTS refers to the 8 30 am different pen size plots The plot tape names WEEKEND and DEFERRED are also recognized Examples T903 1600 FRI 2200 T905 1600 WEEKEND NOTE Use the program PLB described elsewhere in this manual to display information about the format of a plot tape LENGTH OF PLOT SESSION CCT NAME CCT FUNCTION To let the operators know how much time the PLOTTING SYSTEM 432 USAGE plotting session would take and how many user tapes would be used if the plots were dumped on tape and run at the time CCT is run mts cct Response on operator s console might be PLOTTING TO DO 27 MIN
189. LING tasks TMTS NAME FUNCTION PROTOTYPE USAGE PARAMETERS JOBS PROGRAMS 153 TMTS To start a task using a test developmental version of MTS As for MTS This job request is used in testing out a new version of MTS The new version must first be loaded using the TMTS loader It is otherwise similar to MTS As for MTS JOB PROGRAMS TASKS 154 NAME FUNCTION PROTOTYPE USAGE PARAMETER JOB PROGRAMS TN To load the TN 1403 or 3203 print chain or train image into the Universal Character Set buffer on a printer controller Additionally for 3203 type printers the forms control buffer FCB is loaded for the standard MTS carriage control channel assignments TN device i d ECHO FOLD ALLOW NOMSG If one wishes to change print chains or trains on a printer it is necessary to load the UCS buffer in the printer controller with the correct character set If the UCS buffer gets scrambled for any reason one would also want to reload it In addition at system IPL time it is wise to reload the UCS buffers The contents of the UCS buffer on can also be echoed on the associated printer for verification device i d OPTIONS The following options are recognized EXAMPLES JOB PROGRAMS TN ECHO causes a print train verification message to be printed after loading the UCSB FOLD enables folding when loading the UCSB This allows the printing of lower case l
190. MON MON 2PM 1745 MON 2145 MON MON 6PM 2145 MON 0145 TUE MON 10PM 0145 TUE 0900 TUE TUE 2AM 0900 TUE 1345 TUE TUE 10AM 1345 TUE 1745 TUE TUE 2PM 1745 TUE 2145 TUE TUE 6PM 2145 TUE 0145 WED TUE 10PM 0145 WED 0900 WED WED 2AM WED 8 30am Special 0900 WED 1345 WED WED 10AM 1345 WED 1745 WED WED 2PM 1745 WED 2145 WED WED 6PM 2145 WED 0145 THU WED 10PM 0145 THU 0900 THU THU 2AM 0900 THU 1345 THU THU 10AM 1345 THU 1745 THU THU 2PM 1745 THU 2145 THU THU 6PM 2145 THU 0145 FRI THU 10PM 0145 FRI 0900 FRI FRI 2AM 0900 FRI 1345 FRI FRI 10AM PLOTTING SYSTEM 422 1345 FRI 1745 FRI FRI 2PM 1745 FRI 2145 FRI FRI 6PM 2145 FRI 0145 SAT FRI 10PM 0145 SAT 0900 SAT SAT 2AM 0900 SAT 0900 MON WEEKEND If the software senses that the wrong tape has been mounted the following comment will appear on the operators console TAPE IS FOR MON 1000 OK TO USE ves or no or ok A no answer will result in a request for another tape If CCP is run twice in the same plot time frame the following message will appear TAPE LAST USED mm dd yy hh mm ss WILL APPEND NEW PLOTS OK TO USE ves or no or ok If a no is given a new tape will be asked for Occasionally a user tape will be mounted by the plot routines while CCP is running This tape contains the user s file If the plot routines attempt to mou
191. MPER of course and the user FIRE FIGHTERS GUIDE 373 is left in PRINTDUMP command mode The dump will be listed on PRINT by default unless SPRINT is explicitly assigned to something else on the RUN DUMP DUMPER command FIRE FIGHTERS GUIDE 374 HASP EXECUTION AND PRINT PRIORITIES The priority of a job is determined on the basis of the size of the job The priority for execution is computed based on the limits for time given on the SIGNON card Small time requirements produce higher priorities When execution of a job has completed a new priority for printing is computed based on how many pages the job actually produced not how many were estimated times the number of copies required LOW DEFERRED and MINIMUM Priorities These priorities may be specified by the user when the cost of the job is more important than the quickness with which the results can be retrieved The user may specify PRIO LOW OR L for LOW priority or PRIO D or DEFER for DEFERRED priority or PRIO MINIMUM Since the schedule for various priority classes is changed frequently it is not listed here If this schedule needs to be seen it can be found at the beginning of RATES These priorities apply to batch and terminal jobs Notes The commands HOLD DEFER HOLD MINIMUM RELEASE LOW RELEASE DEFER and RELEASE MINIMUM are done automatically by the system These commands could be initiated manually by the operator
192. MTS UMnet AU Configuration 6 April 1992 MTS Jobs on AU10 through AU6F MTS UMnet AB Configuration 6 April 1992 MTS Jobs on AB10 through AB3F MTS UMnet AD Configuration 6 April 1992 MTS Jobs on AD10 through AD3F MTS UMnet AE Configuration 6 April 1992 MTS Jobs on AE10 through AE3F MTS UMnet AF Configuration 6 April 1992 MTS Jobs on AF10 through AF3F MTS UMnet AN Configuration 22 April 1992 MTS Jobs on AN10 through AN17 or current equivalent OPERATOR FILES MNS NAME FUNCTION USAGE NOTE OPERATOR FILES MSD 239 To set the time and date in all PCPs on the UMnet network mts msd This job is normally CLK at midnight run automatically by OPERATOR FILES MSD 240 NAME FUNCTION USAGE REMARKS OPERATOR FILES MSG Cleans up Message System particularly MESSAGES mts msg See REMARKS This program is started up each morning at 2am by the clockwatcher program It should never be started manually A log is kept in the file MAIL MSG LOG OPERATOR FILES MSG OPERATOR FILES 241 MSS Name MSS Function Provides lists of and important information about tape mount instances Usage mts mss oper NOTE In place of OPER a printer gt PTRx may be used x being the number of an available printer Example ENTER R RACK NO STARTING DATE AND ENDING DATE OR U USERID STARTING DATE AND ENDING
193. Motor Stop unless PB I is bypassing NOTE Never press kK k k the Motor Generator Load on Bypass 8 After Load On Bypass indicator is on press Motor Stop METERING SECTION UNIX fil UNIX Filesave 557 les specifically e a and f need to be saved about once per month and most of the others every other month The UNIX filesave can be run either from the DECwriter near the TU80 on UNIX fil lesave cannot be run with users on the A new unitialized tape can be used for the however older filesave tapes will be re used When asked Which Host use XAAD from Addr use 9999 login user types filesave the second floor or from another terminal UNIX system filesave Soon a terminal password user types dogood in lower case source e filesave setup Typing u will Typing df will show the different file finger name of user An example run follows login type filesave password dogood use lower case df shows different file systems source e filesave setup dodump 0 e ok ok Doi kk to do a level 0 y to do backup e y ng file system consistency check dev rhp1b Last mounted on e Kk Kk Kk Kk Kk 442 Phase 1 Check Blocks and Sizes Phase 2 Check Pathnames Phase 3 Check Connectivity Phase 4 Check Refence Counts Phase 5 Check Cyl Groups 1 files systems tell which users on are the s
194. NG ERROR R R 405 ACCTNG STAT ERRROR READ WRITE TSA 405 ADS 170 AES 171 AFS 172 Alphabet 443 Amdahl Disks Interrupting 88 Amdahl Disks Power Down 24 Amdahl Disks Powering Up 86 ANS 173 ANSI Labelled Magnetic Tapes 232 234 ASCII Labelled Tapes 232 234 Attaching Detaching Tape Drives UB VM XA 37 Attach Tapes 31 Attention Interrupts 110 ATTN 110 472 AUS 174 Autohasp 451 Automatic Job Scheduling Program 190 AVS 175 AWS 176 AXS 177 AYS 178 AZS 179 Babysit 180 Backup Console UM MTS 27 Batch 119 121 Batch Jobs 235 Batterys 553 BBS 180 BLACKEN 441 444 BLAST 92 309 472 BROADCST 94 Btchprio 451 Btchprio Values 452 Buffer Print PN 134 Buffer Print P11 135 Buffer Print ON 136 Buffer Print TN 154 Buffer Print T11 155 565 General Index CalComp Plotter 186 CA1Comp 936 445 CAOP 224 CARD READER ERRORS 314 Carriage Control HASP 326 Carriage Control Logical 325 Carriage Control Machine 325 Carriage Control Tape MTS 325 CATALOG EXTENDED 405 CA28 210 CBOP 224 CCD 420 428 CCL 420 430 CCOP 224 CCP 186 190 421 434 CCPFILE 421 CCPPRIO 429 CCP User Tapes 422 CCT 187 420 431 CHANNEL ERRORS 307 Channel Logouts 318 CHARS 441 443 CHK 188 CLB 189 Clear I o 159 CLK 190 231 CLK Restart 192 CLN 191 CLR 192 CMB 193 CMD 191 195 CMDPIKUP 327 CMD Problems 195 CMD Records User 454 CMDSTAT 327 CMDSTAT Prob
195. NSF equipment are located in the NSF machine room where the IBM 4381s are housed Two other EPO buttons on the third floor are located near the two doors leading from the west half of the third floor to the NSF area These can be seen as one enters the NSF area from the west end of the third floor The EPO button on the second floor is located just inside the east facing double doors leading to the Communications area Of the three EPO buttons located on the first floor two are located in the machine room one inside and near the north door and the other inside and near the south east door The third first floor EPO button is located in the Electrical room just inside the door to the left just below a shunt trip breaker The Reset button This button is a green push button and is located just above the shunt trip breaker mentioned in the paragraph above Note Before using this reset button be sure that all of the EPO buttons are pushed back in Also there are some circuit breakers to be reset Several of these are loacted in a large gray cabinet on the same wall as the reset switch but on the east side of the door This cabinet is attached to the wall and is equiped with a door Another breaker which must be reset is located in a breaker box just inside the first doorway leading into the UPS room It is on awest wall There are five vertically positioned breaker switches in this cabinet and the center one must be reset Befor
196. NTER HEADER CCIDDATETIMERECEIPT user enters information shown in the parenthesis that is to appear on the plot or user presses ENTER NUBS if the user wants the plot shipped to NUBS yes is entered PENS COLORS SIZES NEW OR NULL user can press ENTER at this point if the default pen setup is wanted new causes the plot routines to print 1 after which the user can type color size and type e g 1 blak med liq Following this the routines type 2 3 and 4 until all pen positions have been attended to If a null line is entered after an n than the default setup for that position is assumed plot is plotted then WHERE FROM system asks fendfile user types to terminate program If the file to be plotted has been routed to another file for later plotting the where to file can be copied to the plotter copy file gt pltl bin user types The programs SYS PLOT36 and SYS PLOT51 formerly W030 PLOTX and W030 PLOT1X have been modified Nov 1980 to handle pen changes in a different manner than before PLOTTER TESTING PROGRAMS PLOTTING SYSTEM 441 SYS PLTEST36 AND PLTEST51 PLTEST36 is for the CalComp Model 936 and PLTEST51 for the 1051 To run R SYS PLTEST36 or SYS PLTEST51 Tf you get an undefined symbol message from the loader type IGNORE and press enter The program will ask PLOTTER NAME respond with PLTn e g PLT1 The program will ask WHIC
197. Note The underlined portions below represent responses entered by operators or others EXAMPLE RUN FILE DASDI Execution begins MTS DASDI program EB265 Enter input line Dddd 111111 Vx PAGING PRIVATE pars D354 MTS009 VX 9 D354 CURRENTLY LABELLED AS OLDOO09 PLEASE CONFIRM OK PAT TO BE WRITTEN ON PAGES X yyyyyy THRU X zzzzzz Enter next input line Dddd 11 11111 Vx PAGING PRIVATE pars D355 UNUSED V2 PRIVATE LABEL IS UNREADABLE ENTER OK TO CONTINUE OK PAT TO BE WRITTEN ON PAGES X vyyvyy THRU X ZZZZZZ Enter next input line Dddd 111111 Vx PAGING PRIVATE pars D340 SPOOL1 V2 PRIVATE LO OK Dddd 11 D340 CURRENTLY LABELLED AS OLDSPL PLEASE CONFIRM Enter next input line 11111 Vx PAGING PRIVATE pars endfile Execution terminated DISK MANAGER DMGR 354 The program FILE DASDI is used to initialize disk packs 3330 3350 3370 6280 3380 Description of Input Column 1 4 character drive name Column 5 blank Column 6 6 character volume serial number Column 12 blank Column 13 14 VX Column 15 Blank Column 16 public volume number Normally the public volume number woul volume serial number ld correspond to the but not necessaril y Options may follow the Public volume number separated from it by one blank LO just write label IPL reserve 150 pages
198. OC GETSPACE FREESPAC 41 GFDMPMSG FREESPAC GSFS0755 SOURCE GETSPACE FREESPAC 41 GFDMP FREESPAC GSFS0748 gt gt gt HELP ERROR WHILE EXTENDING CATALOG User Explanation Serious trouble call a systems programmer DESTINATION OPER SOURCE FILE SYSTEM 482 23 GETSEG CATALOG SUBR MISCELLANEOUS MESSAGES 408 gt gt gt HELP SNARK IN MTS User Explanation System error see an MTS systems programmer System Explanation This infamous message occurs only on the operators console TE occurs when MTS has received a fatal exit such as a program interrupt from the supervisor and no input line has been requested from the user since the last such fatal exit so we have a potential loop The particular MTS task causing the snark is put into a permanent wait so it can be stopped No other task in the system is affected DESTINATION OPER TEXTLOC MTS 42 HELPMSG OOPS MTS02872 SOURCE MTS 42 OOPSNA OOPS MTS02863 gt gt gt ILLEGAL INPUT User Explanation Printed on the operator s console when after GOOSEing a batch job he enters an invalid response The only legal responses are DUM NEX or ATT DESTINATION OPER TEXTLOC MTS 42 ZTPOPRM ZTPA MTS04530 SOURCE MTS 42 ZTPRDM ZTPA MTS04518 gt gt gt ILLEGAL REPLY User Explanation This error message is printed when the operator has given an improper response to a MOUNT request DESTINATION OPER TEXTLOC MOUNT 101
199. OR THE IPLREADER amp FRIENDS MTS is loaded from an ordinary sequential file when IPL occurs When the system is IPL ed the following message will appear Do you wish to run the current system YES or NO If you reply YES the system will automatically be loaded If you reply NO the IPL program will read further commands from the console You should not give a NO reply unless specifically instructed to do so An explanation of the IPL program follows RAMROD A program called RAMROD Replacement And Maintenance of Resident Object Decks will be mentioned in parts of the IPL section of this manual It is a program which runs under MTS RAMROD replaces the big system object deck of cards which had to be maintained and sometimes loaded by the systems programmers The function of RAMROD is to keep track of one or more systems and their components and to allow the systems programmer to manipulate the systems and components with relative ease SYSTEM START UP PROCEDURES IPLREADER 78 SYSTEM SHUTDOWN AND START UP PROCEDURES Introduction IPLREADER The IPLREADER itself is a stand alone program which when IPLed will load a system usually MTS from an MTS sequential file into memory and then transfer control to the entry point of the loaded system By default it loads the system from the file named IPL 0 Constructing a sequential file containing a system to be loaded is a feature
200. OTE OPERATOR FILES DWB Program which copies command statistics and unit check messages from a file to a tape This program is started automatically by the pickup program when a file fills up Is normally started up automatically Requires only mts dwb Takes no parameters See the CMDSTAT writeup for details and explanations OPERATOR FILES DWB NAME FUNCTION USAGE OPERATOR FILES 213 FIX Update the accounting record of a user to be correct with regard to files and to unlock the user s accounting record To be used if the user had to be BLASTed or if his task SNARKed This is automatically done for all users when the system is initialized after IPL so any users cut off in mid air will automatically be fixed mts fix MTS SIGNON CSIG ODP NDP ODF NDF MTS 0ab2 MTS OAB2 0 47 54 14 15 MTS where the last request for input above was terminated by pressing the cncl button For each i d fixed it prints the signon id the number of current signons the old disk pages new disk pages old disk files and new disk files OPERATOR FILES FIX 214 OPERATOR FILES FLB Name FLB Function To determine whether or not a tape is labeled If labeled displays label owner id and density If not labeled message confirms Usage mts flb Examples mts flb ENTER TAPE NAME t910 TAPE NOT LABELED eof if no additional tapes mts flb ENTER TAPE NAME
201. OWN shutdown requested by operator SHUTDOWN I STOPUSER All user processes will now be stopped SHUTDOWN I REMOVE All installed images will now be removed SHUTDOWN I DISMOUNT All volumes will now be dismounted 6 5 S OPCOM 2 FEB 1988 10 59 44 69 Message from user SYSTEM _XXA1 UMRUAC shutdown was requested by the operator 5 S OPCOM 2 FEB 1988 10 59 45 26 Logfile was closed by operator _XXAl Logfile was SYS SYSROOT SYSMGR OPERATOR LOG 184 5 S OPCOM 2 FEB 1988 10 59 45 59 Operator _XXA1 has been disabled username SYSTEM PM Times 1988 Feb 22 Apr 25 Jun 27 Aug 22 Oct 24 and Dec26 SAN DIEGO SUPER COMPUTER Special Page Printer Forms Occassionally users will request that their ouput be printed on paper of their own choice This paper supplied by users must be the standard 8 1 2 x 11 inches and not be too thick to pass through a page printer It s probably a good idea if the user supplied forms are of doubtful thickness to have a test run will a small number of sheets When a request is made for special forms use the following procedures should be used to give the user access to an SP number 1 Determine the system to be used UB or UM 2 Request the following information from the user a Full name of user b Signon id to be used with the special forms job c The purpose of the prin
202. PROGRAMS BLAST 94 BROADCST NAME FUNCTION PROTOTYPE USAGE LIMITATIONS JOB PROGRAMS BROADCST WARNING This job allows the operator to 1 send a message to all terminals attached to the system or 2 request that only certain terminal tasks receive the message BROADCST ALL nnnnn With the parameter all the message will be sent to all users If any other parameters are present they are assumed to be task numbers After giving this command the BROADCST job program will read a line from the console this is the line that will be BROADCSTed Just issuing BROADCST will only result in printing a list of task numbers which did not receive the last message A message will be sent to only those tasks that were indicated in the parameter list If the operator cancels or enters a null input line nothing will be broadcast The program MTS BROADCST can be used to initiate broadcasts from terminals other than the operators console The message will not be sent to a terminal until the current line at the terminal is terminated i e typed If the terminal is UMnet the 1270 or a 3270 and is in read status the message will not be printed until the line is read If a message is sent to a 3270 display which is being used to edit a file in visual mode the message will not appear until the user leaves visual mode If a BROADCST is given which overrides a previous BROADCST and a termina
203. Press the ASSGN CONS key The ES3090 300J will display the Console Assignment frame IOCDS INSTALLATION 397 19 Select option A3 Program Mode and press ENTER to deselect the program mode console 20 Press the SWAP CONS key one or more times until the legend at the bottom of the screen offers the system mode console Then press END The system console should appear 21 Enter the command CONFIG POR IOCDS sx WP sx and press ENTER where sx is the name of the IOCDS you just built i e AOQ A5 or BO B5 The console alarm will ring the screen may tell you that the POR option is unnecessary simply press ENTER This step will take about 5 minutes When it finishes the ES3090 300J will have been rebooted with the new IOCDS and the new IOCDS will be write protected 22 Reload UM MTS or VM and its guests and proceed with normal operations IOCDS INSTALLATION 398 LIMITED SERVICE STATE LSS This Writeup describing Limited Service State can be seen by COPYING the public file LSS The Writeup contained in the file LSS follows This writeup was last changed on 20 May 1976 when the Computing Center policy on the use of Limited Service State LSS was changed so that the institution of LSS is done without regard to the length of the batch execution queue Formerly LSS was only instituted when the batch queue exceeded 300 jobs Concurrent with this change 12 more dial up terminal ports h
204. Purpose Prototype Parameters JOBS PROGRAMS 103 LSTATUS 01B00005 LSTATUS 27 5 both of the above commands will check the lock status of page 5 on volume 27 Generate a request of disk manager via DISKIO supervisor call REQUEST lt opcd gt lt ea gt lt vma gt OK lt opcd gt is disk manager operation code in character form as defined in DMGR DM MACROS in DRCB copy section lt ea gt is the external address of the page the request is for lt ea gt may be either a quoted hex string or lt volume gt lt page gt Where lt volume gt and lt page gt are integers lt vma gt is a number between 1 and 5 specifying which of the 5 page buffers maintained by this program should be read into or written from lt vma gt may be ommitted if it is not needed but an error message will be issued if it is not present for a request which needs it Description A disk manager request control block is constructed with opcode lt opcd gt The external address field of the request control block is filled in with lt ea gt The virtual memory address field is filled in with the address of the buffer whose number is lt vma gt If lt vma gt is zero the request is checked to make sure it JOB PROGRAMS DMGRSTAT 104 Examples SIGNAL Purpose Prototype Parameters JOB PROGRAMS DMGRSTAT JOB PROGRAMS does not perform I O operations The virtual memory address field is the
205. R 452 TYPE 3350 JOB 3 DEVICE INITIALIZED NO ERRORS DEVICE STATUS CHANNEL END DEVICE END DEVICE INITIALIZED TASK WAITING NO ERRORS I O QUEUE ACTIVE I O OP AWAITING CHANNEL END gt CU gt DEV D115 ADDR 45F TYPE 3350 JOB 3 DEVICE INITIALIZED NO ERRORS DEVICE STATUS CHANNEL END DEVICE END DESCRITIONS OF THE VARIOUS MESSAGES PRINTTED BY UNITS CH lt ch gt The the the The repl the gt CU lt cu gt gt DEV lt dname gt ADDR lt path gt TYPE lt dtype gt string OFF may appear following the channel number control unit number or the device name to indicate that channel control unit or device is logically OFFLINE arrow between the channel aced with an asterisk channel to the control and the control unit may be to indicate that the path from unit is logically OFFLINE It isn t possible to logically OFFLINE the path from a control unit JOB DEV to a device lt j ob gt FREE NOT READY IGNORING INITS JOB PROGRAMS UNITS OLD STYLE Device allocated Device not allocated Task waiting for SPUR DEV END SVC SPWAIT or SVC ATNSU On while device is being freed SENSE HOLD CHAN BUSY CU BUSY FAKE CUE PENDING lt n gt CH OPS lt n gt CU OPS CLRIO FAILED DEV INITIALIZED NO ERRORS INOP CHANERROR UNIT CHK UNIT EXCEPTION I O QUEUE ACT
206. REP will prompt on SERCOM and GUSER for EREP parameters after CNVTEREP has been run if necessary to convert CMDSTAT to MVS history tape format CNVTEREP still prompts for a starting and ending time but this isn t really necessary any more since EREP now accepts a time range as well as a date range Some useful parameter combinations you can use at this prompt are Tape errors THRESHOLD 1 1 Disk errors machine checks and channel checks TYPE MCOT DEV 33XX Example Use upper case run erep sercom ser you don t usually need to see SERCOM sprint x use sprint to direct output to a temporary file etc if you don t want it to go to PRINT 8 00 JUN 20 8 00 JUN 23 ok TYPE MC picks up machine checks and channel errors or TYPE MC PRINT PS if a more detailed printout is desired or TYPE 0T THRESHOLD 1 1 one way of picking up tape errors or NOTE For the time being whenever EREP is used to pick up disk errors use EREP OLD TYPE 0T DEV 33XX picks up disk errors or TYPE OT CUA BXX FXX picks up errors on channels B amp F EREP SYS 320 or TYPE OT VOLID MTS001 MTS004 picks up errors on two disks mentioned or TYPE OT VOLID MTS001 MTS004 PRINT PS same as last parameter but produces expanded printout ENDFILE Remember that SPRINT defaults to PRINT If you must go back more than a couple of weeks or so SCARDS INIT MCFILE or INIT UC3330L0G must
207. RET581E RECORDING RETRIEVAL DISK FULL USER user Message HCPCRC583I RECORDING recording name userid USER NOT RETRIEVING Response These messages indicate that VM XA isn t properly storing system statistics records You should inform a firefighter that this message has been printed This isn t a serious error so walt if no one s around Shutting VM XA down There are two things to keep in mind when scheduling a shutdown of VM XA First VM XA can not be shut down until after UB MTS and DOS VSE are shut down Therefore you must always schedule UB MTS and DOS VSE shutdowns as well as a VM XA shutdown Secondly there may be users on the VM XA system There will probably be Computing Center programmers but you need to give them some advance notice of the VM XA shutdown To shut VM XA down perform the following steps 1 As soon as the shutdown is scheduled put in a signon message called a logmsg in VM XA terms on VM XA To do this type the following command at the VM XA Operator console UB MTS UNDER VM 48 SYSTEM SHUTDOWN AND START UP PROCEDURES WNG All message message where message is the message that you want to be printed For example WNG All message VM XA shutdown at 6pm back up at 7 2 As time approaches the scheduled shutdown send broadcasts warnings in VM XA terms to the users signed on to VM XA To do this type the following command at the VM XA operator console WNG All messa
208. RTUP has tasks none startup ed by the SHUTDOWN command must be re enabled manually meaning only to MTS terminal JOB PROGRAMS STARTUP 144 STAT NAME FUNCTION USAGE JOB PROGRAMS STAT JOB PROGRAMS To collect statistics of the operating system and to record them on tape STAT tapel tape2 This job should be run under the supervision of a systems programmer Data collection proceeds on a tape given as a parameter Persons wishing to use this facility should refer to the University of Michigan Computer Center Memo entitled The MTS Data Collection Facility for further information about this job program JOBS PROGRAMS 145 STATSW NAME STATSW FUNCTION To dynamically change the parameters that a STAT job is using USAGE STATSW As for the STAT job potential users are referred to the University of Michigan Computer Center Memo entitled The MTS Data Collection Facility This job program reads commands from the operators console COMMAND SYNTAX 1 JOBS ALL MTS jobnolist ITEMS ALL itemlist 2 JOBS fALL MTS jobnolist 3 JOBS ALL MTS jobnolist ITEMS ALL itemlist 4 INTS ALL intlistj CHS CHANS ALL chanlist 5 CHS CHANS ALL chanlist where joblist is a list of UMMPS job numbers separated by blanks itemlist is a list of stat job item numbers separated by blanks chanlist is a list of channel numbers sepa
209. S 00 HOLDEX 00 DRAIN 00 WAYT 00 ERRDUMP FF HASPERR 00000000 ERRSAVE 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 IDLELMT 3600 LASTTIME 98618B09 CURNTCLK WCONBYTE 96 WQUEBUSY 00237DD8 WSPOOLQ 00BB0068 00C52B30 00000000 00000000 Since QUEFREE 0 there are no free WTO buffers console DCT to see what s happening d dct type remote_ console DCT at 1E5B78 RMT CON DCTSTAT INUSE DCTPCE DCTPSTAT SINON DCTDCB 24FB18 DCTBUFCT 0 DCTPUNCH 00 DCTPRINC 0 DCTPRLIM DCTCHARS Mx Uc Pn 00000040 DCTPRLEN 126 00000000 00000000 00000000 00000000 00000000 00000000 98618BD6 2COFEEO0 00000000 Display the remote 246948 DCTBUFAD 001ECD40 DCTNO DCTDEVTP REMOTE_CONSOLE DCTPCODE CON MRF CPU TRSP DCTPRINT 00 0F DCTFORMS 23 0 DCTPLINE HARD P20 TRSP BSC DCTEWF 00000000 DCTFLAGS 0 DCTCHAIN 000000 DCTDEVN RMT CON DCTWAYT D9D4F2F9 DCTSINO 19200 DCTLOCK 00 DCTPRNTR 0 DCTTOTPG 0 DCTTOTJB 0 DCTRESCN 00 MDCTRCB 92 MDCTDCT 000000 RM PRIORITIES CLASSES SCHEDULE 382 The important field here is MDCTRCB The record control byte has the value 92 which means the remote console is doing a read If it was 91 then a write to a remote console would be in effect The next thing is to see what the associated PCE is doing The remote console PCE is available via the external symbol MCONPCE It is also saved in the field DCTPCE above
210. SAGE NOTE OPERATOR FILES STI OPERATOR FILES STI This program checks to see whether the statistics work file is proper order If the system crashed during the running of STA some repairs may have to be made These repairs are done automatically mts sti This program STI is automatically run at IPL time and should almost never have to be initiated manually OPERATOR FILES 293 STM NAME STM FUNCTION To monitor the number of lines in STATISTICS during unattended mode and to empty STATISTICS when it nears the number of lines that would cause it to be considered full by MTS USAGE This procedure may be invoked when placing the system in unattended mode It will continue to run until the task is stopped or until the system is reloaded However allowing it to run when the system is not in unattended mode should not cause a problem since it will simply wake up once an hour determine that the system is not in unattended mode and then go back to sleep To invoke the procedure enter MTS STM on the operators console This signs on under the ID SYS and sources to the file SYS STATMONITOR which executes the macro that performs the procedure At the time the procedure is invoked and once an hour thereafter the procedure will check to see if the system is in unattended mode If the system is in unattended mode the procedure will determine if there are more than 6500 lines in
211. SEG2 MONITORS to change the Messagesystem SEND command recipient list to include your name s This command is used automatically to send you a message whenever the system map has changed after a reload so you know that new maps should be printed Three copies of the system map residing in the fileSEG2 S2MAP should be mad and placed in the machine room on the third floor near M Alexander s office and one sent to Huron Towers A copy of each old map should be saved on the old maps shelf in the systems room DWB s old offlice on the third floor Put a note above the operators console informing the operators that you are the first to be contacted in the event of systems trouble Include phone numbers Put a bulletin into MTS COUNSEL announcing that you are now the current fire fighter s and that questions about MTS and the system should be directed to you first FIRE FIGHTERS GUIDE 370 3 HANDLING JOB DUMPS If you are in charge of handling job dumps you most important resonsibility is to make sure that all of the dumps get looked at The dump log book contains a complete history of the dumps that have been taken it should be checked frequently Make certain that it is kept up to date This log book is kept in the second floor machine room near the dump tapes Dumps of non MTS jobs should be routed to the person in charge of the particular component in question For instance M Alexander gets supervisor
212. SER MTS02832 gt gt gt ERROR WRITING PAT OR DSCBS CALL SYSTEMS PROGRAMMER IMMEDIATELY User Explanation When a disk error occurs while extremely sensitive data concerning disk storage allocation is being written this message ensues Under NO circumstances should the message be ignored as it may result in disaster if the file routines use possibly incorrect data resulting from the hardware error Call a systems programmer immediately DESTINATION OPER SOURCE FILE SYSTEM 482 19 GETDSK MISCELLANEOUS MESSAGES 407 gt gt gt FDGTPD ERROR User Explanation System error call an MTS systems programmer System Explanation The FDGET2 pushdown stack underflowed DESTINATION OPER TEXTLOC FSUB 635 HOPPDMSG HOPENIT FSUB0883 SOURCE FSUB 635 HOPEOOPS HOPENIT FSUB0879 gt gt gt FREE RUN nnnnnn xxxx User Explanation The batch job with recipt number nnnnnn and user id xxxx is being signed off without being charged System Explanation For example due to the operator s answering a mount request with RER rerun DESTINATION OPER TEXTLOC MTS 42 TSAFREE TSA MTS06263 SOURCE MTS 42 TSAFREE TSA MTS06269 gt gt gt FREESPACE MOUSE TRAP DUMP User Explanation System error see an MTS systems programmer System Explanation A system component in system storage has issued an invalid FREESPAC request Sorry about that A job dump is given DESTINATION OPER TEXTL
213. SP Priority Schedule cias da 6 dS ens A ere ee dw dee eS 374 Transmitters Receivers HASP Stuck ooooooooooooooooooooo 384 Host Interface Machines HIMS mc rs a a coa 387 Computing Center HOUR cada A a 389 Shurtdowns Scheduled ai te we bbs 389 Holiday Sign Programa aeea A Oa Site Ge hoe BOS ee Dad Stee 390 TOCDS Tnsital Lataon ii A A ba id ae oe eh os 395 Limited Service States SS ei aeni en a Biel Subs aes 398 UMnet Hermes System ns ia a AS Swen a ca a wea S 400 Setda ter Mesias ae o a dd ad E E 401 Messages Miscellaneous Operator o ooooooooo ooo ooo nooo 404 Mylar Ribbon COPTES ss paide ad e A a is 415 ORDS EA a what ates E A at Ma ata 416 Plotting SY SEM is A A A a isis 420 COR stock o A A A a date dt 421 Tame Frames FOr Plot Tapes iii dis 421 User Tapes CCP NDA AS AA A AS A A 422 EP A A AS E IR A a 423 Unusial Conditions PLOT a ove eee 427 Plot Processor Error Messages oooooooooooooooooooooooo 427 DEPSE ERCP TOTS chs a a e a a fod 2 428 ECO Sosa are eG WE eI Ea Bae A A A Bee ele Saas Dae ees DR OS 428 DEFErEING PLOES ooie 8 Rid Shs id Seas 429 Regular PLOGS A O aie ehe aS is 429 Regular PIrotS odeia a Ea E E A eBay 8 SORES se cds a 429 Deterred PHOES RUNNING at a a e 429 Deferred Plots PLOEEIAG ysis see ele eae As 429 COCs 22 CInrtradZes Plot Tapes tis a ete ee ee at eee deal hae 430 CCT Determines Length Of Plot Session
214. SSD 361 Errors 3805 3825 361 ERROR WRITING DSCBS 406 ERROR WRITING PAT 406 ES3090 300J Console As Operator Console 346 ES3090 300J Power Up Down 24 External Damage 311 Fake I o 159 FHF Error Codes 361 FHF Unit Checks 317 File Charge Update 368 FILE DASDI 354 FILE DISKCOPY 355 FILE DS O 211 FILEINFO 358 File Restore 270 280 Files Permanent 19 Private 19 Private Temporary 19 Filesave Daily 283 File Save Merge 217 Filesave Schedule And Tape 567 Strorage 333 Filesave Tape Copies 215 Filesave UNIX 557 Filesave Weekly 297 File Space Retrieval Info 358 File Validation 201 FIO 475 Firefighter Instructions CMDSTAT DWB 328 Firefighters 468 Firefighters System Reliability 468 Fire Fighters Guide DMP 370 372 DUMP DUMP CMDS 372 DUMP DUMPER 371 372 DUMP LOG 372 DUMP TAPES 370 372 Job Description 369 Job Dumps Handling 370 MTS COUNSEL 369 MTS PRINTDUMP 370 371 SEG2 MONITORS 369 SEG2 S2MAP 369 Tour Of Duty 369 FIX 92 213 309 Fixed Head File Errors 361 FLB 214 Floppy Disks Mailing 350 Floppy Disk Submission Retrieval By Mail 500 FLUSH 108 Forms Special 547 FREE RUN N X 407 FREESPACE MOUSETRAP DUMP 407 FSC 215 FSM 217 FTGTPD ERROR 407 GOOSE 110 Graphics Usage Summary 227 GRID 441 442 G11 60 HARDWARE ERRORS 310 HASP Carriage Control Tape 326 HASP Hung 377 HASP Log Collection 221 HASP Priorities 374 HASP Priority Schedule LOW DEFER MINIMUM 374 HASP Status
215. STATISTICS If there are more than 6500 lines the procedure will expand SYS STATISTICS if necessary to a size sufficient for adding the contents of STATISTICS copy STATISTICS to the end of SYS STATISTICS and empty STATISTICS Once this procedure has saved STATISTICS the next time STA is invoked it will issue the message LAST SAVE OF STATISTICS APPEARS UNSUCCESSFUL OR STATISTICS WAS DUMPED DURING UNATTENDED MODE SHOULD THIS BE REPAIRED AND COPIED TO TAPE ENTER Y OR N This is because the result of saving STATISTICS with this procedure appears to STA to be the same as does an unsuccessful save The above message should be answered with Y This will OPERATOR FILES STM 294 OPERATOR FILES cause a repair operation to be performed and when it is successfully completed the message kk N DO YOU WISH TO SAVE CURRENT STATISTICS FILE ENTER Y OR will appear Assuming that the original reason for this invocation of STA was to save the current STATISTICS file this should also be answered with Y If it is not or no longer necessary to save the current STATISTICS file this should be answered with N If there is a problem with the STM procedure e g if SYS STATISTICS cannot be expanded sufficiently SYS STATCOPY will be run saving STATISTICS in a file with a name of the form SYS STATYmmmddi e g STATYAUG17 or STATYAUG17A Once a file bel
216. STEM SHUTDOWN AND START UP PROCEDURES is nothing else to do as part of loading VSE1 and VSE2 A large number of additional messages will be printed by both VSE s The final messages should be F4 004 DFH1500 CONTROL IS BEING GIVEN TO CICS This should appear on each VSE It takes 5 or 10 minutes for VSE1 and VSE2 to load The F4 pertains to Mirlyn the Library s on line catalog system On VSE1 you may see the message F2 025 DFH1500 CONTROL IS BEING GIVEN TO CICS The number immediately following the F2 may not always be 025 as in the example But the F2 means that the part of VSE1 used by programmers is up it does not mean that Mirlyn is up If you miss a particular message on the console you can look back by typing the letter D and then hitting ENTER Each time you hit ENTER the previous screen of console messages will be displayed You can get out of console redisplay mode by typing D E and then hitting ENTER To check the status of VSE1 or VSE2 from the network terminal at the which host prompt type hUbc1 VSE1 status or hUbc6 VSE2 status Shutting VSE2 Down Currently we have promised the library to keep VSE available from 7 AM to 4 AM weekdays so we need to avoid scheduling VSE reloads during that time 1 2 VSE Go to the VSE2 operator s console Type MSG F4 and press ENTER VSE should reply MSG F4AR 015 11401 READY F4 004 The 004 part
217. Seas 66 The ENTERED 3825 ai A A A glories Sees 72 The TPEREADER a rand inane a ad da ia abe gues 77 RAMROD S iia 77 ENTRODUETTON a Ns 78 APTS REVAL A A A AAA a SE AA A E as 78 TPLUREAD 0 nos a a al a a la 78 TPEENTD a A TA a IL ASES A ds 78 TPEBOO DTi A olen Sed A E AA A A A sents 78 TPE ERROR MESSAGES oeart vs iria A A aa 79 What The IPL S quence Did id A atte 84 Power Fluctuations Failures Recovery oooooooooooooooooo 85 Amdahl Disks lf re95 We A Sa ERO Stee aS BOS wea Een gs Boab eb a cabs 86 JOH Programs A Scales SF rete MDa tae Rina E dey SU aaa ee Talked SHEP Saw 91 BASTA A A ma RO Mt 92 BROAD CSG 1d A A Site dat etait ove A A A Ai 94 DESEA A A A A atone AAA A AR A AA See hd eed 97 DMGRSTAT A Sore ed stags AAA A A ts a La 98 ADD EI A A A AS E A AA A A e 98 BLAST A AAA Ad E A IA A EI S 98 NODMGR a ds a A Beds A e o emia Bees Se bela iS me aia 99 DESPLDA a ts sl 00 FORGET A daa 01 HEREDA A A A A gas Tala gales A A a 01 ESTATUS a A de a a A dd 102 REQUEST A td 103 SIGNADA E A Sa eS Ree ooo A jara 104 STATES TEGS 20 ts A A ee A ee a we ai L05 STENONM ia AS A WIFE NE BOL a th aise ce A A a re nae ee a L SOBPTCHK 61d A A A a a L STARTUPS 29h alos ario o td a elo ae a Ne EE ER ANRE AA had wee Ne wen A S als L SPAN 0 E E e e E A A E aula dea hea A a e E aaa L STATSWS yaaa E SP ana E a dina le a a oes el L STOP aopa aa Ea AET Sc enable O A ERE ESAT A TA E A A o L TABIMOD dt aie acta an ANE A A
218. Soe HAS P SELATUS Scene e gee Sabie Mts Seah Sele Reka Sabet Sr cade a aa eos Skee tea sen de 223 HUH UMnet Concentrators State Of o oooooooooooonoooooo o 224 IG SHUFFLE Summarizes IG Usage ee ce ee ee ee ee eee 227 ENT System PG a lization ses a A a A 228 15 TTX Intertask Communication Test UMMPS Jobs oooooooooo o 229 ELAR Reset Line Adapters L2 70 ci a a 230 FLAS Start 1270 And Mis JODS om eis rss es 231 ABH se Tape Dabaling arranda ei a a a aia W a 232 TiBL 4 Tape Labeling Viana tele a id ad a id 233 234 TUM Time Remaanaing Batch JODE ni a whey de ave diia 235 ELMT 2 SROrE Vers ton OF MSS pre dice A A urea gees ite we Bash BONS oars 236 DRE ss Info About Last IPE ti a 237 MNS Normal MERIT Startup vredi a as WAR a a as 238 MSDs Set Tame Date UMMNEE COS ii A E OR EK ow Sade ees 239 MSG Cleans Up Message System o oooooooooooooooo oo et ee eens 240 MSS eo Tape Usage History vico w ew SS eS eee eee Ee 241 MUP Current State Off MERIT Computer oo ooooooooooooooooo 250 NDL Dumps amp Reloads Primary Communications Processor AN A sah SSE hati Sey ated gs adi a a aceite m 258 NST 2 DeadtiMes MESSAGES visits Heke ne HRS el ee eee oboe a ENE S 255 NST NAS Manipulation Of Name Address Spaces oooooooooooo 257 DOT Controls NIM On UB amp UM eera ars a e ala os A O 260 PAGS tests P
219. TABLMOD NAME FUNCTION USAGE JOBS PROGRAMS 147 TABLMOD To display or modify information in the shared file table Takes no parameters Reads input lines until it gets a Cncl Commands available are STATUS Prints out 1 How many files are currently in the shared file table 2 How many matrix computations have been performed since the last reload 3 How many multiple file deadlocks have been detected since the last reload and 4 How many jobs are currently waiting on locked files HOGL Invoked to list large users of the incore file table VERIFY Same as STATUS plus checks the consistancy of the shared file table VERIFY also prints out the number of available entries and elements TRACE Same as VERIFY plus prints out the entire shared file table file by file Each file is listed on a separate line in the form fname entryl entry2 where entry is of the form jobnum ostatus lstatus wstatus where jobnum is the job number ostatus is NOTO OPEN INVLD JOB PROGRAMS TABLMOD 148 JOB PROGRAMS lstatus is NOTL LOCKR LOCKM LOCKD wstatus is WAITO WAITR WAITM WAITD Ex WO45BIGLINE 0003 OPEN LOCKM LSTAT FILE fname prints info as above for single file LSTAT JOB jobnum prints info as above for single job LOCATE FILE fname displays half word offset of file into ICTAB NOTE Don t use CLEAN CLEANF DEQ or DEQF without permission from a senior
220. TALOG SUBR gt gt gt DEVICE BUSY OR NOT AVAILABLE User Explanation This is the system message given to the operator when the operator in response to a mount request specifies a drive which is already in use or is off line DESTINATION OPER TEXTLOC MOUNT 101 DEVBY MOUNTCMD MNT02808 SOURCE MOUNT 101 5T234 MOUNTCMD MNT01685 MISCELLANEOUS MESSAGES 406 gt gt gt DEVICE TYPE NOT ALLOWED User Explanation See an MTS systems programmer System Explanation When starting up an MTS job at the operators console either a device name representing a type of device not supported by MTS was specified to be used as MSOURCE or MSINK or PUNCH output if a batch job was started or else the device specified for MSOURCE was not an input device error return from CFID or the device specified for MSINK or PUNCH was not an output device error return from CFOD DESTINATION OPER TEXTLOC MTS 42 GRRM2 GRRR2 MTS02679 SOURCE MTS 42 OUTNBY GRRR2 MTS02673 SOURCE MTS 42 GTMSOU START MTS00254 SOURCE MTS 42 INITCP START MTS00304 SOURCE MTS 42 INIT67 START MTS00326 gt gt gt DSECT CLOBBERED User Explanation System error see an MTS systems programmer System Explanation TSA thinks that the MTS job is a free run via operator request but the job is not a HASP job This preceeds a Snark message DESTINATION OPER TEXTLOC MTS 42 CHKSM CHKSER MTS02834 SOURCE MTS 42 CHK
221. TIME Time of Mount PROJ Project Number MODE Bits Per Inch MSSCN From a Terminal Heading Explanations FLAGS Description of Tape was labeled had a ring in it e g Where there is a there i s a digit or All instances of that particular position will be printed typing six s would be the same as typing ALL Typing cl2 would give a list of all tapes For instance DATE Date of Mount USER Us er Id JOB Supervisor Number TYPE Type of device tape was mounted on UNIT The particular device that the tape was mounted on STAT Operator Response etc whether or not the tape Msscn can be run from a terminal as well as from the operators OPERATOR FILES MSS OPERATOR FILES 243 console Its proper name is sys msscn and scards is usually sys mountstat although it can be a tape either s0820 or s1091 If scards a tape which at this writing would be either s0820 or s1091 you will want to position the tape according to the dates that you are concerned with Firs mount the tape and then cont j posn eot After you have done this then cont 3 bsr 9999 You will get an error indication the first time that you back space but just do it again After you backed up the tape as shown above a cuople of times you ll want to see where you are on the tape To do this run msscn scards j 1 3 all The above action will print out three lines containing the date and time as well as
222. TROL The printer has a twelve channel carriage control tape which moves synchronously with the paper For each line on the page there is a corresponding line on the carriage control tape which may have a hole punched in one or more of the twelve channels Thus a printer command such as skip to channel 1 has the effect of moving the tape and thus the paper also until a hole is sensed in channel 1 of the carriage control tape For the tape used by MTS this has the effect of postioning the paper at the fourth physical line on the page MACHINE CARRIAGE CONTROL Machine carriage control acts at a lower level than logical carriage control and in fact logical carriage control characters must be converted to machine carriage control characters not a straight translation mind you by system routines The first byte i e character of the output record to be printed is interpreted as the command portion of a CCW channel command word for the printer The reader is directed to IBM form X20 1703 the green card for a description of the legal machine carriage control codes Unlike logical carriage control machine carriage control always spaces after printing rather than before printing CARRIAGE CONTROL MTS LOGICAL MACHINE 326 CARRIAGE CONTROL TAPE FOR HASP Col Punch Col Punch 3 3 69 3 4 LI 70 11 7 1 73 1 8 5 74 5 TI 10 77 10 17 6 83 6 22 4 88 4 27 7 93 7 37 2 103 2 47 6 113 6 52 4 118
223. TS This should only be done after powering up the ES3090 300J or after shutting UM MTS down To load UM MTS perform the following steps 1 Go to the ES3090 300J console for the UM side of the 3090 2 Press the INDEX key 3 Select the OPRCTL frame by typing 9 and pressing ENTER 4 Make sure that the load address on the OPRCTL frame is correct The load address should be given on a note near the UM MTS operator s console If the load address is incorrect a type L1 and press ENTER Your cursor will move to the LOAD UNIT ADDR field b Enter the load address and press ENTER 5 Start the load command by typing 04 and pressing enter That s the letter oh not the number zero 6 After a moment the message Load Complete should UM MTS ON 3090 SYSTEM SHUTDOWN AND START UP PROCEDURES 27 appear on the ES3090 300J console 7 A minute or so later the message Do You want to run the current system should appear on the UM MTS Operator console 8 See the section Loading MTS to proceed from here Setting the TOD clock If you change the setting of the TOD clock during a reload see the section Loading MTS UM MTS may tell you to Enable the TOD clock To do this 1 Go to the ES3090 300J console for the UM side of the 3090 2 Press the alt key and while holding it down the TOD key Under some circumstances you may need to enable the TOD clock twice If UM MTS prompts you a second time fo
224. UIT NO will print this message to the operator if the requested tape cannot be positioned as the user specified either because the specified data set name did not exist or the file number specified was greater than the number of files on the tape The tape will be unloaded and the job terminated DESTINATION OPER TEXTLOC MOUNT 101 POSNCOM MOUNTCMD MNT02430 SOURCE MOUNT 101 ST174 ST225 MOUNTCMD MNT01879 MNT01882 MISCELLANEOUS MESSAGES 412 gt gt gt RATE INITIALIZATION ERROR User Explanation System error see an MTS systems programmer System Explanation The rate vector RATEVEC could not be initialized by GETRATES from RATEFILE during MTS task initialization DESTINATION OPER TEXTLOC MTS 42 INITERR MTS00357 SOURCE MTS 42 INITERR MTS00358 gt gt gt REMOUNT TAPE WITH RING how User Explanation This is the system message given to the operator when the operator mounts a magnetic tape with the file project ring improperly placed e g OUT instead of IN how will be either OUT or IN DESTINATION OPER TEXTLOC MOUNT 101 ST113 MOUNTCMD MNT01770 SOURCE MOUNT 101 ST113 MOUNTCMD MNT01770 gt gt gt SIGNOFF QUIT User Explanation System error see an MTS systems programmer System Explanation This message occurs if ICDEQ in core file table routine gives an error return through DEQIT to SYGNE while a signoff job cleanup is in progress An MTS SNARK fol
225. UMN es LS A A A A a a Suara 400 Holiday SID Program ii id iaa 390 INTED 3O10 5 s it a A A A A a Be ed wee bi 66 TOCDS ANS Cad Vae OR ni A did 395 TiMmtted Service State HSS SA AS ees 398 LogoutsrxPrinting Ofsusing IBM EREP ee ee ee ee ee eee 318 MaintenanGe AGGECMENES arta a OEMS eT StS ws ae a whee 533 Memorex Line HUNG es SSS SSN A Ba WS GS RS ERS RES 322 M ssages MrseCedlaneoOus aif ee eS i kA Se a ee ee he a a aa 8 a 404 My kar Ribbon Copies vs ds fe ees A Sie eels ba Se ge Waele SER EAS A 415 NUBS Terminal Stations LAGOS sesser eels Id a FSS R 531 OLTS on A Diagnostic Program erster esante a A aa a ea ae ig wes a Ea 416 Pager NASE oE A A E E EEA oe etna sd E T o ah 347 PDP Restarting Also See JOBS PDP amp PGET ooooooooooooooo 75 PLB displays Identifying Information About Plot Tapes 432 Plot Processor Error Messages ssn ete bee eS eRe eke We ee Wl tee Bee 427 Plot System Problems 3 3 525 sesso edge ii laid Ween ao ha aoe eet ee E 427 Plot Tap tdentritying ENtOTMAtION di a a eee nh eS ae 420 Plot Time FLAMES A Wee bee ed sta ei eee SR e sees 421 PLOPLANG SYSTEM she has ae he a iaa atlas Saas See Behe cs laa 420 PM2 STC ETror Analys iS os eters Soe ee eid ae Wa hes leh eee BPE ES Se a aes I eS 446 Printing Logable Unit CHECKS rin ai she wate Anes Sheena Bobi Sb en ehe RS 316 Rates Special Holiday voii A Heeb Scale HAS OSS Beals ES es 451 Recoreded System Sta
226. UNIVERSITY OF MICHIGAN COMPUTING CENTER MTS OPERATORS MANUAL Revised Feb 21 1995 NOTE Revision bars will be used to point out information which has been entered changed or deleted at the last two or three printings of this manual For additions or corrections contact Stephen M Donnelly Contents Introduction TORMES seat ER le ices AA we ROLE sane AAA ee wa Reed ad 17 Overview OF The System secs isn e eed A ew is Westone Stee ace die a ees 18 UMMES ze these A A rele A Mey oh civ at RIN Ge ete Rds SiR ele idee RN Ee Oe 18 The Job Program And TASK si jolene eS OR A A A A wend 21 console tn MALI narse A A A ee Sire Shs cosets Se 22 Super DUMP Se Er Ea SENS a E ee Nw BEINN SEND We Se ae SRE BEES 23 Power Up Down ES30907 3000 LA At ek A 24 UM MTS Running Of On 3090 Native eee ee ee ee ees 26 UBE lt MIS Under VM gt ond nure a She Site Sd eevee Side yk BS 29 HOAGING UB MTS AA bie wa es ee or e 32 Loading MTSE iors arse Soles bs ee ete Cerda teers aie eat le ats a dd 39 VS E AON The SOTO aaa hee le ada da 50 Operator AOD Sede yes tt ses as tacos 61 HAG Pip disc ice he ati A tal eee A A cde ei ae Dh Me cde A A A eG AA RD he 61 System Shutdown Power Down Procedures cee cece eee eee eee 62 EMERGENCY POWER OFF MEASURES o ooooooooooooooooooooooooo 65 Virtual Memory inn il A Soe Set el thn hea SE Bae eS 66 Th SEN TEED SOO 5 A a wud oe tole
227. Unit one should see a black dial type knob REMOTE LOCAL and just below it four metal toggle switches Normally the only one of these toggle switches which should be in the OFF position while the system ig running is the upper right one marked RERUN The Amdahl disks should be in LOCAL mode so they must be powered up manually First make sure that green tipped individual power switches inside the back panels of each disk module are switched off In the Control Unit the upper left toggle switch just under the REMOTE LOCAL dial switch is the power switch If the Amdahl disks are in REMOTE they should come up when the system is powered up but again remember to turn off the individual green switches before powering the system up then turn the green switches on before the system is loaded Control Unit Displays At the end of the CUs are a series of switches as well as two displays one for each channel Normally these displays should each show 00 If one or the other isn t displaying 00 open the end panel of the CU and lift the appropriate red tipped toggle switch marked IMPL However before IMPLing all disk activity should have stopped and POWER FLUCTUATIONS amp FAILURES RECOVERY 88 SYSTEM SHUTDOWN AND START UP PROCEDURES either the sytem should be down or the Amdahl disks should be logically offline Each side takes about one minute for an IMPL to complete If these switches are lifted at
228. VS may be used SVS is just about the same as SVW except that when it begins it will ask what series of filesave tapes to use After SVS prompts you type and enter the series number e g 1100 followed by a null line NOTE See TSN NOTE If the message FILE CCID FILENAME DSCB E NOT FOUND AT XXXXXXXXX appears notify a firefighter but don t wake anybody up to do it OPERATOR FILES SVW TNR NAME FUNCTION USAGE OPERATOR FILES 299 TNR Produce a ripple pattern on a printer with a TN train or chain mounted Drain the printer Enter mts tnr ptrn where ptrn is the name of the printer To stop the rippling use the GOOSE or STOP jobs on the task OPERATOR FILES TNR 300 TPD NAME FUNCTION USAGE OPERATOR FILES TPD By default TPD tests 9 track drives in the following way First it tests the drive at 800 BPI by writing a tape mark followed by a random length block consisting of all possible bit combinations It then does a backspace file backing over the block and tape mark followed by a forward space file moving past the tape mark and reads and checks the random length data block If the data checks correctly TPD repeats the entire sequence without rewinding the tape 300 times by default with the MTS read data check recovery disabled RETRY 0 i e any data check will show as a permanent error If no errors are found after the
229. WC 266 OPERATOR FILES sesos Priv jobs waiting for a page in secas Neutral jobs waiting for a page in NAAA SE Total jobs waiting for a page in DASD wait__ To Io Gu Fi l Jobs waiting to lock a file Jobs waiting to obtain a disk unit SVC GTUNIT liiess Jobs waiting for disk I O with or without DMGR is Total jobs waiting for disk or files 3270 wt_ To lo Gu 3270 jobs waiting to obtain the device GRAB FLIP Jobs waiting for I O to a 3270 display od Sree ies Total jobs waiting on 3270s Intertask__ To Ry By Wt Information on the use of UBC intertask communication at UM only the DMGRSTAT job uses this currently Other To Io Ti Unclassified jobs PWC doesn t know about new TASKSTAT codes Jobs in time wait asias Stet see Jobs waiting for I O to something other than a 3270 or a disk ad ido Total jobs in the other group MISC To Do Js Pk Gu Lk Op Hs Jobs waiting for HASP Pl al gt Jobs waiting to read or writ the console SVCs read and write j esas Jobs waiting to obtain a supervisor lock SVC LOCK O a Jobs waiting to obtain a unit other than a disk craneal Jobs in CMDSTAT wait SA A A Jobs waiting for the systemstatus master job SSRTN E ie ne Dormant jobs waiting for an interrupt i a de Total jobs in the MISC group OPERATOR JOBS PWC NAME FUCTION USAGE OPERATOR FILES
230. WILL BE PLOTTED NORMALLY CANCELLED This is of course exactly the program which defers plots It just changes the status of any id you give it as mts ccd ENTER SIGNON ID sys PLOTS FOR SYS WILL BE DEFERRED sys PLOTS FOR SYS WILL BE PROCESSED NORMALLY CANCELLED OPERATOR FILES CCD 184 CCL NAME FUNCTION USAGE REMARKS EXAMPLES OPERATOR FILES CCL OPERATOR FILES CCL This program initializes plot tapes for use by the program CCP mts ccl The program will print ENTER DRIVE DENSITY NAME Type in the name of the drive the density and the name of the plot tape separated by blanks The program will label the tape with the name PLOT and write a file used by CCP to identify the tape The same information may be entered for another plot tape Use an end of file to terminate the program The name of the plot tape must appear in exactly the right format e g MON 1000 but not MON 1000 The weekday names used are MON TUE WED THU FRI SAT and the times are 1000 1400 1800 2200 0200 The plot tape names WEEKEND and DEFERRED are also recognized t903 1600 fri 2200 t905 1400 SUN OPERATOR FILES 185 t914 1400 SAT OPERATOR FILES CCL 186 OPERATOR FILES CEP NAME CCP FUNCTION To produce a tape for the CALCOMP plotter from the plot requests that users have queued CCP should be run by the Operators at the
231. XAMPLE S SYSTEM START UP PROCEDURES IPLREADER SYSTEM SHUTDOWN AND START UP PROCEDURES 81 COMMAND DISPLAY SYNTAX DISPLAY ADDRESS ADDRESS ON ptr PURPOSE to display the memory locations specified If only one address is given then 4 bytes are printed starting with the given address By specifying two addresses a multiple of 4 bytes of memory may be printed By default the output device is assumed to be the console device The output may be directed to a printer by specifying the ON ptr parameter eg ON 008 An address is one of the following 1 a real hexadecimal address 2 the character string IPLREAD without the quotes 3 a hexadecimal virtual address eg V 100000 4 a symbolic virtual address eg V DUMJOB 5 any expression involving additions and or subtractions of the above eg V MAXPVN 2 Addresses of types 3 and 4 are only valid when a system is loaded EXAMPLE DISPLAY 1000 1060 EXAMPLE DISPLAY 1000 1060 ON 008 EXAMPLE DISP 105C EF 105C FF 4 EXAMPLE D IPLREAD 1D8 1E8 IPLREAD ON 8 EXAMPLE D V MAXPVN EXAMPLE D V DSKTAB V DSKTABE EXAMPLE D V 100000 V 100020 EXAMPLE D V OPERDEVS 4 is illegal EXAMPLE D V OPERDEVS 4 is legal COMMAND MODIFY SYNTAX MODIFY ADDRESS HEX STRING PURPOSE to modify memory locations starting at the address specified EXAMPLE MODIFY 1000 070007000700 0700 EXAMPLE M 9D80 BE 4770 E
232. XAMPLE MODIFY IPLREAD DDO 45E0AE4C 12FF EXAMPLE MOD V MAXPVN 00000001 Note if desired the hex string may be segmented with blanks and or a comma appearing between the segments However each segment must contain an even number of digits SYSTEM START UP PROCEDURES IPLREADER 82 SYSTEM SHUTDOWN AND START UP PROCEDURES LPSW PSW LEFT HALF PSW RIGHT HALF rather than keying in programs using the rather awkward facilities provided by the 168 console programs can be entered using the MODIFY command and then started using the LPSW command To get back to command mode press PSW RESTART LPSW 00000000 00001000 Note both the PSW left half and the PSW right half must be exactly 8 digits to enable or disable checks for bad PAT bytes and bad DSCB type bytes during VTOC searches Initially this switch is ON and checking is done Having CHECK ON is obviously an undesirable feature if the VTOC is damaged CALC ADDRESS EXPRESSION same as CALC command in MTS only simpler CALC V DSKTABE V DSKTAB Note the only valid operators are and Note this command also functions as a SYM REPLACE CSECT NAME FROM TAPE ADDRESS to replace a deck in a loaded system with another deck sort of like the RAMROD REPLACE command The usual context in which this command will be used is for replacing the TABLES csect ina loaded system with a DECKGEN d TABLES csect starter syste
233. Y NON APPLICABLE Security Clerk Computing Center Purpose and Duties NOTE At present we do not employ the services of Security ClerkS but when we do this is standard operating procedure for them Purpose The purpose of the Security Clerk position is that of having an employee available to Babysit the Computing Center during the time the Center is not open to the public but the Computer and the Air Conditioning units remain powered on The presence of the Security Clerk is to insure that the integrity of the Computing Center building and all of the equipment inside of the building espcially that of the the Computer and Air Conditioning units will be maintained Duties The Security Clerk will be stationed in the second floor machine room Every 1 2 hour the building should be inpected by the clerk The areas to be inspected include the the three machine rooms the two office areas on the second and third floors the user work area on the first floor and the Electronics shop At 2AM and at 6AM the Clerk should inspect the Boiler room 1012 M and the Generator room 1010 E These two rooms are located on the first floor at the north end of the user work area Keys to all of the areas to be inspected are on a large ring at the west end of the second floor machine room hanging just below the two black recording phones If any problems arise with which the Security Clerk is unable to contend
234. above plus those MTS tasks which are IDLE The format of the status message is nnnnn pgs name jobtbl status parl parn dev where mnnnn is the task number pgs is the number of virtual pages being used if it is a relocatable task name is the name of the command that initiated this task jobtbl is the real core address of the associated job table status a description of the job s current status parl parn are the parameters used to start the task devil devn are the devices that the task currently has For MTS terminal tasks the parameters are user i d and project i d For MTS batch tasks the parameters are user i d project i d and S 8 number For terminal tasks the first device parameter will be the terminal line device i d If the TASKS job program cannot find any job s of the specified type the message JOB NOT FOUND will be typed EXAMPLES tasks 5 gives information about task 5 JOB PROGRAMS TASKS 152 JOB PROGRAMS TASKS JOB PROGRAMS tasks u w025 gives information about the task s associated with user i d W025 tasks d d307 gives information about the task that currently is using the device with device i d d307 tasks t 9tp gives information about all jobs that are using nine track tapes which are of device type 9TP tasks n gives information about all non MTS tasks tasks n hasp ling gives information about all HASP
235. aging DevideSs tia o 261 PLB Prints Identifying Info About Plot Tape 262 Pl 25 BLarts 16 PAG JODS Merit A ies anes de eee 267 ROPA MES SS CALAS A A AS A a a Bad ANS KS 268 RES File Save Status Restoration ooooooooooooooooooo 270 RHB Releases amp Holds HASP Rate Classes o o oooooooooooooooo o 275 RIP sa s PN Ripple Print e A A A e A 8 274 RMC Lists Resource Manager ProCceSSOrs oooooooooo o ooo ooo 276 RMD Shuts Down drains Resource Manager oooooooooo ooo 277 RMP Starts Resource Manager ProceSSOrsS o o ooooooo ooo ooooooo o 278 RMS Starts Resource Manager sredi resa eee Se ele Saree bs aed CAMS oS amp 279 SRST Ss REStores FAVES ti lio seo Se a sa 280 ROV Jaa Reserves DUSK DELVES Vanmeer A a wR Be A es 282 SAV sse Daily File Save cusco ak oe ede a Sew Sans pia 283 SDB UB Shutdown Msg On PERS yo e ce veh ds a Swe eek aa ad 284 SDM PCP Msg Reis UM SHUECOWO ons ecu a a 286 SFT Starts SOFTCHK Checks For Soft MCHKS o oooooooooooo o 288 XOTA Saves Statistics On Tape recono rs du sm a a 289 STI Checks Condition Of Statistics Work File 292 STM Monitor No Lines In Statistics During UAM 293 SUM Remove Network Msg Installed By SDM 295 SVS Special Large Weekly Type Filesave ooooooooooooooo o 296 OVW s Weekly ELLO Save is seb kage SiS Sh Shed Saad Seah BA Gee by ada
236. agram cm m1 NN HP PI 1 mn 7 4 5 6 7 8 9 10 411 12 Loi Ly Leo Ley Let Ly Lt Ly 3 13 NORTH gt rm ro 111 2 Loly by lInterfaced through the Memorex 1270 NUBS TERMINAL STATIONS 532 BUSINESS ADMINISTRATION REMOTE SUB STATION BSAD COMPUTING CENTER Memorex 208B lt Channel 1270 e521 DataSet Etre lt Data AA asiata Data Concentrator Set A ene ee ee eee 4 spears Data Eo E eaegTRE Ss 8 Set E eae eas CR lt Memorex 1270 ROOM 371 24 DecWriters LA36s 12 Data LA36s MUX Set 12 a ee ts LA36s 38 Data ono 2 MUX Set ROOM 371 A DATA 100 208B Computer Data Set Phones 1 Hot Line 1 U M Line 763 4286 Data Set 763 4280 NUBS TERMINAL STATIONS 533 MAINTENANCE AGREEMENTS Update late 1989 Equipment Hours Days per Week Amdahl Disks 6am 5pm Hrs 5 Days M F CalComp Plotters 8am 5pm 5 Days M F Casper Systems Corp See Spacesaver Emergency Power Engineering UPS Contract renewed yearly P M every six months 24 Hrs 7 Days Graham Magnetics Time amp Materials Hewlett Packard Plotter 8am 5pm 5 Days M F IBM ES3090 300J 24 Hrs 7 Days IBM 3097 24 Hrs 7 Days IBM 3092 24 Hrs 7 Days IBM 3089 24 Hrs 7 Days I
237. aking a tape or while the PLAY button is pushed in When a user calls in the recorder will disconnect the lines after ninety seconds uunless the LONG button is pushed in in which case the recorder will terminate the call assuming the user hasn t hung up after three minutes At present this machine is located in the Operations office on the second floor RECORDED SYSTEM STATUS MESSAGE 471 SYSTEMSTATUS For Operators Contributed by S Gerstenberger A set of system status commands is available to provide information about the operation of the system and to allow an operator or systems programmer to modify certain parameters which control the operation of the system When a 3270 is being used as the operator s console any of the system status commands listed below may be entered directly the initial is required in this case If a 3270 is not being used the JOBS job should be used for entering system status commands the initial is optional in this case For Example 3270 USERS Q LCL non 3270 JOBS to start the JOBS job USERS input to JOBS job Q LCL input to JOBS job end of file to stop JOBS job Many of the system status commands listed below are also available to MTS users via the SYSTEMSTATUS command see MTS Volume 1 In addition all of the commands listed below are available to privileged MTS systems programmers via SYSTEMSTATUS Notice that certain parts of the co
238. aleady been written on tape then yuou can do the following goose quickplots type stop r sys plot36 copy 123456 quik STOP This command terminates PLOTQUIK If PLOTQUIK was started with MTS PLQ the job will stop PLOTTING SYSTEM 426 WAKE When PLOTQUIK is napping because it ran out of plots to do this command causes it to check the queue for plots to be done PLOTQUIK wakes itself up periodically every ten minutes at the time of this writing so this command will usually not be necessary OPERATIONAL MATTERS This program is usually started on the 936 plotter each morning and is permitted to run all day Of course it can be stopped at any time It may pause while CCP is running it will start up again when CCP finishes PLOTQUIK can be run on the 1051 Don t run PLOTQUIK on both plotters at the same time or you ll get two copies of everything Tear off the plots about every 30 minutes and send them to the DPAs The current schedule for this service is 9 30 A M to 5 30 P M Monday through Friday CCP PLT and PLOTQUIK PLOTQUIK plots from the queue After PLOTQUIK finishes a plot CCP will dump it on a tape with the normal plots CCP dumps it at the next plot time after the plot is finished not the next time after it is queued CCP marks quick plots with an in front of the plot numbers on the receipt sheet While a quick plot remains in the queue PLOTQUIK can replot it After it ha
239. always queued if the QB response is given The terminal user may interrupt the tape queue wait at any time by sending an ATTENTION If this is done the system will remove the user from the queue print the user s current position in the queue and ask if the user wants to continue to wait If an affirmative answer is given the user is returned to the queue at the previous position if possible Otherwise the mount is abandoned and a message is printed on the operators console as above When a job releases a tape drive not a remount the system will release the first job in the queue that needs a drive of the same density or an unknown density A request will have an unknown density if the user did not specify the mode of the tape i e mmmm where mmmm is 6250 1600 or 800 b p i did not appear on the mount request or the operator did not give a density on the QB reply i e the reply was QB rrrr instead of QB rrrr mmmm The drive should be given to this request TAPES TAPE HANDLING 492 it will be flagged with the character Q indicating that it came from the tape queue before any brand new non queued tape requests If the drive is not compatible with the tape i e no density was specified at the first queueing the request should be returned to the queue using the reply qb rrrr mmmm The actual density mmmm of the tape 6250 1600 or 800 b p i should be given on the reply so that it will be present
240. an conclude that nothing is wrong with the terminal or SCP If the user is still unable to get any response from MTS call the MTS operator and explain that you have determined that a user s terminal and the SCP to which it is attached are both working OK but that MTS still won t respond to the user The MTS operator will then be responsible for checking it out from there If the dead terminal was not in use do the following On Ontels press the ATTN key On DECWriters press the RETURN key On Visual 550s press the RETURN key Within 5 seconds you should get a response something like UMnet Hermes S216 MP16 0P VIS Which host Again if you get this you can conclude that nothing UNYN TERMINALS SCPS ETC 522 is wrong with the terminal or SCP You should try typing UM and RETURN and see if you get something like MTS Ann Arbor CE49 00232 If you don t get this you should call the MTS operator and explain that you have determined that a user s terminal and the SCP to which it is attached are both working OK but that MTS won t respond Don t do anything else unless the MTS operator or somebody else at the Center tells you to Explain to the users that you have reported the problem and somebody at the Center is looking into it If you don t get any response to the SHELLO then the terminal or the SCP to which it is attached is probably dead The last page of this writeup has a
241. and QUEUE queuename WITH attributel attribute2 can be used to show every job in a particular queue along with the value s of the specified attribute s This is as close as you get in the RM to a columnar format queue display like HASP As you will find when you try this it is not completely satisfactory but is often useful Examples QUEUE WAYNESTT WITH DEST QUEUE WAYNESTT WITH JOBNAME QUEUE WAYNESTT WITH DEST SOURCE B PROCESSOR Similarly the PROCESSOR command will print information about active processors P ALL or just P will provide information about all active processors A particular processor can be looked at by giving its name For the case of BITNET links note that there are three processors to worry about usually per link as there are two associated processors one for input and one for output For example P WAYNESTT may yield Processor WAYNESTT 529 is waiting for work TYPE LINK LINK WAYNESTT Doing P ALL LINK WAYNESTT would yield perhaps Processor WAYNESTT 529 is waiting for work TYPE LINK LINK WAYNESTT Processor WSU OUT 533 is processing job RSCS03 TYPE SYSOUT LINK WAYNESTT RESOURCE MANAGER 538 Processor WSU IN 545 is waiting for work TYPE SYSIN LINK WAYNESTT In this case you should check again after a while to make sure the job number the output processor is handling changes Otherwise the link is down or otherwise incapacitated Note that the ma
242. annel Configuration screen chpid xx off offlines chpid channel to which HIM is attached chpid xx on onlines chpid mts him x There is usually a note near the System Console listing HIM addresses At present these adresses are HIM1 17 HIM2 1B HIM3 2C If someone calls for instance from the NOC and mentions that a HIM seems to have failed you can dial in to a HIM using Versaterm At the Which Host prompt type 141 211 144 21 HIM1 or 141 211 144 22 HIM2 or 141 211 144 23 HIM3 depending on which HIM you want If the connection fails you will have to stop and start the HIM as mentioned above HOST INTERFACE MACHINES HIMS 389 Hours Computing Center A statement of regular Computing Center hours can be seen by COPYing NEWS CC HOURS SHUTDOWNS Scheduled Normal scheduled shutdowns for preventive Maintenance and or software testing etc are listed below 6 00 a m 8 00 a m Tues and Fri Emregency shutdowns depending upon their urgency are right now or 6 00 p m In some cases a noon shutdown may be scheduled Weekends from 11 30 p m Sat to 8 a m Sun and from 11 30 p m Sun to 8 a m Mon For regular Computing Center hours copy NEWS CC HOURS COMPUTING CENTER HOURS 390 When Computing Center hours are expected to differ from normal the user public should be made aware of this well in advance Sometimes unusual hours may be anticip
243. ape for quite a while and when they were written they were written out of order ERROR ON ffffffff CALL A SYSTEMS PROGRAMMER This message goes to the console A misformatted line was discovered or an error occurred reading writing or locking fEfLLLFE CMDSTATC CMDDIR or CMDFDIR CMDTAPE will not continue after printing this message la response to the read will cause it to stop Fix the file causing the problem as necessary remove any B flags in CMDDIR and start DWB again CMDSTAT TAPE ERROR PLEASE TRY ANOTHER DRIVE This message is printed on the console followed by a mount request A tape error has occurred on the current tape and CMDTAPE is about to try again If an error occurs on the new drive CMDTAPE will abandon the tape and ask for a new one CMDSTAT 332 ERROR JOBDUMP AND OR CALL SYSTEMS PGMR This message goes to the console Either a program interrupt has occurred or CMDTAPE has gotten some unreasonable return code gre and grf are saved before printing this message In either case the next thing to do is to get out the code If the console read returns CMDTAPE will stop IS ANOTHER COPY OF DWB RUNNING This question is printed on the console CMDTAPE has found the tape marked B line 1 of CMDDIR This is normal if another copy of CMDTAPE is running this happens occasionally during heavy load If there is not another copy running CMDDIR is wrong probably because CMDTAPE terminated abnormall
244. appear mentioning that fact and the program will ask for an IGNORE or RETRY by the operator an end of file To terminate the program use Fixed Head Files are automatically formatted when an IPL occurs JOB PROGRAMS PGET 134 NAME FUNCTION PROTOTYPE USAGE PARAMETERS OPTIONS EXAMPLES JOB PROGRAMS PN JOB PROGRAMS PN To load the PN 1403 or 3203 type print chain or train image into the Universal Character Set buffer on the printer controller Additionally for 3203 type printers the forms control buffer FCB is loaded for the standard MTS carriage control channel assignments PN device i d ECHO FOLD ALLOW NOMSG If one wishes to change print chains or trains on a printer it is necessary to load the UCS buffer in the printer controller with the correct character set If the UCS buffer gets scrambled for any reason one would also want to reload it In addition at system IPL time it is wise to reload the UCS buffer The contents of the UCS buffer may also be echoed on the associated printer for verification 1 The device i d The following opetions are recognized ECHO causes a print train verification message to be printed after loading the UCSB FOLD enable folding when loading the UCSB This allows the printing of lower case letters with an upper case only print train This option should not normally be used with MTS ALLOW allows data check to be pres
245. approximately the same time one will wait for the other to complete its IMPL before commencing Before IMPLing be sure that either the system isn t running or that the disks are logically offline Interrupts Amdahl disks interrupted comparable to popping the plug on other If an Amdahl disk is in a situation where it needs to be disk drives one must open the back panels of the appropriate disk module and and raise the ATTEN switch Stickers on the inside of the back panel doors should help in deciding which ATTN switch to use Emergency Power Switch The Emergency power switches are located on the ouside of the front panels of each module If these switches have been turned off for any reason the panel door must be opened and the rear of the switch must must be lifted with a screw driver or other appropriate implement in order it to resume its normal ON position for a MEMOREX 1270 Go to the MEMOREX 1270 and make sure that it s powered up b IBM amp LEE DATA Terminals See if the CUs for the and LEE DATA terminals are powered up IBM When these CUs are powered up the micro code is reset also C IBM 2671 Removed from system d Data Concentrators DCs Primary Communications Processors PCPs Ef the Concentrators or PCPs were affected go Data to the rear of these machines and power them up They should be left in LO
246. aracter to be displayed on the screen and the character X 1E to be entered in the field This key has little value and should not be OPERATOR CONSOLE 3270 340 used PROGRAM ENTER CURS SEL IDENT and ATTN perform no function when used with the OPERATOR job so they are not used Note The DUP FIELD MARK TEST IDENT CURS SEL ATTN CURS BLINK ALT CURS and PROGRAM ENTER keys are not present on all keyboards Keys which cause program action Many of these keys have a function very similar to their function when the 3270 is used as a terminal All of them require program action and should cause the input inhibited light to come on for a fraction of a second and then go out If these effects do not occur something is probably wrong ENTER This key terminates all input lines The field containing the cursor is the one which is entered Other fields may contain input but they will remain unchanged The position of the cursor in the field is irrelevant If the cursor is not in any field this key is ignored CNCL This key is used to enter a cancel for a message response line The position of the cursor determines the line to be cancelled If the cursor is not in a message response line the CNCL key is ignored The characters CANCELLED will appear on the screen On some keyboards this key must be used in conjunction with the ALT shift key PAl Stops the execution of any System Status command whic
247. are usually released by operators or DPAs during the second half of the midnight shift or when the print queue is very low Print amp Execution Priorities Categories amp Classes Execution Time Seconds lt 1 lt 2 lt 4 lt 8 lt 16 lt 32 lt 64 lt 128 lt 256 gt 256 Priority Print Priority Categories Class 0 X page 0 unused 9 L line 1 minimum 8 P PN 2 deferred 7 T TN 3 low 6 S special 4 normal 5 M mixed 5 unused 4 6 high 3 7 flush 2 I Print Page Limit by rate group Priority Normal Low Deferred Minimum 15 5 14 10 E 13 20 5 12 50 10 5 11 100 20 10 5 10 200 50 20 10 9 350 100 50 20 8 500 200 100 50 7 650 350 200 100 6 800 500 350 200 5 1000 650 500 350 4 1500 800 650 500 3 2000 1000 800 650 2 4000 1500 1000 800 1 6000 2000 1500 1000 0 gt 6000 gt 2000 gt 1500 gt 1000 RM PRIORITIES CLASSES SCHEDULE 377 HASP HUNG Obsolete NOTE This section HASP HUNG is mainly here for firefighters and systems programmers but may be of some use to operators What to do when Hasp won t accept any COMMANDS If HASP does not respond to any operator commands and HASP HASPSTAT as run by the operator file HPS does not report any errors such as no WTO buffers it s probably because HASP has queued the maximum number of reply messages for a particular remote station If HPS reports that there are no WTO buffers then HASP could b
248. arted manually the procedure would be mts ig shuf fle When this program is run the IG statistics are dumped onto tape and the file GRAF IG SHUFFLE is emptied OPERATOR FILES IG SHUFFLE 228 INIT FUNCTION USAGE OPERATOR FILES When the system is IPLed after the time and date have been set the MTS task using INIT as master source is automatically started by the INIT job program See the description of the IPL procedure earlier in this manual Is normally started automatically TE for some reason it has to be restarted enter mts Fama ES te Must NEVER be run once users have been allowed to sign on as it destroys all scratch files OPERATOR FILES INIT ITX NAME FUNCTION OPERATOR FILES 229 ALEX Starts five additional file jobs to exercise and test intertask communication between UMMPS jobs It is normally used by systems programmers OPERATOR FILES ITX 230 OPERATOR FILES FLAR NAME LAR FUNCTION Program to reset the line adapters on the transmission controller Memorex 1270 USAGE mts lar OPERATOR FILES LAR OPERATOR FILES 231 LAS NAME LAS FUNCTION Program to start up the line adapters and miscellaneous terminals and logging functions The results of running LAS appear on the operators console as follows Line configuration 20 April 1987 Jobs on Memorex None started by request Line configuration 25 January 1992 Jobs on Lee Data
249. at the operator s convenience by running CCP using a special plot tape When it is no longer necessary to defer the PLOTTING SYSTEM 429 user s plots a program is run to tell CCP to process the plots normally DEFERRING PLOTS Start the MTS job in the file CCD You will be prompted for a ccid Enter all the ccids for which plots are to be deferred one per line Terminate the list with an end of file CNCL Example 21 10 01 02798 mts ccd 02798 21 10 04 MTS ENTER SIGNON ID 02798 21 10 04 MTS sys 02798 21 10 00 PLOTS FOR SYS WILL BE DEFERRED 02798 21 10 00 MTS mts 02798 21 10 05 PLOTS FOR MTS WILL BE DEFERRED 02798 21 10 11 MTS CANCELLED This program writes information into SYS CCPPRIO telling CCP which ccids to defer plots for All plots for the ccid will be deferred RUNNING REGULAR PLOTS Run CCP in the normal way Deferred plots will be left in the queue and their receipt numbers will be listed on the plot sheet under requeued RUNNING DEFERRED PLOTS Run CCP in the normal way except use the plot tape marked deferred plots This tape is in the rack with the other plot tapes The deferred plots will be added to the logical end of the tape i e after the plots already there The system will place only plots that have been deferred on this tape The deferred plots will remain on the deferred tape until the tape is reinitialized see below
250. atcher program Tf error occurs during running of CWM a message with information on making calls to call appropriate persons appears on operators console OPERATOR FILES CWM NAME FUNCTION USAGE OPERATOR FILES 203 DLG Program which logs disk errors This program is started automatically Is normally started up automatically Requires only mts dlg Takes no parameters OPERATOR FILES DLG 204 DMP NAME FUNCTION USAGE REMARKS OPERATOR FILES DMP OPERATOR FILES DMP Initiates deferred batch jobs which print dump tapes mts dmp ENTER DUMP TAPE OR EOF OR STOP lt operator enters one of these gt ENTER REASON FOR THE DUMP ONLY ONE LINE lt operator does this gt Whenever a dump finishes whether it was initiated by an operator or by the system the program DMP should be started DRN NAME FUNCTION USAGE OPERATOR FILES 205 DRN To issue HASP commands to drain SA00 through SA02 and SA04 through SA19 This is identical to entering the equivalent DRAIN SAnn commands one at a time DRN also drains MNT RD1 RD2 and RD3 as well as MNT MS UM WS WU WM and UB mts drn OPERATOR FILES DRN 206 NAME FUNCTION USAGE NOTE OPERATOR FILES DSD Starts amd runs DASDI mts dsd See DASDI elsewhere in this manual OPERATOR FILES DSD OPERATOR FILES 207 NOTE See the section concerning t
251. ate Classes USAGE Started automatically by AUTOSTART amp HSP OPERATOR FILES RHB 276 OPERATOR FILES RMC NAME RMC FUNCTION Lists RM Processors and permits some manipulating of same USAGE mts rmc p prints all processors p xxxx prints a particular processor and q shows the number of queued jobs and their locations The information produced by rmc may be used to determine which processor s need to be drained if need be such as when a processor name is highlighted on the operators console OPERATOR FILES RMC RMD NAME FUNCTION USAGE RMD Shuts down mts rmd OPERATOR FILES 277 drains Resource Manager OPERATOR FILES RMD 278 OPERATOR FILES RMP NAME RMP FUNCTION Starts Resource Manager Processors USAGE mts rmp OPERATOR FILES RMP RMS NAME FUNCTION USAGE OPERATOR FILES RMS Starts Resource Manager mts rms 279 OPERATOR FILES RMS 280 OPERATOR FILES RST NAME RST FUNCTION Restore user files from filesave or deadfiles tapes USAGE 1 Enter mts rst on console 2 RST will ask WHERE IS A PRINTER 3 When RST asks WHAT FILES ARE TO BE OPERATOR FILES RMS RESTORED do one of the following a To restore one or more files enter filenames one per line Terminate the list with an end of file cnel b To restore all files except some particular one s type EXCEPT
252. ated far enough ahead of time to publish them in our newsletter In any case the Center staff is usually aware of a change in hours enough ahead of time to make signs that can be posted on the doors of the Computing Center A signonmessage which has as part of its contents the special scheduled hours or the same type of message pointing to a file with the hours should be posted HOLIDAY SIGN PROGRAM This program draws door signs for announcing holiday or other unusual hours Currently the implemented formats are 1 title line centered with 4 lines below for hours and 2 title line centered with 3 lines below for hours Examples 1 Thanksgiving Hours Thurs Nov 25 Closed Fri Nov 26 Closed Sat Nov 27 Open at 9 a m Sun Nov 28 Normal hours 2 Thanksgiving Hours Thurs Nov 25 Closed Fri Nov 26 Closed Sat Nov 27 Normal Hours The box around the sign is drawn by the program To run the program enter R W035 HOLIDAY 1000 to use PLT1 R W035 HOLIDAY 2000 to use PLT2 R W035 HOLIDAY 3000 to use PLT3 The first statement that will appear on the terminal is a request for an alphabet in the form COMPUTING CENTER HOURS 391 ALPHABET If RETURN is pressed with no alphabet given at this point the default alphabet SANSERIF 1 will be used Other alphabets can be found in MTS Volume 11 appendix A
253. atus job table address l L project L user id XXXXXXXX nnannnn nnnn nnannn nn nnnn nnnn nnnn nnnnn nnnnn L PSW RO 7 XXXXXXXK XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX R8 F XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX FPRs XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX XXXXXXXX l L page ins L real pages needed l L real pages L virtual pages L supervisor state time secs This command continues to attention interrupt PA1 on a 3270 STOPped or GOOSEd Examples JOB MTS DS06 02181 19 07 SYS 00006 OCX8 LA06 02211 2GDR CB06 02310 SYSTEMSTATUS 13 07 2LFG LA26 02231 8 11 5C44 CB07 02311 L problem state time secs right half until the JOBS 34 62 K3S4 CA00 02282 55 02 2LFM LA29 02234 86 04 SGOA CB08 02312 L file 1 0 produce ouput until interrupted by job 8 7 8 then more detailed information about that XXXXXXXX XXXXX XXXXXXXX XXXXX XXXXXXXX XXXXX an is 15 00 15 477 RSTR 02462 10 00 13 40 1B 76T 3373VP 841RP 13 0 XEC 49 PTR 0 PCH 7 HLD JOB 2376 Job Name Jobtbl Status BTS2 User Proj Par TLC Prob Timt Super Time VP RP RPN Page ins File I O 02376 MTS 06E090 Reading DS40 22 W225 WOPN 4030D120 19 555 4 318 TA 0 1 8491 14511 RO 7 80000001 00E1F670 00004088 00706018 0030B9AC 00000014 0000000F 000014D0 R8 F 8030CEE6 00706000 A0O30F8D2 00
254. ave been made available so that busy signals should be somewhat less frequent but LSS more frequent A full explanation of LSS follows As many experienced users know it sometimes becomes necessary for the Computing Center to limit terminal usage of MTS during periods of very heavy usage or when hardware equipment failures cause the capacity of the system to be reduced Rather than simply reducing the number of terminal users by making some phone lines unavailable we have chosen to use a procedure called Limited Service State LSS which will restrict certain terminal users those in LSS to using a subset of the normal MTS commands i e LSS users are unable to run large programs but they may still run programs requiring less than 25 seconds of CPU time create edit and destroy files mount tapes use the MERIT computer network etc A more complete description of the behavior of the system after LSS is activated follows There are approximately 55 more lines in this file you may prefer to COPY it to PRINT The system will enable the use of LSS only if some hardware component is overloaded and the overload condition cannot be corrected by reducing to number of concurrent batch jobs being executed to 2 The system load will be examined whenever a new terminal user establishes his telephone connection to the system If the system happens to be heavily loaded the user may still sign on but he will be temporarily prohibite
255. be included in the RUN command INIT MCFILE is for machine and channel checks and INIT UC3330LOG is for disk tape etc errors Note the parantheses around the device type e g dev 33xx These are required even if only one device type is given as in the example above Each of these must be a separate run EREP is not capable of making more than one type of report per run You can assign SCARDS to DUMMY and SPUNCH to ACCIN as described above to save time on the second and subsequent runs See the IBM manual mentioned above for more possibilities and examples of each of these reports Most of the parameters that worked with the old EREP also work with this one but often there is a better way to do things Example run ifcacs01 TE you run erep sercom ser sprint x 8 00 jun 20 8 00 jun 23 ok type mc or type mc print ps print ps provides a much much more detailed printout of the error condition s but can be quite voluminous so is used infrequently fendfile you have sent the output to a temporary file e g X and d like to print it you can EREP SYS run pagepr scards x or cont print printer line copy x print rel print 321 EREP SYS 322 MEMOREX LINE HUNG Occasionally the channel to which the Memorex Control Unit is attached may be out of synchronization due to a missed interrupt The Hangup can usually be cleared by generating an interrupt f
256. belt module onto extension shaft unitl it latches Wash and polish PR belt Cleaning the PR belt 1 Use pumice solvent 43p63 or eraser eraser is for small scratches Turn belt counter clockwise only using belt advance crank Pour small amounts of pumice or solvent on foam pad white sponges Apply with light to heavy pressure in small circular motions If cleaning the entire belt turn crank while applying pumice or solvent until entire surface is covered Lightly buff haze from belt using clean foam pad you may use Film Remover if you like it dries faster XEROX 9790 PRB 563 Replacing PR belt 1 2 10 If this doesn t work Push PR belt back in machine Unscrew extension shafts place them back in the machine Replace drive and support shaft nuts Replace Laser lamp Cam up developer module belt Replace Edge Exposure lamp Replace Detrack cortron Lower and latch C cortron Let machine warm up Type offline on the display on top of the machine Press Continue Print 250 samples SAM TEST FRM 250 Check samples to see if the problem is solved if it is Return machine to normal operations and samples run more call XEROX XEROX 9790 PRB 564 Abnormal Conditions OPER 343 ABNORMAL SITUATIONS 307 ABS 169 ACC FILEINFO 358 Accounting Records 213 Account Listing And File Charge Update Monthly 368 ACCT
257. bottom line of the screen is always used to display the list of task numbers of tasks that have requests outstanding pending The input lines and messages scroll upwards to the top of the screen A one page buffer of the these input lines and messages is kept by the supervisor so that they can be logged in a file by a special logging task The support for PRINTER KEYBOARD consoles such as the 3215 is similar to the old 1052 support in that the operator must press the REQUEST button to force the supervisor to hang a read on the operator s console The new method of replying to a request is as follows the operator types an octothorpe followed by the tasks number of the task to which he is replying followed by a blank and then the reply If no tasks number is given after the octothorpe the reply will be given to the task whose task number is first on the list of pending requests i e If there is only one outstanding request no task number need be given but don t forget the octothorpe EXAMPLES PENDING 12 15 15 OK RF9999 T904 this would reply to a tape mount for task 15 OK RF9999 T901 this would be a reply to a tape mount for task 12 NOTE If the first character of a reply is NOT numeric the blank after the task number may be omitted The operator can cause the supervisor to produce its list of task numbers with pending requests at any time by typing R CONSOLE BACKUP SUPPORT ES3090 300J 347
258. brackets in the following three commands should be used to make the TAPE QUEUEING routines aware of the modes of the tapes that are being queued The actual density of the tape should be given if the user didn t specify it See the description of TAPE QUEUEING below QB rrrr 6250 800 1600 Queue this user when all drives are busy etc TAPES TAPE HANDLING 491 QP rrrr 6250 800 1600 Queue this user and displays the top request in the queue pop top job QN rrrr 6250 800 16001 Queue this user and display next request after this one NEXT This is used to return the user to the queue because it was not possible to satisfy the user s request even though there are unused drives TAPE QUEUEING If TAPE QUEUEING is enabled i e TAPEQ 1 a mount request that cannot be fulfilled because there is no available drive of the give density should be given the QB reply instead of BY or RERUN The terminal user will then be promted for approval to be placed in the tape queue e g t rrrr all units busy at this time there are N terminal users waiting for M drives ahead of you in the mount queue do you wish to wait for an available drive If the user responds OK then the user will be placed at the end of the current mount queue If the user s response is NO then the mount request is aborted and the comment CCID RRRR REQUEST ABORTED BY USER is printed on the operators console Batch jobs are
259. buffer on the printer controller Additionally for 3203 type printers the forms control buffer FCB is loaded for the standard MTS carriage control channel assignments QN device i d ECHO If one wishes to change print chains or trains on a printer it is necessary to load the UCS buffer in the printer controller with the correct character set If the UCS buffer gets scrambled for any reason one would also want to reload it In addition at system IPL time it is wise to reload the UCS buffer The contents of the UCS buffer may also be echoed on the associated printer for verification 1 The device i d The following opetions are recognized ECHO causes a print train verification message to be printed after loading the UCSB FOLD enable folding when loading the UCSB This allows the printing of lower case letters with an upper case only print train This option should not normally be used with MTS ALLOW allows data check to be presented when an invalid or an uncomparable character is printed Normally data checks should be blocked with MTS JOBS PROGRAMS 137 NOMSG supress the verification message normally printed on the operators console after the buffer load is completed EXAMPLES qn ptrx qn ptrx echo qn ptrx nomsg JOB PROGRAMS QN 138 NAME FUNCTION PROTOTYPE USAGE PARAMETER EXAMPLE NOTE JOB PROGRAMS SE JOB PROGRAMS SE To print out the first twelve
260. can be used by MTS Any completely blank tape or any tape with an unreadable first record MUST have something written on it by this program before it is attempted to be used in MTS USAGE Enter mts lbl1 The program will prompt for a tape drive and code number Drive Name is the tape drive name eg T901 Code Name is in the form 123 MIKE 345 Example Tapes to be initialized will have a green sticker with the letter I attached to the tape This sticker should be removed after initialization When users submit tapes to be initialized they should have inserted labelling information into a file When LBL is run and the tape and code number are typed in by the operator the tape is initialized with the information that the user entered into the file mentioned above The prompt for input from LBL is ENTER TAPE DRIVE AND RECEIPT CODE Enter the drive name and the receipt code Example t904 123 MIKE 456 At this point the tape will be initialized If the process fails an error message will be printed on the operators console The output on the console will appear as in the example below IBM Labelled DENSITY 1600 VOLUME VOL_2 OWNER JANE Operator Errors OPERATOR FILES LBL 234 OPERATOR FILES In case the tape is mis labelled the wrong tape is mounted the tape must be re initialized Tapes cannot be re initialized using the same code number because once this code has been used a bit i
261. can place a core image of the IPLREADER program in these pages IPLREADER is the program which decides which system to load and loads it into the bare machine The area is DISK MANAGER DMGR 352 reserved by generating the necessary DSCBs to describe the IPL pages writing them onto the disk and marking the DSCB and IPL area pages properly in the PAT If nnn is not specified the maximum number of pages which can be described by a single full DSCB page is allocated 968 38 15 62 IPLSTART nnn gt used in conjunction with the IPL keyword the nnn value determines at what page the IPL pages should begin This defaults to the first available page on the pack if not specified PAGES nnn gt format the pack as if it had only nnn pages PATA nnn or PATSTART nnn gt PAT should start at the given page address CLEARPAT gt construct and write PAT and IPL DSCBs useful only with the LO option NOSLOW gt override the PAR SLOW see below option for this DASDI operation only This is the default SLOW gt perform a 50 millisecond wait between formatting writes This gives other tasks access to the disk s control unit and allows a pack to be DASDI d on a running system The following parameters should only be used for special situations WHA gt read and write home addresses on the volume This is almost never necessary on modern disks and is very dangerous to do since it may cause in
262. ce D200 Type 6280 Density Single Cmd DISPLAY LABEL Label Volume XXX200 Label VOL1 Type 6280 Format VX PVN 1 DISKCOPY 356 PAT start 50400 X C4E0 length 100800 X 000189C0 Catalog pointer X 00000000 Error byte 00 Cmd STOP 4 Check the PAT Page Address Table starting address for the disk you wish to copy Assume we re copying MTS005 Run Sys Dsk list mts005 5 MTS005 on D403 6280 3339 100800 50400 stop In this example the disk s PAT Page Address Table starts at the default location for a 6280 device 50400 The PAT starting address is the last number on the line listed 5 Use DASDI to format the unused disk Name it OLD005 and use the SLOW option if the DASDI is occurring on a running system DASDI takes 10 20 minutes to run using SLOW will double or triple that time gt DASDI MTS DASDI program AR267 Enter input line Dddd 111111 Vx PAGING PRIVATE pars D200 OLD005 VX 5 SLOW If the unused disk has never been formatted before never with our DASDI that is then you should use the WHA Write Home Address option D200 OLD005 VX 5 SLOW WHA If the disk you want to copy has a non default PAT address you should format the unused disk with the PAT starting at the same address D200 OLD005 VX 5 SLOW PATSTART nnnnn The morning of the diskcopy 6 Do a normal system shutdown all the way though the Shutdown All step and also completely shut down the Resource Manager Run t
263. ch jobs per hour Number of batch processor to allow BJPMOD case number Channel activity weight Disk cache size Status of Data Conc C Class 1 load CPU Class 1 static CPU Class 2 load paging Class 2 static paging Class 3 load disk 1 0 Class 3 static disk I O Class 4 load channel operations Class 4 static channel operations Class 5 load drum space Class 5 static drum space Disk I O weight Drum read increment Number of job tables for 3270 grabs BJPMOD CPU idle low threshold CPU idle time too high threshold 100 Lower big job threshold limit real pages LSS flag 0 gt off 1 gt on Least number of privileged jobs threshold LSS flag enabled by operator 0 gt off 1 gt on Maximum execute queue length before LSS aximum number of batch processors to allow Minimum big job time slice multiplier inimum number of batch jobs per hour before LSS Minimum value for drum reads to time slice end Minimum number of batch processors to allow Minimum big job time slice multiplier Maximum real pages for all privileged jobs SSRTN master job sample period 300 s sec Maximum time slice high threshold Maximum slice high value maximum time slice low threshold aximum time slice low value Mean number of non privileged jobs last interval Yo tape mounts 0 gt off 1 gt on Paging load high threshold Mean number of privileged jobs last interval
264. cified all entries contained on that tape are listed If TAPE POOL is specified all unrestricted tapes in the available tape pool are listed If no parameters are specified the entire direc tory is specified MTS This command returns control to MTS command mode The program may be resumed via the RESTART command PURGE type This command purges the master entry and all file entries for the specified type The tapes used for the file entries are returned to the available tape pool Confirmation of this command is required RECREATE type MODE 800 1600 6250 PFMT JX JS JIM JID I LI SFMT JX JS JM JID I LI This command respecifies the master entry for type This allows either the format of the primary or secondary search field to be respecified or the mode of the tape to be changed All files added for this type will use the new format or mode If the MODE parameter is omitted the default mode will be 1600 bpi If either the PFMT or SFMT fields are omitted the corresponding primary or secondary field will become unspecified REMOVE rack label This command removes an empty tape from the available tape pool If the tape currently is in use the tape is not removed TAPE DIRECTORY PROJECT 516 Command STOP Explanation This command terminates the program Tape Directory Subroutines The tape directory subroutines are contained in the file SYS DIRLIB These are also writte
265. ck existence of basic system files Shutdown options NONE REBOOT CHECK This next option is important Must do Reboot check Ef the file does not exist or is damaged the system cannot do automatic reboot SHUTDOWN I OPERATOR This terminal is now an operator s console SAN DIEGO SUPER COMPUTER 545 ooo oe e o OPCOM 2 FEB 1988 10 54 17 73 Operator _XXAl has been enabled username SYSTEM 5 5 5 5 OPCOM 2 FEB 1988 10 54 17 86 Operator status for operator _XXAl CENTRAL PRINTER TAPES DISKS DEVICES CARDS NETWORK CLUSTER SECURITY OPER1 OPER2 OPER3 OPER4 OPER5 OPER6 OPER7 OPER8 OPER9 OPER10 OPER11 OPER12 SHUTDOWN I DISLOGINS Interactive logins will now be disabled SET I INTSET login interactive limit 0 current interactive value gt be shut down ooo eee ee e OPCOM 2 FEB 1988 10 54 20 54 Message from user DECNET DECnet shutting down ooo eee e e OPCOM 2 FEB 1988 10 54 21 64 Message from user SYSTEM PSI I ACCTDISABL PSI accounting disabled 5 5 5 5 OPCOM 2 FEB 1988 10 54 21 77 Message from user SYSTEM PSI I STOPPED VAX PSI V3 2 Multi host Native network device _NWAO dismount ed SHUTDOWN message from user SYSTEM at _XXA1 10 54 24 UMRUAC will shut down in 5 minutes back up shortly via automatic reboot Pleas e log off node UMRUAC Standalone 3 terminals have been no
266. crash and perform repair operations by issuing the command run sys newstatsys dirlib par init If the work file is empty no save operations will be performed If the STA program is run normally when the work file is already full the operator will be notified that the file is currently full and inquire as to whether recovery operations should be commenced This may be necessary for example when the previous run was interrupted by an attention interrupt Errors in running STA If an error occurrs while STA is running the program should be rerun i e MTS STA The program will put forth questions which you must answer such as DO YOU WANT TO REPAIR yes You should answer YES whether the daily or monthly tape is being used Once the repair has been completed the system will ask OPERATOR FILES STA OPERATOR FILES 291 DO YOU WANT TO SAVE THE CURRENT STATISTICS FILE no if the repair was successful If the repair was not successful try using a new tape If this doesn t work some help from a systems programmer is needed When an error is detected while running STA the program will save the data in STATISTICS in a file with a name of the of the form STATmmmdd e g STATAUGO8 When STA is rerun this file is used for the repair operation The file will remain under the ID SYS until manually destroyed by a systems programmer OPERATOR FILES STA 292 NAME FUNCTION U
267. cs tape FCHG File charge update tape Status byte 5 The status bits indicate 01 gt Primary search field present 02 gt Secondary search field present 04 gt Comments information present 08 gt Entry in progress entry unconfirmed by CONENTRY 10 gt 20 gt Tape restricted for particular type marked under POOL master only 40 gt Tape in use marked under POOL master only 80 gt Master entry Tape mode byte 6 02 gt 800 bpi 03 gt 1600 bpi 04 gt 6250 bpi Tape rack name bytes 7 22 Tape volume label bytes 23 28 TAPE DIRECTORY PROJECT 509 Tape file number bytes 29 32 Sequence ID bytes 33 36 The sequence ID orders with respect to the time the tape is written all files on all tapes of a particular function type e g all files on the MSTA tapes will be ordered from 00001 to nnnnn This sequence ID will be useful for finding the next or last tape file when its time range is not explicitly known Primary search field format byte 37 No primary search field Julian microseconds unsigned 8 byte integer Julian seconds unsigned 4 byte integer Julian minutes unsigned 4 byte integer Julian days unsigned 4 byte integer Integer signed 4 byte integer Long integer signed 8 byte integer Duo PWNFR OO ll Primary search field bytes 38 53 This field is composed of two 8 byte integers Normally this field will consist of a Julian time range m
268. current second VM XA will print PRESS TOD ENABLE SET KEY AT DESIGNATED INSTANT Go to the ES3090 300J console for the UB side of the 3090 hold down the ALT key and press the TOD key Go back to the VM XA operator s console VM XA will again prompt NOW hh mm ss tmz weekday mm dd yy CHANGE TOD CLOCK Y N If the time and date are now correct press ENTER If not type YES and press ENTER and go back to step 10a In a moment VM XA will print CLOCK SYNCHRONIZATION PRESS TOD ENABLE SET KEY Go to the ES3090 300J console for the UB side of the 3090 hold down the ALT key and press the TOD key Now go back to the VM XA operator s console After the time is set correctly other messages will be printed on the VM XA operator s console Certain errors could occur at this point Common error messages might include HCPINIO70W MACHINE CHECK DURING SYSTEM INITIALIZATION HCPxxx320W ALTERNATE PROCESSOR INITIALIZATION UNSUCCESSFUL HCPRET581E RECORDING RETRIEVAL DISK FULL USER user HCPCRC5831I RECORDING recording name userid USER NOT RETRIEVING HCPxxx625E PROCESSOR nn TOD CLOCK SYNCHRONIZATION FAILURE VARYING OFFLINE If any of these errors occur or you suspect that any other VM XA operator console message indicates an error see the section VM XA Error Messages UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 45 12 If no errors are detected VM XA will
269. d Then it asks OK ves or y or ok or o or NO NOTE In options b d and e above block or receipt number ranges can be given in increasing order They must be in the same order as they appear on the tape For example one could state block or receipt numbers as follows PLOTTING SYSTEM 435 A LAG TB US 25s 27 or 504223 504227 504230 504233 504236 To plot from a file use plot list list being one of the above options Example r plt51 COPY MTS PLOT STOP copy all WHERE IS A PLOT TAPE t904 PLOTS FROM THU 1400 TAPE TAPE WRITTEN 13 00 26 04 26 79 1 BLAK 4 ok 1 BLAK 4 2 BLUE 4 3 RED 4 ok PLOTTING PLOT 004 NOW WHAT DO YOU WANT bac 5 1 BLAK 4 ok LOTTING COMPLETE COPY MTS PLOT SIG STOP plot 2 LOTS FROM THU 1400 TAPE TAPE WRITTEN 13 00 26 04 26 79 1 BLAK 4 2 BLUE 4 3 RED 4 ok COPY MTS PLOT SIG STOP stop MESSAGE TO USER If it is desirable to print a message to the user at some point e g when lines are being over drawn and the pen is clogging you can use the MSG facility To use this simply GOOSE the supervisor number of the plotter type MSG followed by a space and the message Example MSG There is too much up and down motion of pen Penpoint is overheating Note The message must be entered on one line PLOTTING SYSTEM 436 MULTI PEN PLOTS in PLT Multi Pen Plots are done first beginning with the shorter
270. d be used at least once per week One of the puposes of this function is to determine if any bad partitions were thrown out by the 3805 and replaced by spares and how many spare partitions remain TO USE CODE 90 press CLEAR Displays FCD e key in 90 press ENTER Displays FCC returned press CYCLE Displays DD1 press CYCLE Displays DD2 pages each e press CYCLE Displays DD3 press CYCLE Displays DD4 press CYCLE Displays DD5 00 function code entry mode 0800 means 8 parameters were 0004 number of basic storage modules nnnn number of good partitions 4 nnnn there are nnnn spare partitions remaining if this number is less than 20 the 3805 should be taken offline at the next reload and INTEL should be called nnnn hex number of suspect partitions if this number is greater than 0 INTEL should be told the next time they come in nnnn hex number of unservicable VIRTUAL MEMORY PDP 72 SYSTEM SHUTDOWN AND START UP PROCEDURES partitians if this number is greater than 0 INTEL should be told the next time that they are in press CYCLE Displays DD6 nnnn number of sweeps since last WARM START number varies press CYCLE Displays DD7 nnnn number of big sweeps number varies e press CYCLE Displays DD8 nnnn data transfers since last sweep number varies press CLEAR Displays FCDOO FUNCTION CODE entry mode key in OF
271. d from performing certain computing tasks that require large amounts of CPU time That is he will be put into Limited Service State and will be so informed immediately by a pre signon message THIS IS A LIMITED SIGNON EXECUTION TIME LIMITED TO 25 SEC PER RUN So long as the user is in LSS all commands dealing with the running of programs except RUN are illegal and attempts to use them will produce the message ILLEGAL COMMAND FOR LIMITED SERVICE STATE LIMITED SERVICE STATE LSS 399 The illegal commands under LSS are DEBUG LOAD START and RESTART Except for these four commands all other commands are legal under LSS Attempting to use any of these four commands or the RUN command will cause the user to be removed from LSS automatically if the system load has dropped sufficiently In this case the message LSS MODE HAS BEEN TURNED OFF will be printed unless SET TERSE ON has been given and the requested command will be executed normally Whenever the user is in LSS the RUN command will have a 25 second local time estimate imposed automatically unless a time estimate is explicitly specified on the command in which case it may not be specified as greater than 25 seconds A user in LSS can explicitly request a transfer to normal full service state is by issuing the command SET LSS OFF If the system is able to grant his request it will respond by printing t
272. d situations which the operator is expected to detect and correct Error situations which arise and which are not covered by this section of the manual should in all cases be reported to a Systems Programmer or to a senior member of the Operations Staff Subsequently new problems will be diagnosed and included in this section of the Operators Manual The following sections assume familiarity with the preceding portion of the manual and make free use of terms defined therein Indications 1 Various software detected errors will appear as error messages on the operators console Such errors appearing in the form of messages from the disk support routines the tape handling routines or possibly from individual processors such as HASP 2 Errors or malfunctions in devices may occur which are not immediately apparent from the behavior of the operators console In these cases the operator is expected to recognize and diagnose the error condition and take corrective steps 3 Channel and control unit errors may also occur as above 1 4 A few central processor error conditions are also included in this section of the manual 5 A rather large number of error situations may originate from the HASP processor and these are detailed below under separate sections ABNORMAL SITUATIONS SUMMARY 308 CONSOLE ERROR CONDITIONS Snarks This problem situation is indicated by the following lines on the operators c
273. d they in turn will notify the project director for that ID Unauthorized use is the use of a ccid and password by someone other than the owner of the ccid without the owner s permission Inappropriate use would be when someone uses their ccid to send unwelcome messages run jobs which print out lots of blank pages etc Requests to suspend an ID must be made in person to the Machine Room operator at the Computing Center on North Campus The individual must present some sort of identification that includes a picture e g a drivers license Telephone requests to suspend an ID must be rejected Running CPW will result in the following prompts at the operators console Operators Name Requestors Name OPERATOR FILES CPW OPERATOR FILES 199 Address Telephone Signon ID Re enter ID for verification The operator should respond with his or her name the name address and phone number of the person making the request and the ID of the account for which the password is to be changed The ID must be entered again to reduce the possibility that a typing mistake will cause the wrong ID s PW to be changed If possible the address and phone number of the person requesting the change should be taken directly from the identification presented OPERATOR FILES CPW 200 CTD NAME FUNCTION USAGE REMARKS Note CTD sho OPERATOR FILES CTD To check out the disk drives af
274. daily PM2 run You could use PM2DALY PRINT PRINT PAR P N RET 100 To delete volumes from the tape volume history file put the appropriate DVOL commands as described on page 5 18 of the PM2 manual in the file dvol and execute the commands PM2TVDL TVOLCMDS DVOL This will leave its printed output which will consist of the normal tape volume history report in PRINT To get the monthly summary report for August 1983 you could use the command PM2STAT PAR P N TYPE OT DATE 83213 83243 This command will leave the monthly report in the file PRINT It includes the P N PARAMETER TELLING PM2 to ignore print errors The DATE parameter gives the first and last day in August 1983 in YYDDD format where DDD is the day number in the year PM2 451 Special Rates for Holidays How to Set I Before the Holiday Period 1 Change the times that minimum priority jobs are released by the RM by changing the control file SYS AUTOHASP CTL Signon SYS Duplicate AUTOHASP CTL AUTOHASP lt CTL Edit AUTOHASP CTL make necessary changes to the times and days using the editor Below is the contents of the file sys autohasp ctl Mentioning a specific day such as Monday takes priority over weekday The first line below SUNDAY 00 00 6 59 ON ON ON means that Low Def and Min jobs can run on Sunday from midnight until 6 59 The next line shows that on Sunday from 7 00 to 17 59 Low and Deferred jobs
275. dly executes the program that produced the error Execution is repeated n times up to 32767 n defaults to 32767 MI Manual Intervetion OLTEP bypasses routines which require manual intervention unless MI is specified NOTE The MI and RE options are mutually exclusive RE Remote For sending message to and receiving replys from a remote site when RETAIN 370 is active We don t use this The MI and RE options are mutually exclusive IFD105D ENTER DEV TEST OPT OLTS 418 Message IFD134I warns you that any user pack that is physically on the system and that has the serial number CEPACK will be written upon if the volume is accidentally entered in response to message IFD105D Message IFD107I tells you the OLTEP options which are currently in effect In this case the default options will be listed because you have not enterd a test definition Message IFD105D asks you to define the first test it will be repeated when the test is completed Example 180 184 2400 This will test units with address 180 to 184 Run basic tape test Use standard OLTEP options OLTEP will run the test and then issue messages IFD107I and 00 IFD105D When you receive message 00 IFD105D you can define the next test to be run using devices tests options CANCEL terminates the program When the ENTER DEV TEST OPT message appears you can type one of the following prompt dev prompt test prompt opt You will be given examples a
276. e and also places the label in the catalog USAGE mts clb REMARKS A user may hit enter when asked for a tape label during the running of Tapesubmit in which case the tape label doesn t get entered Using CLB will remedy this When run the program will ask for the drive receipt and tape id OPERATOR FILES CLB 190 CLK NAME FUNCTION USAGE REMARKS OPERATOR FILES CLK Starts the JOB SCHEDULING program Program automatically schedules initiations of other jobs such as CCP mts clk Normally started automatically by LAS This job should not be started manually Use CLR instead OPERATOR FILES CLK CLN FUNCTION USAGE REMARKS OPERATOR FILES 191 Cleans out old information from the files used to hold machine check and unit check information then starts CMD mts cln automatically started by INIT Normally started automatically by INIT OPERATOR FILES CLN 192 OPERATOR FILES CLR FUNCTION Restarts JOB SCHEDULING program CLK USAGE mts clr REMARKS CLR is used if JOB SCHEDULING program must be restarted at a time other than reload OPERATOR FILES CLR OPERATOR FILES 193 CMB NAME CMB FUNCTION Program to combine statistics tapes into tapes of format MTA Tapes to be combined are produced by STA This program produces tapes which are blocked in a rather none standard manner However most system programs which read STATISTICS
277. e HASPLING servicing the device RESTART the device DRAIN the device FLUSH the device NOTE This sequence of commands should never be used with an SAnn device or RJE One should use the STOP SAnn command wait fora drained confirmation and a kill coded stop and then use START SAnn NOTE The order of the above commands is very important FLUSHing a device which has not been DRAINed can cause loss of output This command must be used with extreme care Before using the FLUSHed device again make sure that the FLUSH job is not executing anymore To do this enter TASKS nnnnn where nnnnn is the task number assigned to the FLUSH job If the system responds with JOB NOT FOUND then the FLUSH job has terminated and the associated device is now available for use by HASP Otherwise wait a while and try TASKS again until the JOB NOT FOUND message is typed out The device i d of the device that seems to be failing JOBS PROGRAMS 109 EXAMPLE stop 3 task 3 has device ptrx restart ptrx drain ptrx flush ptrx JOB PROGRAMS FLUSH 110 GOOSE NAME FUNCTION PROTOTYPE USAGE PARAMETER EXAMPLE JOB PROGRAMS GOOSE JOB PROGRAMS GOOSE To simulate an ATTENTION INTERRUPT on the operator s console during an MTS session at the console or to interrupt a batch task GOOSE nnnnn MTS sessions being conducted at the operators console can be interrupted To get the effec
278. e additional paging devices to the PDP PROTOTYPE PGET device i d USAGE To allow additional paging devices disks or fixed head files to be added to an already running PDP without stopping the PDP PARAMETERS device i d of the disk 2301 2305 3805 3825 or 4305 to be given to the PDP Example pget fb00 SYSTEM RESPONSE FB00 ACQUIRED Note While it is true that we don t currently have any 2305 type devices it isn t at all clear that we won t in the future Almost all of the paging devices on the market today are 2305 type devices and when we benchmark machines we are almost always provided with 2305 type paging boxes Note Before adding a 2305 or 4305 to the system via PGET it should be formatted using the program INIT FMT2305 which is permitted read to ccid W035 First make sure that the hardware itself is online Next put the device online using all eight addresses or at least the first and eighth address e g online ff08 ff09 ff0f or online ff08 ff0f The program is invoked with a run command A device will be asked for Devices should be entered one per line Only one device may be added at a time i e per line EXAMPLE JOB PROGRAMS PGET JOBS PROGRAMS r init fmt2305 EXECUTION BEGINS 133 OPFMT ENTER DEVICE NAME TYPE ff16 PFMT FORMATTING FF16 OSECTOR NUMBERS ARE 03 3F 7B fendfile If a device is unavailable i e offline then a comment will
279. e entire device table printed in hex DISPLAY LOCK n where n is a lock number from 1 to 15 Unused was command statistics lock VOLGET lock Unused was 3270 lock Unused was PLOT lock unused was BUFALLOC lock Tape mount queue was ACTIVCNT lock SYSTEM SYMBOL TABLES and ENDSEG2 lock STAT job lock was FRED words lock Open file table lock Catalog extension lock Unused For 0S 360 Queuer HASP CANCEL processor lock Console lock for MOUNT unused at UM Catalog buffer for public files Create scratch file directory lock Unused DNOPFWNPRPOUW WANA PWNE Il DISPLAY name DISPLAY R location disp disp DREAL R location disp disp This command may be used to display the current value of the predefined variables described below The values printed are in decimal For bit flags a zero indicates off and a one indicates on Example DISPLAY UAM n n 1 1 These commands cause n decimal words starting at the memory location specified to be displayed in hex The DREAL command always displays real memory The DISPLAY command displays virtual memory unless the address to be displayed is prefixed with an R Examples D 203140 D 203140 1B4 DR 40 2 SYSTEMSTATUS 475 D R40 C0 FIO dev name dev addr csw This command causes a fake I O interrupt to be generated for the device specified by dev name or for the hex dev addr using the hex
280. e errors will invoke the Machine Check Handler package and possible error messages operator actions etc are outlined in the following pages Some CPU problems prevent anything from running including the machine check handler But occasionally the CPU will just stop with no indication as to why There may may be no response to a request etc The attempt to take a dump or run Serep will usually indicate whether hang was due to Hardware or Software Actions Required of Operator When Machine Checks Occur Machine checks will be indicated by the appearance of one of two messages on the console followed by 0 or more lines of comment ABNORMAL SITUATIONS SUMMARY 311 The following is a brief list of the messages that can be expected to come from the machine check handling code kMCHK INCOMPLETE LOGOUT Insufficient information logged by the machine to recover from the machine check MCHK TIMING FACILITY DAMAGED Failure in the system clock or other hardware MCHK SYSTEM DAMAGE Serious machine check that precludes further processing Printed when the system damage bit is on in the machine check interrupt code MCHK EXTERNAL DAMAGE Hardware failure during channel or other operation outside the CPU itself Printed when the external damage bit is on in the machine check interrupt code MCHK RECOVERED MACHINE CHECK Machine failure that was corrected by the hardware Printed when the system recovery bit is
281. e jobs continue to execute MTS batch tasks are not affected by SHUTDOWN SHUTDOWN ALL has been used a system IPL is just a STARTUP shutdown shutdown all shutdown part For the UMnet MERIT Network PCPs if SDM or SDB cannot be run Go to as the a terminal which can access the network such IBM PC near the operatrs console At which host type net hg net wu signon or if HG can t be accessed w035 If using HG machine do the following JOB PROGRAMS SHUTDOWN 140 JOB PROGRAMS sou mnop sdm and or sou mnop sdb If using WU do the following sou mnop sdm and or sou mnop sdb After you have done the sou enter a one line message as you do when using SDM or SDB from the operators consoles This procedure is used to enter a one line UMnet UMnet message which will be printed for any user attemting to access which ever system is down It will not disconnect any users currently connected to the UB or UM systems There is no shutdown facility used here for the PCPs since even though the system may be down at UM the MERIT network is still in use JOB PROGRAMS SHUTDOWN SIGNONM NAME FUNCTION PROTOTYPE USAGE NOTE PARAMETERS EXAMPLE NOTE JOBS PROGRAMS 141 SIGNONM To print a signon message to all users before they sign on SIGNONM This job program allows the operator to specify that a certain signon message is to be
282. e manual associated with the box for the correct number KEY IN 2B00 HEX NUMBER OF BLOCKS FOR THE DEVICE 2B64 LEAVES 40 SPARES 2BE4 leaves 20 spares 2C5C or any higher number would leave 2 or fewer spares and is invalid anana press ENTER e press CLEAR This re enables FUNCTION CODE ENTRY mode e key in BF the function code in hex for terminating a COLD START press ENTER DEFD should be displayed on the panel indicating VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 69 that the 3805 is defined and a WARM START is permissable WARM STARTs NOTE If the 3805 is enabled a WARM START will automatically be done if the 3805 is powered OFF and back ON An INTEL publication lists the following as reasons to procede with a WARM START power loss hardware reset sevice panel RESET or COLD START diagnostic microcode is reloaded fatal control WARM START al ler error location fails to complete the 3805 depletes all eligible spare memory WARM START procedures e Make sure that the 3805 is physically offline e PRESS CLEAR key in FE this is the FUNCTION CODE for performing a WARM START press ENTER This should result in a display of ALLOC On At the completion of a WARM START FBA 200 should be displayed This takes about 4 minutes Function codes 90 and 95 There are at least two FUNCTION CODES that should b
283. e name of a drive on which a plot tape is mounted e g T911 The program will print several lines of tape information and then prompt for another tape name Enter an end of file CNCL to terminate the program mts plb ENTER TAPE NAME t911 TAPE MON 2000 LAST USED 11 06 78 22 00 11 MIN REUSE TIME 6 DAYS 0 HRS OMIN NEW PLOTS REPLACE OLD ENTER TAPE NAME When running the program from somewhere other than at the Operators console run sys pident When promted for a tape name enter either the name of the drive e g t901 or a pseudo device name e g tape The program reads tape names from GUSER and prints its output on SERCOM ERROR MESSAGES OPERATOR FILES 263 NOT A TAPE The tape name entered was not the name of a drive or proper pseudo device name or the drive was busy NOT A PLOT TAPE either the volume name was not plot or the name of the first data set was not IDENT FORMAT ERROR ON PLOT TAPE The tape was a plot tape but some error was found in the identifying information in the first file on the tape The tape should be re initialized before being written by CCP OPERATOR FILES PLB 264 PL NAME FUNCTION USAGE REMARKS OPERATOR JOBS PLO OPERATOR FILES PLO Runs PLOTQUIK mts plq This program plots which have been queued by the user with PAR ASAP The program was designed to let people get their plots out sooner than
284. e reason for reloading can be optionally specified also This program is normally run from MNS when the MTS job servicing UMnet interactive and terminal jobs is started from xDL after the UMnet PCP has been dumped and reloaded and from MSD at the U of M each night at midnight to set the dates in all UMnet PCPs The program obtains the current date using the TIME subroutine obtains a reason for reloading from the PAR field of the RUN command or by prompting on SPRINT and reading one line from SCARDS mounts a privileged outbound UMnet CC connection resets any reply message for the primary host in which it is running if a reason but no host name was given in the PAR field sets the time and date in the PCP and prints a nicely formatted message which includes the time date and reason for reloading on the PCP operators console If logical I O unit 1 is assigned lines are read and passed unchanged to the MTS CONTROL subroutine These lines will normally begin with a 2 3 or 5 followed by a CC command and could be used for example to reset a Hermes message HMESS after the PCP is reloaded If logiacal I O unit 0 is assigned the file or device is used to log a one line message giving the time date and reason for reloading followed by any 403 commands read from lodical I O unit 1 together with any output that the commands may produce When the program terminates the network con
285. e signonmessage is present type next This will print the number and contents of the next message Note the message number and type ret nnnnnn display full Destroying a Signonmessage If you want to get rid of a Signonmessage for some reason before the expiration date time do the following mess set priv on ret signonmessages Note the message number for example 976431 destroy 976431 destroys signonmessage After this you will be promted for the next message if there is SIGNON MESSAGE MESSAGESYSTEM 465 one You could destroy the next one in the same way if you need to SIGNON MESSAGE MESSAGESYSTEM 466 SUPERVISOR SENSE FLAG VALUES Add Values If Necessary EF 00 No Sense Requested Program Error 01 Good Sense Data 02 Unit Check On Sense Command 04 Device All Fouled Up No Sense Possible 08 Interface Control Check 10 Channel Control Check 20 Channel Data Check 40 Protection Check 80 Channel Program Check SENSE VALUES 467 SUPERDUMPS If the system is completely hung up and nothing can be done short of reloading the Operator should force a Superdump Of course before forcing a Superdump a Systems Programmer should be found and consulted if possible The Systems Programmer may be able to remedy the situation without the necessity of a relaod If however a Superdump must be forced Select the A function of the OP menu of the macrocode The ab
286. e stuck on a remote console read see the section Remote console read stuck for how to recover from that situation Remote console write stuck To avoid running out of buffers HASP imposes a limit of 10 messages that can be queued for any one remote If this limit is exceeded for example if the remote is experiencing hardware or software problems then the WTO write to operator routine in HASP will wait until a buffer is freed This wait causes the console processor to be suspended thus blocking any new commands from being accepted The dump analyzer program HASP D A can be used to detect this problem and to determine which remote is causing the trouble Issue the following commands Run Hasp d a D PCE TYPE 1052 ALL This displays all the console processor control elements PCEs You should see the contents of a message buffer displayed for one of the PCEs If the preceding field PCEEWF is flagged 1052 instead of WORK then the PCE is waiting to send a message and it s likely that the console processor is hung The text of the message in the buffer might include the offending remote number If not you can determine the remote number from Hasp s globals Issue the command D GLOBALS Check the field WCONBYTE The low order 7 bits is the hex value for the remote console number that Hasp is trying to send a message Display the remote table for this RJE via D RTAB where is the remote number in dec
287. e used by the operations and programming staff to check on the configuratio of the INTEL 3805 Two of the FUNCTION CODES are 90 and 95 Unlike COLD START and WARM START the FUNCTION CODES may be used while the 3805 is online and in use by the system Function code 95 The VERSION NUMBER message FBA nnn should be showing on the 3805 display panel press CLEAR VIRTUAL MEMORY PDP 70 SYSTEM SHUTDOWN AND START UP PROCEDURES This should display FCD 00 FUNCTION CODE entry mode key in 95 press ENTER Displays DA1 e key in 0000 press enter Displays FCC 0700 returned press CYCLE Displays DD1 00DO press CYCLE Displays DD2 0000 press CYCLE Displays DD3 1B64 this shows that 7 parameters were channel A address there is no channel B address the number of blocks assigned during COLD START this is machine dependent press CYCLE Displays DD4 0000 press CYCLE Displays DD5 1000 press CYCLE Displays DD6 0004 e press CYCLE Displays DD7 0000 press CYCLE Displays DD8 0000 press CLEAR the size of each block number of blocks per partition the device number entered on input logical device number logical device number Displays FCD 00 FUNCTION CODE entry mode key in OF VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 71 press ENTER Displays FBA nnn FUNCTION CODE 90 FUNCTION CODE 90 shoul
288. e using the green reset button turn off the switches on the wall power panels which control power for the computing equipment This will prevent possible problems when power is restored to everything all at once After the above has been completed the green reset button above shunt trip breaker may be reset SECURITY CLERK 463 SIGNON MESSAGE Using Messagesystem This method of issuing a Signon Message differs form the SIGNONM JOB program in that it is sent through the Messagesystem Only certain privileged CCIDs can issue a SIGNONMESSAGE The message may be placed in the system with a HOLD option that will keep it from appearing until a given time and date Also an expiration time and date may be specified so that the message will stop appearing after a certain time and or date The message may be constructed of more that one line but should be kept as short as possible for the users sake This type of sigonnmessage is usually issued from a terminal rather than form the operators console Example mess Set priv on Send signonmessage to everyone holduntil 3 00 pm jun 8 1994 Note You can use a time holduntil 3 00 pm if it s the same day You can use a date or both a time and date EXPIRATION DATE TIME system asks 6 30 pm oct 5 1983 At this point you may receive a warning message stating that the expiration hold date and time are close together Text system types Next expected sy
289. ea aah Coase A EL ot O LL asd Ae oC ge Te a Sd 560 ae ee AY bar Ea ca ESA RARO AL Ea rA ea a er ee ter et ae ee ee en eee ae PL A a A teresa hs EROS ras A A A AA ees wea Steel Rea 268 FRE Oe sh deh NR NN 270 KREP ae eM A RR OR RE 274 RIRELBE seh ie 2 a a is ia O a e Werk cated av boa eke 275 RRM Ges a Wile cane Bal a a ta 276 PRMD A A A Nine oe ples a Sots tier A at AA AA A A Ning tesa 2h ERMP La A deena ne See head AS ogee eh Fee eee dak re eet ees hee we Bae ead wth 278 PRM GSES A NS A a tans A A tale ghee ts ed ste 279 FROT A A A le hs a ake EB Pl a eve ae eee Beart A le Tet eB At 280 PRSNE A E aucune IS E EI ES anit wise E a Siete BR ee a eens 282 ESAV ist A Moe we hie A A SA he ee eid See hee a 283 ESB sien Saadeh E cent Gnd le 284 ES OM tee See E sate taal tats a RAMA Asa eto A LA Aan e AE CI ls ed Blan A to ROM hee 286 ESET Oo soe A RA A Sao ae ae Greene Gee ae Ces Da Gales Deo 288 ESTA ti a E Beate 289 A O NR agate 292 GTM Fins ae a eas actu tages tds dd Ae O eet E NEE 293 RIM da ta bce eA co ea A la e ta da ee ele Oe 295 EG VEG A O wee Shi ueice ro o des cb Sack ae eee eee 296 ESVIV Sisal it Sete Gui A tals eae Bt Reta A cael A Ste A a uh Sohal eg 297 TN A A A A oa oe Sha tout iria 299 TPD de aak Ad Bove dah A EA e O A hare Loe ed iaa 300 ETEO A e a do 302 PS Tera A E TR RI IO 303 FESN A A aa AAA A A BPC kee ar adenda 304 FUND A A SS A a a oe 306 ABNORMAT SITUATIONS SUMMARY lt a di shee a ae SS 307 TND EP
290. ease it Then it will issue a normal mount request for the new tape DWB will ask for a new tape when it gets to the end of the current tape or when it gets tape errors on the current tape If you think the program is starting new tapes too frequently tell the firefighter DWB is probably also sending messages to the firefighter about the problem but do it anyway If DWB mounts a tape and then shortly thereafter releases it and mounts it again check to see if the message DWB CMDSTAT TAPE ERROR PLEASE TRY ANOTHER DRIVE appears on the console If it does DWB has gotten an error writing the tape this error may or may not show up on the console Mount the tape on a different drive this time If CMDSTAT 328 the message does not appear something is probably wrong tell the firefighter If DWB prints the message DWB IS ANOTHER COPY OF DWB RUNNING Do a t u sys to see if there is another copy Respond Y if there is and N if there isn t If another copy isn t running something is wrong and DWB will probably print another message shortly If DWB prints DWB NO ROOM IN CMDFDIR AFTER nnnn CALL SYSTEMS PROGRAMMER Call someone Don t enter any response to the program just let it be until the programmer can look at it It has successfully written the tape but is unable to update the directory If DWB has trouble reading or writing some file it will print DWB ERROR ON fdname CALL
291. ed again only when an appropriate drive is available The user is returned to the queue in this case with no message printed on the terminal The QN response will cause the next request to be presented since this one cannot use the available drive If the user requests more than one drive but not enough drives are available each request should be queued separately The operator should QB the first request when that is satisfied the system will present the next request which should then be queued if necessary and so on The user will be promted to see if it is OK to queue the first request but subsequent ones will be queued way the operator need not leave several tape drives idle waiting for enough of them to completely satisfy the request Before a request is initially queued the operator should check the rack number and tape i d to make sure that they are correct so that a user won t have to wait several minutes in the queue just to get an IR response If the user did not specify the density the operator should do this on the QB response so that it won t be popped until an appropriate drive is available The following are SYSTEMSTATUS commands and are included in the SYSTEMSTATUS section of this manual They are mentioned here because they are part of the tape mount queueing process tapequeue prints size of queue tapequeue list list the current queue tapequeue pop 6250 800 releases top job in q
292. ed fashion i e there ig a maximum size The limits imposed have no relation to the physical real memory size of the computer MTS has a comprehensive accounting facility as an integral part of the system Only authorized persons who have been assigned a signon id account number and project number can use the system At registration time maximums can be established for such items as disk space terminal time and plotting time In addition an expiration date can be set for a user useful for students taking a programming course as well as a maximum dollar charge The charge is debited from the user s account for each job run Charges for the job are based on such computer resources as CPU time memory usage connect time for terminals cards read lines printed and cards punched Note Cards are all but obsolete Additional charges are made for disk storage Users have access to their own accounting information at any time A facility also exists whereby a project director or course instructor can allocate and manipulate funds disk space and other resources among users within his project OVERVIEW OF THE SYSTEM 21 The Job Program and the Task A job program is a program which can be started from the main operator s console or at the request of any running task A task is the running invocation of a job program There is one job program for each system function depending upon the function that a job program is s
293. ed on the main or backup if active operator s console The system requests that a tape be provided for a superdump a dump of all of real memory Following the superdump the system terminates operation and must be reloaded An operator can force the system to take a superdump at any time by sending a special signal to the CPU Normally this is only done when a system problem occurs and all possible recovery measures have failed On most machines the signal consists of a restart interrupt the RESTART key On the ES3090 300J use ALT RSTRT SUPER DUMPS 24 SYSTEM SHUTDOWN AND START UP PROCEDURES Mainframe and Peripheral Power Powering the System Down After the systems have been shut down it is possible to turn power off and on to the IBM ES3090 300J and the I O equipment To turn IBM ES3090 300J power off 1 Go to the Processor Controller and press the yellow Power Off button The above action turns off the CPUs the processor and the MGs but not any of the I O equipment such as disks and tapes 2 Power down the IBM 3380 disk drives go to the 3880 IBM Disk Controller and turn the Subsystem Power switch off This should power down the IBM disk system 3 Power down the IBM Tape Drives go to the 3803 Tape Control a Turn b Turn Unit the DC switch off first the AC switch off 4 Power down the Amdahl disk equipment a Make sure that the Amdahl disks are offline to the sys
294. ee 297 ENR ne N SRa pple Prine dr A eel ira Ie bak che ah he a Feet 299 TPD Tape Tape Unit Diagnostics ooooooooooooooooooo ooo 300 TPS Displays Amount And Type Of Free Tape Units 302 TSH Add Tapes To Or Remove From UNSP TAPESHARE 303 TSN Weekly Full File Save Slave Stream o oooooooooooo ooo 304 UDP LOock Unlock MSG rocosas e SL Sw Es 305 UNU Enables Terminals On NUBS UNYN SCPs amp Job Stautus Monitor AS arabes A wae A A SS E E ose 306 Overview OE The System A ee ich Be ee Oe soe AS 18 Paging Devices Virtual Memory das Sake cos bees be ae bad 66 PEO RUNS gt BUOTQUIRK aps eink cod abi Seek a e A 423 PEO s RUNS QUICK PIOUS fc ese O 6 Soe he SNS Oe ap Se ha eee es See eae 264 Power Fluctuations Failures ReECOVeEry oooooooooooooo nooo ooo 85 Powering On Loading Procedures 9790 cece eee ec cee eee eee eee eee 560 16 Power Up Down FSS 090 3 000 siose A SS SS ee See SS E S 24 PWCs System Wart COUnESs di td ge he ee eS 265 Super Dumps Most Of System Dumped IPL Necessary After Dump shai A E O Smale SY stabs URS 23 System Shutdown Power Down Procedures cece ec ee eee ee eens 62 System Start Up Procedures EMERGENCY POWER OFF o o ooooooooo o o 65 TPEREADER 204 a a Ga wa vane cos cep Lea reales A A A A haad ae E i 77 System Startup Procedures INTEL 3825 oo ooooooooooooooooooooooo 72 Tape MOUNT QUEUEING a esna ge
295. emely heavy load the CPU time required to manage Paging may rise to as much as thirty or forty percent of the total available time When this happens user pages reside in real memory for relatively short periods of time and thus user programs run for relatively short periods of time until a reference is made to a page that is not in real memory This condition is called Thrashing and is very undesirable INTRODUCTION TO MTS 18 INTRODUCTION TO MTS THE MICHIGAN TERMINAL SYSTEM OVERVIEW OF THE SYSTEM The core of the MTS system is a program formally called the University of Michigan Multiprogramming Supervisor UMMPS UMMPS has the characteristics of a supervisor in that 1 it operates in supervisor state and is almost the only code in the operating system that does so 2 it deals with all interrupts and 3 it allocates resources such as devices CPU time and storage UMMPS executes JOBS which are usually initiated from the operator s console Each job runs in problem state and calls upon the supervisor for its services In the following the terms JOB and TASK are synonymous We shall call a JOB PROGRAM that section of code that can be invoked with the appropriate command from the operator s console Each invocation of a JOB PROGRAM is called a JOB or TASK UMMPS oversees all the computer s multiple user oriented activities including a system called the Michigan Terminal System MTS Although MTS is a complex set of
296. ems I II and III will have to be restarted manually If a power failure occurs 1 Wait 30 seconds and hope that the power comes back on 2 If power does not come back on broadcast UM amp UB to everyone telling them of the power failure then give them two minutes to sign off 3 Run Shutdown to keep new people from signing on 4 If power is still off after 2 1 2 minutes force people to signon off using shutdown all UPS UNINTERRUPTABLE POWER SUPPLY 550 stop 5 When everyone is off press for the machines to stop 6 Do a rs sys on each system xxx From this point on HURRY kk 7 Go to the ES3090 300J 8 Power off the other unneeded lights 9 Wait for the power to 10 Call for help to get consoles and equipment in come back on on the CPUs then wait power the machine off the building as well as things restarted UPS UNINTERRUPTABLE POWER SUPPLY 551 PB I PB I Status Panel The Status panel on PB I consists of four sections A The Alarm section for abnormal functions 1 OVER UNDERVOLTAGE The light will illuminate and the audio alarm will sound if the voltage deviates outside the calibrated settings 10 low 10 high 2 PHASE SEQUENCE LOSS The light will illuminate and the audio alarm will sound in case one of the following conditions occur a Incorrect phase rotation b Loss of any one phase c Extreme low voltage
297. ence obtain the system s attention tell it what to do how to do it and what data to use The command language employed is essentially the same whether the user in the past punched it on a card for batch processing or types it directly on an interactive terminal For example every user starts with the command SIGNON and ends with the command SIGNOFF For convenience these and other commands can be abbreviated Input data and programs may be included in a source deck or they may be stored in the MTS file storage system Three major types of files exist in the MTS system 1 public files 2 private permanent files and 3 private temporary files Public files which usually contain a system program such as the FORTRAN compiler in loadable form are available to all users For the most part they can read but not change the contents of public files some public files can be extended or added to by the normal user in this manner information can be added to an existing file but none of the previous contents will be destroyed Private permanent files are normally available only to the user who has created them to read from write into modify or destroy as he wishes Access to a user s private permanent files is through the use of his i d The user may through the use of the PERMIT command expand access to or restrict access by other users in various ways one may even restrict one s own access to one s own file Tem
298. ent cylinder and head used in the channel program gg is the hexadecimal value of the channel command the drive was trying to perform when the error occurred Typical values for this field are 07 seek 1B seek head 23 set sector 1F set file mask 31 search JD equal DAC UNIT CHECK ERROR CODES 364 05 write 06 read Disk Problems UB If UB hangs or crashes or if the UM system shows a long Standing I O to the disks do the following before doing anything else 1 Write down the number of the disk s which are not ready 2 Write down the status of the system from the console clocks on off process stopped wait 3 Look at the channel activity lights on the disk control units that hold the not ready devices and write them down If the system is not in a clocks off situation try to contact Diane Steve Burling or Kevin Bosley and have them take a look at the problem If the system is in a clocks off situation and none of the above people are available go ahead and reload but run CTD before letting the lines out DAC UNIT CHECK ERROR CODES 365 DUMPS PRINTING PRINTING JOBDUMPS AND SUPERDUMPS The program in the file MTS PRINTDUMP is used to print the jobdump and superdump tapes produced by the supervisor Superdumps are taken automatically by the supervisor when it detects a fatal error within itself all of real memory is written on the tape and system operation term
299. ent 262 SYS PLTEST36 441 SYS PLTEST51 441 System Damage 311 System Reliability 468 System Shutdown Power Down Procedures 62 SYSTEMSTATUS 471 System Status Panel 552 System Status Recorded Message 469 System Wait Counts 265 S8 Cards MTS 268 TABLMOD 147 Tape Commands Operator 149 BSF 149 BSR 149 ERG 149 FSF 149 FSR 149 REW 149 RUN 149 WTM 149 Tape Copies Filesave 215 Tape Delivery 349 Tape Density 232 234 Tape Directory Project 507 ADD 510 CONFIRM 511 CREATE 512 DELETE 513 Directory File 507 Directory Subroutines 516 FILE 513 GET 513 HEX 514 INIT 514 LIST 514 MTS 515 PURGE 515 RECREATE 515 REMOVE 515 STOP 516 Tape Directory Program 510 Tape Drive Diagnostics 300 Tape Drive Error Analysis PM2 STC 446 Tape Drives Attaching Detaching U 37 Tape Drives Attaching 3090 31 Tape Drives Detaching 3090 31 Tape Error Codes 494 Tape Error Message Text amp Explanation 496 TAPE ERROR ON VOLUME 413 Tape Errors 493 Tape Errors CMDSTAT 330 TAPE HANDLING 486 Tape Initialization 232 234 Tape Label 232 234 Tape Label Data Base 189 Tape Label Retrieval 189 Tape Labels 214 Tape Labels Finding 214 Tape Labels Rules For Giving To User 214 Tape Mount Queueing 490 Tape Mount Responses To 488 Tape Mount Statistics 241 Tape Mount Suspension NTM 499 UAM 499 Tape Owner ID 232 234 Tape Queueing 491 Tape Racks Electic Problems 502 Tape Rack Strip Labels 503 Tape Retrieval Subm
300. ented when an invalid or an uncomparable character is printed Normally data checks should be blocked with MTS NOMSG supress the verification message normally printed on the operators console after the buffer load is completed pn ptrx pn ptrx echo pn ptrx nomsg FU NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES JOBS PROGRAMS 135 p11 To load the P11 3211 type print chain train or band image into the UNIVERSAL CHARACTER SET buffer on the printer controller The buffer is echoed after loading to verify its contents P11 device 1 d 8LPI If one wishes to change chains trains or bands on a printer it is necessary to load the UNIVERSAL CHARACTER SET buffer in the printer control unit with the correct character set If the UCS buffer gets scrambled for any reason it should then be re loaded In addition at system IPL time it is wise to re load the UCS buffer The contents of the UCS buffer are echoed on the associated printer for verification Normally the printer operates at 6 lines per inch but this can be changed to 8 lines per inch if the 8LPI parameter is used 1 The device i d of a printer 2 8lpi optional p11 ptrx p11 ptrx 8lpi JOB PROGRAMS P11 136 NAME FUNCTION PROTOTYPE USAGE PARAMETERS OPTIONS JOB PROGRAMS QN JOB PROGRAMS QN To load the QN 1403 or 3203 type print chain or train image into the Universal Character Set
301. er this parameter may be specified as NONE Note that the parameters to OPERATOR are not separated by commas operator ds02 operator operator ds02 none The most usual ways that the OPERATOR command will be used at the U of M Computing Center are as follows JOB PROGRAMS OPERATOR JOBS PROGRAMS 129 Example 1 The OPERATOR job must be stopped Press the INTERRUPT button on the ES3090 300J console to the backup console Type OPERATOR Press enter Example 2 If the 3270 in use as the operator s console won t respond sign on at any terminal sy t n operator 0002 OPERATOR 028140 DORMANT DSO2 ptrx e O stop 2 t n operator JOB NOT FOUND e O operator e t n operator 1627 OPERATOR 029151 DORMANT DS02 ptrx e s signoff JOB PROGRAMS OPERATOR 130 JOB PROGRAMS PDP NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLE JOB PROGRAMS PDP prefix To initiate the Paging Drum Processor This is the task that does all the system paging This is an absolute non reentrant job so there can be only one active use of the PDP job program at any one time PDP The Paging Drum Processor is automatically started in the IPL sequence TE 1E as restarted it will attempt to regain control of the fixed head file and disks it had when it was first started and also will start using any others that are now available If the PDP is started without any type of paging d
302. er on the front of the 3825 that holds the keyboard and write down any error message displayed there 4 Make sure MTS isn t using the 3825 this isn t normally a problem because if MTS is using the 3825 when it dies MTS hangs and must be relPLed 5 Flip the enable disable switch on the 3825 s front panel to the disabled position and wait for the green ready light to go out 6 Flip the reset toggle switch for controller 0 the switch is located in the drawer with the keyboard 7 Open the small door at the back of the 3825 a dime will to the trick and press the black fault reset button this clears the status indicator on the front of the 3825 8 Flip the reset toggle switch for controller 0 again 9 Type FCAO at the keyboard and press return The word DEFINED should appear Note If the above procedure fails to work after two attempts power the 3825 off and back on Warm or SLOW Start 10 Type FCFE at the keyboard to request a warm start The words ALLOC BSM 0 should appear 11 Wait several minutes and when FBA nn n is displayed flip the enable disable toggle swtich to the VIRTUAL MEMORY PDP 74 SYSTEM SHUTDOWN AND START UP PROCEDURES enabled position If all has gone well the green ready light will come on 12 That is all there is to it if MTS isn t running If MTS is running you need to online the 3825 and tell the PDP to start using it From the operator s console
303. er the last totime A statistics record is matched accepted if it satisfies any item thus the items are in affect OR ed The caller of STAT may provide a routine to inspect every accepted record and determine a final acceptance The parameters for STAT are ITEM A beginning of item list EITEM A end of item list lt not the addressing of the last item LENGTH F length of an item ACCEPTOR A routine to call to detrmine final acceptance of a statistics record A 0 not a 0 in the PAR list if routine is not supplied LINES 2F on return beginning and ending line numbers in SSTAT of statistics records accepted The return codes from STAT 0 done 4 no records found 8 unable to complete The routine whose address is in acceptor is called as follows gril a item matched by STAT record gr2 a stat record gre return address return 0 gre to reject the record and 4 gre to accept it grf address called Registers gr0 and gr3 through gr6 may be used All other registers should have the same value on return as on call The routine called must not use any of the routines whose addresses are in the TV except LESPAN and LENUMB The routine should always return STAT may have tapes mounted or files locked Control is not guaranteed to return to a command processor after an attention the command processor may be re entered from the top SCAN 460 The SERCURITY CLERK SECTION IS PRESENTL
304. eral PCPs at the University of Michigan Computing Center The U of M PCPs are connected by leased phone lines to Similar PCPs at MSU WSU and WMU in addition to many SCPs located in Ann Arbor and around the state The PCPS at the Computing Center are also connected to each other by synchronous communications links The function of a PCP is to allow several users to simultaneously transfer data to or from a remote host on the network At the U of M the PCPs are interfaced to the ES3090 3003 via a multi address host interface device attached to a ES3090 300J channel The PCP is connected to the network via two telecommunications transmitter receiver pairs one pair for each line to another PCP or SCP A PCP CPU is at the heart of the PCP The PCP executes the communications computer operating system called CCOS which co ordinates the different devices in the PCPs Some of the major functions performed by CCOS are 1 The multiplexing of information over the host interface and over the phone lines 2 The automatic retransmission of erroneous or lost data over the phone lines 3 The automatic re routing of data over a second phone line if the primary line becomes disabled HERMES SYSTEM The UMnet Communications Computers are equipped with hardware and software which enables terminals to be dialed in directly to them This facility has been named HERMES Hermes appears to a user as a Data Concentrator servicing seve
305. eral facts about the device and about the control unit and channel mentioned in the path specification PART 5 The device status information This part appears only if the S B Wor A options follow the units command AND if the device belongs to some task It gives more specific information about the current state of the device PART 6 The error information This information appears only if the Vv option is in the units command AND if a possible inconsistency has been discovered in the Supervisor tables Subject to the conditions just mentioned parts one through four appear once for every path to every device that meets the criteria given in the units command For instance if there are two paths to device D240 then a units command of D D240 will result in two sets of output one set for each path to D240 Normal output is one line per path If the S B W or A options are used PARTs 4 and 5 appear on additional lines If the V option is used any error messages appear on extra lines The actual output format can best be explained by JOB PROGRAMS UNITS OLD STYLE JOBS PROGRAMS these four examples EXAMPLE 1 Units D D240 161 CH 1 CU 14 gt DEV D240 ADDR 140 CH 2 gt CU 14 gt DEV D240 ADDR 240 These two lines are interpreted as follows First there are two paths to D240 Data can go either through channel 1 and control unit 14 to D240 or through channel 2 and control uni
306. erator to various abnormal conditions followed by a list of four character device names for those devices which are not ready i e which need operator intervention followed by the five digit job numbers of any jobs whose operator console output is being SIGNOREd see below This line will be intensified THE REQUEST AREA This two line area at the bottom of the screen is used for entering four kinds of requests 1 New job requests e g MTS HSP 2 HASP commands e g DISPLAY JOB n 3 System Status commands e g T M 4 Device commands e g IGNORE 17 Requests may be entered in this area at any time see the section describing the keyboard for details on entering requests and message responses When a request has been entered it will be moved to the message area and the request area will be cleared for input of further requests Thus the request area will generally be blank If there is an error in the request which can be detected by the console management program the request will be left in the request area for possible editing and re entering OPERATOR CONSOLE 3270 336 Note only one System Status command may be active at a time If a second command is entered while one is already active the active command will be terminated and the new one executed The time at the last screen update will also appear in the request area at the bottom of the screen MESSAGE AREA
307. ers on the ouput tapes If a second ouput tape is needed and is not filled then this should be used as the input tape for the next CMB run Filled output tapes should have a copy made of them The original copy 1 should be stored at the print building and the copy copy 2 in the racks at the Computing Center The format for monthly statistics tapes is VB 16379 4000 The format for the combined statistics tape is U 4352 32767 Older Version NOTE Generally this program is used on the first or second day of the month When an output tape is asked for a new tape labeled txxx 1600 statc umcmpctr should be used When input tapes are asked for the first one should be the monthly STATM statistics tape for the preceding month e g if it is Oct 2 use the September monthly statistics tape as the first input tape When a second input tape is asked for use the daily statistics tape which carried over from September to October The third input tape would be the Filecharge tape for September OPERATOR FILES CMB OPERATOR FILES 195 CMD NAME CMD FUNCTION Program to write command statistics and unit check messages into a file This program is automatically started by CLN USAGE Is normally started up automatically Requires only mts omd Takes no parameters NOTE See the CMDSTAT writeup for details and explanatiions Problems If the message THE CMDSTAT PICKUP PROGRAM CMD ISN T RUNNIN
308. ers when they attempt to repair a broken tape drive Gggggg is the tape rack number if the tape has been mounted by the MOUNT command If it wasn t mounted this way this field contains a dash Hhhhhh is the tape volume label if the tape is labeled If it is not then this field contains a dash Tiga Is the current dataset name on the tape which only appears if the tape is a labeled tape If not a dash appears here Jj is the file number in which the error was encountered This along with the block number see below identifies precisely where on the tape the error occurred Kk is the block number in which the error was encountered Lill Is the description of the nature of the error This is a line of English text which is terse but describes the cause of the error A list of each of these is given below along with the error code which appears in the first line of the error message associated with each ERROR MESSAGE TEXT AND EXPLANATION Code 0 Program interrupt looking at channel program This error is always caused by a program bug A Invalid or insufficient sense data This error is almost always caused by a tape drive which is broken so badly it can t tell what is wrong with itself It can also be caused by indiscreet usage of SVC CLRIO which is a programming problem TAPE ERROR CODES 497 Channel detected error This error occurs when the tape drive is operating properly or t
309. ertaken if stop failed This is done for both terminal and batch tasks First read the dire warnings under BLAST earlier in this manual Then invoke the BLAST job program with the SNARK ed job s task number The BLAST job program will irrevokably remove the SNARK ed task from this system Irrespective of the results obtained from the indication of the BLAST job program you may proceed to STEP 5 5 When a job has been removed from the system using the BLAST job program or if a message is returned from stop the accounting records have been somewhat disturbed And in most cases it will be impossible for the user with a CSID of the job which SNARK ed to start another task To permit this user to continue running it is necessary for the operator to use the MTS library file FIX to restore the user s account record This program is invoked by the command form MTS FIX The use of this program is described elsewhere in this manual 6 For terminal jobs the successful reinitiation of the terminal device requires that the operator enable the line This is similar to the operation performed by the MTS LAS or DCS at IPL time However the form here is to issue a command to the system of the form MTS xxxx where xxxx is the device name of the terminal in question as determined from the TASKS output obtained above For example MTS LA23 MTS CA11 MTS DS12 ABNORMAL SITUATIONS SUMMARY 310 HARDWARE Most CPU machin
310. es should be handled by the UPS You may want to call Security 764 3131 or even in some cases the Edison company to find out what is causing the power problems and how long they are expected to occur Recovery from Power Interruptions T On rare occasions a power failure will cause loss of all power including general building power no lights except the battery powered emergency ones This would be the case when the power outage lasted longer than the UPS batterys are capable of functioning In this case find out from Security how long the power will be down Of course when no power is available users cannot be informed POWER FLUCTUATIONS amp FAILURES RECOVERY 86 make SYSTEM SHUTDOWN AND START UP PROCEDURES via a recording etc how long they are going to have to wait for system availablility In some cases after power comes up the Air Conditioning won t work and the system hardware will be inoperative or the harware could be inoperative for other reasons In this case an appropriate recording should be made After power has been restored the lights should come on if they had been affected Check the little red neon lights above the power panel doors on the first and second floors to see if power is being fed to these breakers If either set of lights is off equipment power has to be reset To do this go to right of the doors at the north end of the second floor machine room Turn the key
311. es to keep the read there but doesn t want to tend to answer it at the time For example if there are several tape mounts waiting and you know that you can t take care of them for a few minutes but don t want them removed from the screen you can place a using the DUP key where you would ordinalirly answer the request Now when you move the cursor from one read to another the cursor won t stop at the read s with a after them THE KEYBOARD For the following discussion define a field to be either a message response line or the request area Editting keys These keys affect the display and allow editting of input lines without program interaction ERASE INPUT Clears all fields and positions the cursor to the first field This is the recommended method for deleting and retyping a line ERASE EOF Clears the field containing the cursor from the OPERATOR CONSOLE 3270 339 cursor position to the end of the field This may be used in combination with the backtab key to erase a single field or part of one without changing others or moving the cursor RESET CURSOR Positions the cursor to the first field on the screen which may be the same field if there is only one RESET Turns off the input inhibited light thereby unlocking the keyboard Also turns off insert mode TAB Positions the cursor at the beginning of the next field on the screen which may be the one it is in if there is
312. ess and device names is given 6 then the paths between the channel and the control units to which the devices are attached is marked logically offline JOB PROGRAMS OFFLINE 124 JOB PROGRAMS When a control unit a channel or the path between a channel and a control unit is taken logically offline devices may be allocated to and released by jobs but no new I O will be initiated using the offline units If there are other online paths to the devices through other channels and control units then I O to the devices can continue If not any jobs doing I O to the devices will end up in I O wait UNITS will show SIO PENDING until a path to the device is made available or the system is reloaded Even if there are other online paths to a device a job may still end up in I O wait if the channel control unit or path between the units was offlined during error recovery In this case UNITS will show the device with SIO PENDING ON SPECIFIC PATH until the path is made available or the operation is terminated using UNITS CLR The UNITS command may be used to check the status of Channels control units and devices Certain diagnostic programs can use a specific path to a device even if it is marked logically offline Control units should ALWAYS be marked logically offline before they are taken physically offline PARAMETERS AND EXAMPLES To offline devices offline ptrx offline ptrx rdrx To offline
313. et 401 SFT 288 SHUTDOWN 139 ALL Immediate 139 PART OPER file Jobs Continue 139 Shutdown Procedure 62 Shutdowns Scheduled 389 572 Shutting VSE Down 52 SIGNOFF QUIT 412 SIGNONM 141 Signon Messages 141 Signonmessages Messagesystem 463 SIN 441 442 SNARKS 308 SOFTCHK 142 288 Soft Error Recording CPU 142 SOFT MACHINE CHECKS 288 Soft Recording Disabled 311 Software Failure 312 Solid State Disk Errors 361 sos 481 Special Forms Jobs 547 Special Rates Holiday 451 SSD Error Codes 361 SSRTN 335 SSS Startup 231 STA 289 STA Failure 289 Staff Only Status SOS 481 STAR 441 443 STARTUP 143 STAT 144 193 STATC 193 Statistics 289 292 STA 289 Statistics CCID 454 Statistics Command 195 STATISTICS IS FULL 413 Statistics Line Monitor 293 STATM 193 STATSW 145 STC PM2 446 STI 292 STM 293 STOP 146 308 Storage Data Failure 311 Storage Protect Failure 311 Strip Labels Tape Racks 503 Submitting Tapes amp Diskettes By Mail 500 SUM 295 Super Computer San Diego 543 Super Dumps 23 Superdumps Forcing 467 Superdumps Printing 365 Superdump UM MTS 3090 27 Superdump VM XA UB 35 Supervisor Damage 311 SVS 298 See Filesave For Schedule And Tape Storage SVW 217 See Filesave For Schedule And Tape Storage SV1 297 SV2 297 SV3 297 SV4 297 SV5 297 SVW TSN 304 SYMTEST 441 445 Synchronous Lines 205 Sys pid
314. etters with the upper case only print train This option should normally not be used with MTS ALLOW allows data check to be presented when an invalid or an uncomparable character is printed Normally data checks should be blocked with MTS NOMSG supress the verification message normally printed on the operators console after the buffer load is completed tn ptrx tn ptrx echo tn ptrx nomsg H NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES JOBS PROGRAMS 155 T11 To load the T11 3211 type print chain train or band image into the UNIVERSAL CHARACTER SET buffer on the printer controller The buffer is echoed after loading to verify its contents T11 device i d 8LPI If one wishes to change chains trains or bands on a printer it is necessary to load the UNIVERSAL CHARACTER SET buffer in the printer control unit with the correct character set If the UCS buffer gets scrambled for any reason it should then be re loaded In addition at system IPL time it is wise to re load the UCS buffer The contents of the UCS buffer are echoed on the associated printer for verification Normally the printer operates at 6 lines per inch but this can be changed to 8 lines per inch if the 8LPI parameter is used 1 The device i d of a printer 2 8lpi optional t11 ptrx t11 ptrx 8lpi JOB PROGRAMS T11 156 JOB PROGRAMS UNITS UNITS output appears as follows 0850 PTR3 T
315. evice 2301 2305 3805 3825 4305 or disk no paging will be done but PAGE CONTROL BLOCKS PCBs will still be released In this case a condition may occur when all the available real core pages are in use and more are requested for allocation in such situations some tasks using virtual memory will have to give up some of it or the system must be re IPLed If given a 3 character prefix to be used by the PDP when it searches for paging disks to use If no prefix is given the prefix PAG is used PDP JOBS PROGRAMS 131 PDROP NAME PDROP FUNCTION To remove Paging Devices from PDP PROTOTYPE PDROP device i d device i d USAGE To allow paging devices Solid State Disks or Fixed Head Files but not paging disks to to be romoved form an already running PDP without stopping it PARAMETERS device i d of the equipment e g 3805 or 3825 to be removed from the PDP EXAMPLES pdrop dev dev pdrop fb00 fb01 fb02 gets rid of all of our presently owned virtual memory except for the paging disk SYSTEM RESPONSE DEV HAS BEEN DROPPED BY THE PDP NOTE The process may take a while since the all the pages must be removed form the device This job will not work for paging disks only for drums and similar devices One or more devices may be removed at one time i e per line See PGET to re attach VM devices JOB PROGRAMS PDROP 132 JOB PROGRAMS PGET NAME PGET FUNCTION To giv
316. evice numbers This command can sometimes fail for no apparent reason so if you get an error message just try again Attaching a device to UB MTS Once a device has been varied online to VM XA it must be attached to UB MTS before it can be used To attach a device type the following command on the VM XA operator console ATTACH nnnn UB MTS or ATTACH nnnn mmmm UB MTS where nnnn and nnnn mmmm indicates a device number or range of device numbers Loading MTS To start the loading process you should follow the instructions given in the section Loading UM MTS on the 3090 or Loading UB MTS under VM XA Once the message Do you want to run the current system is displayed on the MTS operator console proceed as follows 1 The message Do you want to run the current system is UB MTS UNDER VM 40 SYSTEM SHUTDOWN AND START UP PROCEDURES displayed 2 Normally you should just press ENTER 3 Several lines of text will be displayed There will be a short pause while MTS checks all of the I O devices If any devices particularly the UMnet PCPs are hung this pause may take up to 5 10 minutes 4 If you are running native i e loading UM MTS eventually the screen will clear and MTS will print the message Local time is If you are loading UB MTS this prompt will be suppressed since the time can only be changed by shutting VM down and reloading See the sections Shutting VM XA Down and Loading VM X
317. f a particular station LCL station or RMTn is specified the connect status of that station in addition to the information specified by ALL is printed The print and punch counts and queues reflect only the jobs routed to that station The execution queue reflects the entire execution queue not just the jobs submitted from that station Multiple parameters of the form shown may be entered separated by commas or blanks Examples Q 1234 Q CNTR Q RMT7 171324 SCREEN ALL LCL station RMTn The SCREEN command displays continuously the status of each batch PRINT and PUNCH job for the specified station If ALL or no parameter is specified the status of all jobs in the system is displayed The name SCREEN derives from the fact that a television like display of relevant job status information is generated The format of the display is as follows First the system SIGNON message if any is printed Next a summary line is printed consisting of the date and time the number of terminal users the number of batch jobs executing and awaiting execution and the approximate batch throughput rate Next the status of each job is given by a one character code following the job receipt number The codes are awaiting execution executing awaiting print printing awaiting punch punching being sent on network done page printer awaiting network send awaiting page print page printing done t
318. flag byte DCTFLAGS offset X 1C in the DCT in the line DCT For example SYS MOD addr 1C 80 where addr was determined above and assuming DCTFLAGS was 00 Once the console queue clears up and the SAnn line has successfully stopped it will be necessary to restart the line via START SAnn The following is an editted HASP D A session demonstrating the above steps for remote 22 BSSR In actuality the last few times this situation has occured remote 2 FLNT has been the culprit run hasp d a 12 17 53 why I don t know same as running HASP HASPSTAT d pce type 1052 all RM PRIORITIES CLASSES SCHEDULE 379 PCE AT 228728 1052 PCESAVEA 01000000 00000000 REGISTERS 13 gt 12 00228728 601C7938 901C7AEE 02014114 0021A504 00000000 001A919C 00000000 00000000 00000003 00232858 00000000 00000000 00000000 001A9018 001C770C PCEEPSAV 00000000 PCENEXT 0022877C PCEEWF WORK PCEID 1052 a couple of PCEs not shown PCE AT 230E04 1052 PCESAVEA 01000000 4021E634 REGISTERS 13 gt 12 00230E04 5021ACF8 5021AC38 00000000 00000000 00230FDC 001A91B4 00000001 00000005 A021C3BA 8021C35A 0021CA9C 00000000 001B62D8 001A9018 0021ABFO PCEEPSAV 00000000 PCENEXT 00230F20 PCEEWF 1052 PCEID 1052 230E58 000054 22005BD3 D6C7D5C5 E6D3D6C7 C6C9D3C5 LOGNEWLOGFILE 230E68 000064 40404040 40404040 40404040 40404040 230ED8 0000D4 84858440 40404040 4040F5F1 F2F440E2 ded 5124 S 230EE8 0000E4 D740C3D5 E3D94040 F
319. for restoration will fit in the disk space allocated to the user s CCID Restore operations are aborted if the user s file space limit is exceeded while a file is being restored The save tapes used by RES only contain files which have existed during the most recent past about six weeks However the Computing Center OPERATOR FILES RES OPERATOR FILES 271 maintains a separate set of tapes to provide file restoring service for older files that are not available through RES These older tapes are used with the program RST Commands The following commands are available for RES The underlined portion of each command may be used as an abbreviation RESTORE usid fname mm dd yy hh mm ss NEWNAME usid newfname or RESTORE usid mm dd yy hh mm ss NEWNAME usid or RESTORE any filename pattern e g usid c3m usid old usid x In the above examples usid filename would be specified as w000 file for example If only the file name is specified then the most recent version available of that file is queued for restoration The file will be restored with the same name unless a NEWNAME is given If res usid is specified all files under this USER ID will be restored If res usid newname usid is used all files from the first USER ID will be restored to the second If a patterned filename such as res w035 c3m is used all files owned by CCID w035 and beginning with the characters c3m will be re
320. formation essential to proper error recoverey to be lost This parameter is necessary on brand new disks however to insure that the alternate tracks are correctly formatted If this parameter is not given then the VIRGIN and the ALTERNATES parameters are illegal VIRGIN gt try to read home addresses but don t complain if they can t be found If the label is unreadable when this option is specified DASDI will continue on ALTERNATES nnnnn ALTERNATES nnnnn mmmmm DISK MANAGER DMGR 353 ALTERNATES NORMAL ALTERNATES KEEP ALTERNATES NONE the default gt DASDI will also format tracks which are marked as alternate tracks in the home address even though MTS does not use alternate tracks The first and second forms designate the range of the relative track numbers starting at relative track 0 for tracks to be assigned as alternates Note that if ALTERNATES KEEP is also given then any other track which is already an alternate will be kept as an alternate in addition to the track range explicitly specified You cannot explicitly specify alternates for 3340 3344 3350 3370 devices You can either specify ALTERNATES KEEP or ALTERNATES NONE for these devices You cannot specify anything other then ALTERNATES NONE unless you also specify WHA ALTERNATES NORMAL means that the track range for the alternates is as is found in the relevant component description
321. formatted in the same way as the STC 4305 when the system is loaded Usually when one looks at the operator panel on the 3805 FBA 370 will be displayed are changed periodically by INTEL personnel FBA 370 is the MICROCODE RELEASE number for this machine FB01 amp 2 This RELEASE numbers will change as we receive new versions of microcode All of the numbers displayed on this panel are in hexidecimal When keying information into the service panel do so slowly as there is a pause between registration and display of the information As you key information into the panel it will be VIRTUAL MEMORY PDP SYSTEM SHUTDOWN AND START UP PROCEDURES 67 displayed even before you press enter There are four tasks that operators may have to do on this panel These are COLD starts WARM starts Function 95 Confirms what was entered during a COLD start Function 90 This checks to see how many good partions there are how many spare partitions etc COLD STARTs NOTE Also refer to the Utilities and Diagnostics manual each box has a different manual pp 10 3 for proper COLD START procedures and parameters These manuals are stored in the first floor machine room There is a UNIT RESET button inside the 3805 at the left front bottom RESETS should only be done when the 3805 is physically offline and not in use by the system The following situations call for a COLD START Initial ins
322. ge is one that indicates some definite error either in a command issued at the VM XA operator console or inside of VM XA An information message describes the result of a command or tells you about the status of VM XA A few of the errors that might occur during a VM XA system reload are described below For a complete list of VM XA messages including suggested operator and programmer actions in each case see the manual System Messages and Codes Reference which is located on top of the 3370 disk drives Message HCPINIO70E MACHINE CHECK DURING SYSTEM INITIALIZATION Response This message indicates that a hardware error was detected during a VM XA reload Try to reload again if that doesn t work call a firefighter who will probably tell you to call IBM for service UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 47 Message HCPxxx320E ALTERNATE PROCESSOR INITIALIZATION UNSUCCESSFUL Message HCPxxx625E PROCESSOR nn TOD CLOCK SYNCHRONIZATION FAILURE VARYING OFFLINE Response These messages indicate that VM XA was able to reload successfully but that one of the two ES3090 300J CPUs was left offline To correct this 1 Go to the VM XA operator console 2 Type VARY ONLINE PROC 04 and press ENTER 3 VM XA should ask you to enable the TOD clock 4 Go to the ES3090 300J console for the UB side of the 3090 and press the TOD key 5 If any further errors occur Call a firefighter Message HCP
323. ge message where message is the message you want to send For example WNG ALL VM XA shutdown at 6pm back up at 7 You should send warnings at about the same frequency as you would send broadcasts for a UB MTS shutdown 3 At any time you can display the number of users signed on to VM XA by typing QUERY USERS on the VM XA operator console You can display the userids signed on by typing QUERY NAMES 4 DO NOT PROCEED ANY FURTHER UNTIL UB MTS AND DOS VSE HAVE BEEN SHUT DOWN 5 Clear the LOGMSG you put in earlier by typing WNG All message NULL on the VM XA operator console 6 To actually shut VM XA down type SHUTDOWN on the VM XA operator console VM XA will ring the ES3090 300J alarm once or twice during the shutdown process Each time it does so go to the ES3090 300J console for the UB side UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 49 of the ES3090 300J and press ENTER After the alarm has rung a second time the message SYSTEM SHUTDOWN COMPLETE should be displayed on the VM XA operator console VM XA is down at this point and you can reload or power the ES3090 300J off UB MTS UNDER VM 50 SYSTEM SHUTDOWN AND START UP PROCEDURES VSE on the ES3090 300J Running of Running VSE on the 3090 VSE is an IBM operating system which runs on the 3090 At Michigan VSE runs under control of VM XA on the same half of the machine as UB MTS VM XA must be loaded and running befo
324. gnetic tape or diskette by mail the user must TAPES amp DISKETTES MAILING 501 1 Obtain a mail retrieval form either from the Mail Librarian at the Computing Center or by COPYing the file TAPERETRIEVE 2 Send the completed form the yellow receipt card and a check or money order if this is to be the mode of payment to the Mail Librarian at the above address TAPES amp DISKETTES MAILING 502 Tape Racks Electric Problems There are two sets of electric tape racks in the first floor machine room that lie in a contiuous string As you face west the new ones are at the left and the older ones to the right Older Tape Rack Operation If power is lost to the older ones the fault is usually a blown fuse The fuse is located in a silver colored box on top the north tape rack If the fuse isn t blown then there is probably a tripped circuit breaker If power is lost to the older racks then a socket wrench which should be located on the the rack with the fuse can be used to manually operate the racks The wrench should be inserted in the holes at the bottoms of the racks Newer Tape Rack Operation The newer tape racks have an ON OFF switch on top of their right most unit which is near the center of the row The fuse for for the units is located near this switch If power is lost to the newer tape units there is a battery operated device which can be plugged into the waist high holes
325. gram will ask for a pen number DOTS plots dots ad infinitum Program will ask Y DOTS DISTANCE BETWEEN Respond with 2 numbers separated by commas the second number is a floating point number of inches The program will then ask X DOTS DISTANCE BETWEEN Respond similarily Finally it will ask Enter the number of a pen holder to use from 1 through 4 You might try 100 125 for y and 10 125 for x LINES This is a check of line quality in various directions The program will ask LINE LENGTH Enter a number in inches lt 16 All lines drawn will be this length It will then ask ANGLE INCREMENT DEGREES Enter a number gt 0 but lt 45 for the degrees between successive lines The program will then ask for a pen number The program draws Try a length of 10 and an angle of 5 to start GRID draws a rectangular grid lifting the pen very seldom It tests long lines and drawing with the pen down for long periods PLOTTING SYSTEM 443 The program will ask Y LENGTH INTERVAL Enter the length of lines in the y direction along the drum and the distance between lines separated by commas The program will then ask X LENGTH INTERVAL This is the same question but for the x direction The program will then ask for a pen number Try 20 125 for y and 4 125 for x to start CHARS draws character sets The program will ask ALPHABET Enter the name of any alphabet described in Volu
326. h may currently be running On the 3278 keyboard this key must be used in conjunction with the ALT shift key CLEAR Just rewrites the current screen contents Probably of little use but if the screen gets messed up for some reason try this first On the 3278 keyboard this key must be ised in conjuction with the ALT shift key TEST REQ This is a pause key which can be used to temporarily stop the scrolling of messages off the screen Pushing it again will reset the pause If it is not reset within five minutes it will be reset automatically The request area will contain PAUSE while in pause mode On the 3278 keyboard this key must be used in conjunction with the ALT shift key OPERATOR CONSOLE 3270 341 Program function keys PF1 through PF12 or perhaps through pf24 These can be used to enter various requests into the request area These may be either partial or complete input lines For partial lines the effect would be to enter the corresponding text into the appropriate field and position the cursor following it For complete lines the effect would be the same as the above plus pressing the enter The initial program function key definitions are Complete requests PF1 PF7 MTS OPER BACKLOG ALL Partial requests PF2 MTS PF3 STOP PF4 BROADCST PF5 DISPLAY JOB PF6 RELEASE JOB PF8 BACKLOG PF9 DRAIN Program function keys may be redefined by a as described bel
327. he Disk Manager DMGR in this manual for additional disk pack manipulation information NAME FUNCTION NOTE USAGE DSK To manipulate the table of disks volumes Disks may be dynamicallyadded and removed as well as allowing the drive address to be forgotten This program only handles MTS public volumes It has no knowledge of SPOOL packs or paging packs Enter mts dsk on the operators console or run sys dsk The commands are ADD MTS xxx MOUNT MTSxxx CATALOG MTSxxx to add MTSxxx to the system This should be used to add a new MTS volume to a running system when it becomes necessary to expand the available disks space NOCATALOG MTSxxx DELETE MTSxxx DISMOUNT REMOVE MTSxxx these commands effect whether or not the file system catalog is permitted to expand onto the designated volume to make the volume MTSxxx unavailable to MTS Doing this a running system will cause all jobs which reference the pack to receive Hardware error or software inconsistency OPERATOR FILES DSK 208 DMGR MTSxxx NODMGR MTSxxx EXPLICIT MTSxxx NOEXPLICIT MTSx FORGET MTSxxx LIST DISPLAY OPERATOR FILES messages This should only be used on a running system in cases of extreme emergency these commands control whether the Disk Manager is to be used to manage the specified volume These commands should NOT be used unless the Disk Manager is installed in your version of MTS Also if you are us
328. he Startup job program and start an MTS task either on a terminal or on the operator console 7 Signon and attach the MTS and FILE macro libraries by DISKCOPY 8 9 gt D E E E E E O E 10 11 gt macrolib mts cmdmaclib gt macrolib file cmdmaclib Remove the MTS volume from Run SYS DSK NODMGR MTS005 STOP Copy the disk and swap the 357 the Disk Manager labels This operation takes from 10 20 minutes If there were any errors or relocations during the diskcopy that you do not understand you should answer NO to the prompt to swap labels and get some help ISKCOPY nter FROM device type DISK TAPE DISK nter device and volume label Dxxx MTSyyy D508 MTS005 nter TO device type DISK TAPE DISK nter device and volume label Dxxx MTSyyy D200 OLDO05 nter options SWAP IPL SWAP k to swap labels OK nter FROM device type DISK TAPE STOP Return the MTS volume to the Disk Manager Run SYS DSK DMGR MTS005 LIST MTS005 STOP Bring the system back necessary to re IPL the shutting the system down a disk copy It is only disk activity involving copied The easiest way kick everybody off System IPL s should only there is a new system to up normally It is not system either before or after for the sole purpose of doing necessary that there be no the disk s that are being to ensure this is to just it is by no means the only way
329. he disk the subtask is associated with when the Signal is to be sent toa DISKLING or IOMNGR disk manager subtask Description The event specified by lt event gt will be generated lt event gt may be any of the following DISP generate a signal to the DISPATCH subtask DSKL generate a signal to a DISKLING subtask lt pvn gt must be specified IO generate a signal to an TOMNGR subtask lt pvn gt must be specified Examples SIGNAL dskl 19 the above command will generate a dskl event for the DISKLING subtask associated with volume 19 STATISTICS Purpose To display statistics about the operation of the disk manager Prototype a STATISTICS lt sbt gt lt pvn gt b STATISTICS OVERVIEW Parameters lt sbt gt is the type of the subtask JOB PROGRAMS DMGRSTAT 106 JOB PROGRAMS statistics are to be displayed about lt sbt gt may be any of the following DISPATCH LOCKMNGR DISKLING IOMNGR ITSKMNGR lt pvn gt is the MTS public volume number the subtask is associated with when there is more than one subtask of type lt sbt gt Description If form a is specified statistics about the operation of the subtask specified by lt sbt gt and lt pvn gt are displayed If form b is specified statistics about the operation of the disk manager as a unit are displayed STATUS Purpose To display the status of a disk drive task or the subtasks comprising the disk ma
330. he prefix character asking for the next input line This is the only confirmation the user will receive But if the system load is still too high to grant the user s request he will receive the message CAN T SET LSS OFF LOAD TOO HIGH Over the past few years the functions provided by several system programs which formerly were usable only via the RUN command have been incorporated into the MTS command language and have thus become available even to LSS users In fact it is better i e cheaper for all users to become accustomed to using these commands rather than RUNning the older programs The following is a list of public file utility programs whose functions have been incorporated into the MTS command language Public File Name MTS Command CATALOG FILESTATUS DISMOUNT RELEASE ED EDIT FITLESNIFF FILESTATUS FILEUSE FILESTATUS HBO SYSTEMSTATUS MOUNT MOUNT PERMIT PERMIT PROJECTACCOUNT ACCOUNTING TASKS SYSTEMSTATUS USERS SYSTEMSTATUS LIMITED SERVICE STATE LSS 400 MERIT HERMES SYSTEM The UMnet Communications Computer or CC provides a data link between the University of Michigan s IBM ES3090 300J Host Computer and Host Computers at Michigan State University Wayne State University and Western Michigan University CCs can be Primary Communications Processors PCPs or Secondary Communications Processors SCPs There are now sev
331. hex 01 Return on punch check after message 02 Return immediately on punch check TE Program flag values are 80 Initial rejection of CCW 40 Bus out check occurred 20 Repunching card after punch check 10 EOF unit exception on retry FF Sense flag See description above should be 01 SS Sense data 2 Printer Errors PTRn Machine ERROR PRINTER STATUS UUIIFFSS Program UU Program flags 10 Doing UCS buffer load 08 Return if channel 12 detected 04 Return if channel 9 detected 02 Return on print check 01 Expect invalid character check II Program flags 80 Initial rejection of CCW 40 Bus out check occurred 20 Equipment check occurred 10 Bad character printed 08 Parity check occurred 04 Immediate operation 02 Reprinting last line after print check 01 Return after reprint FF Sense flag as above should be 01 SS Sense data ABNORMAL SITUATIONS SUMMARY 314 3 Card Reader Errors Machine ERROR CARD READER STATUS UUIIFFS Program UU II FF SS ABNORMAL SITUATIONS SUMMARY Program flags 40 20 10 08 04 02 Ordes Return on fatal error Ignore validity check if col 1 has a 7 9 punch Return after validity check after message Return immediately on validity check Unusual command sequence is OK Command was column binary Return in read check Program flags 40 20 10 08 0
332. hich cleans up the file that is removes the extra title lines and leaves only the last mount statistic for each tape in the file This clean up program must be compiled before it can be used The programs as it is is set up to take care of C tapes only run ftn scards w225 w32 tpst 57 This compiles the program and puts the output spunch in the file LOAD run load 5 0 6 00 This actually cleans up the file and puts the output in file 00 The followint sort run will sort the tapes by rack number and then by date Note In order for the sort program to run most efficiently the number of records to be sorted should be determined and then placed in the record field at the end of the sort run To determine the number of records to be sorted in this case f oo lines if the number of lines is for example 12 154 then the record field should reflect this See followint example run sort par sort ch a 2 6 ch a 15 2 ch a 9 6 ch a 18 8 i 00 o 000 r 12 154 If you want the tapes sorted by date only which show you which tapes haven t been mounted recently do the following r sort par sort ch a 15 2 ch a 9 6 i 00 o 000 r 12 154 Following is a break down of the sort command ch a character sort a ascending order 2 6 begin in col 2 and sort 6 columns in from there in this case the rack number If you look in the file o after running msscn you can see what is sorted by count
333. hinks it is but the channel detects a problem which may or may not be the fault of the device This can be caused by either a programming error failing hardware in the tape drive channel or even the CPU To pinpoint the error the bits in the sense flag field dd above must be examined The symptoms are 80 Program check programming bug 40 Protection check programming bug 20 Channel data check hardware problem 10 Channel control check hardware problem 08 Interface control check hardware problem 04 Chaining check either program or hardware problem To decode this value you have to decode the bits indicating each of these symptoms from the sense flag byte For example a sense flag of 18 means channel control check and interface control check were declared Equipment check This error is caused by an ailing tape drive Bus out check Either an error in the channel or an error in the cable connecting the device to the channel can cause this Command reject Usually this is caused by a bad tape drive It can also be caused by programming bugs if the program tries to tell the tape drive to do something it doesn t understand Overrun This condition signifies that there is a mismatch between the speed of the tape drive and the channel to which it is connected It can occasionally be caused by a hardware malfunction but usually signifies an overloading of a low performance CPU such as an IBM 1
334. his it uses the Deadfiles DEADSAVE program MSV will be automatically started when MDP completes It will announce which series of tapes it expects and then issue prompts What tape drive can I use Respond with a four character cartridge tape drive e g TC04 Load up the tapes into the autoloader After you tell it which drive to use it will ask you to mount the first tape in the series Please mount FSxxxx on TCxx Assuming that the tape has been mounted just press the Enter key If the tape has never been used MSV will prompt OK to label tape on TCxx Y N Assuming that you ve got the correct tape there and not some poor user s unlabeled tape type Y and press the Enter key If the tape has been used previously MSV will prompt OK to re use tape on TCxx Y N to which you should respond by typing Y and pressing the Enter key The Please mount and OK to sequence will be repeated for each tape that MSV needs to use currently three cartridges but this will increase In each case you should be able to just press the Enter key in response to the Please mount prompt and type Y and press the Enter OPERATOR FILES MSV DEADFILES MESSAGES 256 OPERATOR FILES key in response to the OK to prompt Remarks The tapes used by MSV come in four series which will be cycled through The tape series are numbered 9000 9099 9100 9199 9200 9299 and 9300 9399 There is an ini
335. hould be indicated by a note near the MTS operator consoles 5 Start the LOAD CLEAR function by typing 07 and pressing ENTER Note that the command requires the letter O oh and not the digit 0 zero 6 After a moment the message Load Complete should appear on the ES3090 300J console 7 Go to the VM XA operator s console In a minute or two the following lines should appear on the VM XA operator console VM XA Systems Facility Release 2 START COLD WARM FORCE DRAIN DISABLE NODIRECT 8 Normally you should press the ENTER key here However if you are reloading VM XA following a VM XA system crash you will need to type FORCE Never type COLD without explicit instructions from a systems programmer 9 If the time of day clock is set the following message will be printed NOW hh mm ss tmz weekday mm dd yy CHANGE TOD CLOCK Y N If the time and date are correct press ENTER and skip to step 11 If the time or date is incorrect type YES and press ENTER 10a VM XA will print SET DATE MM DD YY You should type mm dd yy and press ENTER where mm is the current month dd is the current day and yy is the current year 10b VM XA will print SET TIME HH MM SS UB MTS UNDER VM 44 10c 10d 10e 11 SYSTEM SHUTDOWN AND START UP PROCEDURES Type hh mm ss and press ENTER where hh is the current hour 0 23 mm is the current minute and ss is the
336. ial pen sizes 8 30am Mon Fri Regular plots e Mon 10am 2pm 6pm 10pm e Tues Fri 2am 10am 2pm 6pm amp 10pm e Sat Sun 2pm Plotting When it is time to run plots two procedures must be followed The first of these follows and the second the running of PLOTTING SYSTEM PLT36 421 or PLT51 appears later in this section 1 Type mts ccp on the operators console This procedure has been started automatically by the clock watcher program for some time now now being 4 18 85 The system will ask where a 9 track labelled plot tape can be found T9XX operator types X being the tape unit on which the requested tape is mounted lt lt lt Tape Description First file only 9TP 1600 bpi label PLOT fmt vb 6370 244 gt gt gt The plots listed in CCPFILE will be dumped onto the tape A receipt number will be printed on the console The printer output from this job consists of two lists of plots You can check this list against the plots being plotted This sheet will also have the sizes and colors of the pens to be used The pen size is printed on the side of the pen The clerks will separate the plots and wrap them Note The plot receipt list is also entered at this time into a file named SYS PLOTRECEIPT from which it can be copied if the original list is lost Time Frames for plot tapes TAPE 0900 MON 1345 MON MON 10AM 1345 MON 1745
337. iced by the disk manager will be displayed If form b is specified counts for each volume will be displayed The statistical counts are of the bulk number of channel programs issued the number of pages transferred and the number of pages which were retreived froma and placed into the disk cache If form c is specified statistical counts are displayed for each volume The counts contain information for the Metric system watching program Examples Purpose Prototype Parameters JOBS PROGRAMS 101 DISPLAY C the above command will display the current disk drives under the disk manager s control FORGET To get the disk manager to release and reacquire a disk drive thus forgetting its location momentarily FORGET lt pvname gt pvname is the MTS public volume name of the disk whose location is to be forgotten momentarily Description The disk manager will check to see that it has Examples NOTE See the section Purpose Prototype the mes drive specified If not an error sage will be generated and the call ignored If it does have the drive the dis k manager will wait for any current I O on the drive to be completed and then call DMVOLREL to release the drive disk manager will then immediately DMVOLGET to reacquire the drive the above command will cause the disk manager to forget the location of MTS006 and then re acquire the drive
338. ich can contain intervening plusses so that one can add addresses and displacements to get the wanted address The second parameter string is numeric and denotes the number of words starting at the address given that are to be displayed on the console If not given nn defaults to one 1 1 A hexadecimal expression representing the address and 2 the number of words to be displayed starting at the address stated by the first parameter dis 403a 00403A 90ECD008 dis 403a 1000 2 00503A 90ECD000 18FF1B00 This is not really a JOB request but a SUPERVISOR command DIS may not be issued from the 3270 operator s console the DREAL command should be used instead See the section on Systemstatus JOB PROGRAMS DIS 98 JOB PROGRAMS DMGRSTAT NAME DMGRSTAT FUNCTION Disk Manipulation USAGE DMGRSTAT NOTE The DMGRSTAT job does not process parameters but rather reads commands from the console ADD Purpose To add a disk volume to the disk manager Prototype ADD lt pvname gt Parameters pvname is the MTS public volume name of the disk to be added Description A check will be made to see if the disk Examples BLAST Purpose JOB PROGRAMS DMGRSTAT manager already has the disk associated with lt pvname gt If so an error message will be generated and the request will be ignored If the disk manager does not have the disk it will attempt to acquire it via DMVOLGET If the disk is succe
339. icro seconds etc since March 1 1900 or in character format giving the date and time in the form MMxDDxYY hh mm ss or MMXDDXYY midnight where x is any character For example PFIELD 01 15 78 12 00 00 For integer format the parentheses may be omitted HEX This command prints the entire directory in hexa decimal format Each field is printed with a separating blank The mount message and comments fields are omitted INIT rack label 800 1600 6250 type This command adds tapes to the available tape pool in the directory If the mode parameter is omitted the tapes will be initialized as 1600 bpi tapes If the type parameter is given the tape will be restricted in use to the specified type if the type parameter is omitted the tape will not be restricted If the type parameter is given the mode parameter also must precede it LIST ALL type TAPE rack POOL PAR CURR LAST This command lists the contents of the directory If the ALL parameter is specified the entire directory is listed If type is specified all entries for the speci fied type are listed If PAR CURR is specified only the entry in progress for type TAPE DIRECTORY PROJECT Command Explanation Command Explanation Command Explanation Command Explanation 515 is listed If PAR LAST is specified only the latest entry for type is listed If the TAPE rack parameter is spe
340. icroseconds relating to the data contained in the file However this field may contain integer values 4 or 8 byte other than Julian times if the search is being done on a basis other than time Internally all integers are stored right justified i e 4 byte integers are in bytes 4 7 of each doubleword Secondary search field format byte 54 No secondary search field Julian microseconds unsigned 8 byte integer Julian seconds unsigned 4 byte integer Julian minutes unsigned 4 byte integer Julian days unsigned 4 byte integer Integer signed 4 byte integer Long integer signed 8 byte integer DUR WNEF OO ll Secondary search field bytes 55 70 This field is composed of two 8 byte integers Normally this field will consist of a Julian time range microseconds relating to the data contained in the file However this field may contain integer values 4 or 8 byte other than Julian times if the search is being done on a basis other than time Internally all integers are stored right justified The secondary search field provides an alternative or TAPE DIRECTORY PROJECT 510 secondary searching method for the data contained in the file MOUNT message field bytes 71 166 This field contains the MOUNT message in character format needed to mount the tape containing the desired file The VOL RING and MODE keywords will be presented in the MOUNT message RING IN OUT will be supp
341. ier in this manual POWER FLUCTUATIONS amp FAILURES RECOVERY SYSTEM SHUTDOWN AND START UP PROCEDURES 91 The following pages contain descriptions of the UMMPS job programs and supervisor commands In the PROTOTYPE line of each description the following conventions are used a The first word is the name of the job program It is always in upper case in the description but may be typed at the console in either upper or lower case b Other than the exception in a upper case letters denote parameters which must be typed as shown they may be in lower case however c Lower case letters denote generic parameters which must be replaced by an appropriate substitute For example the sequence nnnnn denotes a number the sequence device i d denotes the symbolic name of an I O device such symbolic names are system dependent an example of a symbolic 1 0 device i d would be RDRx or PTRx d denotes a choice of options e g x y means choose either x or y e denotes optional characters or parameters JOB PROGRAMS 92 BLAST NAME FUNCTION PROTOTYPE USAGE PARAMETER NOTE EXAMPLE JOB PROGRAMS BLAST JOB PROGRAMS BLAST To forcibly kill a job if STOP will not kill it BLAST nnnnn This job forcibly removes a task from the system It should only be used as a last resort since it does not signoff the user properly if it is an MTS task that is to be
342. ig sou sig NOTE The may be dealt with or the output may show up on the operators console w035 getfilechg One or more tapes will be requested by the system The tapels will be Daily MTS Statistics File Charges acc dlist This produces a listing of Computing Center Accounts and will request tapes of the type ACCO1 ACCO2 etc GETFILECHG program should always be sourced before the DLIST prgram ACCOUNT LISTING FILE CHARGE UPDATE MONTHLY 369 FIRE FIGHTERS GUIDE This section is included in the operations manual since the duties described here are closely related to operations though the people mentioned are with the systems group 1 FIRE FIGHTER JOB DESCRIPTION A fire fighter is supposed to handle as much as possible of the day to day fix it busy work jobs which otherwise tend to eat up the time of the systems group people The tour of duty lasts two weeks overlapped with a new person starting each Monday morning Specific duties include being on call 24 hours per day to investigate system problems and whenever possible to resolve them without involving other systems people printing and routing jobdumps keeping current system loader maps posted keeping the systes group advised of any problems that come up 2 TOUR OF DUTY CHANGE The following steps are used to identify offlicial fire fighters for the next tour of duty a Edit the file
343. imal If the field RTABMCNT is 0 then the maximum number of messages are queued for this remote RM PRIORITIES CLASSES SCHEDULE 378 RTABMCNT starts at 10 and is decremented for each queued message To clear up the WTO wait it is necessary to STOP the remote SAnn line However Since HASP won t accept the STOP SAnn command you have to do it by hand First of all you need to know which SAnn line the remote is on and whether it is a leased line or dial up Assuming the remote number is 1 or 2 digits type D DCT RM RD1 One of the fields displayed for the device control table DCT for the first reader RD1 for this remote is DCTDCB addr where addr is the vm address of the line manager DCT for the remote Display this DCT as follows D DCT AT addr Looking at this DCT you can find the SAnn device name in the field DCTDEVN and whether the line is leased signified by the flag LEASE in the field DCTPSTAT If LEASE does not appear then the remote is a dial up and you can get rid of it by hanging up the phone line for SAnn given in DCTDEVN Things should clear up shortly If the remote is on a leased line then you have to stop the HASPLING doing the I O on the SA device and set the stop bit in the DCT for the line Find the UMMPS task number via SYS T D SAnn or look at the field DCTSJNO and stop it via SYS O STOP task Finally turn on the high order bit DCTSTOP X 80 of the
344. imply truncates any output line that will not fit on a single line of the display although the complete line will be printed on the console log printer If the SWRAP ON command is given the longer messages will be displayed using two screen lines An attempt is made to split longs lines OPERATOR CONSOLE 3270 343 at a blank SWRAP OFF This command cancels the affect of a WRAP ON command ABNORMAL SYSTEM CONDITIONS The following keywords may be displayed at the front of the second line on the 3270 operator s console display to alert the operator to various abnormal conditions Depending on the situation these conditions may or may not represent errors BACKUP SHUTDOWN HASP DEAD HASP PCE HELD EX HELD The 3066 or 1052 backup console is in use The 3270 operator s console will continue to operate HASP commands device commands systemstatus commands and job requests may still be entered in the request area Existing input requests may be filled Output from device and system status commands will continue to be displayed on the 3270 screen but output from HASP commands and job requests as well as output from executing jobs and all requests for input will be directed to the backup console The RESTART command may be used to transfer control back to the 3270 operator s console or the OPERATOR job may be stopped and restarted from the backup console The SHUTDOWN
345. in INIT INITCMD which a Checks and verifies the time and date given with respect to the time and date of the last record in the file STATISTICS b Records the time and date and reason for IPL in the system log file c Starts the PDP Paging Drum Processor d Prints out the comments for IPL O or the operators console e Starts MTS CLN Starts MTS FSM to complete any work left unfinished because of a system failure while the file save merge program was running g Starts MTS STI to complete work unfinished because the system failed while STA was running h Starts JOBS MAS i Starts a separate task MTS S2L to load shared segment 2 components such as BASIC the shared part of the PL I Library APL the EDITOR etc Jj Destroys any temporary files which were left if the system was not shut down normally previous to this IPL k Fixes all i d s which need to be fixed as a result of a system failure SNARK or BLASTed job WHAT THE IPL SEQUENCE DID SYSTEM SHUTDOWN AND START UP PROCEDURES 85 Power Fluctuations and Failures Occasionally power will be interrupted by storms worn power lines by someone working on power equipment etc Some of these interruptions will be so short that they won t interfere with the system s performance since the UPS Uninterruptable Power Supply system should take over for up to several minutes Short power spikes should be handled with no interrup
346. in after a short wait or t d xxop to see what job is using the device lt name gt is logically offline Either lt name gt was not up when MTS was IPLed or it has been offlined using the OFFLINE job In either case to online lt name gt sign on to the CCID STRT and SOURCE STRT ONLINExx where xx is AB AD AE AF AN AU AV AW AX AY or AZ Once this is done rerun HUH lt name gt is physically offline switch is in the Switch it to online The online offline offline position and rerun HUH Message Meaning Message Meaning OPERATOR FILES 225 lt name gt does not respond Check and reload if necessary Either lt name gt is powered off ora serious error has occurred If the power is off it should be turned on and HUH rerun If the power is on then the PCP has probably crashed in such a way that the High Core monitor is not running To see if this is the case select lt name gt at the Super Console terminasl and type HELLO T there is no response the machine has crashed and must be restarted from the front panel of the PCP If it says lt name gt monitor then high core is running and xDL should be run If it says name gt PCP MINOS version Sok Or UE pes PCP MCS then the operating system is running but the channel interface is not operating properly If the latter happens do a UNITS D lt name gt OP S on MTS and try to cantact a UMnet UMnet Firefighter
347. in processor for a link will always be waiting for work since it doesnt do anything but direct the other processors Es JOB The command DISPLAY JOB job spec or JOB job spec where job spec is either the job name or the job number will display information about a specific job in the system The amount of information presented depends on the verbosity level see the following section on Messages from the RM for examples The command DISPLAY NETINFO job spec has the same syntax as the DISPLAY JOB command but shows only the source and destination network addresses at verbosity 1 at verbosity 9 it also gives additional information It is often more useful than JOB for network jobs Dz DISPLAY DEF The current definition of all symbols processors and can be displayed at any time For example DISPLAY DEF procname will display the processor definition And DISPLAY DEF ALL LINK linkname will display all definitions for that given link Messages from the RM 1 Messages that come out on the operator s console are printed with a header The header consists of a four character module identification name a dot and a seven digit error message number For example RESOURCE MANAGER 2 3s 539 LNKL 2023210 Link WAYNESTT disconnected due to line protocol error For this one LNKL identifies the message as coming from the LINKLING The first digit of
348. in that you have noticed that an SCP seems to be stuck loading Give the operator the name of the SCP Don t do anything else unless the operator or someone at the Center tells you to All the terminals on that particular SCP will be dead until the problem is corrected If neither of the LOAD lights is on and you previously determined that the SCP wouldn t respond to any of its terminals then you should reload it by pressing the white INIT switch down and then back up DO NOT turn any power switches on or off Whenever you reload an SCP write it in the log book provided One or both LOAD lights should come on within a minute or two and then go off within three minutes If that doesn t happen call the MTS operator on the red phone and explain that you determined that an SCP was dead by trying several of its terminals that you tried to reload it using the INIT switch but that it won t reload Give the operator the name of the SCP Don t do anything else unless the operator or someone at the Center tells you to UNYN TERMINALS SCPS ETC 524 Summary of the above actions 1 2 Figure out if MTS is alive If it is test the dead terminal to see if you Can get a response from it If it was already in use press FIX Ontel DELETE DECWriter or DEL Visual 550 Then type HELLO followed by RETURN If you get a response like 11 48 29 03 31 83 216 LSI 11 MINOS Version AR293 then there is
349. inates Jobdumps are usually initiated by operator request typing JOBDUMP n to force a dump of job number n all of real memory and the job s virtual memory are written on the tape and system operation resumes System dump tapes are 9 track and labeled DUMPxx xx being the dump tape number This program reads the tape on unit 1 or 0 reads commands from SCARDS and prints the dump on SPRINT SPRINT should generally be assigned to a printer or PRINT since large amounts of output are usually produced Output is controlled by commands as follows CORE start_address lt end_adress gt lt ST segtable_address gt This dumps real memory Addresses are hex byte address DISPLAY This is the same as CORE EXPLAIN This is the same as HELP FOLLOW start_loc length lt offset gt lt st segtable_address gt This follows a chain in REAL memory start_loc is the address of a fullword that points to the first item in the list length is how many bytes to dump for each item end_address is the offset of the list link from the start of the item Default end_address is 0 HELP This is the same as EXPLAIN ICTAB OPERAND This displays various aspects of the IN CORE file table Legal operands are LSTAT JOB xxx or LSTAT FILE yyy LOCATE FILE yyy displays ICTAB entry offset for file yyy DUMP to dump the whole thing TRACE to give a file by file trace DUMPS PRINTING 366 VERIFY to
350. ineno s is taken to mean a range ccid is a signonid for this type all records for that signon id are printed prin is a project number all records for signon ids in that project are printed jobn specifies a task number and may only be used with cmdstat records All cmdstat records for jobn are listed SCAN prints out the listing for one item before requesting another A null line or end of file will cause SCAN to request a new command A new list command must be entered to change the fdname or whether statistics or cmdstat records are being printed COPY STAT CMDS fdname The syntax of this command is the same as for LIST COPY copies records to fdname in their internal form as read from the tape All records are in the new FEB 1982 statistics records format MTS This command causes a return to MTS SCAN may be REStarted SET STATSCAN fdname DEFAULT This command changes the location where the STA command searches for statistics records STATSCAN fdname will cause only fdname to be searched for statistics records the monthly tapes and STATISTICS will not be searched SCAN 456 SCAN trusts that you will provide statistics records likely to interrupt you if given bad have the STATSCAN statistics 3 records statistics records scan STATISTICS only Since lines 1 and 2 do not Also command so be judicious STA command to resume searching in the normal places
351. information on running VM XA Real System Operation Contains information on reloading shutting down and controlling VM XA Virtual System Operation Contains information on UB MTS UNDER VM 42 SYSTEM SHUTDOWN AND START UP PROCEDURES running a guest operating system This information will be of more use to systems programmers System Messages and Codes Describes all VM XA messages Vector Processor To check the status of the Vector Processor on the VM UB side Use the VM 3278 and type q vec Response should be Vector facility 03 online If you must put the Vector Processor online type vary online vec 03 To check the Processor status on the VM UB side of the 3090 use the VM 3278 console and type q proc Response should be PROC 04 MASTER PROC 03 DEDICATED TO UB MTS CPU 02 PROC 05 DEDICATED TO UB MTS CPU 01 Loading VM XA This should only be done after VM XA has been shut down or after the ES3090 300J has been powered off and back on again 1 Go to the ES3090 300J console for the UB side of the 3090 2 Press the INDEX key 3 Select the OPRCTL frame by typing 9 and pressing ENTER UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 43 4 Make sure that the load address on the OPRCTL frame is correct If it is not a Type L1 and press ENTER Your cursor will move to the LOAD UNIT ADDR field b Enter the load address and press ENTER The load address s
352. ing in from col 1 Columns 15 2 and 9 6 get the date and 18 8 the time i 00 means oo is the input file and o 000 means that ooo is the output file and r is the number of records to be sorted Since you will probably want a double spaced listing of your output do the following edit 000 col IL this will confine any edit OPERATOR FILES MSS 246 OPERATOR FILES commands to column 1 a a nv opl f 0 this changes any blanks in col 1 to 0 s 0 s in col 1 will cause the lines to be double spaced NOTE The above example will only list C tapes since the clean up program was set up that way To get a listing with just N tapes or with N and C tapes one must edit w225 w32 tpst and change line 2 from INTEGER CEE C OK 0K to INTEGER CEE N OK OK and then compiled specifying an spunch for example copy w225 w32 tpst 57 w32 gedit w32 ala ETIENNE stop gets you out of the editor run ftn scards w32 spunch load 1 run load 1 5 0 6 00 last 1 The last 1 adds the N tapes statistcs to the end of file oo which contains the C tape statistics Now you can run sort as shown above Next to print out your listing To do this run pagepr scards 000 par format landscape onesided or if you want a line printer listing cont print printer line copy 0o00 print release print Below is an example of a procedure for compiling a list of last
353. ing the version of the PDP which pages through the Disk Manager you should not remove the Disk Manager from any volume which contains a paging extent XX these commands affect whether files will be created on the designated volume Tf a volume is designated EXPLICIT nothing will be created on the volume unless explicitly specified via the VOLUME parameter on the MTS CREATE command NOEXPLICIT removes this designation to forget the device address of an MTS volume This causes the file routines to re initialize the disk table entry for the specified volume on the next reference to it which is useful if a program such as VAMREC has been used to change information in the volume s PAT which the file routines need to know about e g the pack s relocation entries to list all currently defined volumes in the table Useful for determining which volumes are on which drives This will also summarize the total amount of space and the total amount of free space in the system LIST MTSxxx DISPLAY MTSxxx OPERATOR FILES DSK to list the DSKTAB information for the volume OPERATOR FILES 209 MISxxx POLR MTSxxx NOPOLR MTSxxx these commands control whether the volume should be considered one of last resort for the creation of files If it is no files will be created on it unless insufficient space is available on the other MTS volumes NOPOLR resets this designation
354. ing up the Amdahl disks for any reason open POWER FLUCTUATIONS amp FAILURES RECOVERY SYSTEM SHUTDOWN AND START UP PROCEDURES 87 the back panels of each disk module and turn off the green tipped power switches for each disk drive then proceed with the rest of the powering up sequence All disk activity should have stopped and either the system should be down or the Amdahl disks should be logically offline before using the green tipped switches Before powering down the Amdahl disk drives turn these same green tipped switches off before doing anything else but after all disk activity has stopped In order to get the Amdahl disks running after a power failure open the front and back panels of the disk drive modules Make sure that all of the circuit breakers on on If these circuit breakers ever go off for some unknown reason something may be wrong the the Amdahl disk system itself in which case an Amdahl F E should be notified before going any further Of course if the C B s went off because of a power failure or some other obvious reason then it s OK for us to reset them There is one POWER Switch and one ATTN switch for each disk drive These are located inside the back panels of each module See the stickers on the inside of the back panel of each module These stickers indicate which switches go with which disk LOCAL REMOTE Switch Amdahl Disks Upon opening the front panel of the Amdahl Control
355. ission By Mail 500 TAPERETRIEVE 350 501 Tapes 486 Tapes Attach 31 Tapes Detach 31 TAPESHARE 303 Tapes Labelled ASCIT 232 234 Tapes magnetic labelled ANSI 232 234 Tapes Mailing 350 500 Tapes Permitting Unsp tapeshare 303 Tapes responses To Mount 493 Tapes runaway 493 TAPESUBMIT 350 500 Tapes VSE 54 Tapes W035 503 TAPE UNIT DOES NOT SUPPORT ETC 413 Tape Usage History 236 241 Task 21 TASKS 150 573 Terminal Stations NUBS 531 Termprio 451 Termprio Values 452 THE FOLLOWING VOLUMES ETC 414 Time Of Day Clock TOD 59 Times Frames For Plot Tapes 421 Timing Facility Damaged 311 TMTS 153 TN 154 TNR 299 TOD Clock UM MTS 27 TPD 300 TPS 302 Transmitters Receivers HASP 384 TSH 303 TSN 304 TU81 543 11 60 155 AM 471 499 519 AM Display 518 AM Modify OFF 518 AM Modify ON 518 Disk Problems 364 Disks 364 MTS Backup Console VM XA 34 TS Loading 32 MTS Under VM Running Of 29 B Shutdown Message On PCPs 284 B System Hung 364 B VM XA Controlling I O Devices 38 JB VM XA Stopping The CPU 36 JB VM XA Superdump 35 UB VM XA Tape Drives Attaching Detaching aa a ww qaqqqqaqaqc aac GGG wW wW w 37 UCS Buffer 134 136 155 UDP 305 UMLOAD 79 UMMPS 18 UMMPS Intertask Communication Test 229 UM MTS Backup Console 27 UM MTS Loading 26 UM MTS Stopping CPU 28 UM MTS Superdump 3090 27 UM MTS TOD Clock 27 UMnet 169 170 173
356. it type and address match DESTINATION OPER TEXTLOC CMDS 636 EXITDMSG ERRRTN CMDS1361 SOURCE CMDS 636 EXITSVCO ERRRTN CMDS1223 gt gt gt NO SPACE FOR RELOCATION ENTRIES VOLGET User Explanation This message may but shouldn t occur after adding a new pack to the system It should be followed by the message THE FOLLOWING VOLUME S HAVE NOT BEEN FOUND The operator should respond with CANCEL and call a systems programmmer DESTINATION OPER SOURCE FILE SYSTEM 482 5 VOLGET MISCELLANEOUS MESSAGES 411 gt gt gt NOT A HASP JOB User Explanation This error message is printed when the operator has given a RER response to a MOUNT request by other than a HASP job TE EOE example all drives of that type e g 9 track tape drives are in use the operator should use BY for busy for a nonHASP job and RER for a HASP batch HB job DESTINATION OPER TEXTLOC MOUNT 101 NOTHSP MOUNTCMD MNT02803 SOURCE MOUNT 101 ST109 MOUNTCMD MNT01518 gt gt gt QUT STACK ERROR User Explanation System error see an MTS systems programmer System Explanation The 3 level stack for the subroutine OUT etc has either overflowed or underflowed during the SINK MSINK error printing recursion DESTINATION OPER TEXTLOC MTS 42 OVERM OUT MTS02238 SOURCE MTS 42 OUTCKSS OUT MTS02231 gt gt gt POSITIONING FAILED User Explanation The batch job not specifying Q
357. job has been run HASP has detected a serious error within itself and has given up It will need help if it is to run again without reIPLing the system and even that may not help One or more HASP PCES are frozen and will remain that way until fixed by a programmer This message while less serious than the HASP DEAD message still indicates an error condition within HASP Either HASP has not been started or the HASP command HOLD has been issued Either HASP has not been started execution has not been released after running HSP or the HASP command HOLD EX has been issued OPERATOR CONSOLE 3270 344 NTM No tape mounts MODify NTM 1 UAM The system has been placed in unattended mode using the system status command MODIFY UAM 1 AB CC AD AE or AN When one or more of these names are displayed it indicates that some abnormal condition in one of the PCPs the data concentrator or the UMnet PCP has been detected by the system status master job Running HUH should explain what the trouble is Following any keywords that may appear may be a list of four character device names of those devices that are not ready and which need operator assistance Following any not ready device names may appear a list of five digit job numbers for those jobs whose output to the 3270 operator s console is being ignored 3066 or 1052 SUPPORT AS BACKUP If the 3270 support fails to respond or dies some h
358. k if you wish the output on your terminal or sprint print EXECUTION BEGINS WHICH DEVICE TYPE 2305 3330 OR BOTH operator types option ENTER STARTING TIME E G 8 20 JAN 13 1 am jan 17 EXAMPLE ENTER ENDING TIME 3 am jan 17 EXAMPLE SUMMARIZE YES OR NO yes PRINT ASSIGNED RECEIPT NUMBER EXECUTION TERMINATED If the option yes was given in the above example the output will be sorted by module and format 6 type information will be accumulated If the option no had been given then the output would be the same as if mta printuc3330 had been run run mta puc sprint print par stc The STC parameter says to assume the disks are STC disks and interpret the physical drive ID accordingly EXECUTION BEGINS PRINT ASSIGNED RECEIPT NUMBER EXECUTION TERMINATED signoff DISK AND FIXED HEAD FILE UNIT CHECKS 318 EREP SYS LOGOUTS CPU AND CHANNEL ERRORS PRINTING OF USING IBM S EREP Note Presently 16Ju190 use SYS ERP 2 rather than EREP e g sou erp 2 after altering the dates or any other material in the file SYS ERP 2 using the MTS Editor This is a complete rewrite of the obsolete version of EREP that we have been using and contains many new features The IBM manual GC28 0772 3 contains a writeup on the output formats and parameters for controlling this version of EREP There are some suggestions for useful combinations below FUNCTION This
359. kette Retrieval Submission By Mail 500 Diskettes Mailing 500 Disk Manager 351 Disk Problems UB 364 Disk Reserve 282 Disks adding 207 Disks Amdahl Interrupts 88 Disks Amdahl Powering Up 86 Disks DSK 207 Disks listing 207 Disk Space Available 211 Disk Space Retrieval Info 358 Disks removing 207 Disk Storage 351 Disk Unit Checks 316 317 DISPLAY Systemstatus 474 DLG 203 DLG Startup 231 DMGR 207 351 ADD 98 BLAST 98 DISPLAY 100 FORGET 101 HELP 101 LSTATUS 102 NODMGR 99 REQUEST 103 SIGNAL 104 STATISTICS 105 STATUS 106 DMGRSTAT 98 351 DMP 116 204 367 DOTS 441 442 DRN 205 D DSD 351 DSD DASDI 206 SECT CLOBBERED 406 DSK 207 351 Add 207 List 207 Remove 207 DSN 210 DSO 211 DUMP DUMP CMDS 367 DUMP LOG 367 DUMPS 115 Dumps Printing 204 365 DUP Key 338 DWB 212 327 DWB Errors 328 EMERGENCY POWER OFF 65 462 EREP SYS 318 Error Codes Disk 361 Error Codes tape 494 Error Codes 3330 3350 361 ERROR MESSAGES IPL 79 Error Messages Plotting Processor 427 Error Messages VM XA 46 ERRORS CARD READER 314 CHANNEL 307 318 CONSOLE 308 CONTROL UNIT 307 CPU 310 318 DAC 361 Disk 361 Disk Unit Checks 316 DWB 328 HARDWARE 310 INDICATIONS 307 Memorex 1270 322 PRINTER 313 PUNCH 313 SSD 361 Tape 493 Tape codes 494 Typeout Formats 75 Unit Check Codes 361 1270 Line 315 2305 361 4305 361 Errors FHF 361 Errors PDP 361 Errors Soft Recording CPU 142 Errors
360. l parameter itself either msink or syspch The first group of optional parameters those preceding msource consists of one or two of the following NDSK FULL BATCH really ATCH MORE and HASP However there are limitations as to the combinations of these optional parameters The optional parameter FULL has meaning only to terminal tasks and denotes whether or not the task is to always have FULL terminal capability if MTS finds that the system is rather overloaded it may allow users to use only limited capabilities of MTS e g EDITing would be allowed but RUNning of programs would be allowed only if the execution time lt 25 sec If a line has been started with the FULL parameter it gets full service no matter what the system loading happens to be The HASP parameter is used internally and should not be used by the operator The next two parameters may be used together but if used together they must be used in the order BATCH MORE If just BATCH is used MTS will read jobs from msource until a hardware end of file occurs JOB PROGRAMS MTS 120 NOTE JOB PROGRAMS MTS JOB PROGRAMS on msource When this end of file is sensed MTS makes a note of it and when it gets ready to read the next job it will STOP itself i e the MTS task will go away If MORE is also used i e BATCH MORE then MTS will read jobs from the msource device until the MTS task is STOPped by the operator o
361. l unit numbers 2 is given then the control units are marked logically online If the string CU followed by device names is given 3 then the control units attached to the devices are marked logically online Form 3 of the ONLINE command may only be used if the devices are attached to a single control unit Control units numbers are not directly related to device addresses but the UNITS command may be used to determine the control unit number for a given unit If the string CH followed by a channel address is given 4 then the channel is marked logically online If the string CH followed by a channel address and control unit numbers is given 5 then the paths between the channel and the control units are marked logically online If the string CH followed by a channel address and device names is given 6 then the paths between the channel and the control units to which the devices are attached is marked logically online Devices control units channels and the paths between channels and control units may be taken logically offline by the OFFLINE PARAMETERS AND JOBS PROGRAMS 127 job they can be marked initially offline in the TABLES assembly or they will be marked logically offline if they are physically offline when the system is IPLed The UNITS command may be used to determine the status of channels control units and their paths EXAMPLES To online devices
362. l PLOTQUIK to continue plotting or napping The commands are CCP receipt Plot receipt will be changed form a quick plot to a normal plot PLOTQUIK will not plot it and PLT will plot it after it has been dumped to tape TE receipt is omitted the plot currently being plotted is assumed Example CCP 716253 CON PLOQUIK will finish plotting the current receipt then stop plotting and prompt for a command Note that PLOTQUIK stops after the receipt is done while PLT stops after the block is done MSG Plot Message plot message will be written onto the end of the plot currently being plotted Use this command to send a message to the user Example MSG redrawing a line many times digs holes in the paper MTS This command has the obvious meaning Note that if you started PLOTQUIK with MTS PLQ MTS will cause it to stop PEN Use this command to change the assignment of pen colors to pen holder slots PLOTQUIK will prompt for the new assignments Enter the holder slot numbers in order for black blue red and green e g the first number is the position of the black pen Use 0 to indicate that a color is not being used Example 213 assigns black to the second holder blue to the first red to the third and green is not used trailing Os may be omitted 2 0 3 assigns black to PLOTTING SYSTEM 425 the second slot red to the third Blue and green aren t used in this
363. l of available tapes If required the new tape will be initialized to the correct mode TAPE DIRECTORY PROJECT Command Explanation 511 as specified in the master entry for type If the entry is to be allocated to a particular tape the parameters TAPE rack and VOL label must be specified to indicate the correct tape If the entry is to be allocated to a new tape the parameter TAPE POOL or TAPE RES must be specified to indicate that the tape tape is to be allocated from the pool of available tapes If RES is specified the tape is allocated from a restricted pool of available tapes The PFIELD and SFIELD parameters specify the primary and secondary search field values to be assigned to the new new entry m is the lower bound value and n is the upper bound value The format of m and n must correspond to the format specified in the master entry for type For Julian time formats m and n may be given in either integer format giving the actual number of microseconds etc since March 1 1900 or in character format giving the date and time in the form MMxDDxXYY hh mm ss or MMxXDDxYY midnight where Wy is any character For example PFIELD 01 01 78 00 00 00 01 31 78 23 59 59 These search field values may be given either with the ADD command or CONFIRM command any value given with the CONFIRM command will override an ADD command value The COMMENTS parameter may be used to specif
364. l user has not yet received the previous BROADCST message the message LAST MESSAGE NOT RECEIVED BY nnnnn will be typed where nnnnn is a list of one or more task numbers At this point a null input line will terminate the broadcast request without affecting the previous JOB PROGRAMS BROADCST JOBS PROGRAMS broadcast 95 If one tries to BROADCST specifically toa non terminal task or to a terminal task which is not in active use the message CAN T BROADCST TO nnnnn will be typed JOB PROGRAMS BROADCST 96 NOTE PARAMETERS EXAMPLES JOB PROGRAMS The operator should be aware that the carriage width of some terminals is normally limited to seventy characters and any message longer than that may become truncated at the carriage width a all send message to all active terminals b mnnnn where nnnnn is a list of one or more task numbers associated with active terminal tasks broadest all shutdown in five minutes broadest 40 please call machine room now broadcst 401 403 405 please sign off we will reload JOB PROGRAMS BROADCST MAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLES JOBS PROGRAMS 97 DIS To display the contents of real memory locations DIS xxxxx yyyyy nn This command allows the operator to display the contents of real memory locations There are really two parameter strings The first is the address string wh
365. ld be used with great care so as not to to damage files This is useful when there are no job tables available CE dev name This command given without a device name generates a table giving the device name and the hex value of RDVBTS from the device table for all devices for which a device table is allocated If a device name is specified more detailed information for that device is printed Both forms of the command continue to print information until interrupted by an attention interrupt PAl on a 3270 or the JOBS job is STOPped or GOOSEd Examples DEVICE CB08 70 CBO9 70 CB10 70 CB11 70 CB12 70 CB13 70 CB14 70 CB15 70 CB16 70 LAO6 70 LAO7 70 LA08 70 LAO9 70 LA10 70 LA11 70 LA12 70 LA13 70 LA14 70 SA02 70 SA03 70 SA04 70 SA05 70 SA06 70 SA07 70 SA08 70 SA09 70 SA10 70 SA11 70 T900 70 DEVICE DS14 Dev 3270DS14 Job 00458 Bits 80000000 CSW 10503048 0C000000 Sns 00 00 00000000 00000000 00000000 00000000 00000000 00000000 SYSTEMSTATUS 473 Atnx 01503050 SNSX 015035DB PCIX 00000000 ENDX 00000000 2101B0 0C0E280E 80200000 00037D90 60008834 00000000 00000000 00000000 2101CC 00000000 0000000A 015035DB 10503048 0C000000 01503050 00000000 2101E8 00000000 101990F0 10000400 00000000 00000000 00000000 00000000 210204 00000000 where DEV is the device type followed by the device name The job number of the task using the device is 458 BITS is the value in hex of four bytes of switches from the device table as follows
366. le locked Then use the T command several times to check the status of that job Usually you will find that CMD is waiting to lock INIT UC3330L0G INIT MCFILE SYS CMDSTATn or SYS HASPLOGn and that this file is in turn locked by a staff member If this is the case try to find the staff member and ask them to unlock the file If the staff member isn t working from a terminal in the Computing Center you can try BROADCSTing to them and asking them to unlock the file If you can t find the staff member in question or a BROADCST doesn t work after waiting several minutes you should go ahead and stop the staff member s job 6 If CMD is in a FILE WAIT but for some file other than the ones listed above or if the job that has the file that CMD is waiting for locked doesn t belong to a staff member you should wait a few minutes and check again If CMD is still waiting for the same file and that file is still locked by the same job as before then you should job dump the job that has the file locked and then stop it If a new job has the file locked then you should wait a few more minutes and check once again 7 If CMD isn t in a FILE WAIT use the J n command to see if CMD is completely stopped or if it is using some CPU time If it is using some CPU time wait several minutes and check again EE set TS completely stopped try to see why If you can t figure it out STOP CMD and start a new version
367. lems 195 CMDSTAT Tape Errors 330 CMDSTAT Tapes 327 Combined Statistcs Tapes 193 Command Statistics 195 212 CONRAC 210 Console Backup 22 Console Backup Support ES3090 300J 346 Console Backup UB MTS VM XA 34 Console Device Commands 341 SACCEPT ALL 341 ACCEPT Jobn 341 SEMPTY 342 SFAST 341 SIGNORE 341 PF 342 566 PFn new 341 SPTR Devn 341 SREFRESH 341 SRESTART 341 SLOW 341 SWRAP OFF 341 SWRAP ON 341 3270 341 CONSOLE ERROR CONDITIONS 308 Console Main 22 Console OPER 335 CONSOLE Operator 128 Control Panel 551 CONTROL UNIT ERRORS 307 CON1 Startup 231 Courier Startup 231 CPU ERRORS 310 CPU Logouts 318 CPU Stopping UM MTS 28 CPW 198 CTD 200 CT3 201 CURSOR 338 CWM 202 DASDI 351 DASDI DSD 206 DASH 441 444 Data Collection Facility MTS 145 Data Concentrator Status 224 DEADFILES 359 Dead Files Information 358 DECK CARDS 268 DEC RX02 543 Deferred Plots 181 420 428 Deferred Plots Plotting 429 Deferred Plots Running 429 DEFERRED Priority 374 Deferring Plots 429 Degradation Reported 311 DELIVERY 349 Delivery Codes 350 Detach Tapes 31 DEVICE BUSY OR NOT AVAIL 405 Device Commands OPER 341 DEVICE TYPE NOT ALLOWED 406 Differential Rate Periods See Rates 374 Direct Access Storage Errors 361 DIS 97 Disk Checking 200 DISKCOPY 355 Disk DASDI 351 Disk Error Anal PM2 STC 446 Disk Error Codes 361 Disk Errors 361 Dis
368. lied automatically depending on whether the tape is to be read or written The POSN keyword is not included in the message Comments field bytes 167 422 This field provides an arbitrary field for use by the calling program The maximum length is 256 bytes Tape Directory Program The program to interact with the tape directory located in the file SYS DIRPROGRAM This is a GOM program and may be run by issuing the command run sys dirprogram scards input sprint output sercom errors par fdname The command input is read from SCARDS and the program output is written on SPRINT Error diagnostics are written on SERCOM The name of the directory file may be specified in the PAR field The asterisk is the program prefix character The command language is described below Command ADD type TAPE rack POOL RES VOL label1 PFIELD m n SFIELD m n COMMENTS string Explanation This command adds a new entry into the directory The parameter type specifies the type of tape entry to be added The tape directory program will allocate a new tape file and assign a sequence ID n to the entry The entry will be marked as being in progress and will remain as such until confirmed by the CONFIRM command Normally the new tape file will be assigned to the current tape the tape containing the entry with the highest sequence ID If there is no current tape a new tape will be allocated from the poo
369. llow the above procedure again Switching to the Backup Console To start the backup UM MTS operator console at any time you must make sure the proper terminal is attached to UM MTS and signal an external interrupt 1 Go to the ES3090 300J console for the UM side of the 3090 2 Press the alt key and while holding it down press the IRPT key 3 The backup console should appear on the terminal Forcing a UM MTS Superdump To cause a UM MTS superdump you must signal a restart interrupt UM MTS ON 3090 28 SYSTEM SHUTDOWN AND START UP PROCEDURES To do this 1 Go to the ES3090 300J console for the UM side of the 3090 2 Press the alt key and while holding it down press the RSTRT key Stopping the CPU You must do this as the last step in shutting down UM MTS Go to the ES3090 300J console for the UM side of the 3090 2 Press the alt key and while holding it down press the STOP key UM MTS ON 3090 SYSTEM SHUTDOWN AND START UP PROCEDURES 29 Note There is no longer a UB system Running UB MTS under VM XA There are two ways to run MTS on the 3090 native and under control of VM XA Currently at UM we are running UM MTS native and running UB MTS under control of VM The following discussion applies only to running UB MTS under VM XA on the 3090 see the section Running MTS on the 3090 for information on running UM MTS native VM XA is an IBM operating syste
370. llowing SIGNON ccid PRINTER PAGE DELIVERY NUBS Because there is currently no TN mixed case printer at BSAD the delivery option can also be used to have TN output printed at the CNTR and then delivered to BSAD DELIVERY OF PLOTS To have a plot delivered to NUBS append the DELIVERY keyword to your plot request when running CCQUEUE For example RUN CCQUEUE plotfile DELIVERY NUBS ENDFILE DELIVERY 350 or issue the following command before you run CCQUEUE SET DELIVERY NUBS DELIVERY OF TAPES AND FLOPPY DISKS To have a tape or floppy disk delivered to NUBS call 764 9109 Be sure to give the tape s rack number and external tape id Tapes and floppy disks can also be retrieved by mail but there is a service charge for this service to cover postage and handling COPY TAPERETRIEVE and TAPESUBMIT for more information DELIVERY TO DEARBORN AND FLINT Delivery of plots and page printer output to the Dearborn DBRN batch station is available A messenger picks output up in Ann Arbor on Mondays Wednesdays and Fridays Delivery of plots and page printer output to the Flint FLNT batch station is planned This service will be on an as needed basis so users should contact the Flint station when they have output they wish to have picked up in Ann Arbor NOTE When a user comes to pick up output with a DELIVERY code all of the output with that code should be picked up i e a user j
371. llowing Permit PAGE F_SP5 READ SXYZ The contents of the file are irrelevant to the user When you wish to un permit the file to the user do this Permit PAGE F_SP5 DEFAULT SXYZ Below is as an example the contents of file PAGE F_SP5 K064 Jennie Rowe 764 0505 CRLT SCXP Ray Humphrys 517 373 1284 MDPH Aug 84 SHJ6 Ray Humphrys MDPH Nov 1984 LAP3 Kathy Bazil Jan 31 1985 STOW Eric Quackenbush 517 373 9340 State of Michigan 3 7 85 K7XB Jim Kulik CRLT 764 0505 6 17 85 leave permitted to K7xb k067 for about a year SKB7 8883 Lewis Yen Fitness Res Center 3 2462 8 15 86 until 7 1 88 LALD Jan Opdyke 28 July 88 5 Aug 88 K450 LGE5 Meryl Klein Bus Ad Finance Research 6 3528 8 24 89 9 24 89 Notice that a couple of the files have been permitted to more than one ccid Note It is very important that you inform the person making the request for special forms that on the outside of the package or top sheet there appears in clear and boldly printed letters 1 the SP number that you ve give them SP5 SP6 etc as well as 2 the CCID which will be used with the job which prints the forms When users are ready to print the speial forms job they should Run pagepr scards file PAR PAPER SP5 optional parameters The PAPER SP5 parameter may only be used during the running of pagepr and not as a CONTROL PRINT SET or SIGNON command Textured letterhead is not rec
372. lows DESTINATION OPER TEXTLOC CMDS 636 SYGNUGN SYGNE CMDS0610 SOURCE CMDS 636 SYGNUGN SYGNE CMDS0611 MISCELLANEOUS MESSAGES 413 gt gt gt STATISTICS IS FULL User Explanation Use STA to dump STATISTICS on tape System Explanation This message is written to the operators console whenever a line is written into the file STATISTICS which is whenever anybody signs off and there are more than 7500 lines in the file already The operator should dump STATISTICS on tape via MTS STA before the file gets really full and jobs start SNARKing DESTINATION OPER TEXTLOC MTS 42 MAXSTM PUTSTATR MTS05269 SOURCE MTS 42 PUTSTATR MTS05237 gt gt gt TAPE ERROR ON VOLUME LABEL User Explanation This is the system message given to the operator when the specified tape drive is unable to read the first block on the to verify the presence or absence of a volume label The operator may then try another drive or give an LE response to give the user an error message DESTINATION OPER TEXTLOC MOUNT 101 VOLERR MOUNTCMD MNT02809 SOURCE MOUNT 101 ST183 MOUNTCMD MNT01816 gt gt gt TAPE UNIT DOES NOT SUPPORT DENSITY OF VOLUME LABEL User Explanation The user specified a MOUNT of a tape and the drive that the operator specified did not support the actual mode of the tape If the user did not specify some density which is unavailable on a drive which also supports the actual densit
373. lso Control Unit 11 Channel 0 and the path from Channel 0 to Control Unit t15 EXAMPLE 4 Units D PTRx V CH 2 CU 3 gt DEV ptrx ADDR 208 REV PATH POINTS TO PTRX This example shows the kind of error printed by use of the V option These error messages point to possible inconsistencies in the Supervisor Tables run mta units o d ds0a CH 7 gt CU 18 gt DEV DSOA ADDR 70A TYPE 3270 JOB 512 d d001 CH B gt CU 19 gt DEV D001 ADDR B41 TYPE 3330 JOB 3 CH 2 gt CU 1 gt DEV D001 ADDR 241 TYPE 3330 JOB 3 CH 4 gt CU 2 gt DEV D001 ADDR 441 TYPE 3330 JOB 3 b41 CH B gt CU 19 gt DEV D001 ADDR B41 TYPE 3330 JOB 3 dds0as CH 7 gt CU 18 gt DEV DSOA ADDR 70A TYPE 3270 JOB 512 JOB INITIALIZED NO ERRORS GTUNIT QUEUED DEVICE STATUS CHANNEL END DEVICE END CU DSOA W CH 7 gt CU 18 gt DEV PTRx ADDR 70F TYPE 3284 JOB 2 2 CH OPS 1 CU OPS DEV INITIALIZED NO ERRORS I O QUEUE ACTIVE TASK WAITING I O OP AWAITING CHANNEL END JOB PROGRAMS UNITS OLD STYLE 164 T 33 CH 2 CH 2 CH 4 CH 4 JOB PROGRAMS 50 A gt CU 1 gt DEV D100 ADDR 250 TYPE 3350 JOB 3 1 CH OPS 1 CU OPS DEV INITIALIZED NO ERRORS DEVICE STATUS CHANNEL END DEVICE END gt CU 1 gt DEV D101 ADDR 251 TYPE 3350 JOB 3 1 CH OPS 1 CU OPS DEVICE INITIALIZED NO ERRORS DEVICE STATUS CHANNEL END DEVICE END gt CU 2 gt DEV D102 ADD
374. m a terminal and you find it necessary to STOP a job make sure that there isn t another person working on the same problem from the console or another terminal There is almost never a need to dump the job that is printing the message about CMDSTAT being hung on the console OPERATOR FILES CMD 198 OPERATOR FILES CPW NAME CPW FUNCTION To change the password for a given account to something unknown USAGE MTS CPW Note User Advocate requests for password changes should be honored CPW may be used to suspend the use of an MTS ID to prevent unauthorized use of the CCID The program would normally be used at the request of the CCID s owner when the Computing Center business offlice is closed the owner suspects that the ID is being used in an unauthorized fashion and the owner is unable to change the password Ater CPW has been run no one including the true owner of the ID may use the ID until a new password has been assigned by the business office Under certain extreme circumstances it may be necessary for the operator to use CPW to suspend an account to prevent unauthorized use of an ID without a request from the owner of the ID When this is necessary the ID involved and the reason for the suspension should be reported to operations supervisors and to the business office CPW should never be used to prevent inappropriate use of an ID instead inappropriate use should be reported to the business office an
375. m context COMMAND LPSW SYNTAX PURPOSE EXAMPLE COMMAND CHECK SYNTAX CHECK ON or OFF PURPOSE EXAMPLE CHECK OFF COMMAND CALC SYNTAX PURPOSE EXAMPLE CALC 1EF560 ECD 10 EXAMPLE CALC IPLREAD EXAMPLE command as in SDS COMMAND REPLACE SYNTAX PURPOSE EXAMPLE REP TABLES FROM OBO Note the FROM keyword must be specified Abbreviations are accepted for both commands and keywords SYSTEM START UP PROCEDURES IPLREADER SYSTEM SHUTDOWN AND START UP PROCEDURES 83 CONSOLE DEVICE SELECTION The console device at the U of M is located at address 3A8 The IPL reader attempts to do I O to the device at 3A8 whether it is a 3270 or a 3215 If you do not have a device which responds either like a 3066 a 3270 or a 1052 at address 3A8 then this program will go into wait state after attempting to do I O unsuccessfully on 3A8 This situation can be identified by a PSW right half of 00000600 To define a console in this situation either cause an attention interrupt from the console by pressing request if the console is a 1052 or by pressing enter if it is a 3066 or 3270 If for some reason this doesn t work then do the following 1 enter the 2 byte console address at real memory location X FE0 2 press PSW RESTART Alternatively pressing the INTERRUPT key at any time will put you into wait state with the PSW right half set to 00000100 signifying an external interrupt and then
376. m that runs on the ES3090 300JU and controls its operation One of VM s functions is to provide an environment in which UB MTS can run For more information on VM see the section Running VM XA In order to run UB MTS under VM XA you need to know how to do the following 1 Load UB MTS 2 Switch to a backup UB MTS console 3 Force a UB MTS superdump 4 Stop the UB MTS virtual machine after a shutdown 5 Attach and Detach tape drives 6 Control I O devices It should be noted that there is never any need to set the TOD clock during an UB MTS reload when running under VM XA The clock must be set when you load VM XA If you notice that the clock is set incorrectly when you are reloading UB MTS you will need to reload VM XA to set it properly Some of the above procedures require that you find a VM XA terminal and sign on to VM XA The 3278 terminal located behind the tape drives is connexted to both UM MTS and VM XA The black box next to the terminal has a switch that indicates whether the terminal is connected to UM MTS or VM XA To switch the terminal to VM XA 1 Turn the switch on the terminal s black box to VM XA 2 Switch the TEST NORMAL switch to the right of the screen to TEST and then back to NORMAL 3 At this point the VM XA logo should appear on the screen After VM has been loaded and is running one can use the UB MTS UNDER VM 30 SYSTEM SHUTDOWN AND START UP PROCEDURES following commands fr
377. me Ti SANSERIF 1 is a good choice Be sure to use uppercase It will then ask LETTERING SIZE This is the size of the characters in inches Use a number lt 25 or so 15 or 2 is a good choice The program will then ask for a pen number as for the other tests POLY draws sets of concentric polygons The program will ask LARGEST amp SMALLEST RADII INCREMENT Enter the radii of the largest and smallest polygons in each set of concentric polygons and the increment in radius between successive ones For example if 2 1 1 is entered each set will have a polygon with a radius of 1 inch inside a 2 inch one The program will then ask SIDES lt 0 FOR PCIRC Enter the number of sides in each polygon A large number of sides will cause the polygons to look like circles this is a reasonable test If you enter a number less than 0 a circle will be drawn using very fine steps Finally the program will ask which pen to use You might try 3 1 5 for the radii and increment and 5 for the number of sides pentagons or perhaps 200 near circles For each repetition the program will plot as many sets of polygons or circles across the drum as will fit STAR draws a group of lines which intersect in a single point PLOTTING SYSTEM 444 The program will ask LINES Enter the number of lines which are to be drawn through one point The program will then ask which pen t
378. mmands are underlined to denote an abbreviation e g DISPLAY UAM could be stated as D UAM ALTER name value MODIFY name value These commands may be used to modify a number of predefined variables that control system scheduling Name must be one of the character identifiers described below and value is an unsigned decimal number When changing bit values zero is used to represent off and one is used to represent on Example ALTER UAM 1 ALTER R location disp disp value value MODIFY R location disp disp value value AREAL R location disp disp value value MREAL R location disp disp value value SYSTEMSTATUS 472 These four commands may be used to alter real or virtual memory The AREAL and MREAL commands always alter real memory while the ALTER and MODIFY commands alter virtual memory unless the location to be altered is prefixed with an R The replacement values are hex digits optionally separated by single commas Examples ALTER 600100 C1 C2 C3 ALTER 600100 C1C2C3 ALTER 600000 100 C1C2C3 MODIFY R100 OF ATTN job BLAS DEVI Gives an attention interrupt to the specified job This performs the same function as the GOOSE job program This is useful when there are no job tables available T Job Blasts the specified job This performs the same function as the BLAST job program and as such shou
379. mount statistics for all N and C tapes on both the UM and UB systems This list would be used to determine which tapes should be pulled due to their lack of usage and sent to storage elsewhere sig w035 UM System run sys msscn scards sys mountstat sprint o Enter R rackno etc or U userid etc or P projno etc or ALL r cHHHHH gets all C tapes r n tH gets all N tapes OPERATOR FILES MSS OPERATOR FILES 247 fendfile run ftn scards w225 w32 tpst 57 This run ftn etc compiles the cleanup program for C tapes The cleanup program extracts the last mount statistic for each tape from the output of msscn which is in O and gets rid of extra titles etc The object from this fortran run automatically goes into LOAD run load 5 0 6 00 Now you need to do the same thing for the N tapes copy w225 w32 tpst 57 w32 This copies the first 57 lines of w225 w32 tpst to file w32 gedit w32 p 13 prints 1st 3 lines of w32 Line 2 should appear as INTEGER CEE C OK OK To change this line so that the cleanup program looks for N tapes AD Cher Nis p 2 Now line 2 of w32 should appear as INTEGER CEE N OK OK stop gets you out of the editor To compile this program run ftn scards w32 spunch load 1 In the example above spunch is specified and has a different file name otherwise the object code would automatically go to load
380. mplete before releasing the volume when a blast request is issued As a result inconsistent disk data may result if an I O operation is prematurely terminated NODMGR Purpose To delete a disk drive from the disk manager Prototype NODMGR lt pvname gt OK Parameters pvname is the MTS public volume name of the disk to be deleted JOB PROGRAMS DMGRSTAT 100 Description If Examples DISPLAY Purpose Prototype b DISPLAY GLOBAL c DISPLAY METRIC_ Description If JOB PROGRAMS DMGRSTAT JOB PROGRAMS from the disk manager the disk manager does not have the drive specified by lt pvname gt an error message will be issued and the command ignored If the disk manager does have the drive the user will be asked for confirmation unless OK has been specified on the command If the user confirms deletion VOLREL will be called and the disk drive removed from the disk manager Most requests will be retried successfully using non disk manager methods NODMGR MTS001 the above command will delete MTS001 from the disk manager s control if the user confirms that it is to be done NODMGR MTS007 OK the above command will delete MTS007 from the disk manager s control without prompting the user as to whether that is ok To display various information related to the disk manager a DISPLAY CONFIGURATION COUNTS COUNTS form a is specified the disk drives currently being serv
381. n E Non sync light Checks generator ouput as opposed to critical bus F Fuses Never pull fuses when Power Bloc is on in either Bypass BYP or generator mode Doing so will drop the load G Audio Reset H Alarm Test Reset PB I1T System Status Panel Alarm Section A H Input Power Failure Means commercial power is gone Becomes active anytime the control logic senses no power or reverse power At this time the inverter automatically goes into operation Battery Breaker Open Indicates the status of the battery breaker if on means batterys not connected to PB II Battery Discharging Batteries being used Emergency Power Off EPO switch has been pulled Fuse Failure Failure of one or more fuses Over Temperature Monitors heat sink in inverter over temperature will cause inverter to shut down Logic Failure system is continuously monitoring logic If something is out of sequence etc the light will come on Audio Reset resets horn METERING SECTION 553 I Alarm Test Reset resets erroneous alarms Meter Panel A DC Bus Voltage Visual indication of the DC value in volts B Charger Amps Charge to battery during recharge C Test Switch to test system D Emergency Power off Indicates whether or not EPO has been pulled Batteries Three Cabinets for each system Powering up Power Blocs Note PB I must be started before PB II
382. n filled in with a zero After the request has been built it is sent to the disk manager via SVC DISKIO Upon return from the disk manager the return code for the request executed will be printed For requests which involve destructive operations e g writes and zaps the user will be asked for confirmation if OK has not been specified upon the command WARNING the DISKIO supervisor call is synchronous and the use of the REQUEST command may result in the hanging of this program in diskio wait REQUEST DRWRITE Q0010001A 2 NORELOCATE the above command will write page 26 of volume 1 from wm buffer 2 with relocation disabled Because OK was not specified on the end of the command the user will be prompted for confirmation before the operation is performed REQUEST DRLKREAD 2 19 4 the above command will lock and read page 19 of volume 2 into vm buffer number 4 REQUEST DRZAPTSK 91 the above command will remove task 91 s locks and waits from the locking tables Note that DRZAPTSK is the only request where the lt ea gt may be given as a single integer rather than a hex string or n m form To generate a signal event for a disk manager subtask or subtasks SIGNAL lt event gt lt pvn gt JOBS PROGRAMS 105 lt event gt is a character string specifying the event to be generated See the description below for a list of the possible events lt pvn gt is the MTS public volume number of t
383. n 551 Michigan Terminal System MTS 18 imic Bus 551 MINIMUM Priority 374 iscellaneous Messages 404 MLD 402 MNET AUTH 250 MNET MUP 250 MNET SETDATE 403 MNOP 224 MNS 238 250 MOD 118 OUNT MESSAGES TO OPERATOR 410 ount Queueing 490 Mountscan 241 MSD 239 402 MSG 240 SG SAVE 255 MSS 241 Msscn 241 MSV 255 MTA HOGS 218 MTA OLTEPCMD 416 MTA PAGELOAD 261 TA UNITS O 157 MTS 119 Batch 119 121 Testing developmental 153 TS DEBUG JOBDUMP 410 MTS IT 114 MTS Loading 39 58 MTS PRINTDUMP 365 569 MTS UB Loading 32 MTS UM Loading 26 MTSUM Running Of On The 3090 Native 26 Multi Tasking 17 MUP 250 Mylar Ribbon Copies 415 Name Address Spaces 257 WAS 257 Native Mode UM MTS 26 NDL 258 NDSK 119 Network Control Blocks 113 Network Nodes 113 NEWS CC HOURS 389 NLO 259 No Log Area Available 312 NO SPACE RELOC ENTRIES VOLGET 410 No Subclass Bit On 311 NOT A HASP JOB 411 VOTE 441 444 NOTL 147 NOTO 147 NST 257 NTM 499 NTM OOT UM amp UB 260 NUBS 531 NUBS Terminals starting 306 NUBS Terminal Stations LA36s 531 OFFLINE 123 Offline Equipment Checking For 188 OK 125 OFF 125 ON 125 OLTEP 416 OLTS 416 ONLINE 126 OOT 260 OOT NTM On UM amp UB 260 OPEN 147 OPER Abnormal Conditions 343 OPERATOR 128 Console 128 1052 128 1403 128 3066 128 3277 128 3284 128 Operator Files 168 Operator Messages Misc 405 OPER Backup 343 OPER Console 335 OPER Device Commands 3
384. n active I O queue terminated with UC UE or other error and an SVC SAVE hasn t been done yet Power on popped plug FAKEIO SVC CEWAIT Svc STIO option Ditto Ditto Another task is waiting for the device CSW status Retry operation every so often even if we don t get the device end following a device busy CHANNEL END DEVICE END UNIT CHECK UNIT EXCEPTION Busy and Status Modifier is long control unit busy Busy Status Modifier and Control Unit end is short control unit busy STATUS NOT OPERATIONAL PCI INCORRECT LENGTH PROGRAM CHECK PROTECTION CHECK JOB PROGRAMS UNITS OLD STYLE CHANNEL DATA CHECK CHANNEL CONTROL CHECK INTERFACE CONTROL CHECK CHAINING CHECK JOBS PROGRAMS 167 NOT OPERATIONAL status is reported as channel status X 3F The remaining device and channel status bits are defined in the 370 Poo JOB PROGRAMS UNITS OLD STYLE 168 OPERATOR FILES Operator Files Operator files are initiated using the MTS command and giving as a parameter the name of a public file which will contain at least an MTS SIGNON command The operator file may CONTINUE WITH another file but there must eventually be a SIGNOFF command to terminate the task Unless otherwise specified when initiating the MTS task all output directed to msink will be funnelled into the great infinite wastebasket It is to your advantage to name your operator files if you create them with names of fo
385. n in GOM For information on these subroutines see Dick Salisbury TAPE DIRECTORY PROJECT 517 UMNET PCPs and The C Data Concentrator CC The C Data Concentrator AB AD AE AF AN UMNET PCPs Operator Files for PCPs and the C Concentrator BDL 1 DDL Used to dump and EDL load UMNET PCPS AB FDL AD AE AF and AN NDL J Startup Files DCS CCS both start CC MNS starts AB AD AE AF and AN using ABS ADS AES AFS and ANS CLO Loads but does not dump the WON Data Concentrator BLO 1 DLO Loads but does ELO l not dump UMNET PCPs FLO AB AD AE AF and AN NLO J Note Use _DL only if a PCP reload is needed due to a power failure or by a system request otherwise use _LO TAPE DIRECTORY PROJECT 518 UNATTENDED MODE From midnight to 8 A M on Sunday and Monday mornings MTS will be available but in Unattended Mode During these periods a Security Clerk when one is in our employment will be present to maintain the integrity of the Computing Center but will in no case be responsible for operating or restarting the system Users will be made aware of the situation by a broadcast and a SIGNONM message The signonm message will mention the fact that MTS is in unattended mode and will ask users to COPY UAM The phones recorded messages will also mention the fact that the system is in unattended mode During unattended mode the doors to the
386. n released row Once a dump tape has been analyzed and is no longer needed it is released by moving it to the top row for re use and marking it released in the operator s dump log book The operators NEVER release tapes that is your responsibility Aside from crisis situations where a dump was printed and perhaps even released right away your job each morning is to pick up the new bunch of dump listings and figure out FIRE FIGHTERS GUIDE 371 what to do with them The BATCH receipt numbers will be in the dump log book For non MTS jobs e g HASP PDP SUPERVISOR usually little more is done than making sure the person responsible for that component knows about the dump and receives the DMP produced listing For MTS jobs our goal is that we will all learn to do most of the detective work in figuring out what went wrong 4 BACKGROUND INFORMATION ABOUT JOBDUMPS AND DUMP TAPES In many situations when something goes wrong with MTS or UMMPS a job dump is taken In producing a dump the supervisor writes all real memory and all virtual memory for the job being dumped in an internal format to a dump tape Other miscellaneous information is also included The rack numbers for the dump tapes are DUMP01 through DUMP36 The volume names and the external label always match the rack number The program MTS PRINTDUMP is used to scan the dump tapes for various types of information and print a formatted listing For
387. nager Prototype a STATUS DISK lt pvn gt b STATUS SUBTASKS c STATUS TASK lt task gt Parameters pvn is the MTS public volume number of the disk whose status is to be displayed task is the task number of the task whose requests are to be displayed Description If form a is specified the disk manager will check to see that it has the disk drive for lt pvn gt If not an error message will be issued and the command ignored Tf the disk manager does have the drive the JOB PROGRAMS DMGRSTAT JOBS PROGRAMS 107 current state of the DISKLING and IOMNGR subtasks for that drive as well as the queues of requests currently being processed for the drive will be displayed If form b is specified the status of all the subtasks which currently comprise the disk manager will be displayed If form c is specified the status of all the requests for the task specified by lt task gt will be displayed JOB PROGRAMS DMGRSTAT 108 JOB PROGRAMS FLUSH This is a HASP command and is obsolete NAME FUNCTION PROTOTYPE USAGE PARAMETER JOB PROGRAMS FLUSH FLUSH To clear requests from HASP for I O operations on a device FLUSH device i d This job serves as an infinite sink for output from HASP for a given device It is used when the device has hung up in such a way that does not allow it to be DRAINed The usual sequence of commands to clear this hang up is STOP th
388. name gt Hicore system is running Start xDL Meaning lt name gt is running but must be dumped and reloaded using ADL BDL CDL or NDL Once this is done rerun HUH Message Unexpected status from lt name gt show this output to Meaning This message should never appear but if it does it isn t clear just what the problem is or how to fix it Rerunning HUH might result ina better message the second time In any case please show the output toa Communications Group member Next if the UMnet PCP was operating normally MUP will count the number of UMMPS jobs using the network and print the following message aa Idle Inbound bb Active Inbound cc Active Outbound dd Hasplings where aa is the number of free MTS jobs serving the network bb is the number of MTS jobs serving inbound network users cc is the number of outbound connections mounted from MTS jobs dd is the number of HASPLINGS acting as network batch transmitters and receivers Next one of the following messages will be printed if anything out of the ordinary appears in the totals Message There are no inbound MTS jobs serving OPERATOR FILES MUP 254 OPERATOR FILES the UMnet network Meaning Either MNS was not run it was run but the UMnet PCP was not available or SHUTDOWN has been run Running MNS should start the necessary MTS jobs Message All inbound MTS jobs serving the UMnet network are in use Meaning This i
389. nblank character unless there are more than four characters given in which case only the last four are used The N descriptor followed by a 1 to 8 character job name e g PDP specifies all jobs with the given name If no name is given all non MTS jobs are printed In the argument for the D U and P descriptors the pound sign is used as a fill character which matches any character SYSTEMSTATUS 483 Examples T T OTP T P WHHH T U 0123 USERS The USERS command displays counts of the current number of terminal users batch jobs executing available terminal lines non MTS jobs running pages of virtual memory in use and pages of real memory in use Example U SYSTEMSTATUS 484 The following is a list of the predefined names that may be used with the DISPLAY ALTER AESTATUS ABSTATUS ADSTATUS ANSTATUS APG AVAILJT BDL BJP BJT BPH BTR CASE CAW CS CCSTATUS C1L c1s C2L C25 C3L C38 C4L C4S C5L C5s DIW DRI GRABS ITL ITH LBJ LLF LPJ LSS MAB MAXBJTSM MBH MDR MIB MINBJTSM MPP MSP MTHITH THIVAL MTLOTH SYSTEMSTATUS and MODIFY commands Status of UMnet PCP AE Status of UMnet PCP AB Status of UMnet PCP AB Status of UMnet PCP AN Real memory page frames available Current number of available unused job tables Batch startup delay secs BJPMOD enabled flag 0 gt off 1 gt on Big job threshold real pages Bat
390. nd other helpful information when you do this IFD147I Examples of Device Field 1FD1471 0181 test device 181 1FD1471 185 187 test device 185 186 187 IFD1471 CHICAGO TP2 TEST SYMBOLIC TP DEVICES IFD1471 CHICAGO1 AND TP2 IFD1471I NDR NO DEVICE REQUIRED FOR TEST IFD1471I CH282 TEST CH2 USING DEVICE 82 IFD1471I SLASH ALONE TEST PREVIOUSLY SELECTED DEVICE S IFD148I EXAMPLES OF TEST FIELD 1FD1481 2400 TAPE TESTS T2400A T2400Z 1FD1481I 2400A SEC A OF TAPE TEST 2400 IFD1481I 2400C 2 IFD1 RIN 2 SEC C TEST 2400 48I 2400A C E G SEC A B C E AND G OF TEST 2400 OLTS IFD1 481 IOEPZZ SEC ZZ OF EDIT PRINT FUNCTION IFD 1481 IFD1 481 R2540AA SEC 419 AA OF READER TEST 2540 SLASH ALONE RUN PREVIOUSLY SELECTED TEST S IFD IFD 1491 IFD1 IFD1 149I 491 491 TESTING LOOP IFD1 IFD1 491 49I ERROR LOOP IFD IFD 1491 1491 ERROR PRINT IFD1 49I IFD1 49I IFD 1491 IFD1 491 CONTROL PRINT PARRALEL PRINT PRINT IFD1 491 FIRST ERROR IFD1 491 COMMUNICATION IFD 1491 MANUAL INTERVENTION IFD1 49I REMOTE FE CONTROL IFD1 49I EXTERNAL DATA IFD1 49I EXAMPLES OF OPTION FIELD IFD1 49I IFD1 49I EP TL 50 FE EXT A B PP NMI RE
391. nd the other on the back DPA Action When users run tapesubmit the information which they type in is entered in a data base When users present tape s at the I O window another label must be filled out with the user s name department phone address etc A tapesubmit program is continuously running on a terminal screen near the I O window A DPA upon receiving a tape enters information as requested by the program initials reel size tape ID receipt The program responds with either tape receipt no assigned to rack number xyyy where x is a letter and y is a digit or That tape ID does not match the record Accepted tapes are then taken to the operators area When users wish to retrieve tapes a program named taperetrieve is run at which time information pertinent to the tape is TAPES TAPE HANDLING 487 entered The running of this program causes a request to appear on the operators console as follows 00525 08 47 01 MTS K4MU Tape Retrieval c0414 Tape ID SE1 Return via NUBS 00525 08 47 02 MTS At this point the operator may type Enter the receipt code of the tape NA receipt code if the tape is mounted or unavailable II receipt code if ID is incorrect If the receipt code is correct it is typed in which removes the tape information from the data base and a delivery code sticker is attached to the reel An operator should note the time and date on this sticker Prese
392. nection is released Examples run mnet setdate par cu an n Al told me to run mnet setdate par NDT run mnet setdate par cu ae n MERIT SETDATE 404 MISCELLANEOUS MESSAGES Following are miscellaneous messages that may appear on the operator s console HASP messages are not included see the HASP Operator s Manual Machine check and similar messages presented earlier in this manual are likewise not included MISCELLANEOUS MESSAGES 405 KKKKKKKKKKKKKEKKKEKKKE OPERATOR MESSAGES kk kkk kkk kkk kkk KEE gt gt gt ACCTNG ERROR READ WRITE SIGNON User Explanation System error see an MTS systems programmer System Explanation During signon processing an error return was received from either GETACCR RC 8 or PUTACCR RC 0 DESTINATION OPER TEXTLOC CMDS 636 SYACCRAR SYGNE CMDS1088 SOURCE CMDS 636 SYACMR SYGNE CMDS1086 gt gt gt ACCTNG STAT ERROR READ WRITE TSA User Explanation System error see an MTS systems programmer System Explanation An error was flagged by the file routines when the TSA TAILSHEET STATISTICS ACCOUNTING subroutine tried to read or write the accounting files or the statistics file A SNARK follows DESTINATION OPER TEXTLOC MTS 42 SYACCWAR TSA2 MTS02841 SOURCE MTS 42 TSA2 MTS02839 gt gt gt CATALOG WAS EXTENDED SUCCESSFULLY User Explanation No problems just informative DESTINATION OPER SOURCE FILE SYSTEM 482 23 GETSEG CA
393. next to the red EMERGENCY knobs to the right a quarter turn as far as it will go and back to it s normal position It is usually necessary to reset the Air Conditioning system To do this you will have to enter the electrical rooms These rooms are just to the left of the elevator on each floor In each room is a large Air Conditioning power panel and to the left of each of these panels is a gray box the bottom of which is about five feet form the floor On the front of these boxes are two rectangular buttons separated by a white rectangular bar The top button is green and is marked START Pressing this button will reset the Air Conditioning on that particular floor The lower red button is marked STOP Make sure that the entire system is up except for some components that may be offline for some reason or other such as maintenance 6 After the IBM ES3090 300J has been powered up you must sure that all of the attached components are up and functioning Magnetic Tape equipment Make sure that all tape units are powered up Rewind and unload all loaded tapes If the tape units aren t powered on go to the rear of the 3800 Contol Units CUs open the doors and power them up using the button located at the bottom The tape drives also have POWER ON switches at the lower rear of each unit DISK Storage Often some of the Disk Drives fail to come up after a power interruption Amdahl Disks Before power
394. nfirm by typing OK DESTINATION OPER SOURCE FILE SYSTEM 482 5 VOLGET gt gt gt VOLUME NAME INCORRECT User Explanation This is the system message given to the operator when the tape that the operator has mounted in response to a mount request does not have the label that was specified in the the request DESTINATION OPER TEXTLOC MOUNT 101 BADVOL MOUNTCMD MNT02806 SOURCE MOUNT 101 ST182 MOUNTCMD MNT01812 MISCELLANEOUS MESSAGES 415 MYLAR RIBBON COPIES The following procedure may be used for producing mylar ribbon copy on the TN printer 1 DRAIN printer to be used a TN printer 2 Change ribbon and print train or chain a Remove cloth ribbon b Be sure a TN train or chain is mounted c Install mylar ribbon 3 Reverse paper to side without lines 4 Set TN print buffer TN PTRX ECHO X being the number of the printer to be used This is a JOBS command and must be entered from Operators console 5 Adjust timing and print density a RUN MTA TNRIP 1000 SPRINT gt PTRX b While TNRIP is running adjust forms thickness and print density The characters and are best used for the adjustment c UNL above program to release the printer The program is stopped by an attention interrupt 6 Adjust paper alignment a Set printer to neutral b Press CHECK RESET and CARRIAGE RESTORE c If TEXT360 manual output move paper down one line d Set printer to DRIVE MY
395. ng Dave Parker 9 1 517 377 3780 or 9 1 517 377 3700 ICS Canon International Calculator Sales Inc 994 5079 or P O C81158 9 1 358 0140 Lee Data Manager B Ernst 9 1 313 647 7840 or 1 708 986 5550 Service See ComputerVision Customer Code No 80079 001 Litho Service Services Developer Photo Typesetter Service amp Sales 9 1 313 593 4333 Megabit Communications Inc 9 1 612 481 0921 MEMOREX Manager Daniel N Bendtsen 9 1 313 444 2511 X232 or Art Carpenter 1 708 955 1020 Sales 9 1 313 354 4511 Tech Support Rep 9 1 800 522 0822 Michiko Nielsen 1 313 524 8800 X212 Service B Miller 9 1 800 331 7410 Site Code 28809150 P O No Tape Drives G34536 Line Ptr G02426 Ontels See Telcom Engineering SORBUS NUBS Site ID 1189391 01 Manager 9 1 313 591 2225 or 9 1 313 591 1707 Service 9 1 313 591 1270 SORBUS Services IBM 2040 1052 2821 2 1403 N1 amp 2701 at NUBS also IBM 2022 amp 2501 at UNYN Spacesaver Electric Tape Racks Casper Systems Corp General Phone No 9 1 517 321 4697 Service Jerry 1 442 9000 If no answer Call general phone no and ask to have Jerry VENDOR INFORMATION 530 paged P O no 84100 StorageTek Mgr Warren Kasting 9 1 313 354 2860 Sales Jane Whitehead 9 1 313 354 0850 Service 9 1 800 525 0369 or If problem is unusual etc have F E call us Site Code 44569 Call Collect 9 1 303 673 4056 TAB Terminals etc D Mazzorana 9 1 313 477 5420 Racks R Smi
396. ngs may not be the appropriate action but it probably won t cause a catastrophe CMDFDIR under SYS at UM contains a line for each tape file That line contains the beginning and ending times for the file in binary and printing forms the mount request for the Mmmm tape the file is on and the posn of the file on the tape On the tape the file DSN contains the current time and the last character of the name of CMDSTAT 330 the file the data was copied from The lines are ordered by beginning times CMDTAPE writes these lines when it copies files It puts each line in its proper place so it does not always write at the end These lines contain binary data so handle with care Programs which read CMDSTAT data from the tapes use this file for an index CMDPIKUP does not use it CMDTAPE and soon CMDPIKUP prints comments on the console and sends messages to the firefighter when it has problems The next section describes these messages CMDTAPE Messages All messages printed sent by CMDTAPE are labeled DWB A message is sent to the firefighter for a problem which needs fixing but will not prevent CMDTAPE from continuing A message is printed on the console for more serious things Console messages require some response mostly to intensify the message If the message does not ask a question a null line is an adequate response FILE ffffffff IS UNAVAILABLE or LOCKED File ffffffff could not be used for c
397. nothing wrong If the terminal was not already in use try pressing ATTN Ontel or RETURN DECWriter and Visual 550 and see if you get the UMnet Hermes message If you do then the terminal and SCP are both OK Try typing UM and RETURN and see if MTS responds with the MTS Ann Arbor message If it doesn t call the MTS operator and tell him that you have a working terminal and SCP but that MTS won t respond If you don t get any response to the SHELLO then the terminal or the SCP to which it is attached is probably dead Look at the map on the last page of this writeup and figure out the name of the SCP to which the terminal is attached Find another terminal on the same SCP and test it using the directions in step 2 If it works then the SCP is OK and you should try the first terminal again If it is still dead power the terminal off and report it as being out of order If the second terminal doesn t work either then the SCP is probably dead You should look at the lights on the front of the SCP Be absolutely sure you are looking at the correct SCP If either of the LOAD lights is on the SCP is reloading and all terminals on that SCP will be dead until it is finished UNYN TERMINALS SCPS ETC 525 If neither of the LOAD lights is on but the terminals are still dead you should reload the SCP by pressing the INIT switch down and then back up Whenever you reload an SCP write it in the
398. nt one of these tapes during CCP and the tape cannot be mounted see the following Reason tape already in use or no tape drives available Action na the tape Reason tape can t be found gets a label error fatal tape error wrong tape id is given Action ab the tape NA will cause another mount attempt after all the plots have been done then if the tape still can t be mounted the plot will be put back in the queue AB causes the user to get no second try and no plot The plot is not requeued When CCP has finished running it will print out a comment PLOTTING SYSTEM 423 containing information on how many minutes of plots there are and what color pens are to be used Example CCP DONE APPROXIMATELY 410 MIN PLOTTING BLUE RED USED QUICK PLOTS PLO PLOTOQUIK This program processes plots for which the user has requested quick or ball point pen service When this program is started it will check the queue for such plots and plot any that it finds It will then wait for more plots When more plots are queued it will plot them and then wait again It will continue to do this until it is stopped Plots processed by PLOTQUIK are put onto the plot tapes by CCP with the normal p lots PLT handles these differently see the PLT writeup Running the program Set up the plotter with the black blue red and optional green ballpoints in holder positions 1 2 3 and 4 re
399. ntly the stickers are printed as NUBS UNYN CNTR MAIL FLNT and DBRN The tape is then placed in a rack near the I O window A user signed on at a terminal or running HASP batch may at any time request a tape to be mounted Users do this by using the MOUNT command or subroutine as outlined below mount c3296 on 9tp t vol restl2 research2 See MOUNT in MTS Volume 1 for complete description If a batch user were to run above mount request the following would appear on the operator s console 00097 09 37 05 MTS SQB1 HB 1 9TP C3297 ON 9tp RING OUT RESEARCH2 00097 09 37 25 MTS This indicates to the operator that the user signed on as task 00097 requests the tape with Rack No c3296 to be mounted with the ring out ona 9 track drive The tape is supposed to have the paper label RESEARCH on it The operator would then either locate the tape and in response to request from task 00097 reply C3296 Txxx to indicate the tape is or will be mounted on drive Txxx or else indicate that for some reason the tape cannot be mounted If the tape is mounted and the rack number the operator types i e C3296 the ring status in or out the device type i e 9TP and the tape volume serial that is written con the tape magnetically i e RESTL2 agree with what the user requested then the operator receives after mounting the tape the messages TAPES TAPE HANDLING 488 00097 09 38 58 MTS Mount check OK
400. o turn NMM off m ntm 0 When NTM is on users will be informed that tape mounting isn t possible but that this is only temporary TAPE MAILING SUBMITTING AND RETRIEVING TAPES BY MAIL Users may now send magnetic tapes and diskettes floppy disks by mail Also upon request the Computing Center will return tapes and diskettes to users by mail To mail a tape or diskette to the Computing Center the user must 1 Obtain a mail submission form either from the Mail Librarian at the Computing Center or by COPYing the file TAPESUBMIT 2 Package all tapes and diskettes securly and send them along with the completed form s to the Mail Librarian at the Computing Center Mail Librarian Computing Center 1075 Beal Avenue Ann Arbor Mich 48109 When tapes or diskettes are received they will be placed in the racks at the Computing Center and a receipt will be mailed to the user There is a charge for retrieving magnetic tapes and diskettes by mail This charge covers the mailing and handling costs The cost for mailing magnetic tapes 7 00 for the first tape and 3 00 for an additional tape mailed in the same package Each package holds only two tapes The cost for mailing diskettes is 4 00 for the first diskette and 1 00 for each additional diskette mailed in the same package Each package holds five diskettes The charge may be paid by check or money order or charged to a sponsored research account To retrieve a ma
401. o use This program can be used to test plotting paper For each repetition the program will plot as many stars as will fit across the drum NOTE prints a character string along the drum The program will ask MSG Enter the text to be printed The program will then ask for the pen number and will print the message BLACKEN draws parallel lines which may overlap It can be used to draw over the same line many times The program will ask LENGTH OF LINE INCHES All lines drawn will be this length It will then ask SPACING OF LINES INCHES Enter the distance between lines Usually you will want the lines to overlap so enter a small number e g 01 Enter 0 to have the same line redrawn over and over Then the program will ask LINES TO A BUNCH Enter the number of lines to be drawn If the spacing is 0 this will be the number of times the same line is redrawn Finally the program will ask for a pen number For each repetition the program will draw the specified number of lines It will then draw the time message and begin a new group of lines This can be used as a test of fiber pick up for various inks pens and papers DASH draws dashes The program will ask DASH INTER DASH LENGTHS Enter the length of dash to be used and the distance between PLOTTING SYSTEM 445 consecutive dashes in inches The program will then ask TIMES Enter the number of dashes to be drawn in a
402. of RAMROD The entire IPLREADER process from the time the LOAD key is pressed to the time the CONFIG initialization message is printed usually takes less than 5 seconds The IPLREADER itself currently has three components 1 The first component IPLINIT is the program which writes the IPL data onto the iplarea of a disk in the first place It is an MTS program which may be viewed as an extension of DASDI 2 The second component IPLBOOT is a stand alone program which is IPLed from the IPLAREA when the LOAD key is pressed and which then loads or bootstrap loads the third component 3 The third component IPLREAD is the stand alone program which does all the work It provides the operator with a repertoire of commands for loading and starting systems displaying and modifying memory etc All three components are designed to run on a 3090 without modification SYSTEM START UP PROCEDURES IPLREADER SYSTEM SHUTDOWN AND START UP PROCEDURES 79 IPLBOOT the IPL Bootstrap when the LOAD key is pressed this program is loaded into real memory starting at address X 8000 to avoid the PSA logout areas The first thing that this program does is to read the core image of UMLOAD from the iplarea into real memory starting at location X 1000 Then it calls UMLOAD to load IPLREAD into the top end of real memory If everything goes OK IPLBOOT passes control to IPLREAD If something goes wrong
403. ointed info every time it is started So when things get really hosed up the preferred thing to do is to kill it off and then restart it 2 For the time being we have both HASP and the RM around Lines entered from the operator s console as well as SY O from a terminal go to HASP if they start with a dollar sign Eventually they will go to the RM But for now you need to RUN SYS RMOPR and give it the RM commands If you are using the operator s console MTS RMC will do this for you 3 The remainder of this document discusses primarily BITNET processing by the RM as this is the only current production use More will be added later dealing with printing remotes and the like How to tell if something is wrong 1 Use SYSTEMSTATUS to enquire about the RM tasks SY Tasks RM or SY T R should print something like on UM 00032 RM LOGR 00076AA0 19 Monitor Idle 00033 RM MAIN 00076D60 361 Monitor Idle 00034 RM ACC 00076ECO 17 Monitor Idle 00035 RM CKPT 00077020 128 Monitor Idle 05318 RM LINK 000AD520 91 Monitor Idle SA17 02297 RM LINK 000ADD60 92 Monitor Idle SA07 There are one each of the logger main accounting and checkpoint tasks The LINK tasks are for the BITNET links and there is one for each link Currently SA07 is the WAYNESTT link to Wayne State which is our connection to the rest of BITNET SA17 is UMICHUB which is the link to
404. older task the one with the smaller task number This will allow the other to finish initialization RESOURCE MANAGER 543 San Diego Super Computer Consortioum The University of Michigan SDSC Remote User Access Computer SDSC UMICH EDU Modem located on second floor The smaller satellite dish is connected to this modem via two coaxial cables There is a larger gray interface cable leading from the modem to the VAX eqiupment on the first floor Modem customer service code MI 011 Customer service phone no 1 800 221 1617 Note LED Light Emitting Diode If any error indicators are on call Hal Marshall or Len HJarding If the Channel BER indicator has a number larger than 20 call Hal Marshall or Len Harding Make a note of the error lights DEC RX02 LEDs that should be on are Run DC On and Process There is a key that must be used to activate the switches This key must be in Local in order to use the Boot switch Turning this key to the left powers the system down VAX 11 750 Location Code 72Q ID 11750 86068326K Service Phone No 1 800 482 0164 After Hours 1 800 323 6612 The proper positions for the switches on this box are Boot Device to D Power On Action to Boot Key in LOCAL This for booting system LEDs POWER RUN ERROR dim DF112 This is for remote diagnostics and should only be powered if requested It is unplugged TU81 LE
405. om cp q names under VM cp attach cp detach cp q tapes UB MTS UNDER VM the UB operators console this shows what is running attach devices to VM detach devices from VM VM tape information SYSTEM SHUTDOWN AND START UP PROCEDURES 31 Attaching and Detaching Tape drives When UB MTS runs under VM it is very important that no tape drives be attached to MTS except for the ones that UB MTS is using at the moment If a tape drive is attached to UB MTS but in use from the UM MTS system it is possible that VM cound cause the tape to be rewound unexpectedly this would obviously cause problems for the UM MTS user of the tape To avoid problems of this sort it is necessary to issue an explicit ATTACH or DETACH command at the VM operator s console to make a tape drive available or unavailable to UB MTS These commands are issued automatically by MTS for tapes that are normally mounted however the commands must be issued by the operator under other circumstances For example if an MTS task that is using a tape SNARKs or is BLASTed MTS will be unable to automatically issue the required DETACH command so the operator will need to issue the DETACH command at the VM operator s console Another example of a case in which it would be necessary to issue manual ATTACH and DETACH commands would be to run job programs like the STAT job or to issue WIM commands to write tape marks UB MTS UNDER VM
406. ommended as toner doesn t adhere to it very well SPECIAL FORMS JOBS 549 Uninterruptable Power Supply General Information The computers and their peripherals the air conditioning and the lights are running through the UPS system that is Edison power flows through the UPS to the equipment The elevator and the emergency lights remain on a direct commercial power link For the time being if problems occur with the UPS system call Lee Mitchell or the Electrical shop 4 9444 Note Don t try to fix things yourself Do not open covers on the UPS equipment If there is a power failure certain lights will be off so whenever you decide to go the the Generater or UPS room always take a flashlight with you As you enter the UPS room just to your left under the light switch there should be a light in a holder This light is plugged into a wall socket where it is being charged You can disconnect this light and use it until the charge runs out You shouldn t go into either the Generator or UPS room when it is dark It s too dangerous Power Failures Following a complete power failure where batterys are completely discharged power through system 0 will come on automatically This will bring on the building lights but will not include the lights in the MG and UPS rooms so take a flashlight downstairs with you Following a power failure where batterys are completely discharged the PB I MGs on syst
407. omment HELP 1 0 TROUBLE ON PLT1 occurs see if someone is using the plotter if not type RES If after using RES the I O trouble still occurs signoff and start again A description of cleaning and filling pens is found on p 11 of the first OPERATIONS MANUAL of two manuals which were supplied by the Vendor When it is time to refill a pen simply put the plotter in manual while it is running and remove the pen from the pen carriage To do this grasp the pen by the top of the pen holder and turn it counter clock wise 1 8 turn and lift it out See p 11 of the Vendor supplied Operators Manual for pen maintenance and filling See p 5 of same manual for paper changing OUT OF PAPER SWITCH MODEL 936 A photocell has been placed at the bottom center of the rear vacuum column of the CalComp 936 plotter The plotter now automatically stops if it runs out of paper as long as the overhead lights are on If these lights are off the plotter will behave as it did without the switch Operating Instructions IMPORTANT Before doing anything else put plotter in manual since the manual auto switch is over ridden by the photocell 1 Be sure that the room lights over the 936 plotter are on 2 Be sure that the plotter is in manual 3 Interrupt the plotter 4 Let concentrator run out plot buffer by putting plotter in PLOTTING SYSTEM 439 auto covering the photo cell with something e g a card or y
408. on 3380 R Seek check S PCI fetch mode and correctable data check 3330 3350 DAC UNIT CHECK ERROR CODES 363 T Intervention required U Channel data check V Chaining check W Environmental data present X Truncation error during data check correction N Program interrupt looking at user s CCW list Z Incremented seek address past a file mark boundry Regular recording of buffered log of activity on this disk eccccccce gives the hexadecimal value of the unit check routines input switches This is only of use for programmers which work on the unit check routines themselves dd gives the hexadecimal value of the supervisor s sense flag This value is interpreted by the unit check routines to determine the cause of some kinds of tape errors It is mostly of interest to programmers working on the unit check routines but also required to interpret channel detected errors described below If this value is anything other that 01 the sense information will be unreliable eeeeeeee is the hexadecimal sense data associated with tape error Again this is interpreted by the unit check routines to determine the type of error and how to recover from it This information may be of use to both tape vendor customer engineers if they are attempting to correct a hardware problem and also to programmers working on the unit check routines themselves ffffffffFF is the hexadecimal value of the first seek argum
409. on cards as described in chapter 3 of the PM2 manual It is normally a file The default is CFIGIN DUMMY SAVEIN save save must be an fdname from which will be read save cards as described on pages 3 16 and 3 17 of the PM2 manual It also is normally a file The default is SAVEIN DUMMY HISTORY and TAPEVOL These parameters indicate to PM2PUSH and PM2POP which of the two sets of history files to push or pop Two consecutive pushes or pops will have no effect All PM2 macros except PM2HDMP automatically push the appropriate history files upon successful completion The default is to push or pop neither set of files All of the following paramaters are useful only when debugging PM2 they should be ignored for normal use of it and are included here only for completeness VSSCMDS cmdfdname PM2 449 Cmdfdname must be an fdname from which VSS commands will be read immediately before PM2 is invoked It may be MSOURCE VSS vssobj Vssobj is the name of the file containing the object for the version of VSS to be used to run PM2 PM2LIB pm20bj Pm2obj is the name of the file containing the object library for PM2 itself It must be a VSS PDS as produced by the MVS link editor CNVTEREP cnvterepobj Cnvterepobj must be the name of a file containing the object for the CNVTEREP program that converts MTS unit check log records into the format accepted by PM2 and EREP CNVTCTL ctlfdname Ctlfdname
410. on in the machine check interrupt code MCHK SOFT RECORDING DISABLED No more recovered machine checks will be reported The SOFTCHK operator command can be used to turn the reporting of recovered machine checks on or off MCHK MACHINE CHECK IN WAIT STATE An otherwise fatal machine check occurred in WAIT STATE the system will try to ignore it MCHK SUPERVISOR DAMAGE Some information crucial to the supervisor was damaged by the machine check RE IPL required MCHK RECOVERED STORAGE PROTECT FAILURE Failure in the storage protect mechanism that has been recovered from by reloading the damaged keys from the relevant PCBs The system will attempt to continue MCHK STORAGE PROTECT FAILURE Solid failure in the storage protect mechanism Printed when the key error bit is on in the machine check in the machine check interrupt code MCHK STORAGE DATA FAILURE Uncorrectable error in main storage Printed when the storage data failure bit is on in the machine check interrupt code MCHK MACHINE CHECK IN M C HANDLER Second machine check while trying to recover from a machine check MCHK DEGRADATION REPORTED The hardware has moved some part of the machine from service because it is malfunctioning Printed when the degradation bit is on in the machine check interrupt code MCHK NO SUBCLASS BIT ON Machine check occurred for a reason that cannot be determined MCHK INSTRUCTION PROCESSING DAMAGE
411. on it SAV can be used without ridding the system of users Usually only one save tape is called for but occasionally a second may be requested If the tape requested by SAV is in use a user is restoring some files from it the operator should type next and a different tape will be requested by the program Tf a tape error occurs the program will ask if it should continue to write on that tape The operator should answer no and a new tape will be requested by SAV If a new FSnnnn tape is requested it should be initialized by writing a tape mark on it first SAV aoutomatically starts the directory merge program which does the same thing as FSM No operator intervention is required by FSM FSM is also started at every system reload in case FSM or SAV were interrupted by a crash If SAV generates any error messages of the form FILE SAVE ERROR they should be reported to a systems programmer SVW will continue to be used for the large weekly file save OPERATOR FILES SAV 284 OPERATOR FILES NOTE SDB should be run after the SHUTDOWN ALL COMPLETE appears If UB has been taken down before SDB has been run SDB can be run from the UM system Also see SDM NAME SDB NOTE If SDB cannot be run for some reason use a terminal which connects to the network and at which host type HG Then signon and sou mnop sdb If HG isn t available net wu sig w035 sou mnop sdb
412. on one phase This also checks for improper phase sequence which could damage hardware If there is a loss of or extereme low voltage on any phase an alarm is actuated and a shutdown will occur if the condition persists after a short time 3 LOAD ON BYPASS Light will illuminate and alarm will sound when the MG is being bypassed 4 OVERTEMPERATURE Causes a TRANSFER Also light will illuminate and audio alarm will sound if there is high temperature in the MG or if in BYPASS mode high temperature in the bypass transformer 5 OVERLOAD Causes a transfer Also light will illuminate and alarm will sound if bypass output contactor or generator contactor is overloaded B MIMIC BUS Mimics input and output power using LEDs in the panel C METERING Displays input voltage A B B C and C A and A B and C to neutral Also shows LOAD Calculates maximum output as opposed to current being supplied etc D CONTROL PANEL and FUSES includes a Motor Stop b Motor Start c Transfer Initiate transfers load to Bypass BYP or MG when used in conjunction with the METERING SECTION 552 Transfer Switch d Generator Time meter runs when Generator is running Note the small square in the upper right corner a red flag should be seen moving e Transfer transfers to Motor Generator or Bypass BYP f Emergency Off For emergencies gt Smoke or fire from Power Bloc or electrocutio
413. on the front of the tape If the tape was being read the premature rewinding would cause incorrect information to be read TAPE ERRORS TAPES TAPE HANDLING 494 TAPE ERROR TXXX AAAA BB C DDEE FF XXXXXX IITII JJJJ KKKK LLLLL MMMMM The following Tape Error information is used by RST TXXX Tape drive name AAAA Path BB Software flags Es Error codes see TAPE ERROR CODES DD Device status from CSW EE Channel status from CSW FF Sense flag should be 01 XXXXXX Sense data up to 24 bytes LELT Rack number up to 6 characters JJJJ Volume Serial number if labeled up to 6 characters KKKK Data set name if labeled up to 17 characters LLLLL File number MMMMM Block number TAPE ERROR CODES The tape unit check routines are a collection of subroutines which are invoked by most programs when the tape drive they are using detects an error The purpose of these routines is to correct the error or if it is uncorrectable to report the error to the operators console All errors are logged in CMDSTAT and may be retrieved by using EREP Recoverable errors do not cause a message to be written to the console but unrecovarable ones do The job programs which use the tape unit check routines are MTS and almost all programs which are run by MTS HASP the STAT job and the collection of tape utility jobs FSF BSF REW RUN ERG and WTM The above tape error information is used by r
414. onging to the ID SYS that has a name beginning with STATY exists whether as a result of this procedure or of SYS STATCOPY having been run previously all subsequent emptying of STATISTICS by this procedure will be done by running SYS STATCOPY Furthermore STA will not save the current STATISTICS file while there is a file belonging to SYS with a name starting with STATY anda systems programmer will have to manually make repairs OPERATOR FILES STM SVS NAME FUNCTION USAGE OPERATOR FILES 295 SUM To remove the network message installed by use of SDM mts sum OPERATOR FILES SUM 296 OPERATOR FILES SVS See SVW for SVS and SVW differences and similarities OPERATOR FILES SVS OPERATOR FILES 297 SVW NAME SVW FUNCTION Weekly file save Saves a copy of all disk files on tape USAGE mts svw Before running SVW there are several things you must do e A BROADCST message must be sent to the users telling them to signoff The phone message must be changed Execution must be held The SHUTDOWN PART command must be issued The system should be taken out of unattended mode mod uam 0 e Make sure that no users are signed on It is all right to leave the public files with user id SYS CMD etc running Behavior and output are pretty much the same as the daily file save see example under SAV except that each stream fills up more tapes and
415. onsole 00023 10 40 13 MTS RAEXKARKARKARKARKARKA HELP SNARK IN MTS KKEKEKKKEKKKKKKKKKKKKEKK do an mts srk at operators console this message indicates that the task whose number is specified on the first line 23 in sample above has encountered an irrecoverable system error In short the job has folded up and is going out feet first A SNARK indicates that a job is terminating because of a failure in MTS It is necessary to take special steps to remove a SNARK ed task from the system These steps are outlined here Step Step Step 1 KKKKKKEEKTAKE A JOBDUMPX XX XX 2 Determine whether the task is associated with a terminal job or a HASP job To do this initiate the TASKS job program with the job number of the job which is SNARK ed as an argument This will type out the standard tasks message format By looking at the devices associated with the job you will be able to determine for terminal tasks the line adaptor it had and for batch tasks the S8 number 3 The next step after having identified the type of job is to initiate the STOP job program giving it as an argument the task number of the job which has SNARK ed If stop fails to ABNORMAL SITUATIONS SUMMARY Step step Step 309 terminate job see STEP 4 However should STOP work OK and should this be a terminal job which has SNARK ed proceed to STEP 6 below then you are finished 4 This step is only to be und
416. ooooon nooo nooo 359 D f rrediPLOtS a aian e a a A uaa lade Bab aaa e a Bho E RE S 420 Delivery ees soe A SA SAS Aw athe EINE SO A wy ESS 349 Device Commands 3270 OPER os is se aoa ee ee tele sd soa ud as 341 DESKCOPRY 02 oes outs ance Bei A ae ae hic coe A cov an idea s a e azist Sheeler 355 DISK ETOR COGESSs is A BAe ty seme Made Sakae B S iia BG 361 DUES ET LOES A Bees We ace wk ee cae a a Bate 361 DESK PES Marling Pn oe eee iii 500 DT USkettes ia MarlIng Aa a a adan 500 Disk amp Fixed Head File Unit Checks Printing ooooooooooo 317 Disk Manager DMGR Disk Manipulation o o o 351 DESK Storage Ai A a A A A A A A lee Bags es 351 DMGR DESK Mana PULA LOM dale 351 D mps P inNtING A as 365 EMERGENCY POWER OFP Asdo 84 as e a e a a ot ele Gee ah tae a a E ee EN ae a a eS LS 462 EREP Print MCHKS amp I O Outboard Errors o oooooooooooo ooo 318 Error Codes DISK Wis wreak ra A a Rodis A aE 361 Error Formats PDP a o 75 FILESAVE Schedule And Tape Storage o o ooooooocoooooooooooooooooo 333 Firefighters oeu ind tte dad A a AA deen od du eee 468 13 Fire Fighters iGuide iS A Sete SAS Oe 369 Fixed Head ELMO ALLE eld ah sees aed eo tess 3 A a Sales Seer ieee eae Oreud ae olen ude ahs 361 Fixed Head Fil e FOrmatt ing insti ee eid ee ee We id ae a ew ee ale Dae eee 132 HASP Priorities Classes Schedule oooooooooooooooo nooo 374 HeErmes
417. ooooooooooooooooo ooo 549 UNIX FP ULES IMA A A a AA date ee ee ae 557 IBM ES3090 300J Water Temp FlOW o oo oo 559 Powering On Loading Procedures 9790 ooooooooooonoomooooo 560 Xerox 9790 Photo Receptor Belt o ooooooooooooooooooooooo 562 14 Quick Reference Guide ABNORMAT ST TUATIONS SUMMARY 000 A a A eet 307 CARD READER ERRORS S A A O A te er less Jo A a 314 CPU ERRORS tution ss be Ra co deh ah ia bos oe dead caren da 310 Disk amp Fixed Head File Errors Printing oooooooooooooooo 316 Disk Unit Checks 1 ase tal Sat Wah ee See eS ed a WAS RE Awe a eet 316 EMERGENCY POWER QEPE ci A Stee Bs gk LS ener ds Seb Sk See SRS eee ee 462 HARDWARE PAT UIURES anane serie ee ve as A Rew a Gece 310 Logouts BU And Chanel iy Ge Se Peat Sd A eet amp 316 achane Check Messages ni fee sesh a kA a t 311 emorex Line HUNG asbie eee Bes ETa E Sees E Aa A ae AAA E E 322 emorex 1270 tine TO ETrorS Geru eu aee a pa aa a a SiG RE aa 315 PRINTER ERRORS och ais aa fete he ahs SS a aa a Sine aae av Sabet e hate eens ne bier eher ar guohe et 313 PUNCH ERROR G s 3 tee ole eee eae Heed ar ne ele Ss ae a sree etd Sak Se oa eas lao ao 313 SNARKS Crit ara eee ees a oe hee we es Ta aie a one the ae eae ta Ge we oe 308 CCE eaP lOt Tape Initia lizatil n sss ene ws vile et eos Se bh ene eee Wa she ob tendo Be oben 430 CCTS Weng th
418. opying when CMDTAPE copied a CMDSTAT file to tape CMDTAPE continued anyway The data now missing from ffffffff can be sifted from the tape file The file should be checked to see what the problem is locked means just that and unavailable means anything else NOTE CMDTAPE does several minutes of waiting before declaring a file to be locked RC n FROM WRITE ON LIOUNIT n of Mmmm OF FILE p LINE nnn CONTINUING WITHOUT IT A non zero return from write occurred while copying to the copy file attached to logical unit n mmm p and nnn give the current CMDSTAT tape position use this information to sift records off the tape The file should be checked to see what the problem is TAPE Mmmm HAD A TAPE ERROR NOW USING NEXT TAPE If CMDTAPE gets a tape error while writing or positioning a tape it will try again on a different drive If that fails it will abandon the tape and send this message Check the tape and make any repairs necessary to make it readable Also check the tape drive s this problem is often caused by drives having difficulty with tape marks TAPE Mmmm FILLED IN lt 5 DAYS PLEASE CHECK This may or may not be an error It does mean than data is being generated at a much faster than normal rate Check the CMDSTAT 3d data for too much of some record type some program going crazy generating it or for something wrong with CMDPIKUP or CMDTAPE CMDSTATC WAS MARKED FULL IN CMDDIR BUT WAS EMPTY
419. or the appropriate device address An Operator may examine and probably solve the problem in the following manner Procedure to present CHANNEL END and DEVICE END to the Channel from the 1270 CE panel Simple CEDE A Set TEST switch OFF in the bottom row of switches B Set TEST ADDRESS switches to desired address NEVER change address switches with TEST switch on Remember that Memorex Address 0 is equal to Channel Address X CO0 C Press the gray CMD switch in the TEST FUNCTION switch group leave it in D Set roller to panel A E Set the COMMAND BIT switches at the right end of roller A to 011111 This is the 1270 internal code for a DISABLE command For a zero bit the switch should be DOWN fora one bit it should be UP F Set TEST switch ON G Push SINGLE CYCLE in the bottom row of switches down momentarily it will spring return to the normal position H Set TEST switch OFF I Repeat the procedure for other addresses as needed CEDE with UNIT CHECK and TIMEOUT status Follow the procedure given above for the simple CEDE but set the COMMAND BIT switches in step E to 00010 This is the 1270 internal code for a READ command After it is entered the READ command will timeout MEMOREX LINE HUNG 323 in 28 seconds or less and present ending status to the Channel MEMOREX LINE HUNG 324 Some Memorex 1270 Information Displaying line information
420. orrible death switch to the backup console by selecting the x function of the OP menu of the macrocode Switch back to OPER mode to use the backup console The job name is OPERATOR and no parameters are required If it is desired to override the default devices a 3270 should be given as the first parameter and a 3284 3286 3287 1052 1403 or NONE as the second parameter If a 1403 is to be used as the console log printer the UCS buffer must have been previously loaded using the PN or TN job If only one parameter is given the default printer list will be used If the 3270 specified by default or otherwise is not available the OPERATOR job will terminate with the message NO 3270 AVAILABLE or devn NOT AVAILABLE If the 3270 is available but the printer is not the OPERATOR job will continue but the message NO PRINTER AVAILABLE or devn NOT AVAILABLE will appear on both the 3270 screen and the backup console In this case the PTR device command may be used when a printer becomes available When the OPERATOR job is successfully started two greeting lines giving the time date machine model and serial the device names of the display and printer and the device type of the printer are displayed at the backup console and on the 3270 screen Before restarting the OPERATOR job all pending requests for input through the backup support console must be satisfied If this is not done it will not
421. other information This will help you to decide whether to re position the tape before running msscn Example Msscn Run run sys msscn scards sys mountstat sprint 0 The program will then wait for you to enter rack numbers CCIDs project numbers or all followed by an endfile u abcd this will list all tapes mounted by CCID abcd that is the last mount and the last mount with the ring in r nl234a this will print the last time s that this tape was mounted as well as the last time with the ring in r c30 H The acts as a wild card that is it will look for and print all numbers and letters that are in its position in this case c30 it will do from c3000 through c30999z The can be used in any or all positions e g n 2 4 or HHHHHH Using would be the same as the all parameter Example If one wanted to pick up all tapes in the range c3650 to c4300 and all tapes used by CCID scd4 they should do the following OPERATOR FILES MSS 244 OPERATOR FILES run sys msscn scards sys mountstat sprint o r c365 r c37 r c38 H r c39 r c40 r C41 r c42 r c4300 u scd4 endfile or you could r c3 r C4HHHE and then delete the lines in ooo after you run the sort which follows that you didn t need OPERATOR FILES MSS OPERATOR FILES 245 After running sys msscn if you expect more than a few output lines you should run a program w
422. our hand 5 Install new roll of paper 6 Set up the plotter to begin a plot 7 Do a BAC check to make sure that only the last plot was lost otherwise backup further NON QUEUED PLOTS USING SYS PLOT36 or SYS PLOT51 The program SYS PLOT36 uses the CalComp 936 and SYS PLOT51 the CalComp 1051 Before a plotfile can be routed to the plotter it must be translated into plotter commands by a program such as SYS PLOT36 or SYS PLOT51 CCQUEUE is a program users usually use to queue their plots for scheduled runs Ccqueue does no translation If one wishes to plot a file which has not yet been modified to be plotted directly or does not want the plot to be queued then the program SYS PLOT36 or PLOT51 may be used Note For net work attached plotters gt pltn doesn t work Instead do the following mou mnet plot d hXxCcb7 where xx is 00 for plt2 and 01 for plt3 Run sys plot51 and enter plot Bin at where to plot Bin for WHERE TO run sys plot36 or 51 user types WHERE FROM program asks w035 asd user types as example of plotfile to be used WHERE TO program asks plot Bin user types if the user wants to route the plotfile directly to PLOTTING SYSTEM 440 the plotter SCALE FACTOR user enters scale factor 0 25 example PRIMIMG TIME SEC user enters time interval in seconds that is wanted before pen priming occurs after last use of pen or user presses E
423. ove prodedure will cause a Superdump tape request to appear on the Operators console on the second floor SUPERDUMPS 468 System Reliability Computing Center System Reliability Project System Operation Reliability Committee This message has three parts 1 to explain what the CCSRP is 2 to exaplin the purpose of the SORC 3 to issue a statement on firefighter paging procedures 1 The CCSRP includes everyone designated as eligible to serve as a Systems Group Firefighter 2 The SORC is a management level committee charged with monitoring overall system reliability 3 Firefighter beeper policy At the last SORC meeting agreement was reached on the proper calling sequence and timing for contacting a firefighter FF When a system problem occurs which requires assistance from a EF Run Sys Fire for list of Firefighters currently Fire Fighter 930 8086 Systems Manager 930 8976 See next Only F F allowed to beep manager Don t call 1 Firefighter has been called but can t fix p 2 Firefighter hasn t responded within 20 minu SYSTEM RELIABILITY FIREFIGHTERS 469 SYSTEM STATUS RECORDING Information concerning the status of the hardware and software is availbale to users of MTS by listening to a taped message The phone numbers for this message are 763 0420 and 763 0421 The tapes are recorded and played on a Model 5001 Multi Line Announcer manufactured by Automation Electronics Co
424. ow PF1O0 the area and any partial input removed When present definitions that duplicate 3278 keyboards the use of ALT shift key 11 and 12 are spares already in the DEVICE COMMANDS Note are not translated into upper case A request line prefixed by a percent sign is device command SFAST Switches into make the OPERATOR which we don t tries to minimize FAST work certain have In FAST mode the number of mode on SLOW Switches into slow mode SLOW mode device Lines which begin with a field mark interpreted The only valid device commands are The OPERATOR defaults key command If pressed spare keys cause the cursor to be moved to the request request PF keys 13 through 24 have initial the definitions for PF1 gt PF12 PF1 gt PF12 requires the use of are is On the as a FAST mode was added to types the OPERATOR erase writes of 3270s done to REFRESH Makes sure that the screen is up to date OPERATOR CONSOLE 3270 342 EMPTY Empties the screen PF Lists PF key definitions IGNORE jobn Causes future message lines from job number Jobn to be ignored i e neither displayed nor printed This is useful if a job is producing large amounts of output which is known to be of no interest Message response requests from jobn will never be ignored The job numbers of any jobs being ignored will appear in the system statu
425. owed by a space and a brief explanation If problems are experinced during the day contact the Computer Communications Group or the MERIT office or if at night call one of the current UMnet system firefighters NLO NAME FUNCTION USAGE NOTE No dump OPERATOR FILES 259 NLO Loads Primary Communications Processor AN mts nlo occurs when this is run OPERATOR FILES NLO 260 OOT NAME USAGE FUNCTION OPERATOR FILES NLO OPERATOR FILES OOT mts oot Automatically starts NTM and offlines all tape drives on the UB system at 10 am on weekdays At 4pm OOT removes NTM and onlines t906 and t907 on the UB system and offlines them tho the UM system One can online more drives on the UB or UM system if necessary OPERATOR FILES 261 PAG NAME PAG FUNCTION Runs the program MTA PAGLOAD which creates an artificial paging load on the system This program is useful for testing the PDP and its paging devices USAGE mts pag REMARKS This job will continue to run until GOOSEd OPERATOR FILES PAG 262 NAME FUNCTION USAGE EXAMPLE OPERATOR FILES PLB OPERATOR FILES PLB This program prints identifying information about plot tapes i e the name e g TUE 1000 when it was last written and the minimum time to re use To run from the Operators console enter mts plb The program will print ENTER TAPE NAME Respond with th
426. perform a validity check on the table INTERTASK JOB lt TASK_NUMBER ALL gt or INTERTASK NETWORK lt NETWORK_NAME gt ALL gt or INTERTASK dumps intertask control blocks JOB lt task_number NONMTS all gt this dumps the job table page tables etc task number is decimal LOADCLASS This prints the status of each task ala taskstat and then prints a categorized table a la loadclass MTS This calls MTS POOLS This dumps supervisor storage pools SUPERVISOR lt FULL CHECK gt This dumps the PSA s and the supervisor error save region sverr and so on FULL causes all supervisor pages to be dumped CHECK invokes a validity check of supervisor management TRACE This prints the trace table UNITS unitspec unitspec is any units job command dump output is the same as from the UNITS job VM first_page lt last_page gt or VM lt MTS gt lt GSFS gt or just VM These dump virtual memory first_page and last_page are hex page numbers e g 400 means address address starting at 40000 MTS dumps storage in a format convenient for mts debuggers GSFS causes a dump of storage control blocks Any input line beginning with a is assumed to be an MTS command and is passed to MTS to be executed All commands begin in column 1 with parameters separated by one or more blanks Attention interrupts are handeled as follows a message is printed on SERCOM acknowledging the interrupt If PRINTDUMP IS doing
427. porary private files may be created by a user during the processing of a job These are automatically destroyed at SIGNOFF time All three types of files may be organized as either line files or sequential files Line files have a very useful indexed sequential structure i e each logical record within a line file has a unique line number associated with it so that a user may easily request access to only one or a range of the lines in the file Perhaps the main advantage of line files is that they are easily edited Sequential files are used for storing large amounts of data that need not be in a form that may be easily edited Note Sequential files are seldom used any longer since the OVERVIEW OF THE SYSTEM 20 INTRODUCTION TO MTS THE MICHIGAN TERMINAL SYSTEM OVERVIEW OF THE SYSTEM lines in a line file can be much longer than in the past Which form a user chooses for a private file depends on the nature of data and the ease of access for editing purposes that one wishes to have with respect to that file When a file is created for a user the user may specify the initial amount of storage to be set aside for it If it is not specified MTS assigns a small amount of storage and will automatically expand the size later if necessary until some limit is reached Virtual machine capablity is available under the MTS system MTS provides the users with memory for problem programs in a restrict
428. program designed to make strip labels for our tape racks It may be altered to change the height length and number of labels etc ina strip After any necessary alterations the program must be compiled and then put into a form acceptable to the plotter more about this later What you must do first is determine the height and length of the strip to be produced The height in the program is stated as STRIPH As you can see in the program below STRIPH is set at 10 0 16 0 If after measuring the strip holder you decide that the strip height should be 11 1 2 16ths change STRIPH from STRIPH 10 0 16 0 to STRIPH 11 5 16 0 using the editor STRIPL is the length of the strip and can be changed the same way NSLOTS is the nuber of rack slots that you want in the strip NSTRIP is the number of strips to be plotted It s best to try 1 or 2 strips at first as a test to make sure that they fit LETRH 0 4 STRIPH FACTOR iS a proportion and you may need to change this 0 4 several times until the characters on the strip appear to be the right size NCHAR is the number of Single digits following the letter e g C2345 kK 0000 is the starting figure in the strip and in the way the program is set up now C000 would be the first number produced on the strip Changing K 0000 to K 0001 would cause the first number to be coool Any of the statements mentioned in this writeup may be chnaged using the editor In the string CALL PNUMBR XLOC YLO
429. programs in itself to UMMPS it is merely one of a number of job programs Functionally then we have a hierarchy of command levels UMMPS is in complete control of the system it allocates the use of all hardware MTS is one of several job programs that run under the supervision of UMMPS The typical user of MTS communicates with the system via MTS commands in his command stream examples of such commands are SIG CRE EMP LIS RUN under normal circumstances such MTS commands are meaningless if entered at the operator s console since it is UMMPS or the OPERATOR job not MTS that services the console The user never sees UMMPS the interface between MTS and UMMPS is hidden from him When he makes a request of MTS MTS asks UMMPS for whatever it needs to handle that request The MTS job program is reentrant Thus there can be many MTS tasks using the same MTS job program A user may operate within MTS in one of two modes conversational mode or batch mode When operating in conversational mode the user communicates directly with MTS via a terminal in real time each command typed by the user at his terminal is immediately processed by MTS and the results of this processing are made known to the user at his terminal The user may decide what needs to be done and enter the appropriate MTS command The user thus carries on a conversation in real time with MTS via his terminal hence the term conversa
430. queue numbers will be listed on under requeued and their receipt the plot sheet Running Deferred Plots Run CCP in the normal way except use the plot tape marked deferred plots This tape is in the rack with the other plot tapes The deferred plots will be added to the logical end of the tape i e after the plots already there The deferred plots will remain on the deferred tape until the tape is reinitialized see below All normal plots will remain on the queue and their receipt numbers will appear in the list of requeued plots on the plot sheet Plotting Deferred Plots Run PLT in the usual way Note that the deferred tape is added to so that you may not want to plot all of the plots The WHICH TAPE message from PLT will say that the deferred tape is being plotted EXAMPLE MTS PLOTS FROM DEFERRED TAPE TAPE WRITTEN 13 46 15 03 11 78 MTS 1 BLAK 4 2 BLAK 3 3 BLAK 6 MTS ok To Stop Deferring User Plots This should be done to allow a users plots to be processed in the normal way after they have been deferred Start the MTS job in the file CCD You should be promted for a ccid Enter all ccids for which you wish plots to no longer be deferred one per line Terminate the list with an end of file CNCL EXAMPLE OPERATOR FILES CCD mts ccd ENTER SIGNON ID OPERATOR FILES 183 mts PLOTS FOR MTS WILL BE PROCESSED NORMALLY sys PLOTS FOR SYS
431. r 15 VTAM should then terminate promptly 8 Once VTAM has stopped type PEND and press ENTER 9 A few more messages will be printed as VSE finishes shutting down VSE 54 SYSTEM SHUTDOWN AND START UP PROCEDURES 11 Hold down the ALT key and press the PAl DUP key This will put you in CP READ mode 12 Type LOGOFF and press ENTER At this point VSE2 is completely shut down Shutting VSE1 Down Note VSE2 must be shutdown before VSE1 1 Go to the VSE1 operator s console 2 Type TC and press ENTER 3 VSE should respond with something like F2 022 K103I ENTER TERMINAL CONTROL OPERATOR COMMAND F2 025 4 The 025 part of F2 025 is known as the reply ID It won t always be 25 Type the following command substituting the reply ID in place of the 25 25 CEMT P SHUT I and press ENTER MSG F4 5 At this point the instructions for shutting down VSE1 are exactly the same as for shutting down VSE2 starting with point number 2 After VSE1 and VSE2 are completely shut down you can reload or shut VM XA down Mounting Tapes VSE1 All tape use under VSE1 will be done from batch jobs VSE1 can only run one batch job at a time so there will never be more than one VSEl tape user at a time although that one user may mount more than one tape at a time For a while at least VSEl users will call before submitting batch jobs that use tapes and leave their name and phone number If
432. r should show 414V and the AC amp meter should momentarily show 20a The 20 amp surge shows that the DC Capacitor Deck is being charged 3 Close Battery Disconnect s on wall cabinets 4 Close individual Battery Disconnect breakers on battery cabinets As each cabinet breaker is closed the AC Amp Meter should jump You can t see this if your e alone 5 Clear Alarms using Audio Alarm Reset and Audio Reset 6 The DC Volt meter should show 400 DC volts The entire Power Bloc s should now have power Powering down Power Bloc system Power down PB IT first 1 Take batteries offline by opening Battery Disconnect breakers on individual cabinets and then the Battery Master Disconnect s on the wall panels Battery Breaker Open alarm will go on This is normal 2 Reset alarm test reset and audio reset 3 Open the Charger Rectifier breaker The DC bus indicator should start dropping very slowly 4 Reset Battery Discharge alarm Power PB I down 5 Depress transfer initiate and hold The non sync lamp should start to go off and on However the off and on states occur very slowly This light should come on and off about 2 times per minute 6 After the non sync light dims rotate transfer initiate to Bypass BYP You should hear contactors open The MG is still running 7 Reset alarms using alarm test reset and audio reset METERING SECTION 556
433. r ungracefully dies NDSK specifies that the MTS job is to run without using the disk subsystem This is useful when installing MTS at a new installation and useless thereafter NMSG specifies that MTS should not try to access the system message file for SIGNON messages or Bulletins STSK is used internally to start an MTS terminal job as a subtask of another job The group of parameters following the msource designation are optional The msink parameter specifies the device that is to be allocated to the master sink This is where in MTS output directed to the terminal would go The second parameter in this group denotes the I O device where output directed to the MTS logical I O unit SPUNCH would go if SPUNCH were defaulted Normally the operator does not have to worry about either of the optional parameters which precede or follow msource If msink does not appear then msink defaults to the msource device unless the msource device is a file in which case msink defaults to the MTS pseudo device DUMMY which is the infinite wastebasket If syspch does not appear then calls to the MTS logical I O unit SPUNCH are illegal if SPUNCH was not assigned in the user s command stream or program The parameters msource msink and syspch must be a multiple of 4 characters long use trailing commas to pad things see example 7 below and must be broken into groups of 4 characters by blanks It
434. ral hosts on the network After a terminal is dialed into Hermes and its type and speed have been determined the user is prompted for the name of the particular host Hermes provides the user with a common front end to all the hosts on the network users can access MERIT HERMES SYSTEM 401 remote hosts as if they were local and can access these hosts with the same set of keyboard editing characteristics To the user Hermes appears to greatly resemble a Data Concentrator Some of the code for the support of Hermes is from the PCP MCS system The terminal identification procedure is almost identical to that of the Concentrator and the keyboard editing characters are essentially the same The hardware line adapters and Modems is similar to that of the Concentrator although there are additional synchronous devices for the Inter Nodal communication From an operational viewpoint it is very important to realize that although MTS may be down locally users may still be accessing other hosts via Hermes Therefore do not power down or halt any of the UMnet PCPs or busy out or power down any of the MERIT modems unless it is essential to do so Currently there is no simple way to broadcast to Hermes users about impending shutdowns etc Such a broadcast will be installed shortly Messages broadcast from the MTS operators console are not seen by Hermes users accessing hosts other than the one at the Universit
435. rameter may be used to specify a comments field for the entry This field is an arbitrary character string of up to 256 characters The COMMENTS parameter must be the last parameter specified CREATE type MODE 800 1600 6250 PFMT JX JS JIM JD I LI SFMT JX JS JM JD 1 L1 This command creates a master entry for type The master entry specifies the mode to be used for tapes allocated to this type category and specifies the formats for the primary and secondary search fields If the MODE parameter is omitted the default mode will be 1600 bpi PFMT and SFMT specify the formats for the primary and secondary search fields respectively The formats available are JX Julian microseconds unsigned 8 byte int JS Julian seconds unsigned 4 byte integer JM Julian minutes unsigned 4 byte integer JD Julian days unsigned 4 byte integer I Integer signed 4 byte integer LI Long integer signed 8 byte integer If PFMT or SFMT is omitted then there is no TAPE DIRECTORY PROJECT Command Explanation Command Explanation Command Explanation 513 corresponding search field assigned to the entry DELETE type TAPE rack VOL label SEQ n PAR CURR LAST This command deletes an entry or series of entries from the tape directory The parameter type specifies the type of tape entry to be deleted If PAR CURR is specified the entry currently in progress is deleted if any
436. rated by blanks intlist is one or more of the following keywords separated by blanks SVC I O IO EXT PGNT or SIO P may be used in place of M may be used in place of JOB PROGRAMS STATSW 146 NAME FUNCTION PROTOTYPE USAGE PARAMETER EXAMPLE JOB PROGRAMS STOP To forcibly terminate a task STOP nnnnn If the task is an MTS task it will be forced into signoff there will be no loose ends as there are with the BLAST job If the task is not an MTS task it may have enabled a trap for the STOP command if it has done so it depends on what that For a terminal task if STOP is successful the job should terminate in a few minutes This allows a terminal task to go through a full signoff procedure If after a reasonable time the task does not terminate with a code of KILL then a BLAST may have to be used The STOP command should not be used more than once on the same job If the task is a terminal task the operator will have to re enable the line with the MTS command if that line is to be accessible to users again As with the BLAST command make sure the task you are STOPping is indeed the one that you wish to STOP and not that of another poor user or a HASPLING task nannn A job number stop 345 NOTE The SYSTEM STATUS command kill nnnnn performs the same function and is especially useful if all job tables are in use JOB PROGRAMS STOP
437. re VSE can be loaded VSE must be shut down before VM XA can be shut down Currently we run two VSE machines VSEl and VSE2 Note Currently we have promised the library to keep VSE available Monday through Friday from 7 AM to 4 AM the following morning so we need to avoid scheduling reloads during that time period VSE tape mount requests have priority over all UM and UB mount requests A VSE mount request should be given the next available tape drive regardless of the UM and UB tape mount queues There are three things that operators need to do to run VSE 1 Load VSE1 and VSE2 2 Shut VSE1 and VSE2 down 3 Mount Tapes Loading VSE1 and VSE2 This should only be done after loading VM XA or after shutting VSE1 and VSE2 down 1 Go to the VSE1 operator s console a 3278 with the label VSE1 CONSOLE on the front The VM XA logo should be displayed 2 At the LOGON prompt type VSE1 and press ENTER 3 The terminal should respond with the message Enter Password 4 Type the password and press ENTER If you don t know the password ask your shift supervisor 5 The terminal should respond with the message LOGON AT VSE SYSTEM SHUTDOWN AND START UP PROCEDURES 51 time and the word MORE should be displayed in the lower right part of the terminal 6 Hold down the ALT key and press the PA2 field mark key The screen should clear and then begin displaying messages as VSE1 loads 7 After
438. repetition Numbers in the hundreds or thousands are quite reasonable here Finally the program will ask for a pen number This can be used as a test of ink globbing as the pen is lowered or as a test of pen block overheating For these purposes use short dashes and short interdash lengths e g 1 and 05 Several thousand to a repetition is also good For each repition the program will draw the specified number of dashes going back and forth along the drum It will then draw the time message and start a new set of dashes LINE draws a continuous line back and forth along the drum The line does not cross itself The program will ask NUMBER OF TIMES Enter the number of times the program is to draw back and forth 60 inches total in each repetition The pen will not be raised within a repetition The program will then ask for a pen number The purpose of this program is to test ink and pen points Usually you will want a large number of repetitions For example 50 times would produce a 250 foot line before the pen is raised to draw the time message SYMTEST draws a character string several times in decreasing sizes Enter the character string to be drawn when the program prints STRING The program will next print LARGEST SMALLEST SIZES Enter the largest and smallest sizes in inches at which the string is to be drawn Then enter the difference in size inches between successive strings The program will
439. rily get global time limit exceeded at the end of that time If the job has no limit set e g a terminal job you will get the message JOB 00427 MTA NOLIMIT If you supply an invalid task number JOB 99999 2277 If you supply an invalid user id JOB 00000 BARF LIM will come back and prompt for another line of input until terminated by an end of file OPERATOR FILES LIM 236 OPERATOR FILES LMT Name LMT Function Same as MSS except more limited Lists no more than last four instances of tape usage Usage mts lmt NOTE Before using see MSS OPERATOR FILES LMT Name Function USAGE OPERATOR FILES LRL 237 Gives information about last IPL i e time in hours minutes and seconds since last IPL as well as the reason reload mts lrl date and time for the OPERATOR FILES LRL 238 OPERATOR FILES MNS NAME MNS FUNCTION To start up UMnet tasks on the AA AB AD AE and AN PCPs after IPL USAGE mts mns Response on operator s console is MTS UMnet AZ Configuration 6 April 1992 MTS Jobs on AZ10 through AZ6F MTS UMnet AY Configuration 6 April 1992 MTS Jobs on AY10 through AY6F MTS UMnet AX Configuration 6 April 1992 MTS Jobs on AX10 through AX6F MTS UMnet AW Configuration 6 April 1992 MTS Jobs on AW10 through AW6F MTS UMnet AV Configuration 6 April 1992 MTS Jobs on AV10 through AV6F
440. rnoff 9 1 313 477 5420 Telcom Engineering Services Ontels amp VISUAL 550s Bill Doty 9 1 483 6669 Terminal Services Small DataSouth printers attached to 9700s Sales Mgr Mark Width 92995 9355 Sales Mike McManus 9 995 9355 Service Kurt Shelling 9 995 9355 UMTel Channel Extenders Phones etc 763 9138 VISUAL 550 See Telcom Engineering Walton Enterprises Dish 2 4 M 1700 W 120V Mr W B Walton 9 1 714 683 0930 XEROX 9790 S Ns CNTR 95600 2695 UNYN 95600 1784 NUBS 95600 3203 Serve Mgr 4045 D Rauser Mgr 9 996 8700 Business Office 9 1 827 3159 Printing Systems Rep Debbie Eastwood 9 1 313 827 3159 Software Acct Rep Judy Freedman S N V31 403615 P O L54271 Service 4045 CNTR 9 1 517 482 2414 9 998 4220 9 1 800 822 2979 or 971 356 5300 1 CalComp After 5pm Answering service takes message service is performed the following day VENDOR INFORMATION See Maintenance Agreements 531 NUBS TERMINAL STATIONS LA36s At present there are twelve Decwriters at NUBS Seven of these are connected to the Data concentrator and five to the Memorex 1270 Below is a simple diagram showing port and station numbers as well as a position diagram of the terminals Stations 10 9 8 7 6 5 4 Pte eee ee MUX Ports 71 64 57 4 3 2 1 C H O A U Z C N E Oo 6 L O O N E T P P R L I E E A O N Stations N N 2 3 C R E 12 11 ee et MUX Ports 16 15 14 13 12 11 10 91 81 Station Position Di
441. rporation To make a tape Make certain that the OPERATE button is pushed in this is the Power button e Snap a tape cartridge into the slot at the upper left of the recorder next to the microphone plug in Push in the Record button and leave it in until the elapsed time equals the time printed on the tape cartridge e g leave the Record button in for 14 seconds to erase a 14 second tape e Plug in the microphone Some microphones have an ON OFF switch which must be turned on The tape cartridges have time limits printed on the outer edge Often one will want to repeat a message several times on a tape with a pause in between Of course this depends on the physical length of the tape and the length of the message Most of the buttons on this machine may be pushed in and will stay in that position until pressed again at which time they will spring back out Press the red RECORD button and record your message e To check you message leave the RECORD button out and press in the PLAY button If you wish you may re record the message e When you are satisfied with the message i e it s length tone content etc push in buttons 1 and 2 on the right front of the machine as you face it This will allow persons calling the two connected phone lines to access the recorded message RECORDED SYSTEM STATUS MESSAGE 470 Comments The two phone line buttons may be left pushed in while one is m
442. s been dumped to tape PLT must be used to replot it When a COPY command is issued to PLT it will copy the quick plots into the file with the normal ones but will not plot them To plot quick plots specify QUIK at the end of a plot or COPY command Example COPY ALL QUIK plots all the quick plots on the tape When QUIK is specified only quick plots are plotted PLOTTING SYSTEM 427 UNUSUAL CONDITIONS Plotting System CCP Dies If CCP dies several registers and remarks will be displayed on the operators console Bring these to the attention of the appropriate systems programmer for the time being Gail Lift Once CCP has been started and the plot tape has been accepted by the system CCP cannot be restarted See Gail Lift if a restart is necessary ERROR MESSAGES FROM THE PLOTTING PROCESSOR This section explains the error messages which the plotting postprocessor prints on the plot receipt sheet The postprocessor will print one of these messages whenever it is unable to produce a plot USER TAPE UNAVAILABLE This tape containing the user s plotfile could not be found by the operator at the scheduled plotting time The most likely cause is that the tape has been picked up by its owner HARDWARE SOFTWARE ERROR IN FILE The postprocessor encountered a hardware error or software inconsistency in reading the file containing the plot Have the user see the
443. s area following any Not Ready device names IGNORE requests may only be entered for currently active jobs and a job number will be automatically accepted if the job terminates for any reason As many as 16 jobs may be ignored at once ACCEPT jobn Message lines from jobn will once again be displayed and printed A ACCEPT is automatically issued for any job that is being ignored upon termination ACCEPT ALL Message lines from all ignored jobs will once again be displayed and printed PFn new definition Redefines program function key PFn New definition is an arbitrary string of length 30 characters or less To define a partial request the new definition should end with an underscore _ Otherwise it will be interpreted as a complete request 3270 devn Switches the console to the 3270 whose device name is devn PTR devn Switches the printer to the 3284 3286 3287 1052 or 1403 specified by devn The devn may also be NONE in which case no printer is used RESTART Directs operators console traffic back to the 3270 This is useful if the console has been switched to the 3066 or 1052 backup support see next section when the 3270 support is still operational usually because of paging problems This command may be issued at any time without harm and will cause two greeting lines to be displayed on both the backup console and the 3270 screen WRAP ON Normally the OPERATOR job s
444. s not an error but any additional users who try to access UofM through UMnet will get the all interactive ports busy message Finally if the PCP is operating normally and outbound access to the network is currently restricted by MNET AUTH see above then MUP will check the state of each host on the network and print one or two explanatory messages if they are not operating normally MUP may be run from any MTS terminal using the following command run mnop mup Ouput is written to SPRINT and or SERCOM PAR WAIT may be given or the RUN command to request a fifteen second wait before MUP starts its checking This is useful in xDL just after the PCP has been reloaded but before it s hada chance to respond properly PAR SHORT may be given and MUP will only check MNET AUTH and the current status of the PCP It won t count the UMnet jobs or attempt to open connections to each UMnet host The string PAR MED may be given which will cause MUP to check MNET AUTH the status of the PCP and to count the UMnet jobs but won t attempt to open connections to each UMnet host The strings PAR WAIT SHORT or PAR WAIT MED may also be used OPERATOR FILES MUP MSV Name Function Usage OPERATOR FILES 255 MSV There is a new 07 08 92 operator file MSV Messagesystem Save that will be run from MDP at the end of the UM full file save To back up to tape the MESSAGES OLD files created by MDP To do t
445. s o sprint FDname PAR command mod or run mta units o SPRINT FDname command mod command mod etc end of file or MTS Units takes the following commands each of which may be followed by an optional modifier see below D devname to display a particular device JOB PROGRAMS UNITS OLD STYLE 158 where JOB PROGRAMS UNITS JOB PROGRAMS D ALL CU CU CU CU CH CH CH devname cu cu ALL chan adr chan adr ALL T devtype adr devname cuk chan adr devtype to display all units in device table order to display all units on the control unit to which devname is attached to display all units on the control unit with the given number to display all units on the CU within the given range to display all units in control unit order to display all units attached to the channel with the given address chan adr to display all units attached to the given range of channels to display all units in channel table order to display all units with the given device type to display the path see below with the given hexadecimal I O address is a device name is a control unit number is a channel address is a device type adr is a physical 1 0 address OLD STYLE JOBS PROGRAMS 159 A null command defaults to D ALL MTA UNITS O also accepts the command MTS and can be terminated with an end of file
446. s set in the system to keep the information associated with code number from being re used If the wrong tape was labelled the information associated with the mis labelled tape is printed on the console shown in the example above as is true with correctly labelled tapes This information can be used to correct the situation To re lable a mistakenly initialized tape do the following Mount the correct tape and signon sys run label Enter the tape information from the console just as it was printed by LBL In the example this was IBM Labelled DENSITY 1600 VOLUME VOL_2 OWNER JANE so when you run LABEL you type in t90x 1600 vol_2 jane fendfile The tape which was mis labelled can be relabelled using LBL since the information associated with its code number hasn t been used yet ANSI Labelled Tapes t90x 1600 vol_2 Jane lbltype ansi OPERATOR FILES LBL NAME FUNCTION USAGE OPERATOR FILES 235 LIM To display in minutes and hundredths of minutes the CPU time remaining to a batch job Enter mts lim at the console When it prompts for input enter either the MTS task number of the batch job in question or the user id for that batch job Output takes the form JOB 00302 W035 3 23 which means that job 302 user id W035 has 3 23 minutes of CPU time left before his next limit expires Note that this will be a local limit if he has one set i e he will not necessa
447. s the userid for which statistics records are requested fromtime and totime specify the range of signon times of statistics records SCAN will add a fudge factor to totime currently 12 hours and scan through signoff times up to that time fromtime and twotime are currently in UBCDATE JVCT format but this will change when a replacement for UBCDATE is written TO is required only when the two consecutive times would be ambiguous A null line or end of file terminates the list After the list has been terminated SCAN will search the SCAN 455 monthly statistics tapes and STATISTICS for the requested records When the search has been completed the records found will be printed in LISTSTAT format The LIST and COPY commands may be used to manipulate these records LIST STAT CMDS fdname LIST prints statistics or cmdstat records which have previously been found in a readable format fdname is optional the defualt is SINK SCAN will prompt for items to be listed one per line The possible items are ALL lineno lineno lineno U ccid P prjn J jobn ALL specifies that all statistics cmdstat records are to be printed not just those found in the most recent search lineno is the sequential number of a record for statiscs records this is printed at the right side of the output in a LIST or STAT command If a single lineno is entered just that record will be printed two J
448. se Alarm Test Reset and Audio Reset 8 To get critical bus power close Main Output breaker located on the upper right as you face the controls eta 9 Check output power A B etc as in 6 above This value should be 277vac 2v To start MG 1 Press motor start and wait from about 90 seconds to 2 minutes for contactors to close 2 Observe Generator Run Time Meter If the red flag located in the small square at the upper right of this meter is moving it s ready to be transferred Wait for contacters to sound 3 Press and hold Transfer Initiate Rotate Transfer Switch to MG the system will tranfer itself Hold the Transfer Switch until you hear contacters click This transfers from Bypass BYP to MG 4 Check generator ouput voltage It should be 277vac 2v 5 Observe Bypass light clear it using the Audio Test Reset and Audio Reset When MG is up to normal speed it should be suppling the critical bus with power 6 Close Output breakers located below Main Output breaker System 3 has a Spare output breaker which usually stays open 7 See if you made any mistakes by checking the status panel Powering up PB II At this point PB II has logic power METERING SECTION 555 1 Clear out erroneous alarms using Alarm Test Rest and Audio Reset 2 Close Main Rectifier Charger breaker This starts the Charger At this point the DC Volt Mete
449. separately by individual components within the system This facility may manage either a single pool of tapes that may be used for all of these functions or a series of separate pools of tapes that may be used for a single function Each function may involve the use of several tapes however each tape will only be used for one particular function which indicates what each tape is being used for and what information is contained on De The directory facility will provide a method for locating the tape and tape file that contains a certain data record e g a statistics record for a certain time or to indicate where the next data file should be written e g on which file and on which tape the next statistics file is to be written The actual mounting of tapes and the reading or writing of files or records on the tapes will be done by individual programs calling the directory facility The directory facility will not be mounting reading or writing tapes except for initializing tapes The directory facility consists of a set of subroutines that may be called directly by system programs to manage the directory These subroutines provide the capability of creating adding and retreiving entries from the directory These subroutines are described later There is also a directory program that will allow the user to conversationally interact with the directory subroutines The Directory File When running
450. situation The assignments take effect at the next pen change which may be sooner than the next plot Plots which require colors that are not in use will remain in the queue until you make those colors available PRI n This command sets the pen priming interval n is in seconds When PLTQUIK changes form one pen to another in a plot it will prime the new pen only if it has been n seconds since that pen was last used This command has the same meaning that it does for PLOTCON If you are using ball point pens the usual case for this program you will probably not need to use this command RIP N Use RIP to set the frequency of the rip off the plots reminder message n is in minutes PLOTQUIK will print the message about every n minutes and wait for a reply After you reply a carriage return will do it will continue plotting Initially n is 30 REP receipt This means replot receipt receipt or the current receipt if receipt is omitted Use this command to replace a defective plot any time before the plot is dumped to tape after that use PLT most likely PLT36 PLOTQUIK will not begin plotting the plot immediately unless it was not doing anything it will continue with what it was doing and do the replot in due course You can enter a REP command without disturbing the plot currently being plotted If the QUIK plot that you are interested in has
451. sole before UB MTS commands such as ONLINE OFFLINE UNITS and so on from the UB MTS operator console can be used One very common problem is that certain devices particularly PCPs may be marked offline to VM XA when VM XA is loaded In this case the device will not be available to UB MTS until it is placed online to VM XA and attached to the UB MTS virtual machine from the VM XA operator console VM XA refers to devices by number rather than by name as does UB MTS UNDER VM SYSTEM SHUTDOWN AND START UP PROCEDURES 39 MTS A list of device names and numbers is kept in the machine room Determining the status of a VM XA device Type the command QUERY nnnn or QUERY nnnn mmmm at the VM XA operator console to determine whether one or more devices are on or offline to VM XA nnnn and nnnn mmmm indicate a device number or a range of device numbers Here are the most common responses where dev is a device type TAPE etc and nnnn indicates the device number 1 dev nnnn OFFLINE device is offline to VM XA 2 dev nnnn FREE device is online to VM XA but not in use by any user including UB MTS 3 dev nnnn ATTACHED UB MTS nnnn device is online and in use by UB MTS Onlining a device To make a device online to VM XA type the following command at the VM XA operator console VARY ONLINE nnnn or VARY ONLINE nnnn mmmm where nnnn and nnnn mmmm indicate a device number or range of d
452. spectively Start PLOTQUIK as an MTS file job enter mts plq on the console PLOTQUIK can also be started by signing on and entering a RUN command see the file PLQ for an example PLOTQUIK will prompt for the name of the plotter to use Enter PLT1 or PLT2 The program will then begin plotting as soon as there are plots to do PLOTQUIK can be left running all day if you like If plq fails to wake up an operator must get it going manually by interrupting the program and issuing a wake The only way to get the program to change plotters is to stop it and start it again To stop the program enter a STOP command to it see the description of commands below or STOP the job It is ok to stop PLOTQUIK while it is in the middle of a plot that plot will be replotted the next time the program is run Every 30 minutes you can use the RIP command to change the interval PLOTQUIK will print a reminder to remove the finished plots and then will wait for a response from you When you enter something a carriage return will suffice plotting will resume COMMANDS PLOTQUIK will prompt for a command if you GOOSE the job or after it has finished executing a CON command It prints PLOTTING SYSTEM 424 a prompting message which contains a list of the commands After you enter a command the program will execute it and then either prompt for another or begin plotting Enter an end of file to tel
453. ssfully acquired the disk manager holds onto it and creates a DISKLING and IOMNGR subtask for it If the disk is not successfully gotten an error message will be printed and the call ignored ADD MTS023 the above command will add MTS023 to the set of disks the disk manager has control over To delete a disk volume from the disk manager without attempting to clean things up Prototype Parameters Description This JOBS PROGRAMS 99 BLAST lt pvn gt OK pvn is the MTS public volume number of the disk to be deleted from the disk manager command does nothing it once was intended to have been implemented according to the following description but was ajudjed to be too dangerous If the disk manager does not have the drive specified by lt pvn gt an error message will be issued and the command ignored If the disk manage does have the drive the user will be asked for confirmation unless OK has been specified on the command If the user confirms deletion a blast disk request will be sent ot the disk manager WARNING the use of this command is quite dangerous and can result in lost requests and inconsistent files Because no attempt is made to post disk requests which are currently being processed for the disk drive requests may be lost The lost requests can be recovered only by stopping the disk manager job and restarting it The disk manager will not wait for an I O channel program to co
454. st TAPE ERROR CODES 495 The following tape error information is pertinent to everything but rst An explanation of the tape error messages which are displayed on the operators console is given below Each error produces a three line report of the error which gives device name the sense data associated with the error which precisely describes the error in machine terms the position on the tape where the error was encountered and a short descriptive message giving the nature of the error A prototypic message is as follows aaaa error b cccccccce dd eeeeeeee eeeeeeee eeeeeeee eeeeeeece eeeeeeee eeeeeeee ffffffff gg aaaa gggggg hhhhhh iiiiiiiiiiiiiii file jj block kk aaaa 1111 Each of the underlined areas vary depending on which device declared the error and what type of error it was These are described below aaaa designates the device which detected the error Notice that this appears at the front of each of the three lines B is a one character code which uniquely describes the type of error These don t need to be memorized because a more readable description of the error is given on the third line of the message See the description of the 1111 part of the message below for all the different errors Ceccccce gives the hexadecimal value of the unit check routines input switches This is only of use for programmers which work on the unit check routines themselves Dd gives the hexadecimal
455. state The drive must be made ready before any further operations may occur on the device Intervention required but not statusb set See the description for error H Unable to find failing CCW in channel program This is usually caused by a programming bug It can be caused by injudicious usage of the FAKEIO function too Load point reached This error is a programming error It arises when the program attempts to read a tape backwards and encounters the front of the tape Unlike the end of the reel the tape cannot be pulled of the front of the reel by a properly functioning drive Data converter check This error only happens with seven track drives when the data converter has not been set to the proper translation mode It is a user or program error Not capable This error occurs when an attempt is made to make a drive switch to a recording density it is incapable of handling For example trying to read a 6250 bpi tape on a drive only rated for 800 or 1600 bpi will cause this error to occur Physically offline This condition is declared by the supervisor when it attempts an operation to a device which returns an indication that it is offline You must online the device before any further operations may successfully complete on the drive tape error message is given below along with its interpretation TAPE ERROR CODES 499 T901 error A 41000000 01 40800000 00000d00 00000000 00c12312 24008000 00021234 0
456. stem shutdown is at 6 30 pm The system should be up by 7pm endfile Send edit help etc system asks At this point you can type post if the message is ready to go or you can edit the message by typing edit When the message is finally ready type stop to get out of the editor post to send the message stop to get out of message system NOTE When you are in EDIT mode the prefix is When you are in the message system the prefix is For more information on the Messagesystem refer to CCMEMOs 425 and 426 While in the Messagesystem and when in doubt one can use the help command at certain points SIGNON MESSAGE MESSAGESYSTEM 464 In the above example the message would not begin to appear until 3pm and would cease to exist at 6 30pm 10 05 83 even if the system was not reloaded More than one signonmessage may be issued since you may want different expiration dates or times so after you have posted one message you may begin another by again issuing the command send signonmessage to everyone etc To check on all aspects of your signonmessage s you may ret signonmessage NEXT HISTORY HELP DELETE REPLACE ETC system responds display full The system will respond with something like the example below To Everyone From J Spratt Expires 8 00 AM Fri Feb 18 83 Current State outgoing incoming seen Then the contents of the signonmessage are printed If more than on
457. stored If a date or date and time is specified then the most recent version saved on or before the date time specified is queued for restoration It should be remembered that since daily filesave only saves files that have been changed since the last filesave specifiying a date actually amounts to requesting that a file be restored to its status as of that date DISPLAY any filename pattern mm dd yy hh mm ss HISTORY ALL x If usid fname is specified then the most recent version of usid fname available for restoring is listed If only the date and time are given on the display command then only versions on or before that date are listed If a patterned file name is specified such as usid xxx is given the latest version of each OPERATOR FILES RES 272 ODISPLAY OPERATOR FILES file available for restoring is listed along with the comment There are x older versions of this file If the is not used then the complete history is printed If the complete history of each file is wanted and when an unknown is used then history all or history x x digit for number of history lines must be appended to the display command If for example Display usid c3m is used then all files under usid beginning with the characters c3m will be displayed will list all of the files queued to be restored OPROCESS STOP along with their new names if
458. systems programmer CLEAN JOB jobnum takes the job off all open or locked and waiting lists CLEANF JOB jobnum FILE fname same as CLEAN but for one file only DEQ JOB jobnum takes job off all waiting lists DEQF JOB jobnum FILE fname same as DEQ except for one file only JOB PROGRAMS TABLMOD JOBS PROGRAMS 149 TAPE COMMANDS OPERATOR CONTROLLED Names RUN REW WTM FSF FSR BSF BSR and ERG Functions To manipulate tapes and write tape marks file marks on tapes using the operators console Prototypes RUN T90X REW T90X WTM T90X FSF T90X FSR T90X BSF T90X BSR T90X ERG T90X NNNNNK Usages REW rewinds tape X to the load point RUN rewinds and unloads tape X WIM writes Y number of tape marks file marks on tape X FSF spaces tape X forward Z number of files FSR spaces tape X forward Z number of records BSF backspaces tape X Z number of files BSR backspaces tape X Z number of records ERG erases Z times 3 inches of tape more or less Examples WTM T902 4 BSR T909 2 NOTE Use the ERG and WTM jobs with caution if you specify a tape drive other than the one intended data may be lost JOB PROGRAMS TAPE COMMANDS OPERATOR CONTROLLED 150 TASKS NAME FUNCTION PROTOTYPE NOTE USAGE JOB PROGRAMS TASKS JOB PROGRAMS TASKS To determine the status of a particular task TASKS M mts jobs TASKS U user id TASKS D device id TASKS T device type
459. t 14 to D240 However the symbol in the first path means that the path through channel 1 is logically offline Consequently the Supervisor won t schedule any I O on the path through channel 1 Note that channel 1 might still be used to communicate with some other device and this is the difference between an offline channel and an offline path between a channel and a control unit Other things we learn from these two lines are the two hardware I O addresses of the two paths 140 and 240 EXAMPLE 2 Units D DS15 CH 0 gt CU 7 gt DEV DS15 ADDR 015 JOB 41 CU BUSY 64 CH OPS 1 CU OPS DEV INITIALIZED NO ERRORS CH 1 CU 7 gt DEV DS15 ADDR 115 JOB 41 CU BUSY 1 CH OPS 1 CU OPS These lines show the effect of the g option In addition to the information described in Example 1 above the general status and device status parts appear The general status the second line gives four facts Ls The device is owned by job i e task number 41 2 The control unit is currently busy By There are 64 operations outstanding on JOB PROGRAMS UNITS OLD STYLE 162 JOB PROGRAMS the channel 4 There is one operation in progress on the control unit The device status the third line gives two other details ae The device has been initialized This means that the Supervisor has allocated an area of control storage a Device Table to record the progress of I O on the device
460. t copying operation If could send a message to the operator if someone thinks that would be useful CMDPIKUP and CMDTAPE communicate through the file CMDDIR under SYS at UM CMDDIR has the format TsffftttxxxxxxxKxKXXXKXKCC CC cccc cc cc FSCMDSTAT1 FSCMDSTAT2 FSCMDSTATCc Sun The first character in each line says whether the line describes a tape or a file The second character s is the status of the file or tape and has values available the file is empty and available for use by CMDPIKUP A active the file currently being written in A available tape is not currently mounted B busy CMDTAPE is using it right now F full the file is full and CMDTAPE should write it to tape Line 1 describes the tape currently in use fff is the next file on the tape to use Mttt is the rack number XXXXXXXXXXXXXXXX s the BINTIME the tape was created and cc c is the printing form of the date Note this is binary data be careful when editing this file The tape line must be line 1 the lines describing files may be in any order thereafter If something is marked B and CMDTAPE is not running CMDTAPE died or the system crashed When CMDPIKUP starts up it changes a B on the tape line line 1 to A and any B ona file line to F Since CMDPIKUP is started when the system comes up this covers the system crash situation but if CMDPIKUP is started at some other time un busying thi
461. t of an attention button one must use the GOOSE command giving as a parameter the job number which is to get an ATTENTION INTERRUPT It is also possible to interrupt a batch job with the GOOSE command In this case after MTS has fielded the attention interrupt that the GOOSE caused it will read a line from the console The operator may enter one of the following ATT abandon the current user command and go to the next user command The attention interrupt that was given will be noted in the user s output DUMP dump user storage as if the MTS subroutine ERROR had been called and then do as for NEXT NEXT stop the current batch job and skip all card reader input until the next S 8 card There is usually little reason to use GOOSE with a batch job initiated by HASP since there are HASP commands available to do the same thing GOOSE should cause a reaction within half a minute however it will wait until all I O devices attached to the task are ready Thus one should check for any NOT READY messages before abandoning GOOSE GOOSEing a non MTS task will cause the task to be stopped with code ATTN if that task did not set itself up to intercept attention interrupts The task number of the task that is to receive the ATTENTION INTERRUPT goose 125 JOBS PROGRAMS 111 NOTE The SYSTEM STATUS command attn nnnnn performs the same function and is especially useful if all job tables are in use
462. t run optional d Phone number of person making request e The name of the department that the user is associated with f Ask for the final date that this user needs the special forms The information gathered from the user making the request is entered into the file which has the special forms number which is to be supplied to them Next do the following to determine which numbers are available Some numbers are reserved for special ccids signon w035 on either UB or UM f page f_sp fa This last command will show you the permit status see list below of each file that you can use You can copy list or gedit any one of these file to see the contents In the contents will be information about users including the expiration date for the permit status which someone has written in the file PAGE F_SP1 U Owner N Others U W035 PAGE F_SP10 U Owner N Others U W035 R K45Q LGE5 PAGE F_SP11 U Owner N Others U W035 PAGE F_SP12 U Owner N Others U W035 R SWCE PAGE F_SP2 U Owner N Others U W035 R TB97 PAGE F_SP3 U Owner N Others U W035 PAGE F_SP4 U Owner N Others U W035 R GC4B PAGE F_SP5 U Owner N Others U W035 SPECIAL FORMS JOBS 547 548 PAGE F_SP6 U Owner N Others U W035 RW DWB PAGE F_SP7 U Owner N Others U W035 PAGE F_SP8 U Owner N Others U W035 R SS83 SKB7 PAGE F_SP9 U Owner N Others U W035 As an example to assign a number to a user do the fo
463. t to install Use only the RETURN and arrow keys to position the cursor in the proper fields don t press ENTER until all of the above fields have been filled in Fill in the Device Address field with the number of the tape drive for our example this value would be 0684 specify Device Type as T enter Y in the Ignore Unknown Card Statements field enter NA in the Output Printer field specify 1 for the File Number field enter N in the LPAR MODE field Press ENTER The tape should spin and the ES3090 300J will read in the IOCDS Eventually the message PROCESSING COMPLETE PROCEED should appear Tf an error occurs at this point return to the normal system screen via steps 14 and 17 through 20 below and reload the system normally Contact a systems group firefighter via Message Type MENU on the command line at the bottom of the screen and press ENTER The IOCP general menu will appear Select option 8 Write IOCDS to the Processor Controller File and press ENTER The screen will shift to the appropriate menu Fill in the WRITE IOCDS field with the name A0 A5 BO B5 of the IOCDS you selected in step 5 above Press ENTER and the new JIOCDS will be written to permanent storage It will take a few minutes wait for the Processing Completed message Type END on the command line at the bottom of the screen and press ENTER The ES3090 300J console alarm will ring
464. tallation of the 3805 The 3805 has been powered off longer than the lifetime of non volatile memory about 96 hours The controller fails and inadvertantly alters non volatile memory The 3805 is forced into COLD START by the service panel function key COLD START PROCEDURES COLD START procedures only take a few minutes so if there is any doubt do a COLD START Take the 3805 physically offline If the system is up and using the 3805 a complete shutdown will be required press CLEAR This should result in a display of FCA 00 which means that FUNCTION CODE entry mode is enabled VIRTUAL MEMORY PDP 68 SYSTEM SHUTDOWN AND START UP PROCEDURES e key in Bl press ENTER This should cause a display of DA1 which is a prompt for a channel A device address key in nnnn the device address for channel A nnnn is different for each box e press the CYCLE key This should display DA2 a prompt for channel B device address key in 0000 this means that no address for channel B is desired press ENTER DA3 is displayed a prompt for allocation block size e key in 1000 hex number of bytes per block press ENTER DA is displayed a prompt for the number of blocks in this allocation Note The number 2B00 in the next line isn t always the same i e it depends on the box Even on the same box it may have been changed since the last time it was observed Check with th
465. tatus message to include information on when MTS will shutdown and when you expect it to be back up 3 Five minutes before the shutdown hold batch execution HOLD EX and run the SHUTDOWN job without any parameters Change the status recording and turn on the flashing green light Run DRN and drain the local printers 3 1 mts rmd Shuts down Resource Manager 4 If members of the systems programming staff are working to finish last minute changes before the shutdown check with them before going ahead with the rest of this step At the scheduled shutdown time broadcast one last warning RERUN any batch jobs that haven t finished execution wait from 30 seconds to one minute and then run SHUTDOWN ALL Turn on the solid SYSTEM SHUTDOWN POWER DOWN PROCEDURES SYSTEM SHUTDOWN AND START UP PROCEDURES 63 red light and change the recorded status message 5 After the message SHUTDOWN ALL COMPLETE is printed at the console run SDB or SDM depending on the system MTS SDM SDB to place shutdown messages in the PCPs This procedure is used to enter a one line UMnet UMnet message which will be printed for any user attemting to access which ever system is down It will not disconnect any users currently connected to the UB or UM systems Note If one or more terminals is still active shutdown won t finish See following text 00002 23 45 49 OPERATOR T M 00002 23 45 50 OPERATOR 00657 MTS 000ABD20 57 I O on AD
466. tem unless the system is already powered down 1 If only one disk is to be powered off b Open c Turn d Open e Make Lf offline it open its door and turn off the associated switch The green switch puts the disk into and out of READY state Power will still be available until the power supply is turned off To properly turn this off call an Amdahl representative each disk drive door off all green tipped switches that are READY the Control Unit door sure that the Local Remote switch is in LOCAL the switch had been in remote the disks should have come down with the main system f Turn the silver switch labelled BOTH to the off ES3090 300J POWER UP DOWN SYSTEM SHUTDOWN AND START UP PROCEDURES 25 positioned which should sequence the disk string down Do this with both control units Note There are two control units called Directors in each Control Unit box 5 Power down the Intel virtual memory equipment Powering the System Up Note Before powering up the ES3090 300J do the following Select B2 from the system definition frame 22 before an LPAR POWER ON RESET POR Type west for the zone definition and the current time in the ZONE OFFSET slot This allows the ES3090 300J to POR LPAR Following this the SYSTEM CONFIGURATION FRAME should be displayed by pressing the INDEX key and typing O5 Now that the SYSTEM CONFIGURATION FRAME is present type
467. ter preventive maintenance has been performed or after a hard crash for example when a storm takes the system down MTS CTD This will do some I O to each of the MTS disk drives If a disk error is encountered no output will be printed The procedure may get hung up or if you re lucky it will generate an error message uldn t be run while users are on the sytem since it really hogs the channel OPERATOR FILES CTD CT3 NAME FUNCTION USAGE REMARKS OPERATOR FILES 201 CT3 Validates file system integrity and does a cursory check of the message system data structures of each mailbox mts ct3 CT3 is usually used after a hard crash It is started after peek completes CT3 runs validatefile which itself validates file system integrity on Userdir Master as well as all of the files that comprise the messagesystem data base If files are in good working order the routine starts the message file mamangers and does a cursory check of the message system data structures of each mailbox T everything is satisfactory the routine enables the message system If the routine finds something wrong it leaves the message system disabled and prints error messages on the console alerting the operations staff OPERATOR FILES CT3 202 CWM NAME FUNCTION USAGE mts cwm REMARKS OPERATOR FILES CWM Message file purge and validation Started by clockw
468. tes 3 indicates low rates 4 indicates normal rates 3 When the holiday period begins the operations staff should specifically set minimum rates before they leave While operators are on duty these two priorities should be set to 0 Using the MTS operators consoles enter MOD BTCHPRIO 1 MOD TERMPRIO 1 RELEASE MINIMUM II After the Holiday is Over 1 When the holiday period ends and regular rate periods are to be back in effect operators should reset them by entering on the MTS operators console MOD BTCHPRIO 0 MOD TERMPRIO 0 hold deferred hold minimum hold low 2 Reset the HASP control file Signon SYS Duplicate AUTOHASP lt CTL AUTOHASP CTL ok Destroy AUTOHASP lt CTL ok SPECIAL RATES HOLIDAYS 453 3 Reset the initial reload commands file Signon INIT Duplicate INITCMD lt INITCMD ok Destroy INITCMD lt ok Modifying BTCH and TERM when operators are on duty and when they leave Changing AUTOHASP and INITCMD should be changed once at the beginnig of the holidays and once again after normal hours begin at the end of the holidays SPECIAL RATES HOLIDAYS 454 SCAN CMD amp STAT Records W030 SCANX Purpose SCAN searches cmdstat and monthly statistics tapes for requested CMD and STAT records A search may be done by the CCID job number date time CMD name etc Subroutines may be appended to do specialized searching with SCAN providing reading writing and tape
469. th the error which precisely describes the error in machine terms and a short descriptive message giving the nature of the error A prototypic message is as follows aaaa error b cccccccce dd eeeeeeee eeeeeeee eeeeeeee eeeeeeece eeeeeeee eeeeeeee ffffffff gg Each of the underlined areas vary depending on which device declared the error and what type of error it was These are described below aaaa designates the device which detected the error Notice that this appears at the front of each of the three lines Io is a one character code which uniquely describes the type of error See list the list below Unit Check Error Codes Stop DAC UNIT CHECK ERROR CODES 362 Code Meaning 0 Insufficient sense data less than the full 24 bytes 1 All relevant sense bits are zero 2 Second equipment check 3 Command reject 4 File mask violation unexpected 9 Bus out check A Uncorrectable data check B Overrun 10 tries Cc Unwanted track overflow CKD or block size exception FBA D Unexpected end of cylinder E Intervention required 2301 or 3805 F Overflow incomplete G Unable to define extent parameters FBA H Check data error FBA I Incorrect length J Invalid device type K No record found L Channel detected error program check protection check channel protect check interface control check or offline control unit M Low spares on 3805 or 3825 P Track condition check 3340 or 3344 Q Intent violati
470. the command is terminated with an attention interrupt If a parameter is given it is treated just as if it were typed on the operator s console Operator commands that begin with a dollar sign are taken to be HASP commands while all other commands are treated as startjob requests or UMMPS commands System Status commands prefixed with a slash or 3270 Operator console device commands prefixed with a percent sign may not be entered using this command Any output that would be produced by the OPER command including error messages will appear on the real operator s console This command if given with no parameters continues to produce output until interrupted by an attention interrupt PAl on a 3270 or until the JOBS job is STOPped or GOOSEd Example OPER OPER MTS DS06 QUEUE receipt ALL LCL station RMTn The QUEUE command displays the status of individual batch PRINT or PUNCH jobs and or the status of any or all remote batch stations If a receipt number is specified the queue status of the specified job e g awaiting execution executing awaiting print printing etc the priority and routing of the job and the relative SYSTEMSTATUS 480 position of the job within its queue are printed If ALL or no parameter is specified counts of the total number of pages waiting to be printed and cards waiting to be punched along with the print punch and execution queue lengths are printed I
471. the new contents all commas in this string are ignored and this string is terminated by the first blank or end of line which ever comes first a xxxxxx yyyyyy the address expression b xxxxxx the hexadecimal modification mod 30fa 90 mod 30fa 1000 90ec 56 There is no response or verification for this command There is no MOD job as such since MOD is a supervisor command MOD is not a valid command from the 3270 operator s console Use the MREAL command which can be found in the Systemstatus section of this manual instead MTS NAME FUNCTION PROTOTYPE USAGE JOBS PROGRAMS 119 MTS to start an MTS task these are the tasks for which this whole system exists MTS isa re entrant job program It provides the capability of loading executing and controlling programs from terminals or through batch streams MTS op par 1 op par 2 msource msink syspch The MTS command takes several parameters It needs at least one parameter i e msource This mandatory parameter denotes an I O device or MTS public file from which input is to come Preceding the mandatory msource parameter there may be zero one or two optional parameters Following the mandatory msource parameter there may be zero one or two optional parameters Notice that the optional parameters which follow the msource parameter consist of a single comma delimited by blanks and the actua
472. the expired ccids df_destroy Now destroy the user file system catalogs and entries in the Restore filesave directory for ccids that have expired df_ucatdestroy Now adjust the disk accounting information for the ccids whose files have been deleted df_fix Now remove any names associated with the expired ccids from the directory used by Userdirectory df_userdir DISKS DEADFILES 361 If error messages are printed show the ouput to a member of the systems group or the person responsible for Userdirectory Even if there are errors in this step you can go on with the next step Show any errors to a member of the systems group or the person responsible for the Messagesystem The message search failed ig not an error 14 If there were no errors or other problems in any of the aboves steps you can issue the following command to destroy the files used and to reduce their size df_end If there were errors skip this step DIRECT ACCESS STORAGE DEVICE ERROR CODES NOTE In another part of this manual are programs which may be used to print listings and summaries of DISK ERRORS and UNIT CHECKS Two of these are MTA PUC and SYS EREP The following Disk Error information always applies An explanation of the error messages which are displayed on the operators console is given below Each error produces a three line report of the error which gives device name the sense data associated wi
473. the recording of soft machine checks SOFTCHK ON n enables the recording of soft machine checks for n errors If the command SOFTCHK is issued without an integer information will be printed concerning how many soft errors have occured and how many more may occur before the limit is reached The limit can be raised by issuing another SOFTCHK n command When a reload occurs a default limit of SOFTCHK 16 is set by the system The largest value for n is 1000 If you use a very large value for n you you may get many errors which if occurring fast enough could tie up the system but no damage will be done The errors that can t be logged will be discarded Normally the count n should be omitted since the default of 16 is satisfactory SOFTCHK OFF is reset each hour and reverts back to the default limit softchk off softcjk on 25 softchk alone will bring a response similar to SOFTCHK 5 SOFT ERRORS HAVE OCCURED THE LIMIT IS 21 JOB PROGRAMS SOFTCHK STARTUP Note If necessary NAME FUNCTION PROTOTYPE USAGE PARAMETERS EXAMPLE JOBS Shutdown All not a STARTUP STARTUP PROGRAMS has been To cancel the effect terminal tas STARTUP ks 143 used a system IPL is of shutdown on MTS The STARTUP job will cancel the effect of the SHUTDOWN job on any currently enabled lines This job will not re enable any lines that were disabl such lines STA
474. the time given Set time hh mm ssapzzz date mm dd yy zone zzz Where hh is the hour 00 23 or 00 11 mm is the minutes 00 59 ss is seconds 00 59 ap is am or pm if omitted 24 hour time is assumed zzz is the time zone as specified earlier mmm is month 1 12 or Jan Feb etc dd is day 1 31 yy is year 0 99 HELP to get information on the above commands 2 VERIFY TIME DATE THEN ENABLE TOD CLOCK After this is done the request for verification of the time and date will be repeated Note 1 and 2 usually aren t necessary and are skipped The PN QN TN P11 T11 G11 SIGNONM OFFLINE ONLINE and UNITS tasks may be run anytime after the message about the date and time being set Not until the POST IPL SYSTEM LOADED OK message should you initiate MTS or HASP tasks however When the IPL procedure is complete the machine will enter the wait state with X FFO in the right half of the PSW The system is now up except you have no terminal or batch capability as yet You may want to load the print buffers now To do this type in the chain or train name there is a corresponding job program with the printer name as the first parameter and ECHO as an optional second parameter e g tn ptrx echo The above example is for 1403 type printers For 3211 type printers ECHO is not necessary and default is 6 lines per inch If 8 lines per inch is desired on 3211 type printers use the parameter 8LPI p11 ptrx 81pi
475. then ask for a pen number This program is useful for checking jittery corner turning on the Calcomp 936 String sizes between about 3 inches and 05 inches are good for this kind of testing Try an increment around 05 PLOTTING SYSTEM 446 PM2 AN STC ERROR ANALIZING PROGRAM NOT PRESENTLY USED AT U OF MICHIGAN Running PM2 in MTS Example run emp pm sep0683 ok Empties output file rename pm sep0683 pm sep0983 ok Renames output file to comply with current date set macros on gt set var maclib 1 etc pm2 mac pm2daly print pm sep0983 Runs program PM2 is a program developed by STC to maintain a history of device errors on their equipment and print reports related to this It is run in MTS by invoking one of several command macros which must first be made available to MTS by executing the following two commands SET MACROS ON gt SET VAR MACLIB 1 ETC PM2 MAC After executing the above commands the usual manner in which we run PM2 can be done Usually PM2 output is sent to a file where the limits of its three sections can be determined via the editor so that they can be printed separately if desired The three sections are Tape Perfomance Report Disk Performance Report amp Tape Volume Report The name of the output file is changed daily i e if PM2 is actually run every day emp pm may0683 ok rename pm may0683 pm may0983 Changes the name of the output file
476. they would by queueing up for the normal plot times Currently the QUICK plots are done in ball point For information on how to use this program see the PLOTTING section later in this manual OPERATOR FILES 265 PWC NAME PWC FUNCTION Prints WAIT COUNTS for the system USAGE mts pwc COMMENTS PWC runs MTS PWC and directs the output to the operators console The job will stop automatically or it may be stopped via STOP or GOOSE OUTPUT Totals Ready Page Wait__ DASD wait__ 3270 wt_ Intertask__ Other MISC MON MAY 25 81 To Sk W To Nu Pr 34P To Nu Pr 34P To Io Gu Fi To Io Gu To Ry By Wt To Io Ti To Do Js Pk Gu Lk Op Hs BQ AB A a a eS See ADR OR oe OE 9 A ce Os BB DD SA 4h OBR a Dl See oS od es 52 0302 52548 AL aa rs Oe LO e BAO go a te Die IR nz See bh Oe 21520205 SS sb ie Ma se nes WR at We OD ee BO AGE SO a hy ee AO Pde eS BS ge SDo SL ee ce oe Cs SEDO S m 52 48 4 amp gh a nee ts 2 ee hy 29 aT nk a 34 30 Lines that start with e are the extremes with m the means and with a blank the currnet count Totals To Sk W estes State unknown remiss Se ses State known Grand Total Ready To Nu Pr 34P IS be Non Priv jobs ready for a CPU l Priv jobs ready for a CPU rada Neutral jobs ready for a CPU Total jobs ready for a CPU Page wait__ To Nu Pr 34P I Non priv jobs waiting for a page in OPERATOR JOBS P
477. tial set of five tapes for each series in the tape racks with the UM Dump tapes on the left side of the isle near the top They re marked as the mail backup tapes If you have problems with this please get in touch with either Steve Rothwell 3 4902 or Steve Burling 3 4905 OPERATOR FILES MSV DEADFILES MESSAGES NAME FUNCTION USAGE REMARKS OPERATOR FILES 257 NST Test program manipultes Name Address Spaces mts nst The purpose of NST is to allow the NAS test program to be run from the console The NAS test program allows manipulation of Named Address Spaces into which commonly used programs are loaded Using the NAS test program one can obtain information about the system s usage of NASes and modify certain aspects of NAS processing Note This program is intended to be a debugging tool for use by systems programmers OPERATOR FILES NST NAS 258 NDL NAME FUNCTION USAGE NOTES OPERATOR FILES NDL OPERATOR FILES NDL Dumps and reloads Primary Communications Processor AN PCP mts ndl The High Core monitor must be running in Primary Communications Processor PCP AN before using NDL To start the monitor follow instructions on the front panel of PCP AN PCP After successfully starting the PCP system the following question will appear on the console ENTER REASON FOR RELOADING PCP The operators initials are typed in foll
478. tic tape USAGE The STA program normally should be initiated whenever the MTS file STATISTCS is nearly full MTS will print the message STATISTICS IS FULL on the operator s console whenever a user signs off if there are more than 7500 statistics records in the file one record is added to the file every time a job signs off The program may be initiated by issuing MTS STA This should be done as soon as it is convenient If the statistics file is allowed to completely fill up SNARKS will begin to occur as jobs sign off if this happens STA should be initiated at once The STA program initially copies the file STATISTICS into the internal work file SYS STATISTICS then immediately empties STATISTICS During the initial copy operation the system will prevent users from signing off Therefore it may be preferable not to run the program if the system is extremely overloaded After the copying is completed the program will request that the operator mount a daily statistics tape i e a mount message will appear requesting a particular tape After statistics are saved on the daily tape the program will request a monthly statistics tape again causing a mount request to appear on the console After statistics are saved on the monthly tape the message STATISTICS SAVE SUCCESSFULLY COMPLETED will be printed on the console At this time the work file will be emptied indicating that
479. tified SHUTDOWN message from user SYSTEM at _XXAl 10 56 25 UMRUAC will shut down in 3 minutes back up shortly via automatic reboot Pleas e log off node UMRUAC Standalone 2 terminals have been notified SHUTDOWN message from user SYSTEM at _XXAl 10 57 26 UMRUAC will shut down in 2 minutes back up shortly via automatic reboot Pleas e log off node UMRUAC Standalone 2 terminals have been notified SHUTDOWN I STOPQUEMAN The queue manager will now be stopped EEE OPCOM 2 FEB 1988 10 58 27 29 Message from user SYSTEM SIMPRINT 1 IMPEXIT imprint symbiont exiting ELITE OPCOM 2 FEB 1988 10 58 27 74 Message from user SYSTEM sDSCnet NetMgr stopping 5 5 5 5 OPCOM 2 FEB 1988 10 58 29 44 Message from user SYSTEM SDSCnet NetMgr stopped SHUTDOWN message from user SYSTEM at _XXAl 10 58 30 UMRUAC will shut down in 1 minute back up shortly via automatic SAN DIEGO SUPER COMPUTER 546 reboot Please log off node UMRUAC Standalone 3 terminals have been notified SHUTDOWN message from user SYSTEM at _XXAl 10 59 32 UMRUAC will shut down in 0 minutes back up shortly via automatic reboot Pleas e log off node UMRUAC Standalone 2 terminals have been notified SHUTDOWN I SITESHUT The site specific shutdown procedure will now be invoked ELE III OPCOM 2 FEB 1988 10 59 33 66 Message from user SYSTEM VAXsim exit status VAXSIM S SHUTD
480. tion in service if the UPS is working porperly See the section in this manual concerning the UPS However some others will cause varying degrees of damage to the system if the UPS failed and power to the system was lost abruptly If system power was lost abruptly service to the users could be unavailable for quite a long period of time as in the case of a disk being partially written over causing files to be damaged or other hardware damage First if power was off for just a short time less than five minutes you should restore the AC power to the ES3090 300J in order to allow the cooling fans to run When you are certain that the AC is functioning let it run for at least ten minutes so that the hardware has a chance to cool off If the power repeatedly goes off then you may have to start the AC each time Next after a power failure which affected the Air Conditioning be sure to alert the Air Conditioning people even if the AC seems to be ok and is running To do this call 764 8176 from 7 30 a m to 3 30 p m Monday through Friday At any other time call Security at 763 1131 and ask them to get the AC people The AC people want to come and check the the entire AC system to be sure that everything is really all right If power interruptions occur frequently every few seconds or minutes it is better to wait until they are over before IPLing the system if the system is down although these short power spik
481. tional mode is used to describe this type of interaction In batch mode the user used to keypunch a card deck which contained all the MTS commands translator source statements data lines and object modules which are needed in order to accomplish his described task The deck was then submitted to MTS for processing note the BATCH MTS pseudo device may be used to submit a Batch job from a user s terminal The user s card deck was then read into the computer by a job program named HASP and saved in a file for later processing The deck and a receipt were returned to the user The user s job is then processed by MTS together with many other batch jobs and eventually the results of this processing the job output may OVERVIEW OF THE SYSTEM INTRODUCTION TO MTS THE MICHIGAN TERMINAL SYSTEM ES OVERVIEW OF THE SYSTEM be retrieved by the user with a receipt The MTS batch function is controlled by the Resource Manager RM The Resource Manager replaces the HASP Houston Automatic Spooling System HASP was originally developed for the IBM OS system has been revised to interface with the MTS system and accept MTS jobs to run in batch IBM OS SVS MVS VM CMS jobs jobs cannot be run directly under the MTS system A user communicates with MTS by means of a command language A typical job in batch mode will have command lines a source program data and possibly an object program The commands placed in a precise sequ
482. tly OPERATOR FILES TSN OPERATOR FILES 305 UDP CONT UDP NAME UDP Sig UD 22 Nov 88 sgr I added the mail udp semaphore file to force UDP to wait unt MDP has finished it s LOCK UNLOCK cycle on MSG It is possi for MDP to exclude UDP if they both signed on simultaneousl So now MDP locks udp semaphore then Locks amp Unlock msg and then unlocks udp semaphore thus releasing UDP to do its Loc Unlock cycle on Msg This is the only reason I am able to so up with for UDP not running for a long time Now we get to s if I m right 23 Nov 88 lbt renamed ID W494 to UD on UM MTS 31 May 93 lbt added a LOCKSTATUS on MSG FF E E E E E HF HF E E E lock mail udp semaphore mod wait Lockstatus Msg Lock msg mod nowait quit Unlock msg unlock mail udp semaphore Source UD UdmasterDup OPERATOR FILES UDP 306 OPERATOR FILES UNU Not presently used NAME UNU FUNCTION Enables all the terminals on the NUBS amp UNYN SCPs and starts the job status monitors at NUBS and UNYN REMARKS If the automatic NUBS and UNYN restart procedure at 8 00 a m doesn t work or is missed because MTS was down then this file can be used to bring the terminals at NUBS and UNYN back up OPERATOR FILES UNU OPERATOR FILES 307 ABNORMAL SITUATIONS SUMMARY Abnormal Situation and Error Conditions This section of the manual outlines abnormal error conditions an
483. tus Message ee ee ee ee te ee ees 469 Resource Manager iS Ges A BN CREED AE BE BS 535 RESPONSES TO MOUNT A Senge tee Sasha hades ERE S 488 Ripple Print See RIP Or TNR San Diego Super COMPUESTA a it i 543 SCAN User STAT And CMD Records oooooooooooooooooooooooooo o 454 securtty Clerk Purpose amp DUET ESS a oia 460 sense Values iaa yes dan ae eee aa A ee A hee Bad Ate a a 466 SETDATE Set Time And Date In MERIT PCPs eee eee ee ee ee 401 Shutdowns Scheduled ni A he e 389 Signon Message Using MessagesysteM ooooooooooooooooooooooo 463 solid State DISK EETOLS aa 361 Special EOrMS ODS Ta a neue is A A wg idhial ae a Weel hie Gate eal e 547 SUPSLFAUMPS a Ai I A a heme BRS 467 System ReTLIADILTES A A e a 468 SYSTEMSTATUS fOr Operators it a A io 471 Tape DUFECTOTY Prog ect oie dd whens das A A Ae od 507 Tape Error Codes naa A o oe elses 494 PADS EECO S PTA LA A A td act 493 Tape Mai king ii AA AS A A A AA IS A A E 500 Tape Mount Suspension NTM UAM o oooooooooooo ooo nooo ooo ooo 499 Tape Racks Electricd Probles ou enis ecaa den is ee 502 Tape Rack Strip Lahpels A A ses 503 TAPES S Feels ge A A AA irae a wie E Sis BE OS BRERA AEE X 486 Tapes MALLING cg even eel a She 8 vB Rico State A Riga O a a Sau Ben tes 500 Tapes responses To Mount cos Gane eras hel WE BUS ae a a Ee hl as Shel Sashes 493 Tapes runaway sa ait sieve fonds eats Rye A A A Saeed A Ai 493 UMNet Hermes Sy S
484. ueue requesting given density If no density is given the top job is popped tapequeue release releases all jobs in the tape queue TAPES TAPE HANDLING 493 tapequeue release job releases job number tapequeue abort works like the ab response aborts all of mounts of the jobs in the queue tapequeue abort job works like the ab response aborts the mounts of job display tapeq 0 off 1 on mod tapeq 0 off 1 on used to enable or disable the tape queueing mechanism Any null or cancelled response is ignored and the read is repeated Many responses allow an optional comment to be specified as the last parameter on the line which will be printed on the users terminal or batch output e g C0045 T902 CALL MACHINE OPERATOR prints T C0045 MOUNTED ON T901 CALL MACHINE OPERATOR Any error messages generated by incorrect replies should be self explanatory The normal problem being operator confusing 2 simultaneous mounts and replying to wrong one In checking for correct volume I O errors are sometimes generated especially 7 track tapes on 9 track drives NOTE Ocassionally a tape will run off the end i e the rear end of the tape is actually pulled off of the feed reel Operators should never re thread and or rewind the tape until a not ready indication appears on the Operators console since this may cause writing over information already present
485. ugh Saturday Users may be actively using the system while SAV is running The tapes used for daily filesave are kept in racks in the machine room They are numbered FS0001 FS0002 etc The daily filesave program asks for a tape by number No copies are made of the daily filesave tapes Usually no filesave of any type is run on Monday Weekly filesave SVW for the UM system is run on Sunday morning at 7 00 A M The complete weekly filesave usually takes about 4 hours to run to completion No users may be on the system while SVW is running There are six sets of weekly filesave tapes These tapes are stored in the machine room and are used over and over in order one set per week When one of these tapes shows a serious error it is replaced with a new tape The program FSC is used to copy the weekly tapes There are usually only two sets of copies leaving several sets of the original filesave tapes with out backups The sets of copy tapes are kept in in the safe Every sixth week a new set of tapes is unpacked and used for the copy set of tapes These copies are saved and stored in the safe for a few days until they can be moved to the Bendix building The tapes sent to Bendix are kept in a special room and are assigned Z type rack numbers if they are UM system tapes The directory of Z tapes is kept in the Computing Center second floor machine room in a notebook near the operators console UB System Filesave The weekly
486. umber second column equals user ID and third column equals number of virtual pages being used by each job Under DR first column equals supervisor number second column equals user ID and third column equals number of DRUM READS Before and after this list a line of general system load information will be printed The program will print out names of any non MTS jobs if any appear in its list mts hgs Response example on operators console is DISK I O VM 3 356 IECN 632 536 SQR7 692 6 313 SPL9 269 583 ID2R 354 Etc Etc DRUM READS 76 1ZBK 292 536 SQR7 267 Etc Etc Etc Etc Etc Etc Etc Etc OPERATOR FILES 93 REAL PAGES 2349 DBMS 54 1N2A 92 219 on RPN 130 SPDL 107 216 SPD5 96 OPERATOR FILES HGS 220 NAME FUNCTION USAGE REMARKS The user OPERATOR FILES HLB Forces all access to the UB host through the Host Locator Also can be used to restrict access to the host MTS HLB This program is used for testing by systems people will be promted as follow For Host UB Enter YES to force Host Locator access or NO to release system to users After the operator enters the appropriate response a confirmation message will be given OPERATOR FILES HLB NAME FUNCTION USAGE REMARKS OPERATOR FILES 221 HLG Starts HASP log collection program mts
487. unt tapes The file UAM contains the following information During the two eight hour periods midnight to 8am on Sunday and Monday mornings when the system is in Unattended Mode there is no operator available so that any action on the part of the user which requires operator intervention such as mounting tapes will be rejected If the mount request eminates from a batch job the job will be put in the EXECUTE queue in HOLD status until unattended mode ceases and the job priority is accepted If the system fails for any reason it will very likely not be restarted unless a systems programmer happens to be doing systems maintenance at the Computing Center There is also always the possibility of a shutdown on very short notice UNATTENDED MODE 520 UNYN Terminals and SCPs etc What is an MTS terminal and where are they We have about 50 terminals here at UNYN Most of these are the Ontels DECWriters and Visual 550s located in the user area But there is also the Ontel in the Consultant s Room the DPA s terminal on the wrapping table and the TV jobs status display These terminals are used to access MTS at the Computing Center on North Campus There are hundreds of other MTS terminals around campus The IBM 360 operator s console is not an MTS terminal How are these terminals connected to MTS on North Campus We have three Secondary Communications each of which links some of our terminals to black boxes
488. upposed to perform there may be one or more invocations one or more tasks of that job program running at the same time Another name for a task is a job the two terms will be used interchangably throughout this manual The most important job program is the MTS job program There is one invocation of this job program one MTS task for every terminal line connected to the system in addition a new MTS task is created each time a batch job is run Other examples of job programs include the OPERATOR job which controls a terminal and accepts commands to manage the system the Disk Manager which manages access to the system s disks and the Paging Drum Processor which manages paging activity A job number or task number is associated with every running task The job number is assigned at the time a task is started and may be used to refer to the job in controlling the system For example the STOP job takes as a parameter the number of some other job which is to be terminated A task is started by a request from the main or backup operator s console which are described later A task request consists of the name of the job program to be started optionally followed by one or more parameters A complete description of each job program is given later in this manual Once started a task may continue execution indefinitely or it may terminate stop running Whether and when a job terminates depends upon the particular job
489. ur characters or less that includes the mandatory at the beginning which denotes the file as being a public one However it is possible to use files that have longer names but then you will be fighting the supervisor command language structure of four characters at most per parameter OPERATOR FILES ABS NAME FUNCTION USAGE OPERATOR FILES 169 ABS starts up ports AB10 AB3F mts abs ABS is normally started by MNS OPERATOR FILES ABS 170 ADS NAME FUNCTION USAGE OPERATOR FILES ADS starts up ports AD10 AD3F mts ads ADS is normally started by MNS OPERATOR FILES ADS AES NAME FUNCTION USAGE OPERATOR FILES 171 AES starts up ports AE10 AE73 mts aes AES is normally started by MNS OPERATOR FILES AES 172 AFS NAME FUNCTION USAGE OPERATOR FILES AFS Starts up ports AF10 AF3F mts afs AFS is normally started by MNS OPERATOR FILES AFS ANS NAME FUNCTION USAGE OPERATOR FILES 173 ANS starts up ports AN10 AN17 mts ans ANS is normally started by MNS OPERATOR FILES ANS 174 AUS NAME FUNCTION USAGE OPERATOR FILES AUS starts up ports AU10 AU6F mts aus AUS is normally started by MNS OPERATOR FILES AUS AVS NAME FUNCTION USAGE OPERATOR FILES 175 AVS starts up ports AV10 AV6F
490. ust doesn t hand someone a receipt for one or two jobs out of the box All or none of the jobs are picked up each time DELIVERY codes 06 04 82 NUBS BSAD ISR ISR1 ISR7 HSRI DBRN EPA VRI NOTE See section on TAPE and DISKETTE mailing later on in this manual DELIVERY 3 91 DISK STORAGE NOTE See DSK and DMGRSTAT earlier in this manual These are disk manipulation programs DASDI The pupose of this program is to label re label or format disk packs according to either the VAM2 or VAMX conventions The following devices can be DASDI ed 2301 2311 2314 3330 7330 3340 3350 3370 3375 3380 6280 Usage mts dsd Promts for input at operators console macrolib mts cmdmaclib macrolib file cmdmaclib dasdi Input data consists of the following operands starting in column 1 Dxxx MTSxxx VX V2 pvn PAGING PRIVATE optional pars Parameter descriptions follow In any place a number is called for a decimal number may be given or X hex number pvn gt public volume number if it is to bea public volume PAGING gt if it is to be a paging volume PRIVATE gt if it is to be a private volume LO or LABELONLY gt if the volume is to only be labelled or re labelled as opposed to being formatted IPL or TPL nnn gt if it is desired to leave nnn pages of IPL area starting at the front of the pack as well as formatting the pack The IPLINIT program
491. ut in the temporary file DIAG which may be copied to PRINT after OLTEP terminates NOTE Some of the following information though re worded was taken from IBM volume C28 6650 6 When OLTEP is running it will send you the message IFD102I OLTS RUNNING IFD134I WARNING DASD VOLUME LABELED CEPACK NOT OLTS 417 PROTECTED FROM WRITE IFD107I OPTIONS ARE FE NMI NPP NEL CP EP PR NRE For amore complete description of these options see the IBM publication C28 6650 6 PR Print sends you messages produced by the OLT programs i e test results requests for operator assistance etc CP Control Print OLTEP sends you a message letting you know when a test section starts and terminates The termination message states whether errors were detected EP Error Print OLTEP sends you messages that diagnose errors detected by OLT programs FE First Error Communication When the OLT program detects a device error OLTEP suspends testing and asks for a new test definition You can then define a new test resume testing from the point at which the error occurred or terminate the job step If you choose to resume testing you can change the OLTEP options that apply to the test PP n Parallel Print OLTEP sends output to you as well as to file TL OR TL N TEST LOOP REPEAT TEST N TIMES UP TO 32767 EL Error Loop When an OLT program detects an I O error OLTEP repeate
492. value of the supervisor s sense flag This value is interpreted by the unit check routines to determine the cause of some kinds of tape errors It is mostly of interest to programmers working on the unit check routines but also required to interpret channel detected errors described below If this value is anything other that 01 the sense information will be unreliable Feeeeeee Is the hexadecimal sense data associated with tape error Again this is interpreted by the unit check routines to determine the type of error and how to recover from it This information may be of use to both tape vendor customer engineers if they are attempting to correct a hardware problem and also to programmers working on the unit check routines themselves f is the hexadecimal value of the channel command the drive was trying to perform when the error occurred Typical values for this field are 01 write tape 02 read tape in a forwards direction OC read tape in a backwards direction TAPE ERROR CODES 496 07 rewind tape OF rewind and unload tape RUN 1F write tape mark on tape 27 backspace block on tape 2F backspace file on tape 37 forwardspace block on tape 3F forwardspace file on tape CB set tape mode to 800 bpi C3 set tape mode to 1600 bpi D3 set tape mode to 6250 bpi There are other values which can be encountered here but should not normally be This field can be useful for customer engine
493. w THE PRINTER All lines in the message area will eventually be printed on the console log printer If no printer is available the behavior of the display will be the same as if a two second per line printer were used The format used there will be the same as the format used for the 3066 console The wider printer carriage allows 100 characters of message text and this many characters will be maintained internally for use in printing although only 65 to 72 will be displayed on the 3270 screen as described above Thus a job may put information of less immediate interest in these printed but not displayed character positions Lines will not be removed from the message area until they have been printed To summarize message area handling when a line is added to the message area a line must first be removed the line removed is the first one which has been printed and is not part of a message response pair When the message area becomes almost full with lines which cannot be removed from the screen no further lines will be added except in response to requests from the request area If the message area becomes totally full as the result of entering several requests while the screen is almost full the message SCREEN FULL will appear in the request area and no further requests will be allowed until a line is printed or a pending message response is entered If the printer becomes Not Ready the 3
494. w on file space Always check with the Business Office before running this procedure DEADFILES is a procedure for removing files user file catalogs and entries in the Restore Message and Userdirectory databases that belong expired CCIDs This procedure at present is usually run on the last or second to last Thursday of each month during the 4pm midnight shift The date used is usually the second to last day of the previous month the Deadfiles routine can be run at other times if the system is low on disk space Always check with the ITD Accounts Office before running this procedure If any problems occur while running the DEADFILES procedure check with a senior member of the systems group Tom Valerio before going on with later steps DEADFILES The Procedure 1 Signon from a terminal or at the operators console using the CCID RSTR 2 Make sure that the permanent files that will be needed exist and are empty by using the following command source deadf_start This will turn command macros on attach the appropriate command macro libraries and then make sure that all the necessary files exist and are empty 3 Use the Scanfiles program to create lists of the ccids files message system mailboxes and user directories to be removed df_scan mm dd yy where mm dd yy is the date specified by the Accounts Office for example June 30 1983 would be entered as 06 30 83 while July 1 1983 would be entered
495. wi The call FORGET MTS006 Disk Packs Moving To display information about the use of this program HELP JOB PROGRAMS DMGRSTAT 102 JOB PROGRAMS Description Presently the only information displayed will LSTATUS Purpose Prototype Parameters be the prototypes for the various commands To display the lock status of a disk page disk volume or task a LSTATUS PAGE lt ea gt b LSTATUS TASK lt task gt c LSTATUS VOLUME lt pvn gt lt ea gt is the external address of the disk page whose locking status is to be displayed lt ea gt may be either a quotedhexstringor lt volume gt lt page gt lt task gt is the task number of the task whose status is to be displayed lt pvn gt is the MTS public volume number whose locking status is to be displayed Description If form a is specified the current locking JOB PROGRAMS DMGRSTAT task and the task numbers associated with the requests waiting to lock the disk page whose external address is lt ea gt are displayed If form b is specified the external address for each page locked by lt task gt is displayed along with the external addresses of pages lt task gt s requests are waiting to lock If form c is specified the locking task and waiting tasks are displayed for each locked page on MTS public volume lt pvn gt The external address is included in the display of each page Examples REQUEST
496. with white in the back of the machine it which a phone line used to transmit the data here the Center panels room hooks at UNYN The SCPs are the located on the table Each has a blue Gandalf modem the SCP to Processors on top of from the terminals to the Center on North Campus Each of our three SCPs is hooked up like this SCPs front UNYN TERMINALS SCPS term term term term vee Reeser ens ada a iad ae ek ca I Sa BRO A A te ae cables from 20 terminals gt Te ee Sista eneas SCP Phone circuit to MTS on North Campus ETC 521 What should you do when the MTS terminal on the wrapping table won t respond or when the users complain that one or more of the terminals is dead 1 Figure out if MTS is alive If either of the printers is actually printing then MTS is probably OK If not you should call the MTS operator on the red phone and find out if MTS is down If MTS is OK you should try to test a dead terminal as follows If somebody was already in the midst of using it and was signed on to MTS do the following On Ontels press the FIX key On DECWriters press the DELETE key On Visual 550s press the DEL key Then type HELLO and press RETURN If the terminal and SCP to which it is attached are working properly you will get something like 11 48 29 03 31 83 216 LSI 11 MINOS Version AR293 and you c
497. y In this case the cause of the problem should be found look back up the console log to the last time CMDTAPE ran and CMDDIR fixed Answer the question Y or N as appropriate A yes response will cause this copy of CMDTAPE to terminate quietly a no response will cause it to complain about a directory error MOUNT NEW CMDSTAT TAPE RING IN WITH RACK Mmmm This message appears on the console It is not an error CMDTAPE has either filled or abandoned the current tape and is labeling a new one Mount a blank tape and enter the drive name also put the sticky labels on the reel CMDTAPE will label the tape release it and issue a mount request for it If CMDTAPE can not get the drive or write on the tape it will print a message and ask again Other Problems The most likely problem is that CMDTAPE will appear not to be running files to be copied will pile up Either CMDPIKUP is not starting DWB DWB has stopped working or CMDTAPE itself is fouled up the most likely explanation CMDTAPE can be run or debugged from any terminal if nothing else this will cause system messages which normally go to the wastebasket to be printed Use the ccid SYS at UM anyway PROT OFF and see DWB for device assignments CMDSTAT 333 FILESAVE SCHEDULE AND TAPE STORAGE See SAV and SVW for Filesave procedure Daily filesave SAV is run on the UM system at 12 30 A M Tuesday thro
498. y It is a stand alone program which provides a small repertoire of commands to allow the operator to interactively specify the name of the MTS sequential file containing the system to be loaded and to display and modify memory locations before or after loading This program begins by writing the following message on the operator s console DO YOU WISH TO RUN THE CURRENT SYSTEM YES OR NO A YES or a null line entered at this point will be the normal response Any other response including CANCEL invokes command mode The following commands are currently available COMMAND LOAD SYNTAX LOAD NAME FILENAME ADDRESS DISKADDRESS PURPOSE to load a system from the given MTS sequential file Both keyword parameters are optional If NAME is not specified the current system filename IPL O is assumed IPL 1 2 and 3 are succeedingly older versions of the system Tf ADDRESS is not specified the address of the IPL disk is assumed The filename should be specified in internal format ie colon less Execution of a system loaded using the LOAD command is achieved using the START command E LOAD NAME MTS TESTSYSTEM ADDRESS 260 E L N IPL TEST A 267 EXAMPI EXAMPI COMMAND START SYNTAX START PURPOSE to start execution of the previously loaded system Note entering YES in response to the initial prompt actually causes a LOAD and a START command to be implicitly issued E
499. y of Michigan However you can broadcast by using the Super Console First select the node name desired Then type BROADCAST message once for each PCP desired SETDATE Contents The object module of a program to set the time and date in UMnet PCPs Purpose To set the time and date in a UMnet PCP obtain a reason for the reload print a nicely formatted message on the PCP operators console optionally log the time date and reason for reloading on logical I O unit 0 and optionally reset any reply message for the primary host from which SETDATE is being run Location MNET SETDATE Use The program is invoked by using the RUN command Logical I O units referenced SCARDS response to the Enter Reason For Reloading prompt MERIT SETDATE 402 Parameters Description MERIT SETDATE SPRINT prompt if the reason for reloading is not given in the PAR field of the RUN command 0 optional if assigned one line giving the time date and reason for reloading is written to this logical I O unit followed by a log of the commands read from logical I O unit 1 if any 1 optional if assigned lines are read from this logical I O unit and passed to the PCP using the CONTROL subroutine PAR cu xx hostname reason for reloading The PAR field may begin with a network control unit name AA AB AD AE or AN It must also include a network host name this is normally N Th
500. y of the tape then the operator should try again on that drive otherwise an IM reply should be given or BY if all such drives are busy DESTINATION OPER TEXTLOC MOUNT 101 BADDENS MOUNTCMD MNT02810 SOURCE MOUNT 101 5ST253 MOUNTCMD MNT01807 MISCELLANEOUS MESSAGES 414 gt gt gt THE FOLLOWING VOLUME S HAVE NOT BEEN FOUND xxxxxx RETURN TO RETRY OR TYPE CANCEL IF NOT AVAILABLE User Explanation Generally means a pack is not mounted or the drive is not ready If preceded by a unit check message try moving the pack to a different drive If it still fails call a systems programmer DESTINATION OPER SOURCE FILE SYSTEM 482 5 VOLGET gt gt gt USER xxxx yyyy FILE 22zzz User Explanation Whenever anyone modifies a public file this message is printed on the operator s console followed by the first line inserted into the file The xxxx and yyyy are the two forms of user id kept in the MTS dsect They should be identical zzzz is the name of the public file being changed DESTINATION OPER TEXTLOC DSRS 637 WOD1 WODISK DSRS2386 SOURCE DSRS 637 WOD1 WODISK DSRS2394 gt gt gt VOLUME vvvvvv PVN xx NOW FOUND ON Dyyy PLEASE CONFIRM User Explanation May result from moving a pack to a different drive but probably means a spurious pack such as a test pack has been found After verifying that the indicated drive contains the right pack do not move any packs co
501. y the comments field for the entry This field is an arbitrary character string of up to 256 characters The COMMENTS parameter must be the last parameter specified The sequence ID assigned to the entry is printed CONFIRM type SEQ n PFIELD m n SFIELD m n COMMENTS string This command confirms that the data for a new tape entry has been successfully written to the tape The parameter type specifies the type of tape entry being confirmed The parameter SEQ n speci fies which entry is being confirmed n is the sequence ID that was assigned to the entry by the ADD command TAPE DIRECTORY PROJECT 512 Command Explanation The PFIELD and SFIELD parameters specify the pri mary and secondary search field values to be assigned to the new new entry m is the lower bound value and n is the upper bound value The format of m and n must correspond to the format specified in the master entry for type For Julian time formats m and n may be given in either integer format giving the actual number of microseconds etc since March 1 1900 or in character format giving the date and time in the form MMxDDxYY hh mm ss or MMxDDxYY where x is any character For example PFIELD 01 01 78 00 00 00 01 31 78 23 59 59 These search field values may be given either with the ADD command or CONFIRM command any value given with the CONFIRM command will override an ADD command value The COMMENTS pa
502. yne State for WAYNESTT to stop and start their end Note that the UM to UB link just runs between two halves of the ES3090 300J and is not at all within the purview or responsibility of the NOC local operations should be able to handle it There is currently a policy debate about whether NOC since they watch all the other links and notify the appropriate operations staffs when things die should also look at this link Keep tuned to your favorite local network to see what today s policy is Normal procedure for getting rid of things is using WAYNESTT as an example a Using SYS RMOPR or MTS RMC enter the RM command DRAIN WAYNESTT b It should tell you Processor WAYNESTT drained Processor WAYNESTT destroyed orn Then check using SYSTEMSTATUS or equivalent to make sure that the SAxx line SA07 for WAYNESTT has gone If it tells you that a task named RM LINK still has SA07 attached then you need to STOP that task If doing this from a terminal use the KILL verb of SYS TASKS to avoid having the operators prompted for an ok Then wait It sometimes takes several minutes until it goes away Note this condition seems to occur mostly on the UM to UB link If you don t make sure the task owning the SAxx line has gone before you do a START again then either on that RESOURCE MANAGER Ss 4 541 START or on the following DRAIN the LINK processor will get a program interrupt and
503. you run into any problems call the user back and describe the problem in addition to taking the steps detailed below VSE SYSTEM SHUTDOWN AND START UP PROCEDURES 55 Since VSE1 runs under control of VM XA you must issue VM XA ATTACH commands from the VM XA operator s console before replying to VSE1 mount requests This is just like mounting tapes for UB MTS except that UB MTS issues the ATTACH command automatically in most cases under VSE the ATTACH commands must always be issued by an operator VSE1 mount commands are not really generated by VSE instead they are comment lines which the VSE1 programmer puts into the VSE1 batch job Since each mount request is just words typed by a VSE1 programmer the exact format of each mount request may vary slightly although the overall request should contain the same information as an MTS mount request VSE1 refers to tape drives by number VSE1 tape drives are numbered 680 681 682 and so on A VSE1 mount request will request that a tape be mounted on a particular drive number however the VM ATTACH command can be used to attach any real tape drive to any VSE1 tape drive number so that any tape drive can be used to satisfy any VSEl mount request A single VSE1 mount request may request more than one tape all of the tape drives must be ATTACHed and all of the tapes must be mounted on the drives before you reply to the VSE1 mount request The following steps describe the VS
504. ype 9700 DevNum 0390 ON LPM 80 Sch 0199 Job 00814 Dev Num A number assigned to a device by a systems programmer They ve been assigned in such a way to give channel and device information SCH Subchannel number This is assigned to a device by the I O system LPM Logical path mask This is an 8 bit mask describing how many paths there are toa device Example 80 gt 1 path path 0 available CO gt 2 paths paths 0 1 available a disk will have 2 paths available JOB PROGRAM UNITS UNITS Old Style JOBS PROGRAMS 157 No longer used NAME FUNCTION USAGE UNITS UNITS is a program to display the configuration of channels control units and devices attached to the system for printing certain aspects of the status of peripheral devices and for doing a limited amount of error checking of the tables that describe these things For operations staff the most frequent uses of UNITS will probably be to verify and clarify the effect s of previous ONLINE and OFFLINE commands to investigate why a device is hung and just to see what is connected to what For systems staff UNITS will also be used to help check new Tables assemblies UNITS is available as a job program which in case it is invoked by entering a job request at the console UNITS units command mod UNITS is also available as an MTS program in which case it is invoked by entering run mta unit
505. ystem If any users are signed on send a message asking them to signoff To send a message type write name of user To find out more about who the user is type 50272 used 12763 free 219 frags 1568 blocks UNIX FILESAVE 558 0 3 fra gmentation OK to dump Onu dev rphib y dump Onu dev rhplb DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP DUMP to go DUMP to go DUMP to go DUMP DUMP DUMP DUMP DUMP DUMP DUMP remounti o 5 UNIX FILESAVE Date of this level 0 dump Wed Jan 28 17 51 12 1987 Date of last level 0 dump the epoch Dumping dev rhplb e to dev rmt8 mapping Pass I regular files mapping Pass II directories estimated 54144 tape blocks on 1 3D9 tape s dumping Pass III directories dumping PASS IV regular files 20 93 done finished in 0 18 38 75 done finished in 0 15 56 24 done finished in 0 11 Tape rewinding Change Tapes Mount tape 2 NEEDS ATTENTION Is the new tape mounted and ready yes or no NEEDS ATTENTION Is the new tape mounted and ready yes or no NEEDS ATTENTION Is the new tape mounted and ready yes or no yes TApe 2 begins with blocks from ino 8435 70 81 done finished in 0 10 89 59 done finished in 0 03 DUMP 54144 tape blocks on 2 tape s DUMP IS DONE lelvel 0 dump on Wed Jan 28 17 51 12 1987 Tape rewinding ng file system
506. ystem should display something like the following ODCT at 2BEA64 MNT RD1 DCTSTAT INUSE DCTPCE 2B7F98 DCTBUFAD 0021EE00 DCTPSTAT 0 DCTDCB 000000 DCTBUFCT 1 DCTNO 43 DCTDEVTP MNETREADER DCTIOTYP WRITE DCTFLAG2 0 DCTPRINC 0 DCTPRLIM 0F DCTFORMS 00 DCTCHARS Uc Pn DCTPRLEN 80 DCTPLINE 0 DCTEWF 002B7FE9 DCTFLAGS 0 DCTCHAIN 2BEAAO DCTDEVN MNT RD1 DCTWAYT 0021EE00 DCTSJNO 5969 DCTLOCK 00 DCTPRNTR Line DCTTOTPG 0 DCTTOTJB 0 DCTRESCN 00 WORK SPACE 000034 00000000 00000000 3 For each item from step 2 look at DCTSJNO underlined above to get the task number 4 Issue the following MTS systemstatus command sy j tasknumber where tasknumber is the task number from step 3 The output will appear as Job Name JobTbl Status BTS2 User Proj Par PSW Prob Time Super Time VP RP RPN Page ins File I O 0 05969 HASPLING 000AD320 I O on AF7E 20 MNAF 072D0000 002C1E44 1 299 5 068 1112 0 RO 7 00000001 006001A0 002C2164 00000000 0021EE00 00600000 0021EE80 802C1E08 R8 F 005F7A80 005F7B00 802C1A86 006000DC 002C19B0 0060004C 802C1E08 402C1B96 FPRs 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 Dev AF7E 70 005969 HASPLING 000AD320 I O on AF7E 20 MNAF 072D0000 002C1E44 1 304 5 083 1112 0 RO 7 00000001 006001A0 002C2164 00000000 0021EE00 00600000 0021EE80 802C1E08 R8 F 005F7A80 005F7B00 802C1A86 006000DC 002C19B0 0060004C 802C1E0

MTS Operators' Manual - Deep Blue

Contents

Download Pdf Manuals

Related Search

Related Contents