&% `$xxuurrDESY &% `$xxuurrDESY - DESY
Contents
1. 8 DESY Computing Newsletter No 2 8 DESY Computing Newsletter No 2 November 1991 A Publication of the DESY User Support Group Contents Editorial 2 ee How to deal with hanging IBM Terminals 2 020202 eee eee Upgrade of the DESY Computing Facilities 0 2 2 02000202 eee Introducing the Central VAX cluster at DESY 2 2 2 ee ee ee DESYNET 1979 f September 19 1991 2 ee eee File Transfer Service extended 2 2 Usage of CERNLIB at DESY 2 2 ee Manuals and Writeups 1 ee es The Future of TeX and METAFONT 2 2 ee ee Locating a DESY Library Book by means of a Database 2 2 2 000 Do it better or doit the same 2 2 2 2 20 ee ee ee Archiving inactive Cartridges 2 ee Questions and Answers fromthe UCO sasssa aa es DESY Computing Newsletter No 2 8 e Editorial You have now finally received the second issue of our Newsletter It took us a few weeks longer than we had hoped to complete this issue but the already forgotten summer holidays and other pressing work delayed the completion Several people besides the editors have con tributed to this newsletter and we would like to thank them for the work they did in preparing their articles Nevertheless most of the articles were pro duced by the editors themselves This is probably what happens to many publications like ours but it is not what we are2. the User Consulting Office UCO RO1UCO 11 MANUALS AND WRITEUPS 8 DESY Computing Newsletter No 2 8 Where can I find what Introduction to the DESY Computer Center The First Three Days DESY Computer Center Primer S How to Get Started on a VAX U CERN Computer Newsletter U L DESY Computer Newsletter S L NEWLIB NEWLIB Reference Manual S L P Einf hrung in NEWLIB deutsch S FORTRAN and Preprocessors VS FORTRAN Version 2 Language and Libr Ref IBM S VS FORTRAN Version 2 Programming Guide IBM VS FORTRAN Version 2 Interactive Debug IBM FORTRAN 77 Language Reference Siemens FORTRAN 77 Programmer s Guide Siemens FORTRAN 77 Library Reference Siemens IBM FORTRAN IV Language Reference IBM FORTRAN IV Programmer s Guide An Introduction to FORTRAN 77 M Metcalf S FORTRAN Error Monitor MORTRAN 3 v rrer ee e Other Programming Languages C Language Manual C Compiler User s Guide Pascal Reference Pascal Programmer s Guide PL 1 Reference PL 1 Programmer s Guide PL 1 Compiler Messages Assembler Language Assembler Programmer s Guide Le en rerrerecerere e Legend for Where to find what S Self service in the user area of the computer room U Available from the UCO office L Available for reference in the DESY library P Available for reference at most terminal pools MANUALS AND WRITEUPS 12 DESY Computing Newsletter No 2 Batch operation IBM G
3. user PMDF Pascal Mail Delivery Facility soft ware is running on all the computers in the cluster and provides a consistent addressing syntax as well as automatic forwarding to remote VAXes even for users who do not have accounts on VXDESY In formation on the correct address format for sending mail from the IBM to a DECnet node or the VAX cluster to a BITnet node can be found under o HELP VXDESY on the IBM o HELP BITNET on VXDESY It is planned that the VAX cluster will also be come a gateway for SMTP Simple Mail Transfer Protocol mail which runs on the TCP IP network The VXDESY will then be DESY s official mail gate way connecting three widely used mail networks Available Software The central VAX cluster is not as fortunate as the IBM in having a whole group at DESY dedicated to its support The two full time system managers of the cluster only just have enough time to install software manufactured by DEC the company which makes VAX computers and have to leave the in stallation and maintenance of all other software to the users themselves As a result it is almost im possible to put together a complete list of all the software running on the VAX cluster will there fore just mention a few of the products which are available DEC Products The following 5 language compilers are installed on VXDESY oC FORTRAN LISP only on VXDSYB PASCAL PL I only on VXDSYA and VXDSYB
4. INTRODUCING THE CENTRAL VAX CLUSTER AT DESY 8 DESY Computing Newsletter No 2 8 You can get a list of all the DEC products in stalled on the VAX cluster with the command SHOW LICENSE or LICENSE LIST Other Software The Theory group has installed a few programs which might be of interest to users from other groups FORM a REDUCE like algebraic program MONGO an interactive graphics program NAGIib a mathematical subroutine package For more information about how to use these programs consult H Wittig WITTIG on VAX TOOWIT on IBM tel 2414 or K Wipf WIPF on VAX RO1IWIP on IBM tel 3222 ZEUS has also installed software on the VAX cluster most notably the CERNLIB The article Us age of CERNLIB at DESY in the last Newsletter gives a brief description of how to use the CERNLIB on the VAX For more details look at the file CERN_ROOT HLP CERNLIB HLP User Support The central VAX cluster is operated and maintained by two full time system managers W Krechlok and V Heynen of F58 To help reduce their work load somewhat they have designated group managers for the various groups of users work ing on the VAX These group managers are re sponsible for registering new users managing disk space and resetting passwords for the users in their group Below is a list of the groups and their managers The userid of the manager is added in parentheses where it differs from the last name Group
5. It is now possible to transfer between the PC and any machine with an internet address via FTP using the standard FTP commands As a result the file transfer service was extended to any machine with a TCP IP implementation i e VAXes Workstations PCs etc The name of the PC is ucope desy de in short ucopc and its address is 131 169 7 38 It can be used during UCO office hours as a file server from any machine on the site or you may al ternatively do the file transfer on the PC to and from your favorite host In file server mode the PC files can in principle be accessed by several users simul taneously As this includes your private diskettes in the PC s diskettes drives you should write protect all source diskettes you put there PC s have no real access control mechanism Thus everyone could copy whatever software is stored on the UCO PC We would like to remind you that most software is copyright protected and that it is illegal to make unauthorized copies Apart from legal considerations it is good practice to re spect other peoples property and not to copy their files unless explicitly authorized In addition we have taken measures to protect the hard disk drive C against write access to main tain the PC s integrity e Usage of CERNLIB at DESY by Harald Keller Current State of CERNLIBs The DESY computer center intends to update the CERN program libraries at the same frequency as at CERN The latest i
6. here you type c ibm GKS based appli cations like private GKS programms GEP and PAW are supported but TEX previewing is not possible e GKS based applications and TpEX previewing works on on IBM graphic terminals 3192 G 3193 on the HP9000 425t workstations and almost on X terminals The current situation with respect to IPS 90 is the following e Some of the new workstations and a number of X terminals have been ordered e The physical placement of X terminals on the DESVY site will be controlled by the management e R2 has to set up the whole system and get it to work reliably 2 X terminals are basically specialized stripped down and therefore cheap workstations which allow graphic access of workstations via the standard X protocol DESYNET 1979 SEPTEMBER 19 1991 DESY Computing Newsletter No 2 8 e File Transfer Service extended by Michael Behrens During the last months the UCO PC has been heav ily used for file transfer to and from the IBM The transfer used an emulation that connected the PC to our central IBM thus excluding other systems and especially the VAXes It it was also not the fastest or most elegant method We could and can trans fer files to and from 35 or 54 inch diskettes IBM compatible using both single and double density diskettes We have recently installed TCP IP on the UCO PC and connected it via Token Ring to the Internet
7. inactive cartridges will simply be removed from the robot and stored in a rack somewhere outside the robot New scratch car tridges will then be added to the robot system so that we will always have a set of scratch cartridges available in the robot For the old round magnetic tapes R1 had de veloped an archiving system for inactive datasets In this system archive tapes A tapes were treated differently from the normal M tapes but this system is not used for the inactive cartridges Later there will be a real archive for cartridges in remote archive rooms and the users will be able to see the location of their data sets This system will be an extension of the present archive system and it will be described in detail in a future issue For the the large amounts of raw data and DST cartridges R1 has together with the HERA exper iments developed a different scheme This scheme will also be described in a later issue Inactive cartridges are such cartridges that contain only datasets that have not been used for a longer time Currently this means not used for more than 6 month e Questions and Answers from the UCO by Katherine Wipf The User Consulting Office answers quite a variety of questions but usually only a few people hear the answers This column tries to bring the answers to more frequently asked questions to a larger audience about FORTRAN Question have two versions of my FORTRAN program
8. one is single and the other is double preci sion The single precision version runs without errors but the double precision program produces an under flow exception in a certain subroutine How can this happen when my programs are identical except for the precision of the variables Answer The subroutine with the underflow ex ception was passed a number of parameters and two of these were zeroes i e the number 0 was passed so that the subroutine call looked like this CALL SUBR X Y 0 0 2Z The single precision version interpreted the zeroes correctly but the double precision routine translated the 0 into 0 000000000000000120E 78 and gave an underflow message This happens because 0 is a single and not a double precision zero When 0 0D0 double precision zero is used the error message disappears NOTE It is dangerous to pass numbers rather than variables in a subroutine call because it can lead to a situation where the number 2 no longer has the value 2 The following short program demonstrates this effect PROGRAM MAIN CALL CHANGE 2 WRITE 6 2 END SUBROUTINE CHANGE K K K 1 RETURN 21 QUESTIONS AND ANSWERS FROM THE UCO DESY Computing Newsletter No 2 8 The program output is the number 3 and instead of the expected value 2 A FORTRAN subroutine does its calculations us ing the original data passed to it and does NOT work with a local copy Thus when the value of the vari able K is change
9. re traed tapes much before 10 30 h or 15 30 h Only if your RETRA job claims that a tape is PRESENT can you use it straight away Question am trying to copy a large parti tioned data sets from tape to fast disk My FAST Job failed with a system completion code B37 What can do QUESTIONS AND ANSWERS FROM THE UCO 22 DESY Computing Newsletter No 2 8 Answer A system completion code B37 means that more space was needed for a data set but was not available generally the disk is full or the maxi mum number of extents have been used If the data set was migrated onto tape or copied to tape by one of the standard backup procedures this error means that there is not enough space on the FAST disks to store the data set The best so lution is to try again early the next morning If the data set was copied to tape by any method other than those mentioned above there is no infor mation on the tape as to how large the data set is The FAST program uses a default size for the disk data set when it cannot find size information on the tape and if the data set is larger than this default size a system code B37 will occur The solution is to allocate a large enough data set prior to fasting it must have the same DCB parameters as the data set to be fasted and then use the PRE option of FAST see HELP FAST Question My job terminated abnormally after a data set was writing had filled 5 cartridges and th
10. Manager Argus S Khan External Users K Wipf F15 A Hoelscher HOEL F51 R Pforte F52 K Rehlich F58 V Heynen W Krechlok Hasylab O Beimgraben T Kracht H Schultz HERMES M Dueren H1 E Deffur S Egli KRYO M Clausen CLAUS M F Lohse OPAL T Behnke J Hagemann R1 E A Knabbe K Wipf R2 M Kuersten Theory H Wittig University H Kreiser W1 H Peters PETERS HB ZEUS l Klein T Poser C Youngman 1 Institute A Puskeppel If you need help with a problem on the VAX you should ask your group manager first You can also bring your problems to the User Consulting Office see HELP CONSULT You should only consult the system managers with serious problems which can not be solved by your group manager or the UCO There is a committee the VCC VAX Com puter Committee where VAX managers and in terested users meet to exchange their experiences The meetings take place at irregular intervals and are announced via electronic mail If you wish to be placed on the mailing list please send a message to VXDESY KRECHLOK INTRODUCING THE CENTRAL VAX CLUSTER AT DESY DESY Computing Newsletter No 2 Configuration of VXDESY VAX 8350 VAX 6410 VAX 9110 CPU CPU CPU 32 Mbyte 64 Mbyte 128 Mbyte VAX 6410 CPU 64 Mbyte Star Coupler HSC 70 DISK TAPE CONTROLLER HSC 50 DISK TAPE CONTROLLER DESY F58 7 INTRODUCING THE CENTRAL VAX CLUSTER AT DESY DESY Com
11. as DESY or the same as yesterday Similar with graphics CERN didn t do the same as DESY with well proven GEP it did better with the new PAW meanwhile of course well proven too e Is DESY doing the same Yes we are NEWLIB has been the same for over 20 years The CERN libraries have been here for a long time We made the Siemens FORTRAN 77 com piler standard in 1984 better than FOR TRAN IV and the same as at CERN We followed CERN and made the IBM VS FORTRAN compiler standard in 1986 not as good as Siemens but the same as at CERN We followed CERN and introduced HISTORIAN better than PATCHY but the same as at CERN PAW is here the same as at CERN So is CMZ better than PATCHY but the same as at CERN FATMEN are coming XFIND is coming hope OOP Object Oriented Programming is coming It s not the same as in the past There has never been anything like it But already people are telling me we should do it the same as HEP namely with C Who is bothering to find out whether Eiffel might not be much better than C Enough Let me sum up I think we understand what Mr McCubbin and the rest of our users mean We try to introduce standards also HEP and CERN standards The empire is there though it does not want to be one We try to cooperate with CERN though it seems to b
12. been chewing on this ever since can t swallow it What does it mean e Do it the same as yesterday e Keep punching cards for FORTRAN IV e Wylbur e No VM CMS no UNIX no NextStep Although see FORTRAN IV and Wylbur score a few yes this is certainly not the aim of Mr McCubbin But then what does he mean e Do it the same as everyone else e Do it the same as many business companies e Do it the same as the HEP community e Do it the same as CERN Much closer eh And who should do it the same rather than better e The physicists Should they still hand wire detector electronics e The computer scientists No CERNlib Fatmen PATCHY SHIFT WYL BUR or ZEBRA e The computer scientists of DESY No NEWLIB no GEP no central catering for magnetic tapes Very close suspect The slogan obviously isn t meant for all times and all people This would keep us back in the caves No it is meant for MY ENVIRONMENT EVERYTHING EXCEPT MY OWN WORK This is quite an acceptable position Let the environment of the high energy physicists evolve but evolve slowly This seems to be a special case of what Peter Kafka calls the cancer of modern life Everything evolves so quickly that it cannot be evaluated be fore it takes over When something is found to be considered harmful e g dioxine GO TO nuclear technology asbestos propellent gases it is diffi cult or
13. for PACKLIB users EPIO FFREAD HBOOK4 IOPACK KAPACK KUIP MINUIT ZBOOK ZEBRA containing ZEBRA ZEMQ ZEFQ ZETEST DZEBRA FZCONV RZEBRA RZTEST and from CNL201 on GRAFLIB HIGZ HPLOT5 PAWLIB PAW COMIS SIGMA Compilers Dynamic COMMON The recent versions of CERNlib are produced for use with the IBM VS Fortran compiler only We no longer provide versions for other compilers e g the Siemens FORTRAN 77 compiler For all versions from CNL200 on and for the PRO and NEW versions the compiler option DC PAWC is used This means that this compiler option e MUST be used beginning with versions CNL200 and PRO e MUST NOT be used with older versions If you use one of the newer versions and omit this option your program will not run as expected and produce strange results or abend without any error message during compile or link The same will happen if you use DC PAWC for versions before CNL200 e g the current OLD version Here are examples for Newlib COmpile member CPRM DC PAWC and for JCL EXEC VFORTCLG CPRM DC PAWC Format of Program Library source code CERN plans to distribute the PAM files in a different format beginning with CNL204 Up to now the CERNlib PAM s were distributed in binary format in future they will be distributed in CARD format This means that PAM files at DESY will be in binary format up to version CNL203 and fro
14. impossible to get rid of it So let s have slower progress and take time to get used to new things find out about their merits and their shortcomings or even dangers and maybe even reject them But now back to practice We seem to have ar rived at COMPUTER SCIENTISTS DO IT THE SAME AS THE REST OF THE HEP COMMUNITY What has been practice e Once upon a time CERN initiated an ingenious code manager PATCHY in order to cope with many copies only slightly changed of piles of punched cards CERNLIB centralized much used routines The same would have meant doing nothing and would have saved DD a lot of manpower Meanwhile there have been and are attempts to do better than PATCHY but so far the commu nity has preferred to have it the same e About the same time DESY started NEWLIB in order to do better than TSO or IEBUPDTE Don t worry if these words are unknown to you The same in those days would indeed have meant more punched cards And NEWLIB has stayed the same ever since Excuse me Harald am not denying that it has also become better all the time e But what did the rest of HEP do The same Oh no CERN e g started WYLBUR did bet ter in 1984 with VM CMS and is now talking 19 Do IT BETTER OR DO IT THE SAME 8 DESY Computing Newsletter No 2 8 of doing it even better with UNIX At least they did the same as SLAC but certainly not the same
15. the time of my death it is my intention that the then current versions of TEX and METAFONT be forever left un changed except that the final version numbers to be reported in the banner lines of the programs should become TeX Version 7 and METAFONT Version e respectively From that moment on all bugs will be permanent features As stated on the copyright pages of Volumes B D and E anybody can make use of my programs in whatever way they wish as long as they do not use the names TEX METAFONT or Computer Mod ern In particular any person or group who wants to produce a program superior to mine is free to do so However nobody is allowed to call a system TEX or METAFONT unless that system conforms 100 to my own programs as have specified in the manuals for the TRIP and TRAP tests And nobody is allowed to use the names of the Computer Modern fonts in Volume E for any fonts that do not produce identi cal tfm files This prohibition applies to all people or machines whether appointed by TUG or by any other organization do not intend to delegate the responsibility for maintenance of TEX METAFONT or Computer Modern to anybody else ever Of course do not claim to have found the best solution to every problem simply claim that it is a great advantage to have a fixed point as a building block Improved macro packages can be added on the input side improved device drivers can be added on the out
16. to print files on the printers attached to the IBM or to log on to the IBM Similar operations are also possible between the VXDESY and other IBM computers running the INTERLINK software for example CERNVM Al though it is not yet possible to access VAX files from an IBM session you can log on to the VAX cluster with the help of INTERLINK For more information about INTERLINK and how to use it refer to the following o HELP INTERLINK on VXDESY o HELP COSY on VXDESY o HELP SNA on VXDESY o HELP VT on the IBM o LIST ROLUTL INTERDOC on the IBM The central VAX cluster can also be reached via TCP IP Internet Protocol DATEX P and DEC net as well as locally over the LAT Local Area Transport server The table below gives an overview of these connections Network Where to How to find the find Help VAX cluster s Address DATEX P help P S I help X 25 DECnet help set host nodename is VXDESY LAT Select LAT in show service vxdesy V 24 Switch within LAT then Tutorial TCP IP TELNET ucx show host vxdesy then Help Electronic Mail Since the DESY VAX cluster is a node on both DEC net nodename VXDESY and BlTnet nodename DESYVAX it serves as gateway between the two mail networks BlTnet mail sent from the VAX is forwarded to the IBM by JNET software running on VAX58B This software also provides a SEND FILE command for sending complete files to a remote
17. EWLIB with the com mand QMF and wait until a panel appears The first usage of QMF within a TSO session may require a bit more time before the QMF home panel is shown allocation of required files All subsequent calls of QMF run much faster since most of the required files are then pre allocated 2 Enter the desired print destination for your QMF session By default the actual NEWLIB destina tion is taken If you want to keep it simply press ENTER Otherwise overtype it and then press ENTER the output will not be printed until you leave QMF If you use QMF from a terminal which is not able to display GDDM graphics you will get the message ADM0873 I IF AVAILABLE PLEASE SELECT PCLK OTHERWISE PRESS ENTER Ignore it press ENTER and proceed The only background of this message is to say that you can t produce graphics 3 Printer Setup When you use FINDBOOK for the first time you may have to change your QMF profile to have the correct printer setup If you use the line printers or the IBM 3820 LI1 you ll use the landscape format and nothing has to be changed 15 LocaTING A DESY LIBRARY BOOK BY MEANS OF A DATABASE 8 DESY Computing Newsletter No 2 8 If you use the QMS printers L1 L6 then it is useful to change your QMF profile to use the portrait format it saves a lot of paper This is done as follows When you see the QMF home panel press func tion key PF11 You will then
18. aiming at We would happily accept many more contributions from other people Maybe you have noticed that certain areas are missing or are not covered properly in this newslet ter If you feel that YOUR main area of interest deserves more emphasis then you should probably try to contribute to the newsletter and we will be happy to assist you The article by Don Knuth was not originally writ ten for this Newsletter The original article was printed in TUGBoat 11 No 4 November 1990 We thought that you like us would appreciate the contents and the style of this article Therefore we reprinted it with kind permission of the editors of TUGBoat We do our best to present the hard data cor rectly but already in our first issue we omitted a complete table in the CERNLIB article This table is now contained in this issue s CERNLIB update The opinions presented in named articles are clearly the opinions of the respective authors and should not be treated as the position of the editors or the position of DESY or any organizational unit at DESY The Editors TUGBoat is the publication of the TEX Users Group Editors of the DCN are Katherine Wipf Michael Behrens Jan Hendrik Peters RO1WIP RO1BEH ROINET ext 3222 ext 2556 ext 2583 EDITORIAL DESY Computing Newsletter No 2 8 e How to deal with hanging IBM Terminals by Michael Behrens Many of you have experienced at on
19. ay the DESY logo but instead show the SAMON screens This can be very confusing to the new user who is not familiar with this SAMON is used to logon to other IBM systems directly from a DESY terminal without using the DESY IBM at all The SAMON screens display the word SAMON in the top left corner as well as a list of possible SNA destinations and their status If you simply wish to logon to the DESY IBM either type in TSO and press the ENTER key or return to the usual DESY logo by typing EXIT and pressing ENTER The CERN case On occasion you may find an IBM terminal display ing the CERN logo In this case a previous user prob ably dit not know how to end his session at CERN properly and has left the terminal in this state If you simply type the command VMEXIT in the command line close to the bottom of the screen then this terminal should return to normal state Tf all else fails If none of the previous methods work then the termi nal is probably broken In this case you are requested to report this to our dispatcher Mr Eichbaum ext 3888 You have to find another terminal then and with Help TKAPUTT you will find out all details on how to report the problem How TO DEAL WITH HANGING IBM TERMINALS DESY Computing Newsletter No 2 8 e Upgrade of the DESY Com puting Facilities by Michael Behrens From now until the end of this year you will see an upgrade of the central computing facilities at DESY t
20. bstrings of the title substrings of the edition the signa ture and the ISBN ISSN number as search criteria whereas in HEP a lot of search criteria can be com bined Since the catalog of books is much smaller than the High Energy Physics Index these search cri teria are nevertheless sufficient to make FINDBOOK a useful tool The underlying database for the FINDBOOK ap plication presently contains all books since 1979 and a fraction of the older ones together 85 of the library stocks The missing older books will be inserted in the future new books are added with the actual updates which will take place every few weeks Before new books appear in the database you can find them as described in the online help help BOOKS FINDBOOK runs under IBM s Query Manage ment Facility QMF which is a user interface for DB2 For this reason you must enter the QMF envi ronment which has a larger complexity than FIND BOOK itself Most people who would like to find a book will not be interested in QMF and DB2 Therefore we give a short description below of how to use FINDBOOK and include a few examples If you would like to know more about QMF and DB2 you should also read the last section of this article More detailed information about the environment used Short description of the procedure of locating a book In this section we give a step by step description of the procedure 1 Start the QMF clist from N
21. command EXIT Together with the recursive NEWLIB call you can enter the first NEWLIB command to be ex ecuted For example If you are in QMF and would like to read the online help for FIND BOOK type NEWLIB H FINDBOOK in the command line of the main panel and press enter You will then immediately see this help and can return to QMF with the command EXIT Function key PF12 REPORT If your FINDBOOK search was successful and you then performed other actions with QMF your FINDBOOK report may still be available if it was not overwritten by your other QMF actions You can display it again by pressing PF12 The table editor function key PF8 In principle you can edit every DB2 table with this table editor The table editor is also another way of searching for special data in a table At your own risk maybe you will waste a lot of time you can enter the editor by pressing PF8 then give the table name RO1DB2 BOOKS and try to find out whether you can use it to your advantage This is an application for specialists only For this reason we ll give no further expla nations here In principle the QMF help panels should be sufficient LocaTING A DESY LIBRARY BOOK BY MEANS OF A DATABASE 18 DESY Computing Newsletter No 2 8 e Do it better or do it the same by Otto Hell In a seminar of E Lohrmann N A McCubbin of RAL told us the other day Don t do it better Do it the same have
22. d in subroutine CHANGE it is re ally the value of the parameter passed into K which changes in our example the constant 2 There fore if you need to pass a subroutine a constant value it is a much better practice to set a variable of the right precision to the value you wish to pass and then use this variable in your parameter list Question I m running a FORTRAN program on the IBM in batch which is supposed to read a member of a partitioned data set and write the con tents to unit 6 The program doesn t run and com plains about an illegal floating point character although no s are in the input file Answer The file that the program should read was a member of a NEWLIB source library and thus in COSY format this is a special compressed format used by NEWLIB because it saves space FORTRAN cannot read the COSY format There are two solutions to this problem 1 The whole NEWLIB source library can be con verted into a fixed block data set using the cata logued procedure LIBOUT see HELP LIBOUT for a sample job After the conversion any mem ber of the library can be read by a FORTRAN program and can also be edited in NEWLIB 2 The member to be read can be copied to a fixed block sequential data set First the sequential data set must be allocated using the ALLOC command do not forget to spec ify DSORG gt PS and CARDIM gt X The member can then be copied with the command PUT membname INTO seqd
23. data will not find titles contain ing Data An example When you are looking for all books containing the substrings database or Database in the title author arbitrary type amp AUTHOR and amp TITLE atabase and press ENTER If you would like to find the books which have data or Data and base or Base at separated positions in the title again with arbitrary author use amp AUTHOR amp TITLE ata and amp ALSO TITLE ase For more examples see below Section More ex amples of how to locate a book After a few seconds you ll get a list of all cat alogued books satisfying your search criteria it can be empty For every book the following is displayed authors ISBN ISSN number may be missing The signature is important for finding the phys ical location of the book in the library if you don t know how ask the library staff You can move around in the list using the func tion keys PF7 backward and PF8 forward as well as the commands BACKWARD HALF FORWARD HALF TOP and BOTTOM At the bottom the number of books found is shown The list will be printed if you press function key LocaTING A DESY LIBRARY BOOK BY MEANS OF A DATABASE 16 DESY Computing Newsletter No 2 8 PF4 When you press function key PF3 you ll return to QMF s home panel 6 Now you can leave QMF by pressing PF3 again or you can start a n
24. e job requested a sixth cartridge to continue writing What can do Answer There is a default value for the num ber of volumes cartridges or tapes a tape data set can span This value is 5 To request a larger num ber of volumes for a particular data set you must use the following parameter on your DD statement DD VOL N where N is the required number of volumes and stand for other DD parameters NoTE If possible you should avoid creating multi volume data sets altogether and then you will never need the solution shown above The chance of your data set being lost due to a cartridge or tape error increases with each additional volume it uses There is a DESYlib routine SETEOT which can help you detect the end of a cartridge or tape so that you can start a new data set on a new volume without stop ping your program please consult one of the yellow R1 INFO binders in the terminal pools for more in formation about Editing and Transfering Files Question was editing a new member and found that all the letters added were changed to upper case when pressed enter How can stop this from happening so that can get lower case letters in this member Answer All you need to do is give the com mand ASIS MIX Question am using the EDIT editor for se quential files and every time press ENTER every thing have typed in is translated into capitals How can stop thi
25. e difficult to get involved in the analysis phase of things But we insist on e not doing it the same when we consider a solution poor e trying to do it better when we consider it impor tant If you don t want your computer centers to do it better then you want them not to follow their insight If you want academically trained people at your computer centers then you should not stop them trying to do it better Do IT BETTER OR DO IT THE SAME 20 DESY Computing Newsletter No 2 8 e Archiving inactive Cartridges by Michael Behrens The cartridge robot in the computing center has a total capacity of about 17000 cartridges today Its capacity will increase to about 23000 cartridges by the end of the year after the upgrade mentioned elsewhere in this Newsletter Nevertheless we will not be able to store all car tridges permanently in the cartridge robot This sit uation already arose some time ago when all the cartridges in the robot were full At this time we removed some 6000 inactive cartridges and stored them in an external rack You should not have noticed this action and this should be the case in the future as well Our statis tics show that datasets that have not been used for a longer period will almost certainly stay inactive So we only rarely need the extra manual effort of mounting them to some external drive or of placing them in the robot again Thus for the time being
26. e time or an other that a free terminal in one of the pools which you wanted to use was in a strange state and you were unable to use it You then usually selected one of the other free terminals but sometimes this was the only one available and so you were in trouble Before LOGON the IBM is extremely stupid and ex pects you to give it exactly the input it waits for If somebody fails to do so the terminal will present an error message and from then on it is in some more or less strange state You like many other people do not know how to reset such a terminal to a normal state All as pects of this problem will be covered in full detail in a future paper on how to access the various ma chines at DESY Until this paper is completed you may use one of the following recipes to recover from this situation There are 2 recipes and 2 special cases that have to be considered The standard method After the following sequence of actions the terminal should be in its normal state again and you may initi ate your session simply by following the instructions on the screen e press the RESET button e press the CLEAR button e press the ENTER button e press the RESET button If the terminal still does not show the accus tomed LOGON screen try the next recipe The brute force method e switch the power OFF e switch the power ON again and wait until the DESY logo reappears The SAMON case Some terminals will not displ
27. ecially when a large number of users are logged on The diagram at the end of this article shows the new hardware configuration after the upgrade A Few Statistics When the central VAX cluster was created by ex panding a cluster run by F58 it had roughly 180 users from various groups at DESY Now a little over a year and a half later the number of users has increased to 1067 The table below shows the com position of the VAX user community on the 27th of September 1991 of users from ZEUS 375 H1 225 Hasylab 146 Other groups 321 Total 1067 The number of registered users is still growing at the rate of about 40 to 50 new users a month During the day there are generally 70 to 100 users logged on at the same time and around 120 processes running this includes subprocesses batch processes and users who are logged on more than once During a month between 75 and 80 of the registered user log on and use the VAX cluster Network Connectivity One of the most important features of the central VAX cluster is its close connection to the DESY IBM computer This connection is provided by the INTERLINK software running on both machines INTRODUCING THE CENTRAL VAX CLUSTER AT DESY 8 DESY Computing Newsletter No 2 8 which allows the VXDESY to interact with the IBM as if it were another DECnet node From a session on VXDESY it is possible to copy files to and from the IBM to submit jobs on the IBM
28. ect extremely serious bugs have put these systems into the public domain so that people everywhere can use the ideas freely if they wish have also spent thousands of hours trying to ensure that the systems produce essentially identical results on all computers strongly believe that an unchanging system has great value even though it is axiomatic that any complex system can be improved Therefore believe that it is unwise to make further improvements to the systems called TEX and METAFONT Let us regard these systems as fixed points which should give the same results 100 years from now that they produce today The current version number for TEX is 3 1 and for METAFONT it is 2 7 If corrections are necessary the next versions of TEX will be 3 14 then 3 141 then 3 1415 converging to the ratio of a cir cle s circumference to its diameter for METAFONT the sequence will be 2 71 2 718 converging to the base of natural logarithms intend to be fully responsible for all changes to these systems for the rest of my life will periodically study reports of ap parent bugs and will decide whether changes need to be made Rewards will be paid to the first find ers of any true bugs at my discretion but can no longer afford to double the size of the reward each year Whenever have created a new version will put it in the official master TEX archive which cur rently resides at Stanford University At
29. eneral JCL Job Control Language Linkage Editor and Loader VS2 System Messages Vol 1 VS2 System Messages Vol 2 VS2 System Codes Principles of Operation Program Libraries CERN Program Library Manual ESSL Vector subroutine library for IBM NAG Manual NAG MiniManual FLOP A FORTRAN Language Oriented Parser Graphics G E P Graphical Editor Program Manual GKS P S User s Guide HBOOK User s Guide Text Formatting Die Textformatierungssprache G M L und ihre Verwendung bei DESY Summary of AMS TEX How to use TEX at DESY Using IATpX at DESY Formatting Tables with TEX Symbolic Computing Standard LISP Report Manual for Standard LISP on IBM system 360 and 370 REDUCE 3 User s Manual Other long Write ups B O S Manual CMZ EPIO F S P The Full Screen Package 2 documents GEANT HPLOT KUIP Minuit PATCHY PAW Valley ZEBRA L P L P L P L P L P L U L P L L L P S S L P U S L P U L P L S L help TEXGUIDE on IBM help LOCALGUIDE on IBM S L wn v Cc cccccca a a m a a a a a a a a uo Sn a a ae a ae ae ae ae ae ae 13 MANUALS AND WRITEUPS DESY Computing Newsletter No 2 8 e The Future of TRX and METAFONT Donald E Knuth Reprinted from TUGBoat 11 No 4 November 1990 with kind permission of the editors of TUGBoat My work on developing TEX METAFONT and Com puter Modern has come to an end will make no further changes except to corr
30. equest from the UCO Old invalid documentation is removed from the pools Valid non standard documentation found in the pools is left there We do not yet have a policy for distributing larger amounts of VAX or Unix documentation under the responsibility of the User Support Group We will no longer distribute the CNL Cern Com puter Newsletter in the terminal pools The CNL is not really a reference document for a pool Besides the copies kept at the DESY library many people from DESY get their personal copies already and whoever wants to be added to this list may make a request in the UCO Available printed documents at the DESY computer centre On the next two pages you will find an overview of what literature is available and where More infor mation on what you can find in the terminal pools is in Help LITPOOL on the IBM Besides the literature mentioned below the User Consulting Office UCO has extra copies for several manuals If you have a request please contact the UCO and see if the consultant is able to help you When we expect that very few people will be inter ested in using a manual we will usually stock a few copies in the UCO instead of distributing it to all pools CERN program manuals will in future be im ported from CERN through the DESY User Support Group USG only An arrangement to this effect has been made with CERN If you need any per sonal copies of CERN documentation please address
31. ew book search by typing the command FINDBOOK and pressing ENTER 7 Depending on the applications used the NEWLIB message ISPSTART ENDED AB NORMALLY appears when you leave the clist QMF Don t worry in this case in general nothing terrible happened It simply means that somewhere inside QMF a non zero return code occured which causes NEWLIB to show a warning More examples of how to locate a book Here we give you a list of examples from which you can get an idea of how to use FINDBOOK e Try to determine which conference pro ceedings of the year 1987 are in the li brary To solve this problem you can take advantage of the fact that up to 10 title lines are stored in the database The subtitles contain the required information A suitable search criterion is amp AUTHOR amp TITLE roceedings and amp ALSO_TITLE 1987 since the titles of all conference proceedings start with Proceedings or proceedings due to typing errors and the year of the confer ence is usually part of the title The search terms onferen Conference conference or Konferenz and ymposi Symposium symposium Symposia symposia Symposion etc could also be useful e Try to list all books with High Energy Physics or High energy physics in the title The problem will be solved with amp AUTHOR and amp TITLE igh _nergy _hys
32. ich can t be found with the FIND command These symbols weren t in my PC file and want to remove them How can do this without searching out each symbol by hand Answer These apparent symbols cannot be located with the FIND command because they are place holders for undisplayable characters and not real s Thus the first step in removing them is to find out their hex codes To do this give the com mand LIST member HEX The member will now appear with the hex codes written vertically below each character Now give the commands SET P1 X2C hex code and SET ESC It is now possible to remove or overwrite the symbols with the command cs P1 ALL TOTAL gt to overwrite with blank cs P1 ALL TOTAL gt to remove QUESTIONS AND ANSWERS FROM THE UCO 24
33. ics amp TITLE igh Ynergy hysics would find some additional books e g those containing high energy and then astrophysics in the same title line Try to list the library s holdings pub lished in 1990 and belonging to the cat egory Computer Solution Simply use amp EDITION 1990 amp SIGNATURE C and for all other variables The different categories of books in the DESY library are A Allgemeines General C Computer Computers E Elementarteilchenphysik Beschleuniger oder Quantenfeldtheoriese Particle Physics Accelerators or Quantum Field Theory K Kernphysik Nuclear Physics Konf Konferenzen Conferences L Bibliothekswissenschaften Library Science ask the library staff Mathematik Mathematics Naturwissenschaften Science Physik Physics Quantenmechanik Quantum Mechanics T Technik Engineering Woe Woerterbucher Dictionary ove amp I The signature of a book always starts with the category the year of publication is usually con tained in the specification of the edition Try to find all books with the authors Feynman and Sands Simply use 17 LocaTING A DESY LIBRARY BOOK BY MEANS OF A DATABASE 8 DESY Computing Newsletter No 2 8 amp AUTHOR Feynman amp ALSO_AUTHOR Sands and amp TITLE to get the desired books For an author search it can be useful to
34. inly be used for the export to and import from other sites Exabytes 4 Exabyte drives have been ordered They will be attached to the IBM The new drives will be able to handle new 5 GB per cassette and old 2 5 GB per cassette media These drives are currently under test and will be purchased if they prove to have the capabilities the manufacturer promises Purchase of central number cruncher In order to fulfill the computing power requirements for the HERA experiments DESY has ordered 5 Sil icon Graphics computers 4D 460 with 6 processors and 256 MByte memory each Three computers are intended for Zeus two for H1 The computers will be connected via Ultranet to the IBM All 5 systems have arrived and have already been installed in the computing center UPGRADE OF THE DESY COMPUTING FACILITIES DESY Computing Newsletter No 2 8 e Introducing the Central VAX cluster at DESY by Katherine Wipf One of the biggest problems for scientists who occa sionally visit other institutes is the tiresome task of having to learn a completely new computing environ ment before they can start to work In order to alle viate this situation for scientists visiting DESY the central VAX cluster was installed in February 1990 The cluster runs under the VMS operating system which provides a uniform user interface all over the world and is comparatively easy to learn DECnet a network connecting VAX computers also make
35. m then on in card format What does it mean for the user Users who maintain their own code with PATCHY and access CERNLIB PAM files will have to change the PAM line in their cradle to PAM 11 T CARDS for every PAM file in card format Own PAM files which are still in binary format can be attached in the usual fashion with PAM 12 CMZ users will not be affected because CMZ already uses PAM files in CARD format USAGE OF CERNLIB at DESY 10 DESY Computing Newsletter No 2 8 e Manuals and Writeups by Michael Behrens In the last few weeks the User Support Group has tried to improve the supply of manuals and other documentation needed by the users of the Comput ing Center We tried to identify what written or printed documentation would be needed by our users who should get what documentation and how we could distribute it efficiently and cost effectively Finally we arrived at the following guidelines e The full set of documentation should be available in the DESY library All terminal pools should contain the same stan dard set of documentation e Documentation for a limited user community can be obtained from the UCO it will not be dis tributed to all pools Extra copies of some manuals can be obtained from the UCO on request for group personal use e DESY documentation is available in the self ser vice area of the computing center e CERN documentation is available on r
36. nstalled version at DESY corre sponds to the publication of the CERN Computer Newsletter 201 or short CNL201 There was no formal release of binary PAM files in CNL202 that s why we plan to install CNL203 as the next version at DESY which is expected to be distributed during the next weeks Current state of CERNLIBs on the IBM NOTE You should use the command help CERNLIB under NEWLIB to verify the current state of the CERNLIBs on the IBM because by the time this in formation is printed it might already be obsolete CNL LINE OLD PRO NEW Line from the CNL Line 3 generations are maintained Every set of libraries corresponds to a CNL version r lutl cern suffix rO1utl cern OLD suffiz rO1utl cern cnl200 lib suffiz r lutl cern PRO suffiz rO1utl cern cnl201 lib suffiz rOlutl cern NEW suffiz Version of November 1989 Up to and including the version CNL200 DESY had supplied only KERNLIB GENLIB and PACK LIB Starting with CNL201 GRAFLIB and PAWLIB are supplied as well The contents of the various libraries is as in the following table By mistake it was not complete in the DESY Computing Newsletter No 1 USAGE OF CERNLIB at DESY 8 DESY Computing Newsletter No 2 suffiz containing KERNLIB KERNIBM KERNFOR KERNBIT UMON since CNL201 KERNNUM GENLIB GENLIB PACKLIB KERNLIB is included here to save one extra concatenation
37. o prepare for the HERA experiments This comprises the IBM the VAX cluster and the number crunchers for the experiments Details on the VAX upgrade are contained in a separate article in this issue and are omitted here Upgrade of the IBM filestore Early this summer the processing power of the cen tral IBM was doubled to cope with the increasing demands together with a moderate increase in disk space Now the disk space for the IBM will be in creased by another 120 GB by the end of this year The new disks will be of the new 3390 type with shorter access times and a higher transfer rate than the 3380 disks that are already installed at the com puting center Details on track capacity etc can be found in Help Capacity on the IBM Despite the greater track length for the new disks R1 does not propose changes for the DCBs currently used Most DCBs in current use will give satisfactory utilization of the new disks Cartridge Robot Upgrade A fourth library unit for the ACS cassette robot has been ordered as well and it will probably be in stalled between Christmas and New Year This up grade will extend the capacity of the cassette robot system to about 23000 cassettes This unit will come with 8 drives as usual Besides this 4 extra units for the robot system and 4 additional external cartridge units have been ordered This will then give a total of 36 drives for the robot system plus 8 external drives The external drives will ma
38. on in June 90 In particular users were advised to put their private graphic applications on a GKS basis in preparation for the new system It was announced that IPS and DESYNET would be supported officially only up to the end of 1990 since it was clear that the old hardware would not last forever An internal R2 paper of January 91 described a schedule to replace the old IPS terminals step by step by HP Apollo workstations HP9000 425t for which all necessary software had already been pre pared Work was in progress to eventually add X terminals to the workstations to allow for more than one user per workstation After the purchase and installation of a number of workstations mainly for use at H1 we could not get more of them to make the next steps for two reasons 1 shortage of money 2 the appearance of new and more powerful workstations with a very attractive price performance ratio This implied a modification of our concept Con nect a large number of X terminals to only relatively few workstations Unfortunately this requires addi tional work The situation with respect to IPS on the main frame after the final breakdown of DESYNET on September 19 1991 is the following e The old IPS terminals Tektronix 619 and Con rac are out of operation and will be removed e Ataris and Falcos can access the IBM mainframe via MICOM destination IBM and Falco termi nals in addition with the MICOM destination LAT
39. put side welcome continued research that will lead to alternative systems that can typeset documents better than TEX is able to do But the authors of such systems must think of another name That is all ask after devoting a substantial por tion of my life to the creation of these systems and making them available to everybody in the world sincerely hope that the members of TUG will help me to enforce these wishes by putting severe pres sure on any person or group who produces any in compatible system and calls it TEX or METAFONT or Computer Modern no matter how slight the in compatibility might seem Donald E Knuth Department of Computer Science Stanford University Stanford CA 94305 THE FUTURE OF TEX AND METAFONT 14 DESY Computing Newsletter No 2 8 e Locating a DESY Library Book by means of a Database by Peter Dobberstein General By means of IBM s relational database system DB2 we have created an application to locate a DESY library book if author or title are known This only applies to the BOOKS NOT to the PREPRINTs If you would like to find a preprint please use the High Energy Physics Index HEPI or HEP on the IBM For information about the High Energy Physics Index ask in the library or use the online help help HEP on the IBM mainframe Compared with HEP the FINDBOOK applica tion described here is very simple You can use two substrings of the list of authors and two su
40. puting Newsletter No 2 8 e DESYNET 1979 f Septem ber 19 1991 by Peter Schilling After over twelve years of successful operation DESYNET the packet switching local area network created by R2 has passed away due to unrecoverable hardware failure DESYNET was an important part of the In teractive Plotting System IPS because it served to connect the IPS terminals equipped with Tek tronix 619 storage tube and Conrac raster displays used for interactive graphic work to the IBM main frame Since DESYNET was also connected to the V 24 switch MICOM it was possible to use Atari PCs with R2 software as low cost graphic terminals with full screen capability too In 87 R2 started to develop a new generation of graphics for the growing demands at DESY Prices for powerful workstations had come down so much that it was no longer reasonable to extend the old IPS further The concept was based on workstations using standards like UNIX TCP IP X and GKS The main features of the new system called IPS 90 are e support of the old IPS applications i e full screen and interactive graphics on the IBM mainframe e in addition the possibility to run applications lo cally on the workstations which could not be done on the old dumb IPS terminals high connectivity of the new stations via TCP IP After successful prototyping R2 presented IPS 90 to the public at a CUC Computer Users Commitee sessi
41. s Answer There is a command CAPS in this ed itor which switches the translation to capital letters on and off CAPS ON gt _ All entered text is transposed to capital letters CAPS OFF gt The text on the screen remains as entered Question What is the best way to copy my VAX files onto the IBM Answer The best method is to use INTER LINK see HELP INTERLINK because the neces sary commands are very similar to those you would use for copying files over DECnet It is also possible to copy NEWLIB source files via INTERLINK this is not possible using other methods with the new COSY command see HELP COSY However you must have accounts on both the VXDESY and the DESY IBM in order to use INTERLINK The IBM userid must be defined under RACF Resource Ac cess Control Facility a standard IBM security pack age and you must have a proxy definition linking the two accounts Please consult the RACF administra tor of your group about RACF and proxy definition If you cannot or do not wish to meet the re quirements for using INTERLINK it is still possible 23 QUESTIONS AND ANSWERS FROM THE UCO DESY Computing Newsletter No 2 8 to transfer files with the help of FTP give the com mand FTP on the VAX and then FTP gt HELP or HELP FTP on the IBM Question have just transfered a file from a PC diskette to the IBM and now find that it contains unusual symbols which look like but wh
42. s it possible for a guest physicist to work comfortably on the VAX system of his home institute The Hardware of the Central VAX Cluster The Current Configuration A VAX cluster consists of a number of central pro cessing units CPU s all sharing the same disks The cluster can be accessed via network by its cluster alias for our cluster this name is VXDESY and each CPU can be addressed separately by its own node name Because all the disks are shared between the computers a user only needs to be registered once in order to use any of the computers in the cluster At present the central VAX cluster at DESY consists of the following computers and workstations sharing 18 Gigabytes of disk space 3 Computers VXDSYA VAX 6410 64 Mbyte Memory VXDSYB_ VAX 6410 64 Mbyte Memory VAX58B VAX 8350 32 Mbyte Memory 4 Workstations VSF58A_ V lt AXstation 3100 VSF58B VAXstation 3100 VSF58C_ VAXstation 3100 VSR02A VAXstation 3100 SPX Future Expansion The VAX cluster is going to be enlarged at the end of this year by the addition of a VAX 9110 equipped with a 128 MByte memory a second HSC Hierar chical Storage Controller controls and coordinates disk access and 5 Gbytes of disk space This up grade increases the CPU power from 14 VUPs VAX Unit of Performance to 54 VUPs The new HSC will reduce disk access times and improve reliability The users should notice an improvement in perfor mance and response times esp
43. s name SEQ Note File transfer programs like FTP and INTERLINK also have trouble with NEWLIB COSY data sets For INTERLINK there is a new command COSY on the VXDESY which solves the problem NEWLIB members which are to be transfered via FTP must be treated according to solution 2 above Question Can use the FORTRAN OPEN statement to open an IBM file using the dsname rather than a ddname Answer Yes You can indicate the dsname in the FILE parameter of the OPEN statement For the VS FORTRAN compiler it is necessary to put a in front of the name to indicate that it is a dsname It would look something like this OPEN UNIT 2 FILE userid a b about Tapes and Cartridges Question My FAST job ended with a system completion code 222 and my data sets were not copied to disk The data sets were on archive tapes and submitted a RETRA job for these data sets This job claimed the data sets had been ADDED so assumed that they were now available What went wrong Answer The FAST job abended because the tapes were not yet in the machine room The word ADDED on a RETRA job only means that the tape has been added to the list of tapes to be retrieved from the archives The dispatcher fetches tapes from the archives at the following times Mon Thur 9 00 and 14 00 Fri 9 00 and 13 30 To be sure that your tapes have arrived in the ma chine room you should not start your jobs using
44. see a panel for setting up your profile Overtype the printer WIDTH with 80 default 132 and the printer LENGTH with 87 default 60 and press func tion key PF2 to save the values DON T IN SERT anything in the empty field PRINTER After pressing function key PF3 you ll return the QMF home panel and you can proceed When you see the QMF home panel type in the command FINDBOOK and press ENTER 5 You will now see a panel where values for the variables amp AUTHOR amp ALSO AUTHOR amp TITLE amp ALSO TITLE amp EDITION amp SIGNATURE and amp ISBN are requested All variables other than amp AU THOR and amp TITLE are preset to the string Now give substrings of the list of authors and the title obeying the following rules and hints a The strings have to be enclosed in apostro phes b percent is a substitution for an arbi trary number of unknown characters c _ underscore is a substitution for exactly one unknown character d both amp AUTHOR and amp TITLE have to be entered e if the variables amp ALSO AUTHOR and amp ALSO_TITLE are overtyped they will be used as additional restrictions logical AND with amp AUTHOR amp TITLE and the other variables f SEDITION amp SIGNATURE and amp ISBN are also taken as additional restrictions if over typed logical AND with the other vari ables g the database is case sensitive E g a title search for
45. termi nate the desired name with a comma For ex ample AUTHOR Schulz will find only the books with author Schulz not those with authors Schulze or Schulz Weidenbaum etc without typing errors in the database all names should be terminated with a comma e Try to display the library s holdings published by Springer Solution Simply use amp EDITION Springer and for all the other vari ables e Try to display all books contained in the catalog which is not recommended Use amp AUTHOR and amp TITLE which means that author and title strings are ar bitrary The generated list is so huge that your TSO session could crash More detailed information about the environment used As mentioned above FINDBOOK uses the IBM products DB2 and QMF If you would like to have more information about the environment also read the corresponding online helps help DB2 and help QMF the help panels in QMF or the available lit erature which is present in the UCO the titles are listed in help DB2 Of special interest for the FINDBOOK user could be e Recursive NEWLIB call When you are in QMF you can write NEWLIB instead of FINDBOOK in the command line of the main panel You will then enter a recursive NEWLIB session That s often easier than leav ing QMF with a subsequent re loading of QMF A recursive NEWLIB session will be left with the NEWLIB
Download Pdf Manuals
Related Search