Home

HP SureStore E Disk Array 12H System Administrator`s

1. Additional notes e problems detected by the DELETLUN command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 170 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions DESCRIBE The DESCRIBE command allows the user to display the configuration and warning states of the currently selected array Information displayed includes replaceable FRUs and their states subsystem and physical drive data and LUN configurations Synopsis DESCRIBE Output ARDIAG DESCRIBE Configuration information for the controller in slot X Component Instance State Fan EL Good Fan F2 Good Fan F3 Good Power Supply P1 Good Power Supply P2 Good Power Supply P3 Good Controller X Good Controller Y Good Controller X Battery 1 Good Controller X Battery 2 Good Controller Y Battery 1 Good Controller Y Battery 2 Good Memory Instance Size State Controller X DRAM 1 32MB Good Controller X NVRAM 1 32MB Good Controller X NVRAM 2 32MB Good Controller Y DRAM 1 32MB Good Controller Y NVAM 1 32MB Good Controller Y NVRAM 2 32MB Good Total Memory 192MB Test Level FRU Slot State Warning Indications 0 CNTRL Y Warning Redundancy Loss 1 MECH Al Included 2 Bl Downed 3 MECH A2 Failed 4 MECH B2 Included 5 MECH A3 Included 171 Using the ARDIAG Offline Diagnostic MPE Opera
2. DELETELUN The DELETELUN command allows the user to delete a LUN from the specified disk array Synopsis DELETELUN Output ARDIAG DELETELUN Enter the number of the LUN to delete default 0 Ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck ck Ck kk Ck Ck ck kk Sk ck ck ck ck kk kk kk ko ko ko ko kk DATA LOSS WARNING x kk ck Ck ck ck Ck ck ck kk Ck Sk ck Ck ck kk ck kk ck kk kk KKK ck kk ck kk ko kk kk Sk kk Sk kk ck kk ko kc k ko KKK KKK Continuing with DELETELUN WILL DEFINITELY RESULT IN DATA LOSS 11 information stored on the LUN to be deleted will be lost and the physical storage previously used will be freed Dk ck ck ck Ck ck Ck ck Ck ck ck Ck ck ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck kk ck kk ck ko ck ko Sk ck Sk ko ko ko Lun 2 selected Delete selected lun default n ly Lun 2 was successfully deleted Additional notes e problems detected by the DELETLUN command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 78 DESCRIBE The DESCRIBE command allows the user to display the configuration and warning states of the currently selected array Information displayed includes replaceable FRUs and their states subsystem and physical drive data and LUN configurations Using the ARDIAG Offline Diagnosti
3. the disk array Yes arraymgr Yes Yertremar Creating a disk array alias Yes arraymgr Yes arrayrecover Yes 117 Managing the HP SureStore E Disk Array 12H on MPE Installing the Disk Array Management Software Installing the Disk Array Management Software The AutoRAID Manager ARM disk array management software is distributed on the World Wide Web NOTE At the time of printing the following ARMServer patches were available The latest patches are available from the HP Patch web site www hp com go support which will indicate if the above patches have been superceded Bi Operating System Support The ARMServer software is supported on MPE iX 6 0 C 60 00 and requires the following patches MPEKXU3 e ARMKXWS5 118 Managing the HP SureStore E Disk Array 12H on MPE Tips for Configuring the Disk Array On MPE Tips for Configuring the Disk Array On MPE Before installing a new disk array you should determine what is more important for your operating environment performance or capacity This decision will influence how you configure the disk array hardware There are several configuration options that impact the performance of the disk array Table 15 identifies the various configuration options their impact on disk array performance and any considerations regarding disk array capacity NOTE Overall system performance is a complex issue
4. 155 Simplified Resiliency nt n rei oett ee eoe een ete edel 161 CHAPTER 7 USING THE ARDIAG OFFLINE DIAGNOSTIC ON MPE 165 OPERATING ENVIRONMENT etie n RR RN RR 165 Support Software cini Re Oi de neto gere etx ERU ERN 165 Minimum hardware ene Sede a Bia tete cete this Haven cute e et ee be eS 165 Minimum SOftwalte RIBERA e IERI RRR 165 DEFINITION OF 8 2 12 4 enhn 166 ARDIAG OPERATIONAL COMMANDS 1 1224400000002110010000000000000000000 0 167 ARDIAG INTERFACE COMMANDS assesses setate assesses ette tensa sess estesa naa 168 OPERATIONAL COMMAND 8 0 000111 169 DIVI eite i e 169 H D A E EI E M 170 DESCRIBE ns Ee oie tecnica net a c 171 DOWNELEOAD aset eet e IA OOS 173 BORNDAT etel steer Rue pei 177 INQUIRY RET 179 RBADLOQG 5 iere b reed m bertus rir 180 RECOVER OU ERREUR EIE 184 REOSENSE inocente NI Uit DL ie eee ec a 186 E 187
5. Contents nenne cene trepida eet t Rte 257 Displaying Hardware dto RE edu n reete pride i cre icr e E Oed 257 Formatting the Disk Array 258 Changing SCSES t ngs ei eee e eH EI MR E REIR UU RUE ere 259 11 Changing the Controller SCSI ID 259 Switching Primary Controllers eese eere enne ener tene nren tren nest 259 DOWNLOADING FIRMWARE hr IP REE A NERA I sue 260 Downloading Firmware to a Disk 260 Downloading Firmware to the Disk Array Controllers 261 Copying Firmware From the Primary Controller to the Secondary Controller 262 VIEWING THE DISK ARRAY GENERAL CONFIGURATION SETTINGS 263 Simplified Resiliency Setting iion nr eee tend ede eu e ere e ir re needs 269 Setting EE 272 Creating Disk Array A lias 2 erede tene e RE Eee MI UH EI oe CER eee 273 1 9 2 275 12 Chapter 1 Managing the HP SureStore E Disk Array 12H on HP UX This chapter introduces the tools available for managing your HP SureStore E Disk Array 12H on HP UX These tools provide complete control over all aspects of disk array operation The following information is included in this chapter e brief description of each management tool e
6. 13 DISK ARRAY MANAGEMENT 18 1 2 2 2 0 100 0000000000000000000000000000000000000 0 14 System Administration Manager 5 22 20 0 02 0 00000000000000000000 14 AutoRAID Management ARM 14 Diskartay Control panel sei tre RR 14 WHICH TOOLS TO USE FOR EACH TASK or Snesere heeri nnn asset nette nnns seen 15 INSTALLING THE DISK ARRAY MANAGEMENT SOFTWARE esee ener enean nsns enne 16 Operating System SUpport rer ere ris een eee nennen trennen nns 16 IPR Program ien e nep CDD DOT n DEOR HERO er DER sages RP 16 TIPS FOR CONFIGURING THE DISK ARRAY ON 7 2 2 0 000000000000 17 CHAPTER 2 USING THE HP UX SYSTEM ADMINISTRATION MANAGER 19 SELECTING HARDWARE PATHS ON HP UX 10 20 20 CONFIGURING A NEW DISK ARRAY ccsessssecececsesssaececececcsessaaececececeesesecesececseneseaeceeccseseaaaeseeeceesenssaeeeeees 23 CHECKING DISK ARRAY STATUS ccccesssssscecececsenseaececeececeeseseaececececseseaececececsessaaeceeceeesesaaeceeeeseseneaaeeeeess 24 CHANGING DISK ARRAY CONFIGURATION SETTINGS cccsessssecccecsesssecececcecsenenseseceeeceesenseaeeeeseeesenteaeeeeees 24 MANAGING LUNS LOGICAL DRIVE
7. Information as an int 0 0x0 Command Specific Information field bytes 0x0 0 0 0 0 0 0 Command Specific Info as an int 0 0x0 Additional sense code 0 0x0 and qualifier 0 0x0 Translates to No additional sense information Field replaceable unit code 0 0x0 Sense Key Specific field is NOT valid Sense key specific field 0x0 0 0 0 0 odule Identifier is 0 0x0 Error Number is 0 0x0 186 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions RESTART The RESTART command allows the user to bring the specified disk array out of the Shutdown state Synopsis RESTART Output ARDIAG RESTART This command will issue a hard reset to the array controller s Ready to restart the array default n The array is starting up Please wait while the array performs a reset A two minute time out is imposed here but 90 second reset times are mor typical Array has been successfully restarted Additional notes e problems detected by the RESTART command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 187 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions REVISION The REVISION command allows the user to display the internal drive s firmware revisions on the specified SureStore E AutoRAID Disk Array Syn
8. Number Display Option ck ckck kckck T Event Typ ables 2 Chronological 3 Both Event Type Tables amp Chronological Not Event Type Tables and Chronological are the same information they are just displayed differently default 3 Both 1 Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions DRIVE ERROR EVENT TABLE ck ckck ck ck ckck ck ck ckck ck ck ckck ck ck ck k ck k ck k ck k ck k ck k ck k ck k ck k ck k ck k ck KKK KKK Time Module Event Event Slot Sense ASC ASCQ LBA Stamp ID Code Count Key Dk ck ck ck Ck lt 28279 0x4 0 0 Ox1f 30 A2 0x5 0x24 0x0 N A CONTROLLER ERROR EVENT TABLE kk ck Ck ck ck Ck ck ck ck ck ck Ck ck Ck ck ck Ck ck ck Ck KKK KKK KKK ck kk ck kk ck kk ck kk ck kk kk ck kk ck kk Sk kk Sk kk Sk kc KKK KKK Time Module Event Event Component FRU Abterm Stamp ID Code Count Specific ID Code Dk ck ck ck ck ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck cock Ck ck ck ck ck kc ck ck ck ck ck ck ck ck ck ck ck koc ck kk ck ko ck kk ck Pk Sk Mk kx kv Mk ko ko 30367 0x40 0x0 0x87 1 0 7 0x81 NORMAL SYSTEM CHANGE EVENT TABLE kk k
9. 5 en aded 188 rne aae ce ERE 189 iate tabs dis 191 SHUTDOWN teet eee 194 MRM T uM 195 INTERFACE COMMAND 8 22 220000000 00100100000000000000000 0 0000 197 DISPNI AD 197 RI aba N 198 enn ie unie 199 SHOWENWY spectet uiii dn 200 TES PIS Kis ee AE tir 201 C u 202 PA S e vae ime os cash yeh dae etaed tae uiis 203 ARDIAG INTERFACE TO etr er eere R 205 CHAPTER 8 MANAGING THE HP SURESTORE E DISK ARRAY 12H ON WINDOWS 207 AUTORAID MANAGER COMPONENTS assai seen eite stein nasse 208 HP OPENVIEW INTEGRATION cech reas sese esee E E anas sees setate nans nen 209 Disk Artay Vents eere et ie UR Rem ster bet 209 SYSTEM REOUIREMENTS ete t
10. Disable NVRAM on UPS absent Force Unit Access Response Disable Read Hits SingleController Mode Parameter Warning Lock Write Cache On Disable NVRAM on WCE False Disable NVRAM with One Disable NVRAM on UPS absent Force Unit Access Response Disable Read Hits Resiliency Threshold Resiliency Threshold Single Controller Default Setting ENABLED 700 Series Workstation Simplified Resiliency Setting Results 800 Series Multi Purpose Single Dual Single Dual Controller Controller Controller Controller FALSE Not Data map Not Data map Supported nd write Supported nd write TRUE ache ache nformation nformation FALSE s scheduled s scheduled lo flush to lo flush to 2 he disk he disk very four very four FALSE econds econds 4 Setting Default AB 700 Series Workstation FALSE Single Dual Single Dual Controller Controller Controller Controller TRUE Data map ot Data map ot and write Supported write upported TRUE cache cache information information FALSE is scheduled is scheduled to flush to to flush to 2 the disk the disk once every once every FALSE second 55908 71 XN dH Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration
11. 118 Operating System Support 118 TIPS FOR CONFIGURING THE DISK ARRAY ON 0022 2 0 001000000000 119 CONFIGURING AUTORAID SCSI LUNS FOR cccccccccccecsesssecececeeceesenseaeseeeceesenssaseeeeeeesenseaes 121 Configuration Requirements oie ere diat ten t e Heec e eie ie eee te e e beers 121 SYSGEN Configuration ennemi E K enne trennt tenete tenerent 122 Device Adapter Configuration eessessssseeseeeseeee nennen nennen nne trennen 123 SCSI Target Configuration tier erede 123 LDEV SCSI LUN Configuration sees eene nennen nenne ener 124 CHAPTER 6 USING THE ARM COMMAND LINE UTILITIES FOR MPE eee 125 SETIINGUP MPE POSIX SHEL Deore hree eera tete a ie Pea evi ete been ene 125 INFORMATION N MAN PAGES er arae eite e 126 COMMAND SYNTAX CONVENTIONS I HERE a ERTES aE vs suee 126 THEARMSERVER PROCESS ten i er i i ipee te to eere eren enin 127 SELECTING A DISK ARRAY enne nnn 128 CONFIGURING NEW DISK 129 CHECKING DISK ARRAY STATUS cis cose IARE ER TREE 131 DISPLAYING DISK ARRAY S
12. c d slot Clear the specified logs If c is specified clear the disk array controller usage and event logs Both logs will be cleared when using this option It is not possible to clear just one of the logs If d is specified clear the log for the disk installed in the cabinet slot identified by slot To display the contents of a hardware log type arraylog u e d slot array id To clear the hardware logs type arraylog C cl d slot array id Formatting the Disk Array Should it become necessary to do so you can format the entire disk array or a single LUN Formatting destroys all data on the array or LUN involved Formatting an array first requires that all LUNs be deleted To format the entire disk array type arrayfmt F h array id To format a LUN type arrayfmt L LUN lt array id gt 146 Using the ARM Command Line Utilities for MPE Performing Disk Array Maintenance Tasks Changing SCSI Settings The SCSI settings control the transfer of information over the SCSI channel connecting the host and the disk array The default SCSI settings listed in Table 21 have been chosen to work with all supported SCSI adapters and in most cases should not be changed NOTE Before changing any SCSI setting you should understand what the SCSI setting does and what effect changing it will have on disk array operation Be aware that using an incorrect SCSI setting may make it i
13. eeke 224 CHANGING DISK ARRAY CONFIGURATION SETTINGS cscsscscccceceesessscecececceesesseaeeeeeceesesssaececeeseeensaaeeeeess 225 MANAGING LOGICAL DRIVES retten retener eU ere ree Pre ero eese Enea ves 226 Creating Logical Drive nitore ripeto arn ete eir DERE HRS 226 Renuimbering a Logical Diyes ioien e ier eere i ac PRO mere ipe vt eee 227 Deleting Logical tete ehe ive eee dete 227 ADDING A entere heec tte sleet vel vcs tibiae deve eet EE ET ERI ees 228 Including Disk Manually neo tante ee eo tete PRU De DH nre Dp eeu 229 REBUT DNG TAE DISK eher e ciere ehe ve eee P 230 Rebuilding the Disk Array Automatically 230 Rebuilding the Disk Array Manually 231 Checking the Progress of a Rebuild esisiini triir sesi resore r ie r r E 231 Canceling a Rebuild i eR A E 232 ANALYZING DISK ARRAY PERFORMANCE ccessessssccececeesenaecececeeceesesaaeceeececsesaaeceeeceeseseaeseeeeeesensaaeaeeess 232 Selecting a Time Period for 1 8 232 Displaying Performance Metric Thresholds essent nennen 233 PERFORMING DISK ARRAY MAINTENANCE TASKS cccssessscecececsesssaececeececeeseaeeecceeceesenssaeeeeeceesentaaeeeeecs 234 Shutting Down the Disk Array 234 Resta
14. p Begin the recovery in non interactive mode If there are multiple volume 5 v volume set ID Starts a recovery on the volume set indicated by volume set id This option is intended for use only in environments where there are multiple volume sets on the disk array 63 XN dH Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Viewing the Disk Array General Configuration Settings Checking the general configuration settings allows you to view the current settings for all the disk array operating parameters The settings include the current status of the array the array configuration and the SCSI configuration settings A description of each setting is included in Table 9 The default settings have been selected to optimize disk array operation on HP UX NOTE Before changing any setting you should understand what the setting does and what effect changing it will have on disk array operation Be aware that using an incorrect setting may make it impossible for the host and the disk array to communicate properly To view the general configuration settings type arraydsp s array id A complete list of the settings will be displayed Table 9 General Configuration Settings Default Description Setting Overall State of Array READY Specifies the current state of the array ENABLED Indicates whether the array should reserve space withi
15. 149 Using the ARM Command Line Utilities for Performing Disk Array Maintenance Tasks Creating a Disk Array Alias An alias can be created to identify the disk array The alias provides an alternative to the disk array serial number and raw device file name that can also be used to identify the array Aliasing can be used in a variety of ways to help identify disk arrays in large systems For example by assigning numbers to racks and to the shelf positions within the racks each disk array can be uniquely identified using an appropriate alias If a rack is assigned number 12 the disk array installed on shelf 3 of the rack could be identified using an alias of R12 S03 This technique simplifies locating the disk array should it need service To create a disk array alias type arraymgr D alias name array id alias name can be up to 12 characters in length and can include upper case letters numbers spaces number sign underscore _ and period Recover If the disk array is not shutdown properly it is possible that the data maps in NVRAM memory will be lost For this reason the disk array allows the data maps to be periodically written to the disk drives If the maps are lost an error code such as Address Table will appear on the display If your disk array is a boot device you may have to recover the maps by using the front panel command called Recover under the Cntrl Change
16. HP UX Operational Command Descriptions RESTART The RESTART command allows the user to bring the specified disk array out of the Shutdown state Synopsis RESTART Output ARDIAG RESTART This command will issue a hard reset to the array controller s Ready to restart the array default n The array is starting up Please wait while the array performs a reset A two minute time out is imposed here but 90 second reset times are mor typical Array has been successfully restarted Additional notes e problems detected by the RESTART command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 95 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions REVISION The REVISION command allows the user to display the internal drive s firmware revisions on the specified SureStore E AutoRAID Disk Array Synopsis REVISION Output ARDIAG REVISION Test Level FRU Slot Vendor ID Product ID Rev 0 CNTRL Y HP Arrays R Us YPR1 1 ECH A1 HP 1 050 GB 3rd 0256 2 B1 HP 2 13 GB 2nd 0256 3 ECH A2 HP 1 050 GB 3rd 0256 4 2 2 13 GB 2nd 0256 5 ECH A3 HP 1 050 GB 3rd 0256 6 B3 HP 2 13 GB 2nd 0256 7 ECH A4 HP 1 050 GB 3rd 0256 8 ECH B4 HP 2 13 GB 2nd 0256 ARDIAG gt 96 ROMT Using the ARDIAG Offline Diagnostic on HP UX Operational Command D
17. There are several tasks that you may have to perform in the ongoing management of the disk array These maintenance tasks are typically performed infrequently and may involve taking the disk array off line Shutting Down the Disk Array CAUTION When an array shutdown is performed the disk array becomes unavailable to the host system An array that is shutdown appears to the operating system as if its power has been turned off The disk array must be Shutdown prior to performing any maintenance The Shutdown process copies vital data mapping information from the controller NVRAM to the disks This protects the data mapping information should the contents of the NVRAM be lost or corrupted due to battery failure Shutdown then takes the disk array off line making all data unavailable to the host The disk array can still be managed and tested but all data is inaccessible while the disk array is Shutdown Shutdown is initiated automatically each time the disk array is turned off using the power switch so it is usually not necessary to initiate a Shutdown using the ARM utility To Shutdown the disk array type arraymgr s shut array id Restarting the Disk Array Following Shutdown the disk array can be brought back on line by performing a restart This makes the data on the disk array available to the host once again To restart the disk array type arraymgr s start lt array id gt 254 Using the ARM Command Line Utilities
18. x00 0x00 0x00 0x04 0x80 0x02 DEUS C 1 24 0x80 0x02 32 0x00 0x00 Page Code 3e Addr Hexadecimal ASCII 0 0x0 Ox3e 0x00 0x00 0x08 0x00 0x00 0x00 0x04 SERIE S dou 8 0x8 0x00 0x00 0x97 0 1 91 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions RECOVER The RECOVER command allows the user to restore controller map information in the specified SureStore E AutoRAID Disk Array Synopsis RECOVER Output ARDIAG gt testdisk 1 Array state is No Address Table The following warnings are currently in effect Disk Warning Capacity Warning ARDIAG gt testlevel 0 ESTLEVEL 0 ARDIAG recover y 1 he logs should be cleared befor xecuting this command Do you want to abort to clear logs y n no here is only one volume set in the array 0 125 0 Disks Al 2 Proceed with recover operation y n yes Recovery Progress 1 Percent Recovery Progress 2 Percent Recovery Progress 3 Percent 92 Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Recovery Progress 97 Percent Recovery Progress 98 Percent Recovery Progress 99 Percent Wait This will take several minutes Attempting to check completion status of the recover operation The array state has changed The previous state was No Address Table The new state is
19. 259 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Downloading Firmware Downloading Firmware The download command copies new firmware code to the disk array controller s or individual disk drives in the disk array Firmware can also be copied from a primary array controller to a secondary controller CAUTION In multi host configurations other hosts must not access the disk array while the download is in progress Data can be lost if write requests are made to the disk array while a download is in progress Downloading Firmware to a Disk Module This procedure updates the firmware code on the specified disk module To download new firmware code to a disk module 1 Gather the following information e The ID of the disk array The disk array ID can be displayed using the arraydsp i command e The location and name of the file containing the new firmware code slot number of the disk module Slot numbers are marked on the cabinet and B6 2 Froma command prompt type in the following command download D slot codefilename array id slot identifies the cabinet shelf containing the disk drive codefilename identifies the file containing the firmware code 260 Using the ARM Command Line Utilities for Windows NT Downloading Firmware Downloading Firmware to the Disk Array Controllers This procedure updates the firmware on both controllers simu
20. MPE 116 Recover 58 150 274 Windows NT 207 controller switching HP UX using ARM 56 HP UX using SAM 32 MPE using ARM 148 NT using ARM 260 NT using AutoRAID Manager for Windows 238 copying firmware HP UX using ARM 59 MPE using ARM 151 NT using ARM 263 creating a logical drive NT using ARM 247 NT using AutoRAID Manager for Windows 226 creating a LUN HP UX using ARM 42 HP UX using SAM 26 MPE using ARM 134 creating an alias HP UX using ARM 58 274 MPE using ARM 150 D daemon array monitor 35 data recovery map 58 150 274 data resiliency settings HP UX using ARM 56 273 MPE using ARM 148 deleting a logical drive NT using ARM 248 NT using AutoRAID Manager for Windows 227 deleting a LUN 276 HP UX using ARM 43 HP UX using SAM 27 using ARM 135 Detail pane 215 diagnostics ARDIAG 73 165 disk array events on NT 209 disk array maintenance HP UX using ARM 50 MPE using ARM 142 NT using ARM 254 NT using AutoRAID Manager for Windows 234 disk array status HP UX using ARM 39 HP UX using SAM 24 MPE using ARM 131 NT using ARM 244 NT using AutoRAID Manager for Windows 224 disk testing HP UX using ARM 52 MPE using ARM 144 NT using ARM 256 NT using AutoRAID Manager for Windows 237 displaying disk array serial numbers HP UX using ARM 40 MPE using ARM 132 NT using ARM 245 downing a disk HP UX using ARM 51 HP UX using SAM 30
21. arraymgr R array id Downing Excluding a Disk Downing or excluding a disk is typically done in preparation for testing the disk After the disk is downed testing can be done without impacting disk array operation If testing reveals that the disk is good the disk can be included back in the array configuration Downing a disk has the same effect as if the disk failed or was physically removed from the cabinet If Auto Rebuild is enabled the disk array will immediately begin a Rebuild when a disk is downed The v option identifies the down operation as either destructive or nondestructive This determines whether the disk array will assume there is any valid data on the disk if it is returned to the array configuration If a destructive down is performed not specified the disk array will assume no data on the disk is valid If the down is nondestructive v specified the array will assume any data on the disk that was not updated in the disk s absence is valid To down a disk type arraycfg D slot d v R 2 array id To protect data availability the disk array will not let you down a disk if doing so would result in loss of data redundancy or data unavailability However you can override this protection by specifying the appropriate ARM options The R option allows the disk to be downed even if a loss of redundancy would result but not data unavailability The Z option allows the disk to be downed ev
22. 46 HP UX using SAM 31 MPE using ARM 138 NT using ARM 251 NT using AutoRAID Manager for Windows 230 settings viewing HP UX using ARM 64 MPE using ARM 155 NT using ARM 264 shutting down the disk array HP UX using ARM 50 MPE using ARM 142 NT using ARM 254 NT using AutoRAID Manager for Windows 235 simplified resiliency settings HP UX 70 270 MPE 161 SNMP agent 208 software installing on HP UX 16 installing on MPE 118 solving NT installation problems 220 status checking HP UX using ARM 39 MPE using ARM 131 NT using ARM 244 NT using AutoRAID Manager for Windows 224 NT using SAM 24 switching primary controller HP UX using ARM 56 HP UX using SAM 32 Index MPE using ARM 148 NT using ARM 260 NT using AutoRAID Manager for Windows 238 system requirements HP UX 16 MPE 118 Windows NT 210 T testing a disk HP UX using ARM 52 MPE using ARM 144 NT using ARM 256 NT using AutoRAID Manager for Windows 237 tools and tasks HP UX 15 MPE 117 V View pane 215 viewing configuration settings HP UX using ARM 64 MPE using ARM 155 NT using ARM 264 W What is a view In Windows NT 216 WinDownload 240 working set metric 49 141 234 write read test HP UX using ARM 52 MPE using ARM 144 NT using ARM 256 NT using AutoRAID Manager for Windows 237 279
23. Auto Include Balance in Progress Optimize in Progress Migra ting Write Destination Log Full Warning Rebuild Priority Pari SDTR WDTR ty selected value y n y Value Subsystem Parameter Value Ready 15 Termination Power ON 16 Unit Attention ON 17 Controller X Address 0 OFF 18 Controller Y Address 1 19 Enable Manual Override OFF 20 Manual Override Destination OFF OFF 21 Format Pattern Fill ON 22 Disable Remote Reset OFF OFF 23 Language English 24 Capacity Depletion Threshold 0 OFF 25 Rebuild Progress 0 O 26 Write Working Set Interval 0 27 Subsystem Identifier 1 3 Additional notes SCSI Specification Ready Warning Not Enough Drives No Quorum No Map Shutting Down Shut Down Starting Up No Code RESERVED Shutdown Warning RESERVED Mismatched Code Controller Mismatch Using the ARDIAG Offline Diagnostic on HP UX Operational Command Descriptions Any problems detected by the SETOPTIONS command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED Table 13 contains a list of the subsystem states displayed by the SETOPTIONS command and the naming convention used in AutoRAID documents Several of the subsystem state names displayed by the SETOPTIONS command have been abbreviated Table 13 Subsystem States Ready Warning Not Enough Disks No Quorum No Address Table Shutdown Started Shutdown Complete Initializing
24. Configuration The next and last item to configure is the LDEV number If this is the first time the user is performing this to an AutoRAID then LUN 0 should be configured As mention in the first note in this chapter the user is required to have LUN 0 be the lowest LDEV number NOTE The SYSGEN product ID for ALL AutoRAID devices is HPDARRAY To configure the LDEV number to the AutoRAID as follows 1o ad 30 id hpdarray path 10 16 4 1 0 io ld 30 LDEV 30DEVNAME OUTDEV 0 MODE ID HPDARRAY RSIZE 128 DEVTYPE DEVC PATH 10 16 4 1 0 MPETYPE 4 MPESUBTYPE 2 CLASS DISC SPOOL Once LUN 0 has been configured adding more LDEVs is done by using only the add device command ad The following is an example of adding LDEV 31 and LDEV 32 io ad 31 id hpdarray path 10 16 4 1 1 io ad 32 id hpdarray path 10 16 4 1 2 For more information on the SYSGEN configuration refer to the System Startup Configuration and Shutdown Reference Manual part number 36650 90042 124 Chapter 6 Using the ARM Command Line Utilities for MPE The AutoRAID software includes a set of AutoRAID Manager ARM command line utilities These commands provide the ability to manage the disk array from the MPE command prompt The ARM commands can be used to perform a number of tasks These tasks are intended for advanced users and involve procedures such as diagnostics performance monitoring setting SCSI val
25. Definition of Terms The following terms are used in this chapter HP UX SIO WSIO SERIES 700 SERIES 800 ISL ODE LIF SOM 74 Hewlett Packard s version of the UNIX Operating System Server I O The driver environment for the current Series 800 HP UX system Proprietary non reentrant port server drivers Communication is via message passing Work Station I O The driver environment for the current Series 700 HP UX system Reentrant procedure based drivers Communication is via procedure entry points HP UX PA RISC workstations These systems utilize the WSIO subsystem HP UX PA RISC multi user systems These systems utilize the SJO subsystem Initial System Loader This is the first piece of software loaded from outside the SPU and executed during the boot process Offline Diagnostic Environment The software architecture consisting of several modules to provide diagnostic functionality via ISL support media for Series 700 and Series 800 HP UX 10 04 systems Logical Interchange Format A standard disk format that may be used for interchange of files among various HP computer systems LIF volume contains a header identifying it as a LIF volume and a directory that defines the contents of the volume A boot device with a valid LIF is assumed to have bootable media Spectrum Object Module A SOM is the smallest unit that may be generated by a compiler and it may exist as a single entity or as part of a co
26. ENABLED Indicates whether the disk array should provide power for the SCSI bus terminators Disabled indicates the disk array will not provide termination power Enabled indicates the disk array will provide termination power 66 Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Unit Attention ENABLED Indicates whether the disk array should signal a Unit Attention condition immediately following power on or reset Disabled indicates the disk array will not signal unit attention Enabled indicates the disk array will signal unit attention Disable Remote Reset ENABLED Controls the bus reset behavior of the second controller hen a SCSI reset reset signal BDR or Reset Disk array command is received in the first controller When this bit is disabled the second controller will assert the bus reset signal to indicate that all outstanding requests were cleared in response to the reset When this bit is enabled the second controller will not assert the SCSI reset signal to indicate the commands were cleared Hosts that cannot tolerate target bus resets should enable this setting Secondary Controller DISABLED Controls the behavior of the secondary controller with Offline respect to bus selection When disabled any secondary controller present will respond to host selection When enabled the secondary controller will re
27. Items enclosed are required 126 Using the ARM Command Line Utilities for The ARMServer Process The ARMServer Process The ARMServer process is the server portion of the ARM software It monitors the operation and performance of the disk array and services external requests from clients executing disk array commands The ARMServer process monitors disk array performance and status maintains disk array logs initiates diagnostics and allows clients to examine and change the disk array configuration The ARMServer process must be running to allow management of the disk array using the ARM command line utilities Because of its importance in managing the disk arrays the ARMServer process should be launched automatically when the system is booted Host I Os to the disk array are not dependent on the ARMsServer process and are serviced regardless of whether the ARMServer process is running or not 127 Using the ARM Command Line Utilities for MPE Selecting a Disk Array to Manage Selecting a Disk Array to Manage When using the ARM utilities the array id field is used to identify the disk array The array id field can contain the disk array serial number or an alias text string assigned to the array using the arraymgr command For example assume a disk array has a serial number of 00786B5C0000 is defined for this array and it has an alias of autoraid4 To check the available unallocated capacit
28. Testing block O0 percent complete 1005 Test complete 189 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Testing subrange 1000 1200 percent complete 10 oe oe oe oe oe oe OL Or OY O1 iS CO PO 0 est complete esting block 45000 percent complete 1005 Test complete 1 1 1 ARDIAG gt 190 SETOPTIONS The SETOPTIONS command allows the user to view and set specific options for the array controller Synopsis SETOPTIONS Output ARDIAG gt 51 ETOPTIONS Using the ARDIAG Offline Diagnostic on MPE Operational Command Descriptions Current information for the controller in slot X Subsystem Parameter Value Subsystem Parameter Value 1 Subsystem State Ready 15 Termination Power ON 2 Active Hot Spare OFF 16 Unit Attention ON 3 Volume Set Partitioning 17 Controller X Address 0 4 Rebuild in Progress OFF 18 Controller Y Address in 5 Auto Rebuild 19 Enable Manual Override OFF 6 Auto Include 20 Manual Override Destination OFF 7 Balance in Progress OFF 21 Format Pattern Fill ON 8 Optimize in Progress 22 Disable Remote Reset OFF 9 Migrating Write Destination OFF 23 Language English 10 Log Full Warning 24 Capacity Depletion Threshold 0 11 Rebuild Priority OFF 25 Rebuild Progress 0 12 Parity O 26 Write Working Set I
29. The serial numbers of all disk arrays connected to the host can be displayed by typing arraydsp i Missing Disk Arrays If you know there are disk arrays connected to the host but they are not displayed in response to this command check the following Make sure the AutoRAID Manager service is running AutoRAID Manager must be running to execute this or any other ARM command Rescan for disk arrays by typing arraydsp R This will update the ARMServer information to reflect the current system configuration Make sure all disk arrays are turned on and operating properly Make sure all disk arrays are properly connected to the host This includes proper termination of the SCSI bus Make sure there is at least one logical drive created on the disk array A logical drive must be created on the disk array before the host can access the array During the boot process the host server must recognize a logical drive on the disk array to ensure the HBA device driver is loaded properly If there is no logical drive on the disk array the host may be unable to access with the array A logical drive can be created using the disk array control panel as described in the HP SureStore E Disk Array 12H User s and Service Manual 245 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Changing Disk Array Configuration Settings Changing Disk Array Configuration Settings A number of configuration settings control the
30. number of LUNs is 4 to 6 Capacity To allow for future capacity expansion avoid creating the maximum number of LUNs 8 on the disk array New capacity is made available through the creation of a new LUN If the maximum number of LUNs have already been created it will be necessary to delete and recreate an existing LUN to increase capacity Number of disk Performance Limit the number of disk arrays per host adapter to arrays per SCSI improve performance The recommended configuration for maximum host adapter performance is NIO adapter 3 disk arrays GSC adapter 8 disk arrays Capacity If maximum capacity is more important than performance connect the maximum number of disk arrays to each adapter 17 XN dH Managing the HP SureStore E Disk Array 12H on HP UX Tips for Configuring the Disk Array On HP UX 18 Configuration Option Impact on Performance and Capacity Number of disk drives Performance Increase the number of disk mechanisms in the disk array for maximum performance As the number of individual disks is increased the number of potential I Os that can be performed simultaneously increases as well This will improve performance in environments that place heavy I O demand on the disk array Capacity The capacity of a disk array that is fully populated with disk mechanisms can only be increased by replacing lower capacity disks with higher capacity disks Unall
31. the Create option will not be available E In the Number field select the number of the new logical drive 6 Inthe Capacity field enter the desired size of the logical drive 7 Click OK NOTE After creating a new logical drive you must perform the necessary steps to configure the logical drive into the operating system This is typically done using the Windows NT Disk Administrator Refer to Windows NT Help for information on performing this task 226 Using AutoRAID Manager for Windows NT Managing Logical Drives Renumbering a Logical Drive Renumbering changes the logical drive number assigned to the logical drive by the disk array The logical drive number is the means by which the operating system identifies each logical drive Consequently renumbering a logical drive will impact your operating system s ability to access the data on that drive You should only consider renumbering a logical drive in specific situations For example assume you are using logical drive 0 as your boot drive and you later decide you want to boot from a different drive You will have to assign the current boot drive a new number and then renumber the new boot drive to 0 NOTE Before renumbering a logical drive consult your Windows NT documentation for information on what steps you will have to take to make sure the host can access the new logical drive number To renumber a logical drive Right click the disk array from the Deta
32. 88 iudei eae b ME 92 auem eene trn ERIS 94 IRE Ss PARE C estie ticis tt Dette dite A Eb ir 95 REVISION 96 e n NE be I P UM e 97 SETOPTIONS oru ere eR pu icouuuunseuduoenen quae 99 tope deest Mice esce celis 102 MRM tte iminven WU 103 INTERFACE COMMAND DESCRIPTIONS ssssscccsecsesesseceeecececseensaaececececeeseseaesececseseseaeceeeesesenseaeeeeeeseseneaaees 105 DISPMALD 35 wetter tree eee tete eco beet e RI RII 105 DIS PME GH TE EM MC UM 106 IRAN GE m r ET 107 Nisl ARP 108 TESTDISK 5 RERO ERGIBT ier 109 ugue mbi ee eee 110 ODB INTEREAGE tire tete true E rte ee uice o voee RECRUIT RT ERE 111 ARDIAG INTERFACE TO ODE eR e id 113 CHAPTER 5 MANAGING THE HP SURESTORE E DISK ARRAY 12H ON 115 DISK ARRAY MANAGEMENT TOOLS 116 AutoRAID Management ARM utilities ee ceeeeesseceereececeeeeecaeceeceecsaeceneecaaecesneecsaeeeeeeecsaeeenes 116 Disk array control panel reete ee hte ere eerte etr ER dn 116 WHICH TOOLS TO USE FOR EACH TASK essere en eeth nnn arent terea assess enti reas 117 INSTALLING THE DISK ARRAY MANAGEMENT SOFTWARE eese ener seen enne
33. Be aa aaa es 48 Selecting a Time Period for Analysis etie te ee ere Dre Ir 49 Checking the Working Set Metrics reete precepit le eee cete eee 49 PERFORMING DISK ARRAY MAINTENANCE 50 Shutting Down the Disk Array 50 Restarting the Disk Array tte Peer ORE Gi repre e REUS DER e ipe bre i ents 50 Resetting the DISK ATAY cp toten oet een tet 51 Downing Excluding a Disk 51 Testing a th a eR E bum inen Camel aep eee ium 52 Displaying Test Results nhe phentermine ertt DRE DE eser 52 Cancelmg Disk Test eoe e eo Re e ern e e tod 53 Printing ARMServer Log Contents trennen trennen nee 53 Displaying Hardware Logs eee enne rennen entere tenen nen 53 Formatting the Disk Array intet erre EE RR eb EEES ret 54 Changing SC Settngs siste teet et etta Hed xb net eren e IE tive Eie Unete 55 Changing the Controller SCSI ID 55 Switching Primary Controllers sess 55 Setting Data Resiliency oec ER eE oes Dm 56 Creating Disk Array Alas ico depicted ae evceblecsevadedl dct dp eg etie d deed 58 DOWNLOADING FIRMWARE esee trennen enne enin en
34. Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck kk Ck kk kk Sk Sk Ck Ck Sk kk kk kk ko ko NOTICE Use of this command is not recommended except where a full subsystem backup of user data is available or when a new uninitialized subsystem is being prepared for use a precaution this command will NOT allow a format to an array containing any LUNs The DELETELUN command must be used to remove all existing LUNs Dk Ck ck ck Ck ck Ck ck Ck ck ck 0k ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck kk ck ck ck ko Sk ke kx kv Mk ko ko ko ko ko Do you want to continue y n y Checking the array subsystem for LUNs There are no LUNs presently configured on the selected array selected array subsystem is allowed a format Do you want to continue y n y FORMATTING kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk ck Ck ck ck kk ck kk ck kk Ck kk Ck ck ck kk ck kk Ck kk kk Sk kk ck kk ck kk Sk kc k ko KKK KKK FORMAT Successfully Completed Ck ck ck Ck ck ck kk Ck Sk Ck Ck ck ck kk ck kk ck kk Ck Sk ck kk ck kk ck kk Ck ck ck kk Sk kk Sk kk ck kk Sk kc KKK KK KKK ARDIAG 85 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Example 2 Internal disk drive ARDIAG TESTLEVEL 1 ARDIAG FORMAT kk ck Ck ck ck ck ck ck kk Ck KKK KKK KKK KKK KKK ck kk ck kk ck kk Ck kk kk ck kk ck kk ck kk Sk kc k
35. M array id 59 XN dH Using the ARM Command Line Utilities for HP UX Downloading Firmware Firmware Download Procedure With LVM When LVM has access to the disk array the procedure described below should be followed to perform the download This procedure works for single or multi host configurations CAUTION In multi host configurations other hosts must not access the disk array while the download is in progress Data can be lost if write requests are made to the disk array while a download is in progress If the arraymond process is running it can disrupt the download Make sure you kill the arraymond process before starting a download If the root or opt file system is contained on the array the offline diagnostic ARDIAG must be used to download the firmware If the stand var usr tmp file system or swap partition are contained on the disk array the download must be performed in single user mode or off line using ARDIAG NOTE The download process automatically shuts down the array The array Shutdown will not succeed unless all file systems have been unmounted and LVM access to the array has been deactivated To download firmware to the array controllers or to the disk mechanism 1 Unmount all logical volumes mapped to the disk array Deactivate all LVM volume groups containing logical volumes mapped to the disk array Download the controller or disk mechanism firmware Reactivate the LVM gro
36. Ready The array warnings has changed The previous warnings were Disk Warning Capacity Warning Now there are no warnings Additional Notes Read controller logs to assess the completion status of the command based on the number of ECC errors or event counts 93 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions REQSENSE The REQSENSE command allows the user to view SCSI Sense data from the specified target Synopsis Output 94 REQSENSE ARDIAG gt REQSENSE Performing Request Sense at TestLevel Indx 0 KKKKKK KKK KKK KK KKK KKK KK KKK KK KKK KKK KKK KKK KKK KKK KK ck ck ck KKK ko ck kk ck Sk Sk KKK KKK KKK Error Code 112 0x70 Segment Number 0 0x0 File mark bit is OFF End of medium bit is 1349672 The Incorrect Length Indicator is OFF Sense Key NO SENSE 0x0 The information field is NOT SCSI 2 compliant Information field bytes 0x0 0 0 0x0 0 0 Information as an int 0 0x0 Command Specific Information field bytes 0x0 0 0 0 0 0 0 Command Specific Info as an int 0 0x0 Additional sense code 0 0x0 and qualifier 0 0x0 Translates to No additional sense information Field replaceable unit code 0 0x0 Sense Key Specific field is NOT valid Sense key specific field 0x0 0 0 0 0 odule Identifier is 0 0x0 Error Number is 0 0x0 Using the ARDIAG Offline Diagnostic
37. Settings Secure Mode Default Parameter Setting Simplified Resiliency Setting Results Single Controller ENABLED 700 Series Workstation 800 Series Multi Purpose Warning Lock Write Cache On FALSE Single Dual Single Dual Controller Controller Controller Controller Disable NVRAM on TRUE Data map Not Data map WCE False Supported and write Supported and write Disable NVRAM with TRUE ache ache One nformation nformation Disable NVRAM on FALSE s scheduled s scheduled UPS absent o flush to o flush to Force Unit Access 2 he disk he disk Response nce every nce every Disable Read Hits FALSE goang Resiliency Threshold 1 HighPerformance Mode Parameter Default Setting Simplified Resiliency Setting Results Single Controller ENABLED 700 Series Workstation 800 Series Multi Purpose Warning Lock Write Cache On TRUE Single Dual Single Dual Controller Controller Controller Controller Disable NVRAM on FALSE Data Not Data WCE False Supported Resiliency Supported Resiliency Disable NVRAM with TRUE isabled isabled One Disable NVRAM on FALSE UPS absent Force Unit Access 0 Response Disable Read Hits iv Resiliency Threshold x03 f RN CS 72 ho JL Chapter 4 Using the ARDIAG Of
38. The following are examples of the READLOG command ARDIAG READLOG Example 1 Controller 180 ARDIAG readlog Contents of array controller log at TestLevel Indx 0 Dk kc ck ck CC ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ok kk kk ck Pk ck kv kx Sk Mk ko ko ko ko ko Vendor ID HP Product ID C3586A Usage Log ECC Error Count 0 0x0 Select number to indicate display option for Event Log Information Number Display Option ck ckck kckck T Event Typ ables 2 Chronological 3 Both Event Type Tables amp Chronological Not Event Type Tables and Chronological are the same information they are just displayed differently default 3 Both il Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions DRIVE ERROR EVENT TABLE ck ckck ck ck ckck ck ck ckck ck ck ckck ck ck ck k ck k ck k ck k ck k ck k ck k ck k ck k ck k ck k ck KKK KKK Time Module Event Event Slot Sense ASC ASCQ LBA Stamp ID Code Count Key Dk Ck ck ck KK KKK Sk ck ck Ck ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck KK ck ck ck kk ck ko ck ck ko ck kk ck Mk Sk kv kx kv Mk ko ko 28279 0x4 0 0 Ox1f 30 A2 0x5 0x24 0x0 N A CONTROLLER ERROR EVENT TABLE Kk ck Ck ck ck Ck ck ck Ck ck Ck kk Ck ck ck
39. UX server you must configure the HP UX server as follows to allow access to the array 1 Verify that the NT userid is valid on the HP UX server and that the userid is a member of the adm group 2 Ifyou re not using the root userid add group write access to the var opt hparray admin sub directory 3 Add host and userid to the etc hosts equiv file NOTE To administer a disk array connected to an HP UX server from an AutoRAID Manager client the necessary patches must be installed on the HP UX server Check the README file included on the HP 213 LN SMOpUIM Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT Setting Up AutoRAID Manager for Windows NT AutoRAID on Windows NT CD ROM for information on which patches are required If you will be using a Windows 95 management client it will be necessary to restart the client before you can manage disk arrays connected to an HP UX server Changing the Log File Directory Location The default directory for the AutoRAID Manager log files is x ProgramFiles AutoRAID log If you want to use a different directory for the log files it will be necessary to edit the registry entry for the file location The log file directory is defined by the following registry key HKEY_LOCAL_MACHINE SOFTWARE HEWLETT PACKARD AUTORAID ARMSERVER LOG DIRECTORY Run the Registry Editor and edit this entry to reflect the new directory location Make sure the proper permissions are set on the
40. Working Set value less than or equal to 1 A Working Set value greater than 1 indicates that the Write Working Set is larger than the available RAID 0 1 space and the disk array is servicing writes from RAID 5 space If the Working Set consistently exceeds 1 the amount of RAID 0 1 space available should be increased to improve performance This can be accomplished in several ways as described in the following section If the Working Set is consistently much less than 1 some of the RAID 0 1 capacity can be allocated to a new logical drive without impacting performance The remaining RAID 0 1 space should be adequate to accommodate the Write Working Set Performing Disk Array Maintenance Tasks There are a number of tasks that may have to be performed in the ongoing management of the disk array These maintenance tasks are typically performed infrequently and may involve taking the disk array offline Shutting Down the Disk Array Shutting down the disk array makes the array unavailable to the server When a shutdown is initiated the disk array completes any I Os in progress with the server performs internal housekeeping functions and then takes itself offline Shutdown is intended primarily as a maintenance operation and does not need to be done on a regular basis To shutdown the disk array 1 Alert all users that the disk array is being shutdown and any data on it will not be available Right click the disk array from the Detail pane C
41. are used in the command descriptions and examples Symbol Meaning lt gt Integer value whose units are not defined Exclusive OR Exactly one of the parameters displayed will be used Items enclosed are optional Items enclosed are required The ARMServer Process The ARMServer process is the server portion of the ARM software It monitors the operation and performance of the disk array and services external requests from clients executing disk array commands The ARMServer process monitors disk array performance and status maintains disk array logs initiates diagnostics and allows clients to examine and change the disk array configuration The ARMServer process must be running to allow management of the disk array using the ARM command line utilities Because of its importance in managing the disk arrays the ARMServer process is launched automatically when the system is booted Host I Os to the disk array are not dependent on the ARMServer process and are serviced regardless of whether the ARMServer process is running or not 34 Using the ARM Command Line Utilities for HP UX The Array Monitor Daemon arraymond The Array Monitor Daemon arraymond The Array Monitor Daemon known as the arraymond process runs at regular intervals every fifteen minutes to retrieve disk array status information The arraymond process uses the arraydsp ARM utility and therefore requires the ARMServer process to be running By
42. array For a detailed explanation of the log contents output see the on line help logprint description To print the contents of the ARMServer logs type logprint d log directory name s start time e stop time t record type a array serial number log directory name identifies the location of the log files start time and stop time limit the output to events between the specified times record type identifies the type of record s to print Records include system usage log usage disk error log disk controller error log ct x1r system change log change and performance log perf array serial number limits the output to only those entries associated with the specified disk array Displaying Hardware Logs In addition to the logs maintained by ARMServer hardware logs are also stored on the disk array The arraylog command provides access to the controller and disk logs maintained by the disk array These logs contain information useful for diagnosing and troubleshooting the disk array The logs can also be cleared using arraylog The arraylog options for accessing the disk array hardware logs are listed in Table 29 257 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Performing Disk Array Maintenance Tasks Table 29 arraylog Options for Displaying Log Contents Option Description Display the contents of the disk array controller usage log
43. ck Ck ck ck ck ck ck ck ck kk ck Sk ck ko ck kv k ko ko ko ko i Firmware Download Successful kk ck Ck ck ck Ck ck ck Ck ck Ck Sk ck Ck ck kk ck kk ck kk Ck Sk ck kk ck kk ck kk ko kk kk Sk kk ck kk ck kk Sk kc k ko kc KKK KK ARDIAG Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Example 2 Internal disk drive ARDIAG testlevel 1 TESTLEVEL 1 ARDIAG download Enter the firmware file name displays the files Default quits File name Intended Product ID Rev Size Z1CS HPC3586A disk array Z1CS 1048576 SEA5400 SEAGATEST31230N disk drive 0284 261632 5267200 Generic Disk 1208 261632 5467200 ST15150W disk drive 1207 261632 Legend File name name of the firmware file Intended Product ID firmware file s intended product name Rev firmware Revision of the firmware file Size exact byte size of the firmware image Enter the firmware file name displays the files Default quits sea5400 WARNING Firmware file SEA5400 was made for a SEAGATEST31230N disk drive TESTDISK selects a Unknown Product STOP CONTINUING MAY PERMANENTLY DESTROY ALL DATA ON DISK Do you still want to continue y n Y 175 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions 176 kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk Ck Ck ck ck kk ck kk ck kk Ck Sk ck kk ck kk ck kk ko kk kk ck k
44. disk array as quickly as possible The Auto Rebuild feature does this automatically so it is enabled by default For optimum data protection it recommended that Auto Rebuild remain enabled If you would like more control over the Rebuild process you can disable Auto Rebuild using the ARM utilities This will allow you to manually start a Rebuild at the time you choose A manual Rebuild is initiated using the ARM utilities See Rebuilding the Disk Array Manually in the next chapter for more information A Rebuild impacts disk array performance while it is in progress so before starting a Rebuild make sure the appropriate rebuild priority is set For convenience and maximum protection against disk failure Auto Rebuild is enabled by default on a new disk array Setting Rebuild Priority The rebuild priority determines how quickly a Rebuild completes It allows you to balance the servicing of host I Os with the rebuilding of the disk array The same rebuild priority is used for both Auto Rebuilds and manual Rebuilds A high rebuild priority ensures the Rebuild will be completed at the same priorities as host I Os A low rebuild priority gives priority to host I Os in relation to the Rebuild To ensure that a Rebuild completes without disrupting data storage the rebuild priority is set to high by default on new disk arrays To set the rebuild priority 1 Select the desired disk array controller from the list Select Actions on the men
45. e Display the contents of the disk array controller event log d slot Display the contents of the log for the disk installed in the cabinet slot identified by slot Slot numbers must be of the form or Bn where A or B correspond to a cabinet column and n corresponds to a shelf position 1 6 C slot Clear the specified logs If c is specified clear the disk array controller usage and event logs Both logs will be cleared when using this option It is not possible to clear just one of the logs If d is specified clear the log for the disk installed in the cabinet slot identified by slot To display the contents of a hardware log type arraylog u e d slot array id To clear the hardware logs type arraylog C cl d slot array id Formatting the Disk Array Should it become necessary to do so you can format the entire disk array or a single logical drive Formatting destroys all data on the array or logical drive involved Formatting an array first requires that all logical drives be deleted To format the entire disk array type arrayfmt F h array id To format a logical drive type arrayfmt L LUN lt array id gt 258 Using the ARM Command Line Utilities for Windows NT Performing Disk Array Maintenance Tasks Changing SCSI Settings The SCSI settings control the transfer of information over the SCSI channel connecting the host and the disk arr
46. for Windows NT Performing Disk Array Maintenance Tasks Resetting the Disk Array The disk array can be reset if there is a problem with SCSI channel communication A reset will interrupt access to the disk array temporarily so it should be done only when attempting to solve a problem with the disk array To reset the disk array type arraymgr R array id Downing Excluding a Disk Downing or excluding a disk is typically done in preparation for testing the disk After the disk is downed testing can be done without impacting disk array operation If testing reveals that the disk is good the disk can be included back in the array configuration Downing a disk has the same effect as if the disk failed or was physically removed from the cabinet If Auto Rebuild is enabled the disk array will immediately begin a rebuild when a disk is downed The v option identifies the down operation as either destructive or nondestructive This determines whether the disk array will assume there is any valid data on the disk if it is returned to the array configuration If a destructive down is performed not specified the disk array will assume no data on the disk is valid If the down is nondestructive v specified the array will assume any data on the disk that was not updated in the disk s absence is valid To down a disk type arraycfg D slot d v R 2 lt array id gt To protect data availability the disk array w
47. impact it may have on performance 56 Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks To set the data resiliency mode type arraymgr J SingleController Secure Normal HighPerformance array id The data resiliency settings are described in Table 7 Table 7 Data Resiliency Settings Setting Description SingleController Secure HighPerformance This setting should only be used if the disk array is operating with one controller This suppresses the single controller warning messages that are normally generated when the disk array is operating with one controller This setting will affect I O performance This is the default setting for single controller mode Continually updates the disks with any changes in the controller maps This is the highest level of data protection but it may result in decreased I O performance Updates the maps on the disks at regular intervals typically 4 seconds This option offers both data protection and good performance This is the default setting for dual controller mode Updates the disk maps only during shutdown of the disk array This is the lowest level of data protection but it offers the highest level of performance 57 XN dH Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks Creating a Disk Array Alias An alias can be created to identify the disk array The alias pro
48. influenced by many factors The configuration options described here will increase the potential performance of the disk array However the actual performance of the disk array will be determined largely by host demand AutoRAID technology is particularly suited to l O intensive application environments such as OLTP and NFS It is in these environments that the performance benefits offered by AutoRAID will be fully realized Table 15 Configuration Options Configuration Impact on Performance and Capacity Option Number of LUNs Performance Increase the number of LUNs per disk array to improve per disk array performance More LUNs increases the size of the command queue allocated which increases throughput The recommended number of LUNs is 4 to 6 Capacity To allow for future capacity expansion avoid creating the maximum number of LUNs 8 on the disk array New capacity is made available through the creation of a new LUN If the maximum number of LUNs have already been created it will be necessary to delete and recreate an existing LUN to increase capacity Number of disk Performance Limit the number of disk arrays per host adapter to arrays per SCSI improve performance host adapter Capacity If maximum capacity is more important than performance connect the maximum number of disk arrays to each adapter 119 Managing the HP SureStore E Disk Array 12H on MPE Tips for Configuring the Disk
49. information or steps that may be required to delete a LUN To delete a LUN 1 Qv voe OS ABS From the list of disks and arrays select the LUN to be deleted Select Actions from the menu bar Select Disk Array Maintenance Select Unbind LUN Select OK to confirm the request Note the removal of the LUN from the list of disks and arrays 27 XN dH Using the HP UX System Administration Manager Adding a Disk Adding a Disk At some time you will probably want to add another disk to your array Features such as hot pluggable disks and Auto Include simplify the process of adding a disk to the array even while it is operating A disk can be added to the array without disrupting host operation After you have added a new disk you have three options on how to use it e Increase capacity you can use the disk to increase the capacity available to the operating system by creating a new LUN e Improve performance you can use the disk to improve disk array performance by simply leaving it as unallocated capacity e Enable Active Spare you can use the additional capacity to enable Active Spare if the disk array does not currently have the capacity to implement this feature To add a disk to the array 1 Make sure the new disk has been inserted into the array cabinet 2 If Auto Include is on the disk is automatically added to the array and you can skip to the next step If Auto Include is of
50. information to reflect the current system configuration e Execute the ioscan fn C disk command and ensure that the software state for all active disk arrays is CLAIMED Changing Disk Array Configuration Settings A number of configuration settings control the operation of the disk array These settings are usually established during installation and once set should rarely need to be changed The default settings have been selected to provide the best operation for most systems However if you determine that any setting does not meet your needs you can easily change it Table 5 lists the various settings including factors you may want to consider when changing them 40 Using the ARM Command Line Utilities for HP UX Changing Disk Array Configuration Settings Table 5 arraymgr Disk Array Configuration Settings Setting Default Command Comments and Considerations Option Active On h Active Hot Spare provides optimum Spare protection against disk failure Disabling Active Hot Spare will make additional capacity available to the host but at the expense of maintaining full data redundancy Auto On a Auto Rebuild provides optimum protection Rebuild against disk failure by rebuilding a failed disk as quickly as possible Disabling Auto Rebuild gives you more control over the rebuild process but it can leave the disk array vulnerable to a second disk failure until a Rebuild is performed manually Auto On i Au
51. installed on the server only if SNMP is installed The AutoRAID Manager SNMP agent communicates with the HP OpenView client to provide the mechanism for broadcasting event information to network administrators and establish a communication link between the server and the client Failover Driver this componentenables the operating system to exploit the performance and redundancy benefits provided by dual disk array controllers During normal operation the failover driver manages the transfer of host data to achieve optimum performance If one of the controllers fails the failover driver automatically stops using the failed controller and directs all subsequent I Os to the remaining controller The failover driver provides the same functionality regardless of whether the two disk array controllers are on the same SCSI channel or on separate SCSI channels Managing the HP SureStore E Disk Array 12H on Windows NT HP OpenView Integration HP OpenView Integration An OpenView client for AutoRAID is included with the AutoRAID Manager software enabling full integration with HP OpenView This allows you to manage the SureStore E Disk Arrays on your network from an OpenView console If OpenView Network Node Manager is detected on the client during installation of the AutoRAID software the OpenView client for AutoRAID will be installed The OpenView client for AutoRAID communicates with the AutoRAID Manager SNMP agent running on the server Chan
52. is deleted its capacity is returned to the pool of unallocated capacity space Deleting a LUN is a good way of freeing up capacity for the Active Hot Spare or for simply adding more unallocated capacity to improve disk array performance CAUTION data on a LUN is lost when it is deleted Make sure you backup any important data on the LUN before deleting it NOTE Before deleting a LUN check your operating system documentation for any additional information or steps that may be required to delete a LUN 43 XN dH Using the ARM Command Line Utilities for HP UX Adding a Disk To delete a LUN type arraycfg L LUN d array id LUN is the LUN to be deleted Adding a Disk At some time you may want to add another disk to your array Features such as hot pluggable disks and Auto Include simplify the process of adding a disk to the array even while it is operating A disk can be added to the array without disrupting current I O operations After you have added a new disk you have three options on how to use it e Increase capacity use the disk to increase the capacity available to the operating system by creating a new LUN e Improve performance use the disk to improve the disk array performance by simply leaving it as unallocated capacity Enable Active Spare use the additional capacity to enable Active Hot Spare if the disk array does not currently have the capacity to support this feature Thi
53. kk Sk kc k ko KKK KKK FORMAT Successfully Completed Ck ck ck Ck ck ck kk Ck Sk Ck Ck ck ck kk ck kk ck kk Ck Sk ck kk ck kk ck kk Ck ck ck kk Sk kk Sk kk ck kk Sk kc KKK KK KKK ARDIAG 177 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Example 2 Internal disk drive ARDIAG TESTLEVEL 1 ARDIAG FORMAT kk ck Ck ck ck Ck ck KK KKK KKK KKK ck Ck ck ck kk Ck ck ck kk ck kk ck kk ko kk kk Sk kk ck kk ck kk Sk kc k ko KKK KKK Internal Disk Drive FORMAT Dk ck ck ck Ck ck kk ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck kk ck ko ck ck ko ck ko ck ko Sk Sk Mk Sk ko kx ko ko WARNING kk ck Ck ck ck ck ck ck kk Ck Sk KKK ck kk ck kk ck kk Ck Sk ck kk ck kk ck kk ko kk Ck ck ck kk ck kk ck kk Sk kc k ko KKK KKK FORMAT cannot continue without a loss of redundancy on the disk array selected Failure of a disk before redundancy is restored will result in DATA LOSS Although continuation of this FORMAT will NOT result in direct data loss it may take several hours before redundancy is restored and a disk failure can be tolerated without data loss Dk ck ck ck Ck ck ck Ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck kk ck ck ck ko ck ko Sk Sk Mk ko ko ko ko ko Do you want to continue y n y Dk ck ck ck Ck ck kk ck Ck ck ck Ck ck Sk Ck ck Ck ck c
54. ko KKK KKK Internal Disk Drive FORMAT Dk ck ck ck Ck ck kk ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ko ck ck ck ck Sk ck ko Sk Sk kv Sk ko ko ko ko WARNING kk ck Ck ck ck Ck ck ck kk Ck KKK KKK KKK KKK KKK KKK ck kk ck kk ko kk Ck ck Sk kk ck kk ck kk Sk kc k ko KKK KKK FORMAT cannot continue without a loss of redundancy on the disk array selected Failure of a disk before redundancy is restored will result in DATA LOSS Although continuation of this FORMAT will NOT result in direct data loss it may take several hours before redundancy is restored and a disk failure can be tolerated without data loss Dk ck ck ck 0k ck Ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ko ck Pk Sk ko Sk Sk kv ko ko ko ko ko Do you want to continue y n y Dk ck ck ck Ck ck kk ck Ck ck ck Ck ck Sk Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck kk ck ck ck ck ck ck Ck ock ck ck ck ko ck ok Sk Sk Mk Sk ko ko ko ko NOTICE Ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck ck Ck ck Ck Ck Ck ck Ck ck ck ck ck ck ck Sk kk kk kk ko ko kock ok The following FORMAT will require approximately ONE HOUR to complete During this period you will be unable to perform any other diagnostic activities Dk ck ck ck Ck ck Ck ck Ck ck ck Ck ck Sk Ck ck
55. management tasks have been divided between SAM and the command line utilities In general the more common tasks are available from SAM while the more advanced tasks are performed from the ARM utilities Most tasks can also be performed using the disk array control panel Table 1 Tools and Tasks Tasks 5 SAM ARM Control Utilities Panel Including a disk Yes Yes Yes Checking disk array status Yes Yes arraydsp Yes Renumbering LUNs Yes arraycfg Yes tari Ye Downing excluding a disk Yes Yes arraycfg No no Yes tran Ye Shutting down the disk array Yes arraymgr Yes arog opting sotngs Yoo Changing SCSI settings No Yes arraymgr Yes Monitoring performance No Yes arraydsp No Switching primary controller Yes Yes arraymgr Yes ne ties No Displaying disk test results No Yes dteststat Displaying disk array serial Yes Yes arraydsp Yes numbers E neam Resetting restarting the disk array Yes arraymgr Yes _ Creating disk array alias Yes arraymgr 15 XN dH Managing the HP SureStore E Disk Array 12H on HP UX Installing the Disk Array Management Software Installing the D
56. new directory NOTE Use caution when modifying the registry and make a backup before making any changes Setting Up AutoRAID Manager for Windows NT After installing the software there are a few initial steps that must be performed before you can begin managing your disk arrays This involves scanning the local network for any host that has an HP SureStore E Disk Array 12H connected to it Once a host has been located the disk arrays connected to it must be added to the Array List window before they can be managed using AutoRAID Manager 214 Managing the HP SureStore E Disk Array 12H Windows NT Setting Up AutoRAID Manager for Windows NT The HPAutoRAID Manager GUI The HPAutoRAID Manager s main screen is divided into two main sections View pane and Detail pane The View pane left side contains the views that are available and the Detail pane right side contains the list of hosts visible to the selected view To run the HP AutoRAID Manager From the Start menu select ProgramslAutoRAID ManagerlAutoRAID Manager AutoRAID Manager immediately begins scanning the selected hosts for the current view searching for all operating disk arrays with AutoRAID As disk arrays are identified they are added to the Detail pane right side When the scanning is complete each HP SureStore E Disk Array 12H in the view should be shown in the Detail pane HP AutoRAID Manager oO xi File Device View Help Array Serial Normal
57. of the entire disk array to validate the accuracy of the maps and to correct any drive parity inconsistencies This process can take up to several hours depending on the amount of data on the disk array To start the map recovery process type arrayrecover s 1 c p v volume set ID c array id 62 Using the ARM Command Line Utilities for HP UX Recovering Data Maps The command options for arrayrecover are listed in Table 8 Table 8 arrayrecover Command Options Option Description Used with other options this will cause continuous polling during the recovery The recovery is monitored and the current progress is displayed at regular intervals l Displays the array controller logs on standard output lc Displays the array controller logs on standard output and then clears the logs sets on the disk array a list of the volume set numbers will be displayed If there is only one volume set on the disk array the recovery will be done on it This option provides a mechanism to pass volume set information to a script designed to perform a recovery Returns the recoverability status of the disk array as well as the status of a recovery in progress The status indicates if recovery is needed and what percentage of the entire recovery has been completed When used with the c option status will be returned at regular intervals allowing continuous monitoring of the recovery progress
58. operation of the disk array These settings are usually established during installation and once set should rarely need to be changed The default settings have been selected to provide the best operation for most systems However if you determine that any setting does not meet your needs you can easily change it Table 28 lists the various settings including factors you may want to consider before changing them Table 28 arraymgr Disk Array Configuration Settings Setting Default Command Comments and Considerations Active Hot Spare provides optimum protection against disk failure Disabling Active Hot Spare will make additional capacity available to the host but at the expense of not maintaining full data redundancy Auto Rebuild disk failure by rebuilding a failed disk as quickly as possible Disabling Auto Rebuild gives you more control over the rebuild process but it can leave the disk array vulnerable to a second disk failure until a rebuild is performed manually Auto Include Auto Include simplifies the task of adding a new disk to your array Disabling it will require you manually to include each disk you install in the array Rebuild High p Rebuild priority determines how quickly a rebuild Priority operation will complete a Auto Rebuild provides optimum protection against To change Active Spare Auto Rebuild or Auto Include settings type arraymgr h a i on
59. or standalone e NT cluster In NT cluster installations logical drive 0 must be reserved for exclusive use by the AutoRAID Manager and should not be used for data files This unused logical drive is essential for maintaining proper operation should one of the cluster servers fail Because logical drive 0 is used only for management purposes you should make it as small as possible e Standalone noncluster In standalone installations logical drive 0 can be used as any other logical drive on the disk array Its size will depend on your overall logical drive strategy for the disk array Use the disk array control panel to create a logical drive Refer to the HP SureStore E Disk Array 12H User s and Service Manual for control panel information Installing AutoRAID Manager Software Installation Tips Remember the following tips as you install the AutoRAID Manager software V Select the appropriate system configuration to ensure the proper software components are installed e Select Client Installation to install Windows GUI and HP OpenView client softwareon remote computers used to manage network resources no command line utilities e Select Server Installation to install the clients server and failover driver components 211 1 SMOpUIA Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT Installing AutoRAID Manager Software e Select Custom Installation only if you want to control which components are in
60. select a period of time that represents normal system operation Avoid any unusual events such as a Rebuild or changes made to array capacity If you select a time period that includes an event that may distort the analysis the utility will alert you and will not provide any recommendations You can also control the display interval using the int option This allows you to control how much detail you get when displaying the metrics The display interval is the number of 15 minute increments Checking the Working Set Metric A key factor in monitoring and maintaining optimal performance of the disk array is the Working Set metric To ensure that disk array performance is maintained you should access the performance metrics regularly and check the Working Set value The Working Set performance metric is derived from the Write Working Set parameter It indicates the ratio of the Write Working Set size to the amount of RAID 0 1 space available For a detailed explanation of the Write Working Set and its impact on performance refer to the HP SureStore E Disk Array 12H User s and Service Manual To maintain performance the amount of RAID 0 1 space should equal or exceed the Write Working Set resulting in a Working Set value less than or equal to 1 A Working Set value greater than 1 indicates that the Write Working Set is larger than the available RAID 0 1 space and the disk array is servicing writes from RAID 5 space If the Working Set consist
61. the disk array will not initiate SDTR Enabled indicates the disk array will initiate SDTR WDTR ENABLED Indicates whether the disk array should initiate Wide Data Transfer Request Disabled indicates the disk array will not initiate WDTR Enabled indicates the disk array will initiate WDTR Terminator Power ENABLED Indicates whether the disk array should provide power for the SCSI bus terminators Disabled indicates the disk array will not provide termination power Enabled indicates the disk array will provide termination power 157 Using the ARM Command Line Utilities for Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Unit Attention ENABLED Indicates whether the disk array should signal a Unit Attention condition immediately following power on or reset Disabled indicates the disk array will not signal unit attention Enabled indicates the disk array will signal unit attention Disable Remote Reset ENABLED Controls the bus reset behavior of the second controller hen a SCSI reset reset signal BDR or Reset Disk array command is received in the first controller When this bit is disabled the second controller will assert the bus reset signal to indicate that all outstanding requests were cleared in response to the reset When this bit is enabled the second controller will not assert the SCSI reset signal to indicate the commands w
62. to read the specified target s internal logs Allows the user to restore controller map information in the specified SureStore E AutoRAID Disk Array Issues a Request Sense on the specified target Allows the user to bring the array out of the Shutdown state via the specified controller path Displays controller drives and software revisions Read Verify test to the specified internal mechanism Allows the user to view and set specific options for the array controller Allows the user to put the array into a Shutdown state via the specified controller path Destructive write read and verify test to the specified internal mechanism 167 Using the ARDIAG Offline Diagnostic MPE ARDIAG Interface Commands ARDIAG Interface Commands Table 12 lists the ARDIAG test module interface commands and their descriptions Table 24 ARDIAG Interface Commands Command Description DISPMAP Displays a list of AutoRAID controllers DISPMECH Displays the disk mechanisms within the selected disk array RANGE Environment variable that represents a range of blocks to test SHOWENV Displays environment variables ui TESTDISK Points to the disk array on which to execute commands TESTLEVEL Points to a physical mechanism within the disk array on which to execute commands 168 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Operational Command Descr
63. using the HP UX System Administration Manager or SAM Using SAM you can e Check disk array status e Change configuration settings e Manage the disk array LUNs e a disk to the array configuration e Setthe rebuild priority You must login as superuser or root to use SAM or the disk array utilities LUNs and Logical Drives To maintain consistency with HP UX terminology the term LUN is used to refer to a disk array logical drive The two terms are used interchangeably and refer to the same logical entity on the disk array NOTE Before SAM can be used to manage the HP SureStore E Disk Array 12H the AutoRAID Manager ARM utility software must be installed as described in Installing the Disk Array Management Software To run the SAM disk array management utilities 1 Run SAM by typing sam at the system prompt 2 the main SAM screen select Disks and File Systems 3 the Disks and File Systems screen select Disk Devices A list of disk devices including arrays will be displayed You are now ready to work with the disk array The remaining procedures in this chapter assume that you have already performed these steps 19 XN dH Using the HP UX System Administration Manager Selecting Hardware Paths on HP UX 10 20 Selecting Hardware Paths on HP UX 10 20 On dual controller disk arrays each controller provides a separate hardware path to the disk array The host identifies one o
64. value for the LUN Creation Limit Creation Limit 156 Using the ARM Command Line Utilities for Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Array SCSI N A N A configuration Controller X SCSI Indicates the SCSI bus address to be used by controller Address X Changes in this field will take affect only after controller X is reset Controller Y SCSI Indicates the SCSI bus address to be used by controller Address Y Changes in this field will take affect only after controller Y is reset Write ENABLED Indicates whether the disk array should cache write data This field may be ignored depending on the map resiliency mode Series 800 systems normally disable this field Generally write cache is used even though a host system automatically disables this field Read Cache DISABLED Indicates whether the disk array should cache read data Note This field does not represent what is actually happening The disk array uses read cache algorithms This field is put in place to facilitate some third party operating systems SCSI Parity Checking ENABLED Indicates whether the disk array should check SCSI bus parity Disabled means that bus parity checking is disabled Enabled indicates that bus parity checking is enabled SDTR ENABLED Indicates whether the disk array should initiate SDTR Synchronous Data Transfer Request Disabled indicates
65. will be done on it This option provides a mechanism to pass volume set information to a script designed to perform a recovery P S Returns the recoverability status of the disk array as well as the status of a recovery in progress The status indicates if recovery is needed and what percentage of the entire recovery has been completed When used with the c option status will be returned at regular intervals allowing continuous monitoring of the recovery progress v volume set ID Starts a recovery on the volume set indicated by volume set id This option is intended for use only in environments where there are multiple volume sets on the disk array 154 Using the ARM Command Line Utilities for MPE Viewing the Disk Array General Configuration Settings Viewing the Disk Array General Configuration Settings Checking the general configuration settings allows you to view the current settings for all the disk array operating parameters The settings include the current status of the array the array configuration and the SCSI configuration settings A description of each setting is included in Table 9 The default settings have been selected to optimize disk array operation on MPE NOTE Before changing any setting you should understand what the setting does and what effect changing it will have on disk array operation Be aware that using an incorrect setting may make it impossible for the host a
66. 40 1 0 HPC3586 AUTORAID disk array ZPRF 1 0 GB PHYSICAL MECHANISMS WITHIN THE SELECTED ARRAY Test Level FRU Slot Drive State Initialization State 0 CNTRL X 1 ECH A1 Ready Ready 2 ECH B1 Ready Ready 3 ECH A2 Ready Ready 4 ECH B2 Ready Ready 5 ECH A3 Ready Ready 6 ECH B3 Ready Ready 7 ECH 4 Ready Ready 8 ECH B4 Ready Ready Legend TestLevel Indx of the FRU listed This is flagged with a if it is marked for testing 198 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions RANGE The RANGE environment variable allows the user to set the desired range of blocks for testing Synopsis RANGE start end start end Output ARDIAG gt RANGE 300 500 1000 2000 Range 300 500 1000 2000 RDIAG gt RANGE ange 300 500 1000 2000 RDIAG gt RANGE 10 20 30 50 ange 10 20 30 50 Arn Dp ARDIAG gt 199 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions SHOWENV The SHOWENV command allows the user to view the current values of the environment variables TESTDISK TESTLEVEL and RANGE Synopsis SHOWENV Output ARDIAG SHOWENV TESTDISK 0 TESTLEVEL 3 RANGE 200 600 200 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions TESTDISK The TESTDISK environment variable allows the user to reference a particular disk array for test Syno
67. ARM utilities to perform functions such as formatting or diagnostics Disk array control panel An alternative to the online management tool is the disk array control panel Although you can use the disk array control panel to perform most of the management tasks available through the ARM utilities the added convenience and functionality provided by the ARM utilities make them better tools for managing the disk array 116 Managing the HP SureStore E Disk Array 12H on MPE Which Tools to Use for Each Task Which Tools to Use for Each Task The various management tasks are performed using the command line utilities In general the more common tasks are available from the ARM utilities Most tasks can also be performed using the disk array control panel Table 14 Tools and Tasks Tasks 5 Control Utilities Panel Including a disk Yes arraycfg ves Checking disk array status Yes arraydsp Yes Yes mae ve Renumbering LUNs Yes E Yes Downing excluding a disk l Yes araa No Yes anam ves Shutting down the disk array Yes aam Yes Yes eramo Yes Changing SCSI settings araymg Yes Yes Switching primary controller Yes arraymgr Yes Yor ote Wo Displaying disk test results Yes dteststat Displaying disk array serial Yes arraydsp Yes numbers Yes arama Yer
68. Array General Configuration Settings Description Specifies the maximum supported value for the Queue Full Threshold parameter Attempts to set the QFT parameter to values higher than the MQFT will fail with CHECK CONDITION status and ILLEGAL REQUEST INVALID FIELD IN PARAMETER LIST sense indications Determines the values of the following eight parameters This field regulates the mode of data resiliency that the disk array will operate in The Normal mode is set at the factory when dual controllers are ordered with the array Indicates whether the disk array should assert a warning when only one controller is present When enabled the disk array will assert a Single Controller Warning when there are not two controllers present in the array disk array When disabled no Single Controller Warning will be indicated Controls the modification of the Write Cache setting When LCWO is set to true then the host cannot change the Write Cache setting When LWCO is false the host ill be able to change Write Cache setting Enables the disk array response to the Write Cache setting If false NVRAM use does not depend on the Write Cache setting and will be enabled if no other condition inhibits it If true then NVRAM is disabled when Write Cache is disabled Couples the use of NVRAM to the presence of an operational dual controller If true and no operational dual controller is present then NVRAM use is disabled Otherwis
69. Array On MPE Configuration Option Impact on Performance and Capacity Number of disk drives Performance Increase the number of disk mechanisms in the disk array for maximum performance As the number of individual disks is increased the number of potential I Os that can be performed simultaneously increases as well This will improve performance in environments that place heavy I O demand on the disk array Capacity The capacity of a disk array that is fully populated with disk mechanisms can only be increased by replacing lower capacity disks with higher capacity disks Unallocated disk array capacity 120 Performance Unallocated capacity is used as additional RAID 0 1 space The amount of RAID 0 1 space required to maintain optimal performance is determined by the write working set parameter Capacity To create the maximum amount of storage capacity allocate all available capacity to LUNs Managing the HP SureStore E Disk Array 12H MPE Configuring AutoRAID SCSI LUNs for Configuring AutoRAID SCSI LUNs for MPE iX This section will explain how to configure the AutoRAID using the SYSGEN along with a list of configuration requirements Configuration Requirements There are two configuration requirements has that the user should be aware of before starting the SYSGEN configuration e SCSILUN 0 on the AutoRAID should be configured first LUN 0 will be use
70. CSI cabling is installed and connected properly e SCSI bus is terminated properly using differential terminator part number C2905A NOTE f the disk array will be used in a stand alone noncluster environment it is recommended that the A5252A SCSI HBA setting Reset SCSI Bus at IC Initialization be enabled Refer to the Installation Guide included with the HBA for instructions on changing HBA settings 210 Managing the HP SureStore E Disk Array 12H on Windows NT Installing AutoRAID Manager Software Disk Array SCSI Settings The following disk array SCSI settings must be checked and set to the indicated value if necessary set to enabled e SDTR set to enabled Use the disk array control panel is to view and change SCSI settings Refer to the HP SureStore E Disk Array 12H User s and Service Manual for control panel information Logical Drive Configuration NOTE Atleast one logical drive must be created on the disk array before the array can be accessed by the host server During the boot process the server must recognize a logical drive on the disk array to ensure the device driver is loaded properly If there is no logical drive on the disk array the server may hang during the boot process Logical drive 0 must be created on the disk array to ensure that the host recognizes the array Your strategy for creating logical drive 0 will be influenced by the type of installation NT cluster
71. Ck Sk ck Ck KKK KKK KKK KKK ck kk ck kk ck kk Ck kk kk ck kk ck kk ck kk ck kk Sk kc k kk kc KK KKK Time Module Event Event Component ERU Abterm Stamp ID Code Count Specific ID Code Dk kk ck Ck ck ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck Ck ck ck ck ck koc ck ck ck ck ck ck ck ck ck ck ck ck ck ck kk ck ko ck kk ck Mk Sk Mk kx kv Mk ko ko 30367 0x40 0x0 0x87 1 0 7 0 81 NORMAL SYSTEM CHANGE EVENT TABLE ke kk kk ck kk kk ee kk ck kk kk ck ee ee ee ee ck kckckck ck Time Module Event Event FRU Device ID Number Stamp ID Code Count KKKKKKKKKK KKK KKK KKK KKK KK KKK KKK KKK KKK KKK KKK ck ck ck ck kck ck ckck k ck ck ck ck ck k ck ck kck ck k ck 28254 Oxle 0 0 0x56 1 0 0 0 1 30345 Oxe 0x0 Oxbl 1 0x0 0 1 30348 0x30 0x0 0x8a 1 0x0 0 0 30350 Oxe 0x0 Oxbl 1 0x0 0 1 30510 Oxe 0x0 Oxbl 1 0x0 0 1 30515 0x30 0x0 0x8b 1 0x0 0 0 30515 0x30 0 0 0x89 1 0x0 0 0 30515 0x30 0x0 Oxb5 1 0x0 0 0 181 Using the ARDIAG Offline Diagnostic Operational Command Descriptions Example 2 Internal disk drive 182 ARDIAG testlevel 4 TESTLEVEL 4 ARDIAG readlog Contents of drive log at TestLevel Indx 4 Dk ck ck ck Ck ck Ck ck Sk ck ck Ck ck ck Ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck C
72. Ck ck Ck ck ck ck ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck KKK ck ck ck ck ck ck cock KKK kk kk ck ko ck ko Sk kx Mk ko ko ko ko ko Peripheral Qualifier Requested LUN is supported Peripheral Device Type Direct access devic Medium is NOT removable Device type Modifier 0 0 0 ISO Version 0 ECMA Version 0 ANSI Version 2 Device complies to present SCSI 2 Device does NOT support TERMINATE I O PROCESS message INQUIRY data format is as specified in SCSI 2 Other supported features 16 bit wide data transfer Synchronous data transfer Tagged Command Queuing This device responds to a RESET condition with a SOFT RESET alternative Vendor Identification HP Product Identification C3586A Product Revision Level ZPRF Requested Lun is Supported Product Serial Number is ffffffffffffffffffffffffffffffff Controller Pair Serial Number is TITfTtft amp tftfttffftfffftffffttftttftfttftfifftffftfftf Backplane Serial Number is 007870ca0000 Requested Lun is Supported Manufacturing Product Code is ManProdCod Firmware Revision is 5D03159518 Number of Supported Logical Units 32 Lun Capacity Logical Blocks 0 2097152 1 4194304 2 6291456 179 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions READLOG The READLOG command allows the user to read the internal logs of the specified target Synopsis Output READLOG
73. Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck kk kk ck ko Sk ko Sk Sk KKK ko ko ko Do you want to continue y n y FORMATTING Dk ck ck ck Ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ko ck ck ko ck ko ck ko Sk Sk kv k ko ko ko ko FORMAT Successfully Completed ck ck Ck ck ck kk Ck Sk ck Ck ck Ck ck ck kk ck kk Ck kk kk ck kk ck kk KKK Ck ck ck kk ck kk ck kk Sk kc k ko kc KKK KK ARDIAG gt Additional notes 86 Any problems detected by the FORMAT command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions INQUIRY The INQUIRY command allows the user to view SCSI Inquiry data from the specified target Synopsis INQUIRY Output ARDIAG INQUIRY Performing Inquiry at TestLevel Indx 0 Ck ck Ck ck Ck ck ck ck ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck KKK ck ck ck ck ck ck cock KKK kk kk ck ko ck ko Sk kx Mk ko ko ko ko ko Peripheral Qualifier Requested LUN is supported Peripheral Device Type Direct access devic Medium is NOT removable Device type Modifier 0 0 0 ISO Version 0 ECMA Version 0 ANSI Version 2 Device complies to present SCSI 2 Device does NOT support TERMINATE I O PROCESS message INQUIRY data format is as specified in SCSI 2 Other supported featu
74. Descriptions WRTMT The WRTMT command allows the user to perform a destructive write read and verify test on the specified target Synopsis WRIMT Output ARDIAG gt wrtmt You must set the range variable before performing this task SOFTWARE ERROR S DETECTED BY ARDIAG ARDIAG range 0 1000 1200 45000 The selected disk has blocks from 0 to 4165271 RANGE 0 1000 1200 45000 ARDIAG gt wrtmt KKK KK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KK KKK KKK KKK KKK KKK ko ko KKK KK WARNING ki Dk ck ck ck Ck ck ck Ck ck Ck ck ck Ck ck Ck ck Ck ck ck ck ck Sk ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck Ck ck ck ckock kk ck ko ck ko Sk ck Sk Sk kx kx kx Xo This test cannot continue without a loss of redundancy on the AutoRAID Array being tested Failure of a disk before redundancy is restored will result in DATA LOSS Although continuation of this test will NOT result in direct data loss it may take several hours before redundancy is restored and a disk failure can be tolerated without data loss ck ck ck Ck ck Ck ck kk ck Ck ck ck Ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck kk ck ck ck ko ck ck ko ck kk ck Sk Sk kx kx ko KK Do you want to continue y n y The following data pattern options are available 1 Random pattern 0x00 to OxFF 2 User defined 3 Abort the test Choose 1 2 Enter the p
75. E Operational Command Descriptions Any problems detected by the SETOPTIONS command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED Table 13 contains a list of the subsystem states displayed by the SETOPTIONS command and the naming convention used in AutoRAID documents Several of the subsystem state names displayed by the SETOPTIONS command have been abbreviated Table 25 Subsystem States Ready Warning Not Enough Disks No Quorum No Address Table Shutdown Started Shutdown Complete Initializing No Code N A Shutdown Warning N A Firmware Needed Cntrl Mismatch Front Panel Display ARDIAG Set Options Display Ready Warning NtEnDsk NoQuorm Nomap Shtgdwn Shutdwn Strtgup Nocode Resrved E we or Resrved FWneed CTLmis 193 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions SHUTDOWN The SHUTDOWN command allows the user to put the specified disk array into the Shutdown state Synopsis SHUTDOWN Output ARDIAG SHUTDOWN Are you sure that you want to shutdown the array default n Shutting down the array The array at testdisk 0 is now in the shutdown state Additional notes e problems detected by the SHUTDOWN command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 194 Using the ARDIAG Offline Diagnostic MPE Operational Command
76. ERIAL 132 CHANGING DISK ARRAY CONFIGURATION SETTINGS 132 MANAGING LUNS LOGICAL DRIVES essere eren nnne enne rer rennen nenne 134 Checking EUN Configuration te tete t ote EURO DIRE IR FREE EREMO 134 Creating a LUN Ie ener e euet eae e e tie edet dote edes 134 Ren mbermg a LUN ntt eee EEE REI pU RITU Pr EUREN RE Io ea 135 Deleting Edd neue eene 135 ADDING A nent ERE ER eren dates 136 Including Disk UP OO pe eiie ette 136 REBUILDING THE DISK ARRAY v cgcvdaeteoubsute 138 Rebuilding the Disk Array Manually 138 Setting Rebuild Prionty n enonccitonmee ete ue ere p he eren 138 Checking the Progress of a Rebuild 139 Canceling a Rebuild o ote OI ee D uie amt ed 139 ANALYZING DISK ARRAY 1 4 een nennen enne rennen innen ennt 140 Command Examples netter P e ERR eee eee PERRO EHE 140 Selecting a Time Period for Analysis dieere e IE ORTI 141 Checking the Working Set Metric 141 PERFORMING DISK ARRAY MAINTENANCE 5 86 142 Shutting Down the Disk Array EOR
77. HP SureStore E Disk Array 12H System Administrator s Guide for HP UX MPE and Microsoft Windows NT D HEWLETT PACKARD with AutoRAID Technology HP Part Number 5445 90902 Printed in U S A September 1999 Notice Copyright 1997 1998 1999 Hewlett Packard Company The information contained in this document is subject to change without notice Hewlett Packard Company makes no warranty of any kind with regard to this material including but not limited to the implied warranties of merchantability and fitness for a particular purpose Hewlett Packard shall not be liable for errors contained herein or for incidental or consequential damages in connection with the furnishing performance or use of this material This document contains proprietary information which is protected by copyright rights are reserved No part of this document may be photocopied reproduced or translated to another language without the prior consent of Hewlett Packard Print History Edition 1 May 1997 Edition 2 January 1998 Edition 3 May 1998 Edition 4 August 1998 Edition 5 February 1999 Edition 6 June 1999 Edition 7 September 1999 Typographical Conventions NOTE Notes contain important information CAUTION Caution messages indicate procedures which if not observed could result in damage to your equipment or loss of your data WARNING Warning messages indicate procedures or prac
78. Information Changing Disk Array Configuration Settings There are a number of configuration settings that control the operation of the disk array These settings are usually established during installation and once set should rarely have to be changed The default settings have been selected to provide the best operation for most systems However if you determine that any setting does not meet your needs you can easily change it Table 3 lists the various settings that can be changed with SAM including factors you may want to consider before changing them 24 Using the HP UX System Administration Manager Managing LUNs Logical Drives Table 3 Disk Array Configuration Settings Setting Default Rebuild Priority Auto On Include NOTE The Rebuild type automatic or manual is displayed but cannot be changed using SAM Comments and Considerations Active Spare provides optimum protection against disk failure Disabling Active Spare will make additional capacity available to the host but at the expense of maintaining full data redundancy Rebuild Priority determines how quickly a Rebuild completes It allows you to balance the servicing of host l Os with the rebuilding of the disk array The same rebuild priority is used for both Auto Rebuilds and manual Rebuilds Auto Include simplifies the task of adding a new disk to your array Disabling it will require you to manually include each disk you install i
79. KKK KKK ARDIAG KKK KKK KKK KKK KKK KKK EARRAS C copyright Hewlett Packard Co 1996 ARR ERRER E All Rights Reserved ROCK KE kkk kx KKK kk KKK kk KKK kk kk kk WORDEN UE Version XX XX XX ARE KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KEK KKK KKK KKK KK KKK KKK KK Please wait while I scan the device busses Test Disk Path Product String Rev Size 0 56 40 1 0 HPC3586 AUTORAID disk array 2 1 0 Legend TestDisk Indx of the array listed This is flagged with a if it is marked for testing NOTE The size of disk may not match the Information Specified by the vendor due to difference in calculation methods ARDIAG help 205 Using the ARDIAG Offline Diagnostic MPE ARDIAG interface to ODE ARDIAG Utility Help Menu UTILINFO Shows commands that do the most common ARDIAG tasks HELP This menu or use HELP help item for more detailed help DISPMAP Display the disk arrays found DISPMECH Display the selected disk array and internal disks DISPFILES Display the firmware images found DOWNLOAD Download the image file to the desired disk array controller or internal disk TESTDISK Select the array to be tested TESTLEVEL Select the FRU within the disk array to be te
80. KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK Array Controller DOWNLOAD id KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK he following options are available for DOWNLOAD to the array controller 1 Download a firmware image file to the primary controller with an automatic update of same firmware to the secondary controller 2 Update the secondary controller with the same firmware as the primary controller Choose 11 2 The primary controller is at XXXX Do you want to do the update y n y Please be patient The firmware download process may take several minutes WARNING IF YOU INTERRUPT THIS PROCESS THE DEVICE COULD BE RENDERED INOPERATIVE Shutting down the array Mirroring the microcode image Please wait while the array performs a reset Dk ck ck ck Ck ck Ck ck Ck ck ck 0k ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ok kk ck kk ck ko ck ko Sk kv kx kv KKK ko ko ko Firmware Download Successful kk ck Ck ck ck Ck Sk ck Ck ck Ck KKK KKK KKK KKK KKK KKK ck kk ck kk ko ck ck kk Sk kk ko kk ck kk Sk kc k ko kc KKK KK ARDIAG Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Example 2 Internal disk drive ARDIAG testlevel 1 TESTLEVEL 1 ARDIAG download Enter the firmware file name displays the files Default quits File
81. M utilities for HP UX 33 for MPE 125 for NT 241 AutoRAID Manager for Windows NT client 208 failover driver 208 server 208 setting up 214 SNMP agent 208 software components 208 system requirements 210 C canceling a rebuild HP UX using ARM 47 MPE using ARM 139 NT using ARM 252 NT using AutoRAID Manager for Windows 232 changing configuration settings HP UX using ARM 40 HP UX using SAM 24 MPE using ARM 132 NT using ARM 246 NT using AutoRAID Manager for Windows 225 changing SCSI settings HP UX using ARM 55 MPE using ARM 147 NT using ARM 259 checking disk array status HP UX using ARM 39 HP UX using SAM 24 MPE using ARM 131 NT using ARM 244 NT using AutoRAID Manager for Windows 224 checking logical drive configuration NT using ARM 247 checking LUN configuration HP UX using ARM 42 HP UX using SAM 26 MPE using ARM 134 command line help for HP UX 33 for MPE 126 for NT 241 configuration settings HP UX using ARM 40 HP UX using SAM 24 MPE using ARM 132 NT using ARM 246 NT using AutoRAID Manager for Windows 225 configuration settings viewing HP UX using ARM 64 MPE using ARM 155 275 Index Index NT using ARM 264 configuration tips HP UX 17 MPE 119 configuring a new disk array HP UX using ARM 37 HP UX using SAM 23 MPE using ARM 129 NT using ARM 243 NT using AutoRAID Manager for Windows 224 control panel HP UX 14
82. N All information displayed for options l d c 5 v and h c l Array controller status l h Hardware status m Display performance metrics R Rescan for SureStore E AutoRAID disk arrays recognized by the host S Generate raw output used in combination with other options General configuration information For a complete description of all the configuration settings see Viewing the Disk Array General Configuration Settings at the end of this chapter y Capacity information l 39 XN dH Using the ARM Command Line Utilities for HP UX Displaying Disk Array Serial Numbers Displaying Disk Array Serial Numbers The serial numbers of all disk arrays connected to the host can be displayed by typing arraydsp i Missing Disk Arrays If you know there are disk arrays connected to the host but they are not displayed in response to the arraydsp command check the following e Make sure all disk arrays are properly connected to the host This includes proper termination of the SCSI bus e sure all disk arrays are turned on and operating properly e Make sure the ARMServer process is running ARMServer must be running to execute this or any other ARM command You can easily check to see if the ARMServer process is running using the following command ps ef grep ARM e Rescan for disk arrays by typing arraydsp This will update the ARMServer
83. NT using ARM 257 problems NT installation 220 R read verify test HP UX using ARM 52 MPE using ARM 144 NT using ARM 256 278 NT using AutoRAID Manager for Windows 237 rebuilding the disk array HP UX using ARM 46 HP UX using SAM 31 MPE using ARM 138 NT using ARM 251 NT using AutoRAID Manager for Windows 230 recovering maps HP UX using ARM 62 MPE using ARM 153 remote host managing NT using AutoRAID Manager for Windows 219 remote management SNMP agent 208 renumbering a logical drive NT using ARM 248 NT using AutoRAID Manager for Windows 227 renumbering a LUN HP UX using ARM 43 MPE using ARM 135 rescanning for arrays NT using AutoRAID Manager for Windows 219 resetting the disk array HP UX using ARM 51 using ARM 143 NT using ARM 255 NT using AutoRAID Manager for Windows 235 resiliency data HP UX using ARM 56 273 MPE using ARM 148 restarting the disk array HP UX using ARM 50 MPE using ARM 142 NT using ARM 254 NT using AutoRAID Manager for Windows 235 S SAM 19 SCSI HBA 210 SCSI settings checking HP UX using ARM 55 MPE using ARM 147 NT using ARM 259 security on HP UX servers 213 selecting a disk array to manage HP UX using ARM 36 using 128 NT using ARM 242 NT using AutoRAID Manager for Windows 223 serial numbers displaying HP UX using ARM 40 MPE using ARM 132 NT using ARM 245 setting rebuild priority HP UX using ARM
84. No Code N A Shutdown Warning N A Firmware Needed Cntrl Mismatch Front Panel Display ARDIAG Set Options Display Ready Warning NtEnDsk NoQuorm Nomap Shtgdwn Shutdwn Strtgup Nocode Resrved E wwe o Resrved FWneed CTLmis 101 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions SHUTDOWN The SHUTDOWN command allows the user to put the specified disk array into the Shutdown state Synopsis SHUTDOWN Output ARDIAG SHUTDOWN Are you sure that you want to shutdown the array default n Shutting down the array The array at testdisk 0 is now in the shutdown state Additional notes e problems detected by the SHUTDOWN command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED 102 Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions WRTMT The WRTMT command allows the user to perform a destructive write read and verify test on the specified target Synopsis WRIMT Output ARDIAG gt wrtmt You must set the range variable before performing this task SOFTWARE ERROR S DETECTED BY ARDIAG ARDIAG range 0 1000 1200 45000 The selected disk has blocks from 0 to 4165271 RANGE 0 1000 1200 45000 ARDIAG gt wrtmt KKK KK KKK ck Ck KKK KKK KKK KKK KK KKK KKK KKK ck ck ck KKK KKK KKK KK KKK KK KKK KKK KKK KKK KK KKK KK WARNING Dk ck ck ck Ck
85. OS 0 1o 1 10 16 PATH 10 16 LDEV ID A2372 60003 TYPE BC PMGR BUS CONV MGR PMGRPRI 2 LMGR MAXIOS 0 122 Managing the HP SureStore E Disk Array 12H MPE Device Adapter Configuration Configure the HP PB Fast Wide SCSI adapter using the SYSGEN ID of HP28696 io ap 10 16 4 io 1 10 16 4 id hp28696a Configuring AutoRAID SCSI LUNs for PATH 10 16 4 LDEV ID HP28696A TYPE DA PMGR FWSCSI DAM PMGRPRI 6 LMGR MAXIOS 0 SCSI Target Configuration The next item to configure is the SCSI Target ID TID Since the AutoRAID is a multi lun device the TID will be the same for each LDEV connected to any single AutoRAID The SCSI Target will represent the path to the controller that the SCSI cable is connected to NOTE The setting of the initial SCSI TARGET TID and the SCSI LUN values on the AutoRAID can be set by the front panel control See User s and Service Manual for information on how to use the front panel The SCSI Target ID of the controller is represented within SYSGEN using the ID of PSEUDO io 10 16 4 1 id pseudo io 10 16 4 1 PATH 10 16 4 1 LDEV ID PSEUDO TYPE DA PMGR TRANSPARENT MGR PMGRPRI 6 LMGR MAXIOS 0 123 Managing the HP SureStore E Disk Array 12H on MPE Configuring AutoRAID SCSI LUNs for MPE ix LDEV SCSI LUN
86. QUEST INVALID FIELD IN PARAMETER LIST sense indications Determines the values of the following eight parameters This field regulates the mode of data resiliency that the disk array will operate in The Normal mode is set at the factory when dual controllers are ordered with the array Indicates whether the disk array should assert a warning when only one controller is present When enabled the disk array will assert a Single Controller Warning when there are not two controllers present in the array disk array When disabled no Single Controller Warning will be indicated Controls the modification of the Write Cache setting When LCWO is set to true then the host cannot change the Write Cache setting When LWCO is false the host ill be able to change Write Cache setting Enables the disk array response to the Write Cache setting If false NVRAM use does not depend on the Write Cache setting and will be enabled if no other condition inhibits it If true then NVRAM is disabled when Write Cache is disabled Couples the use of NVRAM to the presence of an operational dual controller If true and no operational dual controller is present then NVRAM use is disabled Otherwise NVRAM use does not depend on the presence of the second controller and will be enabled if no other condition inhibits it Couples use of NVRAM to the presence of an operational UPS or BBU At the time of printing this field is reserved for fut
87. R ERR 142 Restarting the DISK ATAY rte reete diee ep eee eee e eade tle ee Nees 142 Resetting the Disk Array sc ineo eR e qs Rete 143 Downing Excluding a Disk eese eene teen enne nnne 143 Testng a Disk E ERE ERR ERE 144 Displaying Test Results t e ert tt e ie E pete 144 Canceling a Disk Test ee emn ene ata ett otio coe tee ie erdt eene 145 Printing ARMServer Log Contents sees 145 Displaying Hardware Logs ether ee Carta t REIR Dp F e 145 Formatting the Disk teniente ti deli re t tpe 146 Changing SCSI Settings eee etie eter e e niae etie ote a ee eire toe 147 Changing the Controller SCSI ID nne nennen trennen eren rene enne enne 147 Switching Primary Controllers eene Er EE nennen nennen 148 Setting Data oe et EE eee avere 148 Creating a Disk Array Alias cipit rt reae Ie ERE HET Lope Ete Ere ARMAR 150 DOWNLOADING FIRMWARE scssscccececsessssscecececeeceessaaececececsensseaecececseseseaecececeesesaaececeescsensaaeceeeeeeseneaaees 151 Firmware Download Procedure etre ette eret eere tee eec che rae Cae E Ce ERES 152 RECOVERING DATA MAPS ete rere cite inre er lentes er a ite 153 VIEWING THE DISK ARRAY GENERAL CONFIGURATION SETTINGS
88. S sr en renes 25 Checking LUN Configuration ote itm Seo ERE SU ERU eoe A E UE ues 26 Creating a LUN eei eoe eter iine Ire ee cio et baise lestie cuoc 26 Deleting a LUN se ione AG ehe edet eben EIC 27 asia eis 28 Including a Disk hie ee eere er e oe end 29 Downing Excluding a Disk nenne ene 30 REBUILDING THE DISK ARRAY eite erret en 31 Setting Rebuild Priory innare tet Hirn DER Rr RE rre ie ti 31 SWITCHING PRIMARY CONTROLLERS eese nennen eene ne rh nn nnne eret tete 32 CHAPTER 3 USING THE ARM COMMAND LINE UTILITIES FOR HP UX ees 33 INFORMATION IN MAN PAGES eee ei ree peace ee eee ed eee 33 COMMAND SYNTAX 8 2 2 212 240400 40020 000000120000000000000000000 34 THEARMSERVER PROCESS tiet Verre 34 THE ARRAY MONITOR DAEMON ARRAYMOND 2 2 2 2 2 0 2 86001 000000000000000000500 35 Routing arraymond Error 35 SELECTING ADISK ARRAY TO MANAGE bee ei e eb ee ee ae Tee 36 CONFIGURING A NEW DISK ARRAY ccsssssscccceesesssaececececc
89. State of Array Overall State of Array READY READY Specifies the current state of the array Active Hot Spare E Indicates whether the array should reserve space within Desired which to perform a rebuild process Auto Include ENABLED Indicates the action to be taken when a drive is physically inserted into the array Auto Rebuild ENABLED Indicates the action to be taken when a drive becomes unusable Indicates the priority the rebuild process is given with to host I O Rebuild Priority Priority HIGH 263 LN Windows NT Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Capacity Depletion 096 Indicates the amount of space below which the disk array Threshold should signal a Capacity Depletion warning For example if this field is set to 9996 then when the disk array reaches 99 capacity a Capacity Depletion warning will be indicated 096 means that Capacity Depletion warnings ill not be issued Write Working Set 8640 ndicates the period ten second intervals over which Interval seconds write performance measurements should be gathered For example 8640 X 10 86400 seconds 24 hours Indicates the language used when displaying information on the front panel Log Full Warning DISABLED Indicates whether the disk array should assert a warning hen some portion of t
90. Status tab NOTE Double clicking on a disk array entry in the Array List window has the same effect as right clicking the array Changing Disk Array Configuration Settings A number of configuration settings control the operation of the disk array These settings are usually established during installation and once set should rarely need to be changed The default settings have been selected to provide the best operation for most systems However if you determine that any setting does not meet your needs you can easily change it Table 26 lists the various settings including factors you may want to consider before changing them Table 26 Disk Array Configuration Settings Setting Default Comments and Considerations Active On Active Hot Spare provides optimum protection against disk Spare failure Disabling Active Hot Spare will make additional capacity available to the host but at the expense of not maintaining data redundancy in the event of a disk failure Auto On Auto Rebuild provides optimum protection against disk failure by Rebuild rebuilding a failed disk as quickly as possible Disabling Auto Rebuild gives you more control over the rebuild process but it can leave the disk array vulnerable to a second disk failure until a rebuild is performed manually Auto On Auto Include simplifies the task of adding a new disk to your Include array Disabling it will require you manually to include each disk you install i
91. These settings include Active Hot Spare Auto Rebuild and Auto Include Change the configuration settings by typing arraymgr h array id Active Spare arraymgr a array id Auto Rebuild arraymgr i array id Auto Include NOTE Only one setting can be changed on each command line 4 Check the available unallocated capacity on the disk array by typing arraydsp SID The total unallocated capacity available for creating LUNs will be displayed Make sure there is adequate capacity to create the LUN structure you need 129 Using the ARM Command Line Utilities for MPE Configuring a New Disk Array 5 Create each LUN on the disk array by typing arraycfg L LUN a capacity array id Example arraycfg L 0 a 1000 00786 5 0000 This command creates LUN 0 with a capacity of 1000 Mbytes on the array identified by serial number 00786B5C0000 This step makes disk array capacity available to your operating system and it must be repeated for each LUN to be created Make sure you observe any operating system limitations on LUN size or number For more information see Creating a LUN in this chapter 130 Using the ARM Command Line Utilities for MPE Checking Disk Array Status Checking Disk Array Status One of the most important management tasks is monitoring the status operation and configuration of the disk array It is important to know how w
92. Using the ARM Command Line Utilities for MPE Recovering Data Maps Recovering Data Maps If the data maps stored in the disk array controller NVRAM are lost the arrayrecover command can be used to reconstruct the maps The data maps are reconstructed using the latest recovery maps stored on two of the disk drives The recovery operation involves performing a parity scan on the contents of the entire disk array to validate the accuracy of the maps and to correct any drive parity inconsistencies This process can take up to several hours depending on the amount of data on the disk array To start the map recovery process type p v volume set ID c array id arrayrecover s 153 Using the ARM Command Line Utilities for MPE Recovering Data Maps The command options for arrayrecover are listed in Table 20 Table 20 arrayrecover Command Options Option C Description Used with other options this will cause continuous polling during the recovery The recovery is monitored and the current progress is displayed at regular intervals Displays the array controller logs on standard output lc Displays the array controller logs on standard output and then clears the logs Begin the recovery in non interactive mode If there are multiple volume sets on the disk array a list of the volume set numbers will be displayed If there is only one volume set on the disk array the recovery
93. Which tools can be used for each management task e Instructions for installing the disk array management software 13 XN dH Managing the HP SureStore E Disk Array 12H on HP UX Disk Array Management Tools Disk Array Management Tools Three disk array management tools are available for managing the disk array two online tools and the disk array control panel System Administration Manager SAM Most of the common tasks involved in managing the disk array can be performed using the HP UX System Administration Manager or SAM This allows you to manage the host system and the disk array using the same tool AutoRAID Management ARM utilities Included with the disk array is a set of ARM utilities which provide advanced capabilities for managing the disk array Although you will primarily use SAM to manage the disk array you will need to use the ARM utilities to perform functions such as formatting or diagnostics Disk array control panel An alternative to the online management tools is the disk array control panel Although you can use the disk array control panel to perform most of the management tasks available through the ARM utilities the added convenience and functionality provided by SAM and the ARM utilities make them better tools for managing the disk array 14 Managing the HP SureStore E Disk Array 12H HP UX Which Tools to Use for Each Task Which Tools to Use for Each Task The various
94. actors e Any size limitations imposed by the operating system for example LVM e Your backup strategy If you do unattended backup to a device such as a tape you may want to avoid creating a LUN that is larger than the capacity of the tape media This allows you to back up an entire LUN without changing tapes e Configuring the LUN into LVM for maximum performance as described in Table 2 NOTE Before creating a LUN check your operating system documentation for any additional information or steps that may be required to create a LUN To create a LUN type arraycfg L LUN a capacity array id LUN must be an unused value between 0 and 7 Capacity must be less than or equal to the currently available unallocated capacity 42 Using the ARM Command Line Utilities for HP UX Managing LUNs Logical Drives NOTE The following commands require LUN exclusive access This means nothing including LVM can access the LUN while the command is running If LVM is used the volume group containing the LUN must be deactivated for the command to run Delete LUN Format LUN Renumber LUN Renumbering a LUN NOTE Before renumbering a LUN check your operating system documentation for any additional information or steps that may be required to renumber a LUN To renumber a LUN type arraycfg L LUN r newLUN array id LUN is the LUN to be renumbered newLUN is a new available LUN number Deleting a LUN When a LUN
95. ad C codefilename array id codefilename identifies the file containing the firmware code To download firmware from the primary controller to the secondary controller type download M array id 151 Using the ARM Command Line Utilities for Downloading Firmware Firmware Download Procedure The procedure described below should be followed to perform the download This procedure works for single or multi host configurations CAUTION In multi host configurations other hosts must not access the disk array while the download is in progress Data can be lost if write requests are made to the disk array while a download is in progress If the system volume set is contained on the array the offline diagnostic ARDIAG must be used to download the firmware NOTE The download process automatically shuts down the array The array Shutdown will not succeed unless all file systems have been unmounted access to the array has been deactivated To download firmware to the array controllers or to the disk mechanism 1 Quiet quiesce or stop all I O traffic to the array 2 Download the controller or disk mechanism firmware 3 Load the controller firmware download C firmware file 000000123456 If the disk array has LUNs that are being used as raw devices for example a database system doing raw I O the array should be disconnected from all systems except the one doing the download 152
96. ailure Auto Rebuild is enabled by default on a new disk array For information on disabling Auto Rebuild see Changing Disk Array Configuration Settings in this chapter Rebuilding the Disk Array Manually If you have chosen to disable Auto Rebuild you will have to start the Rebuild manually The Rebuild will begin immediately and continue to completion If no Rebuild is necessary the command will be ignored To start a Rebuild manually type arrayrbld r array id Setting Rebuild Priority The rebuild priority determines how quickly a Rebuild completes It allows you to balance the servicing of host I Os with the rebuilding of the disk array The same rebuild priority is used for both Auto Rebuilds and manual Rebuilds To set the rebuild priority type arrayrbld P high low array id 46 Using the ARM Command Line Utilities for HP UX Rebuilding the Disk Array Checking the Progress of a Rebuild You can easily check the progress of a Rebuild This allows you to determine approximately when a Rebuild will complete To check the progress of a Rebuild type arrayrbld p array id Canceling a Rebuild A Rebuild can be canceled only if it was started manually A Rebuild that was started by Auto Rebuild cannot be canceled If a Rebuild is canceled it must be started over again and any progress made during the first Rebuild will be lost When canceling a Rebuild the Rebuild may not stop immediately This o
97. ange SCSI settings type arraymgr W T m r onloff array id Changing the Controller SCSI ID Changing the SCSI ID directly impacts the operating system s ability to access the disk array Before performing this task check your operating system documentation for information on how to change the SCSI ID of a disk subsystem To change the controller SCSI ID type arraymgr C X Y addr array id addr is the new address 0 15 for the indicated controller Switching Primary Controllers In dual controller disk array configurations the array automatically switches to the secondary controller if the primary controller fails However you can switch controllers manually if necessary This will cause the secondary controller to assume the role of primary controller To switch primary controllers type arraymgr c X Y array id 55 XN dH Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks Setting Data Resiliency The data map contents stored in the disk array controller NVRAM is copied to two disks on the array at regular intervals to protect against map loss The interval at which the disks are updated with recovery map information is controlled using the data resiliency setting Selecting a data resiliency setting involves making a tradeoff between data protection and performance The more frequently the recovery maps are updated the more
98. any maintenance The Shutdown process copies vital data mapping information from the controller NVRAM to the disks This protects the data mapping information should the contents of the NVRAM be lost or corrupted due to battery failure Shutdown then takes the disk array off line making all data unavailable to the host The disk array can still be managed and tested but all data is inaccessible while the disk array is Shutdown Shutdown is initiated automatically each time the disk array is turned off using the power switch so it is usually not necessary to initiate a Shutdown using the ARM utility To Shutdown the disk array type arraymgr s shut array id Restarting the Disk Array Following Shutdown the disk array can be brought back on line by performing a restart This makes the data on the disk array available to the host once again To restart the disk array type arraymgr s start lt array id gt 50 Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks After the array completes initialization file system and LVM logical volumes mapped to the disk array must be mounted for HP UX to access the disk array Resetting the Disk Array The disk array can be reset if there is a problem with SCSI channel communication A reset will interrupt access to the disk array temporarily so it should be done only when attempting to solve a problem with the disk array To reset the disk array type
99. arameter Setting Simplified Resiliency Setting Results Single Controller ENABLED Single Controller Dual Controller Warning Lock Write Cache On TRUE Not Supported Data map and write cache Disable NVRAM on FALSE information is scheduled WCE False to flushed to the disk Disable NVRAM with TRUE every four seconds One Disable NVRAM on FALSE UPS absent Force Unit Access 2 Response Disable Read Hits FALSE Resiliency Threshold 4 SingleController Mode Default Parameter Setting Simplified Resiliency Setting Results Single Controller DISABLED Single Controller Dual Controller Warning Lock Write Cache FALSE Data map and write cache Not Supported Disable NVRAM on TRUE Jinformation will be flushed WCE False to the disk once every second Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Secure Mode Default Parameter Setting Simplified Resiliency Setting Results KT CE Single Controller ENABLED Single Controller Dual Controller Warning Lock Write Cache On FALSE Not Supported Data map and write cache Disable on TRUE will be 24 WCE False to the disk once every Disable NVRAM with second One Disable FALSE UPS absent Force Unit Access 2 Response Disable Read Hits FALSE Resiliency Threshold 1 HighPerformance Mode Default Parameter S
100. are Downloading Firmware The download command copies new firmware code to the controller s or individual disk mechanisms in the disk array Firmware also can be copied from a primary array controller to a secondary controller The ARMServer and arraymond processes continue to run while the download is in progress However these processes may not be able to access the array during the download because the download operation shuts down the disk array This may cause warning messages such as the following to be displayed access error Unable to get status from disk array on raw disk file name at machine name These warning messages can be ignored CAUTION The firmware code is copied from the primary controller to the secondary controller so before performing this task make sure the controller that has the desirable code is designated the primary controller See Switching Primary Controllers for instructions on changing the state of the controller To download firmware to a disk type download D slot codefilename array id slot identifies the cabinet shelf containing the disk drive codefilename identifies the file containing the firmware code To download firmware to the disk array controllers type download C codefilename array id codefilename identifies the file containing the firmware code To download firmware from the primary controller to the secondary controller type download
101. array from the Detail pane 228 Using AutoRAID Manager for Windows NT Adding a Disk 3 If Auto Include is on the disk is automatically added to the array and you can proceed to step 4 If Auto Include is off manually include the disk as described in Including a Disk Manually 4 Check the disk array configuration and verify that the disk has been added to the unallocated capacity To add a disk to enable Active Hot Spare 1 Make sure the new disk has been installed in the array cabinet 2 Click the disk array from the Detail pane 3 f Auto Include is on the disk is automatically added to the array and you can proceed to step 4 If Auto Include is off manually include the disk as described in Including a Disk Manually 4 Click Properties 5 Click the Configure tab 6 Select the Active Spare Enable setting Including a Disk Manually A disk must be included before it can be used by the disk array There are two ways to include a disk e Enable Auto Include which automatically includes disk when it is installed in the array cabinet e Disable Auto Include and manually include each new disk A blue status indicator in the Status window identifies any disks that are not included Because of the convenience it provides Auto Include is enabled by default However you can disable Auto Include and include new disks manually if you choose To manually include a disk This procedure assumes that Auto Include has be
102. ary controller See Switching Array Controllers for instructions on changing the state of the controller To copy the firmware code from the primary controller to the secondary controller 1 Gather the following information e The ID of the disk array The disk array ID can be displayed using the arraydsp i command 2 Froma command prompt type in the following command download M lt array id gt 262 Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Viewing the Disk Array General Configuration Settings Checking the general configuration settings allows you to view the current settings for all the disk array operating parameters The settings include the current status of the array the array configuration and the SCSI configuration settings A description of each setting is included in Table 30 The default settings have been selected to optimize disk array operation NOTE Before changing any setting you should understand what the setting does and what effect changing it will have on disk array operation Be aware that using an incorrect setting may make it impossible for the host and the disk array to communicate properly To view the general configuration settings type arraydsp s lt array id gt A complete list of the settings will be displayed Table 30 General Configuration Settings X Parameter Default Description Setting Overall
103. ata on the disk when it is downed To test a disk 1 Right click the disk array from Detail pane 2 Click Properties 3 Click the Diagnostics tab 236 Using AutoRAID Manager for Windows NT Performing Disk Array Maintenance Tasks Click on the disk you want to test Click Test Select Read Verify or Write Read Enter the percent of the disk media you want to test m XS Su If you are performing a Write Read test you must first Down the disk The diagnostic test will begin and a progress indicator will be displayed in the Test Status box When the test concludes the results will be displayed To continue using a disk following a Write Read test it must be manually included back in the array configuration See Including a Disk Manually for more information Downing a Disk If you must remove a disk from the array for testing or replacement the disk should be downed before removing it from the array enclosure In some situations downing a disk may cause a loss of data redundancy or even a loss of user data You must decide if you are willing to accept either of these conditions before downing the disk In addition if Auto Rebuild is enabled the disk array will begin rebuilding the data on the downed disk immediately If you do not want this to happen you must disable Auto Rebuild before downing the disk See Rebuilding the Disk Array for information on disabling Auto Rebuild To down a disk Right click t
104. attern in hex format 0 to FF A5 dd The write data pattern will be 0xDD Testing block 0 percent complete 1005 Test complete 195 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Testing subrange 1000 1200 percent complete 1 oe oe oe oe oe oe F2 o0 OY O1 iS CO PO DTO GO 5 1 complete Testing block 45000 percent complete 1005 Test complete ARDIAG 196 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions Interface command descriptions DISPMAP The DISPMAP command allows the user to display all available disk arrays on the system Synopsis ARDIAG Output ARDIAG dispmap Test Disk Path Product String Rev Size 0 56 40 1 0 HPC3586 AUTORAID disk array ZPRF 1 0 GB Legend TestDisk Indx of the array listed This is flagged with a if it is marked for testing NOTE he size of disk may not match the Information Specified by the vendor due to difference in calculation methods a 1 197 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions DISPMECH The DISPMECH command allows the user to display physical mechanisms within a selected disk array Synopsis DISPMECH Output ARDIAG DISPMECH Path Product String Rev Size 56
105. available to the host but at the expense of maintaining full data redundancy Auto On a Auto Rebuild provides optimum protection Rebuild against disk failure by rebuilding a failed disk as quickly as possible Disabling Auto Rebuild gives you more control over the rebuild process but it can leave the disk array vulnerable to a second disk failure until a Rebuild is performed manually Auto On i Auto Include simplifies the task of adding Include a new disk to your array Disabling it will require you manually to include each disk you install in the array Rebuild High p Rebuild priority determines how quickly a Priority Rebuild operation will complete To change Active Spare Auto Rebuild or Auto Include settings type arraymgr h a i onloff array id To change Rebuild Priority setting type arraymgr high low array id 133 Using the ARM Command Line Utilities for MPE Managing LUNs Logical Drives Managing LUNs Logical Drives An important part of managing the disk array involves defining and maintaining the optimal LUN structure for your system Your system requirements and limitations will influence the LUN structure you choose Managing LUNs is a part of the overall task of managing disk array capacity For more information on managing disk array capacity to meet your system needs refer to the HP SureStore E Disk Array 12H User s and Service Manual C
106. ay The default settings listed in Table 30 have been chosen to work with all supported SCSI adapters and in most cases should not be changed NOTE Before changing any SCSI setting you should understand what the SCSI setting does and what effect changing it will have on disk array operation Be aware that using an incorrect SCSI setting may make it impossible for the host and disk array to communicate properly In this case it will be necessary to use the disk array control panel to return the SCSI setting to its original value to reestablish communication To change SCSI settings type arraymgr W T m onloff array id Changing the Controller SCSI ID Changing the SCSI ID directly impacts the operating system s ability to access the disk array Before performing this task check your operating system documentation for information on how to change the SCSI ID of a disk subsystem To change the controller SCSI ID type arraymgr C X Y addr array id addr is the new address 0 15 for the indicated controller Switching Primary Controllers In dual controller disk array configurations the array automatically switches to the secondary controller if the primary controller fails However you can switch controllers manually if necessary This will cause the secondary controller to assume the role of primary controller To switch primary controllers type arraymgr c X Y array id
107. ay id Setting Rebuild Priority The rebuild priority determines how quickly a rebuild completes It allows you to balance the servicing of host I Os with the rebuilding of the disk array The same rebuild priority is used for both auto rebuilds and manual rebuilds To set the rebuild priority type arrayrbld P high low array id 251 LN Windows NT Using the ARM Command Line Utilities for Windows NT Rebuilding the Disk Array Checking the Progress of a Rebuild You can easily check the progress of a rebuild This allows you to determine approximately when a rebuild will complete To check the progress of a rebuild type arrayrbld p array id Canceling a Rebuild A rebuild can be canceled only if it was started manually A rebuild that was started by Auto Rebuild cannot be canceled If a rebuild is canceled it must be started over again and any progress made during the first rebuild will be lost When canceling a rebuild the rebuild may not stop immediately This occurs if the disk array is busy servicing higher priority I O requests from the host The rebuild will be canceled when the disk array has serviced all higher priority commands To cancel a rebuild type arrayrbld c array id 252 Using the ARM Command Line Utilities for Windows NT Analyzing Disk Array Performance Analyzing Disk Array Performance Disk array performance can be checked using the arraydsp command Howev
108. aymgr command provides additional capability for managing some aspects of disk array behavior e The arraydsp command allows you to check all aspects of disk array operation and configuration Each of the command line utilities is described in this chapter LUNs and Logical Drives To maintain consistency with ARM command line terminology the term LUN is frequently used to refer to a disk array logical drive The two terms are used interchangeably and refer to the same logical entity on the disk array Online Help Detailed online help is available for all the command line utilities To view the help type armhelp from the NT command prompt or from the START menu select Programs AutoRAID Manager Command Line help 241 1 Windows NT Using the ARM Command Line Utilities for Windows NT Command Syntax Conventions Command Syntax Conventions The following symbols are used in the command descriptions and examples Symbol Meaning lt gt Integer value whose units are not defined Exclusive OR Exactly one of the parameters displayed will be used Items enclosed are optional Items enclosed are required Selecting a Disk Array to Manage When using the ARM utilities the lt array id gt field is used to identify the disk array The array id field contains the disk array serial number For example assume a disk array has serial number of 00786B5C0000 To check the available unallo
109. be accessed by the host During the boot process the host server must recognize a logical drive on the disk array to ensure the HBA device driver is loaded properly If there is no logical drive on the disk array the host may be unable to access with the array A logical drive can be created using the disk array control panel as described in the HP SureStore E Disk Array 12H User s and Service Manual DiscoveringDisk Arrays If the disk array configuration on your network changes the AutoRAID Manager may be unaware of it For example if a disk array is added to or removed from a server the change may not be reflected in the Array List In this case it will be necessary to search the network to update the information displayed by the AutoRAID Manager To discover disk arrays 1 From HP AutoRAID Manager s View menu click Select Hosts 2 Click Discover 3 Click Automatic Discovery 4 The AutoRAID Manager will search the local network for disk arrays and update the Known Hosts list with the new information 5 Check the Known Hosts list and add any new disk arrays to the Selected Hosts list if the view is not configured for All Known Hosts Managing Disk Arrays on a Remote Host The discovery feature described in the preceding procedure locates disk arrays connected to hosts on the local server Local Hosts view as well as remote hosts Known Hosts view If a specific host is not displayed or you do not wish
110. ble in SAM can also be performed using the ARM utilities These tasks are described on the following pages It is recommended that these tasks be performed using SAM to ensure predictable results and proper operation of the disk array You should consider using an ARM utility for these tasks only if you clearly understand how the utility works and what effect it has on disk array operation Improper use of the ARM utilities can cause undesirable results including loss of data LUNs and Logical Drives To maintain consistency with HP UX terminology the term LUN is used to refer to a disk array logical drive The two terms are used interchangeably and refer to the same logical entity on the disk array Information in man pages The procedures in this chapter summarize the use of the ARM utilities Detailed information about the ARM command line utilities and their proper usage is included in the HP UX operating system man pages A man page also exists for ARMServer the server portion of the disk array management software To access HP UX man pages information type man lt commandname gt Substitute one of the following ARM utility names for commandname ARMServer arraycfg arraydsp 33 XN dH Using the ARM Command Line Utilities for HP UX Command Syntax Conventions arrayfmt arraylog arraymgr arrayrbld arrayrecover download drivetest dteststat logprint Command Syntax Conventions The following symbols
111. c on HP UX Operational Command Descriptions Synopsis DESCRIBE Output ARDIAG DESCRIBE Configuration information for the controller in slot X Component Instance State Fan F1 Good Fan F2 Good Fan F3 Good Power Supply P1 Good Power Supply P2 Good Power Supply P3 Good Controller X Good Controller Y Good Controller X Battery 1 Good Controller X Battery 2 Good Controller Y Battery 1 Good Controller Y Battery 2 Good Memory Instance Size State Controller X DRAM 1 32MB Good Controller X NVRAM 1 32MB Good Controller X NVRAM 2 32MB Good Controller Y DRAM 1 32MB Good Controller Y NVAM 1 32MB Good Controller Y NVRAM 2 32MB Good Total Memory 192MB Test Level FRU Slot State Warning Indications 0 CNTRL Y Warning Redundancy Loss 1 MECH Al Included 2 Bl Downed 3 MECH A2 Failed 4 MECH B2 Included 5 MECH Included 79 XN dH Using the ARDIAG Offline Diagnostic on HP UX Operational Command Descriptions 6 MECH B3 Included ECH A4 Included 8 MECH B4 Included Lun Capacity GB GB GB GB GB MB oun UU oO 80 Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions DOWNLOAD The DOWNLOAD command allows the user to download firmware to the specified target s Synopsis DOWNLOAD Additional information DOWNLOAD to the array controller The DOWNLOAD process must be performed using the primary co
112. cated capacity on this particular disk array you would use the following command arraydsp V 00786B5C0000 The serial number of all arrays connected to the host can be obtained using the command arraydsp i For more information see Displaying Disk Array Serial Numbers in this chapter Many commands affect the operation of the entire disk array Commands that involve only a specific logical drive on the array will include an option L LUN for identifying the logical drive involved For example to format logical drive 3 on a disk array with serial number 00786B5C0000 use the following command arrayfmt L 3 00786B5C0000 242 Using the ARM Command Line Utilities for Windows NT Configuring a New Disk Array Configuring a New Disk Array After installing a new disk array you can perform the initial configuration using the ARM utilities This establishes the operating environment for the disk array To configure a new disk array 1 Plan your capacity management strategy and logical drive configuration Decide how you want to use the disk array capacity Factors such as data redundancy and performance influence how you manage the capacity See Managing the Disk Array Capacity in the HP SureStore E Disk Array 12H User s and Service Manual for help in planning your strategy 2 Display the serial number of all disk arrays by typing arraydsp i The serial number provides the means to identify disk arrays when using the ARM uti
113. ccurs if the disk array is busy servicing higher priority I O requests from the host The Rebuild will be canceled when the disk array has serviced all higher priority commands To cancel a manual Rebuild type arrayrbld c array id 47 XN dH Using the ARM Command Line Utilities for HP UX Analyzing Disk Array Performance Analyzing Disk Array Performance The disk array monitors and stores a number of metrics that reflect how the disk array is performing The disk array management software periodically retrieves these performance metrics and stores them for your viewing The ARM software can also analyze the performance metrics to identify any potential performance problems Based on this analysis recommendations are made on how to improve disk array performance Checking the metrics regularly is a quick and easy way for you to monitor the performance of the disk array and identify any problems that may be developing You may choose to display only the recommendations or you may want to view the metrics for further analysis For a complete description of each performance metric refer to the HP SureStore E Disk Array 12H User s and Service Manual You can also view arraydsp man page for a complete explanation of all the performance command options To analyze disk array performance type arraydsp r stime etime 4 m stime etime int array id Command Examples The following command display
114. changing tapes NOTE Before creating a logical drive check your operating system documentation for any additional information or steps that may be required to create a logical drive To create a logical drive type arraycfg L LUN a Capacity array id LUN must be an unused value between 0 and 7 Capacity must be less than or equal to the currently available unallocated capacity 247 LN Windows NT Using the ARM Command Line Utilities for Windows NT Managing Logical Drives Renumbering a Logical Drive NOTE Before renumbering a logical drive check your operating system documentation for any additional information or steps that may be required to renumber a logical drive To renumber a logical drive type arraycfg L LUN r newLUN array id LUN is the logical drive to be renumbered newLUN is a new available logical drive number Deleting a Logical Drive When a logical drive is deleted its capacity is returned to the pool of unallocated capacity space Deleting a logical drive is a good way of freeing up capacity for the Active Hot Spare or for simply adding more unallocated capacity to improve disk array performance CAUTION AII data on a logical drive is lost when it is deleted Make sure you backup any important data on the logical drive before deleting it NOTE Before deleting a logical drive check your operating system documentation for any additional information or steps that may be
115. ck ck Ck ck CC ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ckock ck ko ck kk ck ko Sk ke kx Sk Mk ko ko ok This test cannot continue without a loss of redundancy on the AutoRAID Array being tested Failure of a disk before redundancy is restored will result in DATA LOSS Although continuation of this test will NOT result in direct data loss it may take several hours before redundancy is restored and a disk failure can be tolerated without data loss ck ck ck Ck ck Ck ck kk ck Ck Ck ck Ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ko ck ko Sk ck Sk Sk kx Do you want to continue y n y The following data pattern options are available 1 Random pattern 0x00 to OxFF 2 User defined 3 Abort the test Choose 1 2 Enter the pattern in hex format 0 to FF A5 dd The write data pattern will be 0xDD Testing block 0 percent complete 1005 Test complete 103 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Testing subrange 1000 1200 percent complete 1 oe oe oe oe oe oe F2 O0 OY O1 iS x 1 complete Testing block 45000 percent complete 1005 Test complete ARDIAG gt 104 Using th
116. ck kk Sk kc KKK KKK KK ARDIAG Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions FORMAT The FORMAT command allows the user to format the specified target Synopsis FORMAT Output Example 1 Controller ARDIAG TESTLEVEL 0 ARDIAG FORMAT kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk ck Ck KKK KKK KKK KKK ck kk ck kk ck kk Ck kk kk Sk kk Sk kk ck kk ko kc k ko KKK KKK p Array Subsystem FORMAT ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck kk Ck kk kk Sk Sk Ck Ck Sk kk kk kk ko ko NOTICE Use of this command is not recommended except where a full subsystem backup of user data is available or when a new uninitialized subsystem is being prepared for use a precaution this command will NOT allow a format to an array containing any LUNs The DELETELUN command must be used to remove all existing LUNs Dk Ck ck ck Ck ck Ck ck Ck ck ck 0k ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck kk ck ck ck ko Sk ke kx kv Mk ko ko ko ko ko Do you want to continue y n y Checking the array subsystem for LUNs There are no LUNs presently configured on the selected array selected array subsystem is allowed a format Do you want to continue y n y FORMATTING kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk ck Ck ck ck kk ck kk ck kk Ck kk Ck ck ck kk ck kk Ck kk kk Sk kk ck kk ck
117. create a disk array alias type arraymgr D alias name array id alias name can be up to 12 characters in length and can include upper case letters numbers spaces number sign ff underscore and period Recover If the disk array is not shutdown properly it is possible that the data maps in NVRAM memory will be lost For this reason the disk array allows the data maps to be periodically written to the disk drives If the maps are lost an error code such as Address Table will appear on the display If your disk array is a boot device you may have to recover the maps by using the front panel command called Recover under the Cntrl Changes menu The Recover command is only supported in controller firmware versions later than HP40 and any patch delivered after IPR9808 release 273 IN SMOpUIM Index A adding a disk HP UX using ARM 44 HP UX using SAM 28 MPE using ARM 136 NT using ARM 249 NT using AutoRAID Manager for Windows 228 alias creating HP UX using ARM 58 274 MPE using ARM 150 analyzing disk array performance HP UX using ARM 48 MPE using ARM 140 NT using ARM 253 NT using AutoRAID Manager for Windows 232 ARDIAG 73 165 ARMServer log contents printing HP UX using ARM 53 using ARM 145 NT using ARM 257 ARMServer process 34 127 array monitor daemon 35 arraymond 35 arrayrecover 58 150 274 AutoRAID Manager access on HP UX servers 213 AutoRAID Manager AR
118. d This is flagged with a if it is marked for testing NOTE The size of disk may not match the Information Specified by the vendor due to difference in calculation methods ARDIAG help 113 XN dH Using the ARDIAG Offline Diagnostic on HP UX ARDIAG interface to ODE ARDIAG Utility Help Menu UTILINFO Shows commands that do the most common ARDIAG tasks HELP This menu or use HELP help item for more detailed help DISPMAP Display the disk arrays found DISPMECH Display the selected disk array and internal disks DISPFILES Display the firmware images found DOWNLOAD Download the image file to the desired disk array controller or internal disk TESTDISK Select the array to be tested TESTLEVEL Select the FRU within the disk array to be tested RANGE Select range of blocks to be tested by ROMT and WRTMT SHOWENV Display the current settings for TESTDISK and RANGE READLOG Read the internal logs of the desired disk array controller or internal disk CLRLOG Clear the internal logs of the desired disk array controller or internal disk REQSENSE View SCSI Sense data of the desired disk array controller or internal disk INQUIRY View SCSI Inquiry data of the desired disk array controller or internal disk DESCRIBE View Configuration data of the d
119. d by the ARM utility for communications with the AutoRAID e SCSILUN 0 should have the lowest LDEV number configured compared to the LDEVs assigned to SCSI LUN 1 through 7 For example LDEV 30 SCSI LUN 0 LDEV 32 SCSI LUN 1 LDEV 33 SCSI LUN 3 LDEV 35 SCSI LUN 4 LDEV 37 SCSI LUN 5 LDEV 38 SCSI LUN 6 LDEV 39 SCSI LUN 7 NOTE The LDEV assignment does not have to be contiguous but only the lowest LDEV should be assigned to the lowest number SCSI LUN Configuration to a non existent SCSI LUN is not recommended 121 Managing the HP SureStore E Disk Array 12H on MPE Configuring AutoRAID SCSI LUNs for MPE ix MPE iX SYSGEN Configuration The following example illustrates how to configure an MPE iX logical device LDEV to a SCSI LUN It assumes the configuration of an LDEV to a physical I O path of 10 16 4 1 0 NOTE To obtain the path use the offline utility ODE and run mapper To configure any device an system use the SYSGEN configuration from the MPE iX system prompt as follows SYSGEN Within the IO area of SYSGEN configure the upper level hardware first For example a 969K 5 100 has two levels of bus converters which appear in the path prior to specifying the HP PB slot number The following is a SYSGEN list path Ip from the io prompt io 10 PATH 10 LDEV ID A2372 6003 TYPE BC PMGR BUS CONV MGR PMGRPRI 2 LMGR MAXI
120. drive The AutoRun feature will automatically start the setup program if AutoRun is not enabled you must run setup exe manually As you proceed through the setup select the following options when prompted e Select the appropriate system configuration option tis recommended that you use the default locations for the AutoRAID Manager files View the README file for the latest information on AutoRAID Manager e Share the AutoRAID Administrator folder to allow remote clients to manage the disk array For more information see Controlling Access to the Disk Arrays e The install program will also run the Firmware Download Utility Make sure firmware revision HP54 is installed on the disk array If another version of firmware is installed on the disk array download HP54 to the disk array Managing the HP SureStore E Disk Array 12H Windows NT Installing AutoRAID Manager Software CAUTION Firmware version HP54 is required for proper operation on Windows NT Using another version of firmware could result in improper system operation The README file contains information regarding current firmware for the disk array When you have performed all the tasks required to complete setup continue with Setting Up AutoRAID Manager for Windows Controlling Access to the Disk Arrays CAUTION Access to AutoRAID Manager should be restricted to those involved in managing network resources AutoRAID Manager provides capabilities that if
121. e Changing Disk Array Configuration Settings in this chapter Including a Disk A disk must be included in the disk array configuration before it can be used by the disk array There are two ways to include a disk 136 Using the ARM Command Line Utilities for MPE Adding a Disk e You can enable Auto Include which will automatically include a disk when it is inserted into the disk array enclosure e You can manually include each new disk For convenience Auto Include is enabled by default on a new disk array For information on disabling Auto Include see Changing Disk Array Configuration Settings in this chapter After including a disk you must decide how you want to use it For more information see the preceding section Adding a Disk To manually include a disk type arraycfg D slot a array id slot is the cabinet shelf containing the disk drive A1 through A6 or B1 through B6 137 Using the ARM Command Line Utilities for MPE Rebuilding the Disk Array Rebuilding the Disk Array To maintain data redundancy in the event of a disk failure it is important to rebuild the disk array as quickly as possible Auto Rebuild does this automatically and it is enabled by default For optimum data protection it recommended that Auto Rebuild remain enabled If you want more control over the Rebuild process you can disable Auto Rebuild This will allow you to manually start a Rebuild at the
122. e ARDIAG Offline Diagnostic HP UX Interface command descriptions Interface command descriptions DISPMAP The DISPMAP command allows the user to display all available disk arrays on the system Synopsis ARDIAG Output ARDIAG dispmap Test Disk Path Product String Rev Size 0 56 40 1 0 HPC3586 AUTORAID disk array ZPRF 1 0 GB Legend TestDisk Indx of the array listed This is flagged with a if it is marked for testing NOTE he size of disk may not match the Information Specified by the vendor due to difference in calculation methods a 1 105 XN dH Using the ARDIAG Offline Diagnostic on HP UX Interface command descriptions DISPMECH The DISPMECH command allows the user to display physical mechanisms within a selected disk array Synopsis DISPMECH Output ARDIAG DISPMECH Path Product String Rev Size 56 40 1 0 HPC3586 AUTORAID disk array ZPRF 1 0 GB PHYSICAL MECHANISMS WITHIN THE SELECTED ARRAY Test Level FRU Slot Drive State Initialization State 0 CNTRL X 1 ECH A1 Ready Ready 2 ECH B1 Ready Ready E3 ECH A2 Ready Ready 4 ECH B2 Ready Ready 5 ECH A3 Ready Ready 6 ECH B3 Ready Ready 7 ECH A4 Ready Ready 8 ECH B4 Ready Ready Legend TestLevel Indx of the FRU listed This is flagged with a if it is marked for testing 106 Using the ARDIAG Offline Diagnostic HP UX Interface command de
123. e NVRAM use does not depend on the presence of the second controller and will be enabled if no other condition inhibits it Couples use of NVRAM to the presence of an operational UPS or BBU At the time of printing this field is reserved for future functionality with UPS and is disabled in all resiliency modes 159 Using the ARM Command Line Utilities for MPE Viewing the Disk Array General Configuration Settings Parameter Force Unit Access Response Disable Read Hits Resiliency Threshold Default Setting 2 FALSE 4 Description Controls the array s response to the FUA bit The FUA bit is a command from the host when Write Cache is enabled It gives the host an opportunity to flush write cache on command If this field is 0 then the FUA bit is ignored If this field is 1 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed If this field is 2 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed along with the map journal before completing the write request If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored Controls the array s ability to satisfy read commands from rite cache contents during FUA processing If NVRAM use is disabled due to the chosen map resiliency mode
124. e a disk e You can enable Auto Include which will automatically include a disk when it is inserted into the disk array enclosure e You can manually include each new disk For convenience Auto Include is enabled by default on a new disk array For information on disabling Auto Include see Changing Disk Array Configuration Settings in this chapter After including a disk you must decide how you want to use it For more information see the preceding section Adding a Disk To manually include a disk type arraycfg D slot a array id slot is the cabinet shelf containing the disk drive A1 through A6 or B1 through B6 45 XN dH Using the ARM Command Line Utilities for HP UX Rebuilding the Disk Array Rebuilding the Disk Array To maintain data redundancy in the event of a disk failure it is important to rebuild the disk array as quickly as possible Auto Rebuild does this automatically and it is enabled by default For optimum data protection it recommended that Auto Rebuild remain enabled If you want more control over the Rebuild process you can disable Auto Rebuild This will allow you to manually start a Rebuild at the time you choose A Rebuild impacts disk array performance while it is in progress so before starting a Rebuild make sure the appropriate Rebuild Priority is set See Setting Rebuild Priority in this chapter For convenience and maximum protection in the event of a disk f
125. e disk array will normally not let you down a disk if doing so would result in loss of data redundancy or data unavailability However you can override this protection by specifying the appropriate Exclusion Restriction You can choose to down the disk even if a loss of redundancy would result but not data unavailability Or you can down the disk even if data unavailability would occur NOTE Two disks on the disk array are used to store the information for recovering data maps if they are lost The disk array will not let you down one of these RDM disks To down a disk 1 Select the desired disk array controller from the list on the screen Select Actions on the menu bar Select Disk Array Maintenance Select Exclude Disk Select the disk to exclude Select the desired Exclusion Restriction cA SON coms COM Select Assume Valid Data Next Time to make the exclude nondestructive If this option is not selected the exclude will be destructive 8 Click OK to effect the change or click Apply to select another disk to exclude After testing a downed disk can be returned to the disk array configuration by manually including it For more information see the preceding section Including a Disk 30 Using the HP UX System Administration Manager Rebuilding the Disk Array Rebuilding the Disk Array To maintain data redundancy in the event of a disk failure it is important to rebuild the
126. e is a problem with SCSI channel communication A reset will interrupt access to the disk array temporarily so it should be done only when attempting to solve a problem with the disk array To reset the disk array type arraymgr R array id Downing Excluding a Disk Downing or excluding a disk is typically done in preparation for testing the disk After the disk is downed testing can be done without impacting disk array operation If testing reveals that the disk is good the disk can be included back in the array configuration Downing a disk has the same effect as if the disk failed or was physically removed from the cabinet If Auto Rebuild is enabled the disk array will immediately begin a Rebuild when a disk is downed The v option identifies the down operation as either destructive or nondestructive This determines whether the disk array will assume there is any valid data on the disk if it is returned to the array configuration If a destructive down is performed not specified the disk array will assume no data on the disk is valid If the down is nondestructive v specified the array will assume any data on the disk that was not updated in the disk s absence is valid To down a disk type arraycfg D slot d v R 2 array id To protect data availability the disk array will not let you down a disk if doing so would result in loss of data redundancy or data unavailability However you can overrid
127. e is submitted to the scheduler for implementation Depending on the priority of other events in the scheduler the desired time may not be implemented If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored The value of this field is not the same as disabling NVRAM use A value of Oh indicates that no maximum time will be enforced and that no recovery image will be available 69 XN dH Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Simplified Resiliency Setting The simplified resiliency setting is derived from several other fields Collectively these settings control how data resiliency is managed Table 10 identifies the field settings for each of the resiliency modes There are four modes of data resiliency Normal SingleController Secure and HighPerformance The default setting is Normal for dual controller disk arrays If the disk array has been ordered with a single controller the default for this settings is SingleController For information on changing the data resiliency mode see Setting Data Resiliency in this chapter 70 Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Table 10 Simplified Data Resiliency Settings Parameter Single Controller Warning Lock Write Cache On Disable NVRAM on WCE False Disable NVRAM with One
128. e kk ck ke kk ee ee ee ee ckck ck kck ck kck ck ck ck ck ck kckckck ck Time Module Event Event FRU Device ID Number Stamp ID Code Count KKKK ck Ck ck KKK ckck ckckck ckckck kckck ckckckckck kckck ck ck ck ckckck ckck ck ckck ck ckck ck ck ck ckck ck ckck k ck ck ck ck ck k ck ck kck ck k ck kk kk 28254 Oxle 0 0 0x56 1 0 0 0 1 30345 Oxe 0x0 Oxbl 1 0x0 0 1 30348 0x30 0x0 0x8a 1 0x0 0 0 30350 Oxe 0x0 Oxbl 1 0x0 0 1 30510 Oxe 0x0 Oxbl 1 0x0 0 1 30515 0x30 0x0 0x8b 1 0x0 0 0 30515 0x30 0 0 0x89 1 0x0 0 0 30515 0x30 0x0 Oxb5 1 0x0 0 0 89 XN dH Using the ARDIAG Offline Diagnostic on HP UX Operational Command Descriptions Example 2 Internal disk drive 90 ARDIAG testlevel 4 TESTLEVEL 4 ARDIAG readlog Contents of drive log at TestLevel Indx 4 Dk ck ck ck Ck ck kk ck kk ck Ck ck ck Ck ck Ck ck ck ck ck Sk ck ck Ck ck ck cock Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck KKK ko ck kk ck Mk Sk Mk kx kv kx ko ko ko oko Vendor ID HP Product ID 1 050 GB 3rd Page Code Supported Log Sense Pages Page Code 2 Error Counter Page Write Page Write Errors Page Code 3 Error Counter Page Read Page Read Errors Page Code 5 Error Counter Page Verify Page Verify Errors Page Code 6 Non Medium Error Page Non Medium Page Code 37 Vendor Unique or Non SCSI 2 Standard Vendor Uniq
129. e point where it left off This means any portion of the rebuild completed before you canceled it will have to be repeated To cancel a rebuild Right click the disk array currently undergoing a rebuild from the Detail pane 2 Click Properties 3 Click the Configuration tab 4 Under Rebuild click Stop The rebuild will stop immediately Analyzing Disk Array Performance The disk array controller monitors and stores a number of performance metrics that reflect how the disk array is performing AutoRAID Manager periodically retrieves the metrics and stores them for viewing AutoRAID Manager can also analyze the performance metrics to identify any potential performance problems Based on this analysis AutoRAID Manager will make recommendations on how to improve disk array performance Checking the metrics regularly is a quick and easy way to monitor the performance of your disk array and identify any problems that may be developing You may choose to display only the recommendations or you may want to view the metrics for further analysis For more information on performance metrics see the Concepts and Management chapter in the HP SureStore E Disk Array 12H User s and Service Manual Selecting a Time Period for Analysis When analyzing performance you must define the time period over which the analysis will be performed For the best results select a time period when performance may be a concern This will produce the most mea
130. e recommendations The performance metrics may help you understand how your disk array is performing but you should avoid making performance planning decisions based solely on them AutoRAID Manager uses other metrics not displayed in this window when performing its analysis Consequently AutoRAID Manager can make a more accurate analysis and arrive at the best recommendations for improving performance To display the performance metric thresholds Right click the disk array from the Detail menu 2 Click Properties 3 Click the Performance tab 4 Click Thresholds to display the Performance Thresholds window Checking the Working Set Metric A key factor in monitoring and maintaining optimal performance of the disk array is the Working Set metric To ensure that disk array performance is maintained access the performance metrics regularly and check the Working Set value The Working Set performance metric is derived from the Write Working Set parameter It indicates the ratio of the Write Working Set size to the amount of RAID 0 1 space available For a detailed explanation 233 IN Windows NT Using AutoRAID Manager for Windows NT Performing Disk Array Maintenance Tasks of the Write Working Set and its impact on performance refer to the HP SureStore E Disk Array 12H User s and Service Manual To maintain performance the amount of RAID 0 1 space should equal or exceed the Write Working Set resulting in a
131. e this protection by specifying the appropriate ARM options The R option allows the disk to be downed even if a loss of redundancy would result but not data unavailability The Z option allows the disk to be downed even if data unavailability would occur A complete description of the options is included in the arraycfg man page NOTE Two of the disks in the disk array are used to store recovery map information The disk array will not allow you to down either of these drives unless you use the R or Z option 143 Using the ARM Command Line Utilities for Performing Disk Array Maintenance Tasks After testing a downed disk can be returned to the disk array configuration by manually including it For more information see Including a Disk in this chapter Testing a Disk Diagnostics allow you test the operation and integrity of a disk Three different types of testing can be performed e Write Read Verify a destructive test that will destroy data on the disk being tested The disk must be downed before beginning the test Read Verify a nondestructive test that will not alter any data on the disk being tested It is not necessary to down the disk before performing a read verify test Self test a nondestructive internal test that checks the operation of the disk To perform a write read verify test of a disk type drivetest D slot w percent array id To perform a read verify t
132. ecovery Progress 1 Percent Recovery Progress 2 Percent Recovery Progress 3 Percent 184 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Recovery Progress 97 Percent Recovery Progress 98 Percent Recovery Progress 99 Percent Wait This will take several minutes Attempting to check completion status of the recover operation The array state has changed The previous state was No Address Table The new state is Ready The array warnings has changed The previous warnings were Disk Warning Capacity Warning Now there are no warnings Additional Notes Read controller logs to assess the completion status of the command based on the number of ECC errors or event counts 185 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions REQSENSE The REQSENSE command allows the user to view SCSI Sense data from the specified target Synopsis REQSENSE Output ARDIAG gt REQSENSE Performing Request Sense at TestLevel Indx 0 Dk ck ck ck Ck ck KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK ko ck kk ck Pk Sk KKK KK ko ko KK Error Code 112 0x70 Segment Number 0 0x0 File mark bit is OFF End of medium bit is 1349672 The Incorrect Length Indicator is OFF Sense Key NO SENSE 0x0 The information field is NOT SCSI 2 compliant Information field bytes 0x0 0 0 0x0 0 0
133. ection against disk failure consequently it will typically not be necessary to perform this procedure unless you want to change the rebuild priority To enable Auto Rebuild to rebuild the disk array automatically Right click the disk array from the Detail pane Click Properties Click the Configuration tab Under Rebuild click Automatic C 232719 Select the Priority you want the disk array to use when it performs rebuild The disk array will now automatically rebuild the array if a disk fails NOTE If there is no Active Hot Spare or not enough unallocated capacity available Auto Rebuild will be enabled even though a rebuild cannot be performed if a disk fails In this situation you must increase the amount of unallocated capacity so there is enough capacity to perform a rebuild You can increase the amount of unallocated capacity by adding another disk or deleting an existing logical drive 230 Using AutoRAID Manager for Windows NT Rebuilding the Disk Array Rebuilding the Disk Array Manually You can manually start a rebuild if a disk drive has failed and Auto Rebuild is disabled To perform a Rebuild the array must have enough capacity available either in the form of an Active Hot Spare or unallocated capacity To disable Auto Rebuild Right click the disk array from Detail pane 2 Click Properties 3 Click the Configuration tab 4 Under Rebuild select Manual With Auto Rebuild disabled yo
134. ed By Applying Retries Errors Corrected Times Correction Algorithm Processed Bytes Processed Uncorrected Errors Page Code 6 NonMedium Errors 37 Addr Errors Corrected Without Substantial Delay 3 Hexadecimal 34 x0000000000000000 0 8 16 24 32 Page Code 3e Addr 0x37 0x00 0 01 Ox3e x00 Oxle 0x00 0x10 0x00 0x02 x00 0x01 0x80 x80 0x02 0x00 x00 0x00 0x00 Moor Sook 0x80 0x02 0x00 0x00 0x00 0x02 0x00 0x04 Hexadecimal 0x80 Oxff 0x00 0x80 0x02 Oxff 0x03 0x02 0 8 0x0 Ox3e 0x00 0x00 0x08 0x00 0x00 0x00 0x8 0x00 0x00 0x97 Oxfl 0x04 183 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions RECOVER The RECOVER command allows the user to restore controller map information in the specified SureStore E AutoRAID Disk Array Synopsis RECOVER Output ARDIAG gt testdisk 1 Array state is No Address Table The following warnings are currently in effect Disk Warning Capacity Warning ARDIAG gt testlevel 0 ESTLEVEL 0 ARDIAG recover 1 he logs should be cleared befor xecuting this command Do you want to abort to clear logs y n no here is only one volume set in the array 0 125 0 Disks Al 2 Proceed with recover operation y n yes R
135. ee ren o ee oes eie 210 Checking the System Hardware Configuration 210 INSTALLING AUTORAID MANAGER SOFTWARE censent enne 211 Installation dn redolet ete dre et e eee eee tee inrer edet dtes 211 Controlling Access to the Disk 44 4484000 213 SETTING UP AUTORAID MANAGER FOR WINDOWS 214 The HPAutoRAID Manager GUL ressante ier es Ea ERR rh PR EHE Te REPE Reit chs 215 W Wat 185A I 216 How do I add a customized esea a a riae a e i 217 DISCOVERINGDISK ARRAYS rr eter eee ee Ee ose puede ven 219 MANAGING DISK ARRAYS ON A REMOTE 219 EDITING THE DISK ARRAY e eee ree p RU Wet re FR Ee vts 220 SOLVING COMMON INSTALLATION 65 2 2 2 4 02202020 02220001000000000000000000000000000 220 CHAPTER 9 USING AUTORAID MANAGER FOR WINDOWS 2 1 223 ONLINE HELP 5 ies on eo Rt ar 223 SELECTING A DISK ARRAY re Gab ERE eee en EE UE 223 CONFIGURING A NEW DISK 2 0 2 2 000000000000000000000000000000000000000 224 CHECKING DISK ARRAY STATUS
136. election When enabled the secondary controller will remain off line until a failure of the primary controller is detected at which point it becomes primary Only the primary controller will go on line Very Early Busy DISABLED Controls the behavior of the disk array with respect to SCSI bus selection during the early stages of the initialization sequence i e from about three seconds after reset until about fifteen seconds prior to initialization completion When disabled the disk array will ignore SCSI bus selection until the late stages of initialization When enabled the disk array will accept selection during early initialization and will return BUSY status until the late stages of initialization Queue Full Threshold 1952 Specifies the target queue depth beyond which the disk array will return QUEUE FULL status to subsequent host requests Maximum Queue Full 1952 Specifies the maximum supported value for the Queue Threshold Full Threshold parameter Attempts to set the QFT parameter to values higher than the MQFT will fail with CHECK CONDITION status and ILLEGAL REQUEST INVALID FIELD IN PARAMETER LIST sense indications Simplified Resiliency Normal Determines the values of the following eight parameters Setting This field regulates the mode of data resiliency that the disk array will operate in The Normal mode is set at the factory when dual controllers are ordered with the array 266 Pa
137. ell the disk array is operating and if any problems or failures have occurred Using the ARM arraydsp command you can easily check all aspects of disk array operation and configuration The arraydsp command options summarized in Table 16 allow you to display information about each disk array hardware component as well as information about the logical configuration of the disk array Table 16 arraydsp Options for Displaying Disk Array Status Option Status Information Displayed 1 LUN Information for the specified LUN All information displayed for options l d c 5 v and h c l Array controller status l h Hardware status m Display performance metrics R Rescan for SureStore E AutoRAID disk arrays recognized by the host S Generate raw output used in combination with other options General configuration information For a complete description of all the configuration settings see Viewing the Disk Array General Configuration Settings at the end of this chapter y Capacity information l 131 Using the ARM Command Line Utilities for MPE Displaying Disk Array Serial Numbers Displaying Disk Array Serial Numbers The serial numbers of all disk arrays connected to the host can be displayed by typing arraydsp i Missing Disk Arrays If you know there are disk arrays connected to the host but they are not displayed in response to t
138. em documentation for any additional information or steps that may be required to renumber a LUN To renumber a LUN type arraycfg L LUN r newLUN array id LUN is the LUN to be renumbered newLUN is a new available LUN number Deleting a LUN When a LUN is deleted its capacity is returned to the pool of unallocated capacity space Deleting a LUN is a good way of freeing up capacity for the Active Hot Spare or for simply adding more unallocated capacity to improve disk array performance CAUTION data on a LUN is lost when it is deleted Make sure you backup any important data on the LUN before deleting it NOTE Before deleting a LUN check your operating system documentation for any additional information or steps that may be required to delete a LUN To delete a LUN type arraycfg L LUN d lt array id gt LUN is the LUN to be deleted 135 Using the ARM Command Line Utilities for MPE Adding a Disk Adding a Disk At some time you may want to add another disk to your array Features such as hot pluggable disks and Auto Include simplify the process of adding a disk to the array even while it is operating disk can be added to the array without disrupting current I O operations After you have added a new disk you have three options on how to use it e Increase capacity use the disk to increase the capacity available to the operating system by creating a new LUN e I
139. en disabled Right click the disk array from the Detail pane 2 Click Properties 3 Click the Diagnostics tab 4 Click the disk you want to include from the graphic or list 5 Click Include 229 LN SMOpUIM Windows NT Using AutoRAID Manager for Windows NT Rebuilding the Disk Array Rebuilding the Disk Array To maintain data redundancy in the event of a disk failure it is important to rebuild the disk array as soon as possible The disk array is not data redundant until a rebuild is performed If another disk fails before the rebuild is complete data on the disk array may be lost AutoRAID Manager provides two rebuild options Auto Rebuild which allows the disk array to begin rebuilding immediately if a disk fails Manual Rebuild which requires you to initiate the rebuild This option is provided if you want more control over the rebuild process such as delaying the start of a rebuild until the system is less busy A rebuild impacts disk array performance while it is in progress so before starting a rebuild make sure the appropriate Rebuild Priority is set Rebuilding the Disk Array Automatically You can enable Auto Rebuild to allow the disk array to immediately begin rebuilding if a disk fails To perform a Rebuild the array must have enough capacity available either in the form of an Active Hot Spare or unallocated capacity Auto Rebuild with high priority is enabled by default to provide the greatest prot
140. en if data unavailability would occur A complete description of the options is included in the arraycfg man page NOTE Two of the disks in the disk array are used to store recovery map information The disk array will not allow you to down either of these drives unless you use the R or Z option 51 XN dH Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks After testing a downed disk can be returned to the disk array configuration by manually including it For more information see Including a Disk in this chapter Testing a Disk Diagnostics allow you test the operation and integrity of a disk Three different types of testing can be performed e Write Read Verify a destructive test that will destroy data on the disk being tested The disk must be downed before beginning the test Read Verify a nondestructive test that will not alter any data on the disk being tested It is not necessary to down the disk before performing a read verify test Self test a nondestructive internal test that checks the operation of the disk To perform a write read verify test of a disk type drivetest D slot w percent array id To perform a read verify test of a disk type drivetest D slot r percent array id To perform a self test of a disk type drivetest D slot s array id percent is the percent 0 to 100 of the disk to be tested slot is the cabin
141. ently exceeds 1 the amount of RAID 0 1 space available should be increased to improve performance This can be accomplished in several ways as described in the following section If the Working Set is consistently much less than 1 some of the RAID 0 1 capacity can be allocated to a new LUN without impacting performance The remaining RAID 0 1 space should be adequate to accommodate the Write Working Set 49 XN dH Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks Performing Disk Array Maintenance Tasks There are several tasks that you may have to perform in the on going management of the disk array These maintenance tasks are typically performed infrequently and may involve taking the disk array off line Shutting Down the Disk Array CAUTION When an array shutdown is performed the disk array becomes unavailable to the host system An array that is shutdown appears to the operating system as if its power has been turned off As with any disk subsystem it is essential that file system and LVM access to the disk array be correctly removed before shutting down or powering off the disk array Before an array Shutdown is performed all mounted file systems and LVM logical volumes mapped to the disk array must be unmounted The umount operation synchronizes data in the HP UX internal buffers with the data stored on the disk array The disk array must be Shutdown prior to performing
142. er the AutoRAID Manager for Windows provides a graphical representation of performance data that you should more useful than the information returned using the arraydsp command For information about using the Auto RAID Manager for Windows and details on performance topics in general see Analyzing Disk Array Performance in the preceding chapter To check disk array performance type arraydsp r stime etime 4 m stime etime int array id The starting time st ime and ending time et ime values establish the analysis period The format of the stime and et ime arguments is mmddhhmm yy The int option controls the display interval The display interval is the number of 15 minute increments Command Examples The following command displays performance recommendations for disk array serial number 00786 5 0000 Performance is analyzed for the time period starting at 8 00 AM and ending at 5 00 PM 1700 on March 15 arraydsp r 03150800 03151700 00786 5 0000 The following command displays the performance metrics for disk array serial number 00786B5C0000 Metrics are displayed for the time period starting at 11 00 AM and ending at 6 00 PM on April 6 A display interval of 30 minutes is specified arraydsp m 04061100 04061800 2 00786B5C0000 253 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Performing Disk Array Maintenance Tasks Performing Disk Array Maintenance Tasks
143. er controller X is reset Controller Y SCSI Indicates the SCSI bus address to be used by controller Address Y Changes in this field will take affect only after controller Y is reset This field may be ignored depending on the map resiliency mode Series 800 systems normally disable this field and series 700 systems enable this field Generally rite cache is used even though a host system automatically disables this field Read Cache DISABLED Indicates whether the disk array should cache read data Note This field does not represent what is actually happening The disk array uses read cache algorithms This field is put in place to facilitate some third party operating systems SCSI Parity Checking Parity Checking ENABLED Indicates whether the disk array should check SCSI bus parity Disabled means that bus parity checking is disabled Enabled indicates that bus parity checking is enabled SDTR ENABLED Indicates whether the disk array should initiate SDTR Synchronous Data Transfer Request Disabled indicates the disk array will not initiate SDTR Enabled Write Cache ENABLED Indicates whether the disk array should cache write data indicates the disk array will initiate SDTR X WDTR ENABLED IIndicates whether the disk array should initiate WDTR Wide Data Transfer Request Disabled indicates the disk array will not initiate WDTR Enabled indicates the disk array will initiate WDTR Terminator Power
144. ere cleared Hosts that cannot tolerate target bus resets should enable this setting Secondary Controller DISABLED Controls the behavior of the secondary controller with Offline respect to bus selection When disabled any secondary controller present will respond to host selection When enabled the secondary controller will remain off line until a failure of the primary controller is detected at which point it becomes primary Only the primary controller will go on line Very Early Busy DISABLED Controls the behavior of the disk array with respect to SCSI bus selection during the early stages of the initialization sequence i e from about three seconds after reset until about fifteen seconds prior to initialization completion When disabled the disk array will ignore SCSI bus selection until the late stages of initialization When enabled the disk array will accept selection during early initialization and will return BUSY status until the late stages of initialization Queue Full Threshold 1952 Specifies the target queue depth beyond which the disk array will return QUEUE FULL status to subsequent host requests 158 Maximum Queue Full Simplified Resiliency Single Controller Lock Write Cache On Disable NVRAM on Disable NVRAM with Disable NVRAM on Default Setting 1952 Normal ENABLED TRUE FALSE TRUE FALSE Using the ARM Command Line Utilities for MPE Viewing the Disk
145. escriptions The ROMT command allows the user to perform a read only media test on the specified target Synopsis Output ROMT ARDIAG romt Dk ck ck ck Ck ck ck Ck ck ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck kk ck ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ock ko ko ck ko Sk ke kx kv A ko ko o WARNING KKK KK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KK KKK KK This test cannot continue without a loss of redundancy on the AutoRAID Array being tested Redundancy will be restored when the test completes However failure of another disk during the test will result in data unavailability until the test completes Dk ck ck ck KKK KKK KKK KKK KKK KKK KKK KKK KKK KKK KK KKK KKK KKK ck ck ck ok ck ck ck ko ck kk ck ko ko ke Sk Sk A ko ko o Do you want to continue y n y Testing block O0 percent complete 1005 Test complete 97 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Testing subrange 1000 1200 percent complete 10 oe oe oe oe oe oe OL O C C O1 iS 0 est complete esting block 45000 percent complete 1005 Test complete 1 1 1 ARDIAG gt 98 SETOPTIONS The SETOPTIONS command allows the user to view and set specific options for t
146. esired disk array REVISIO View revision information of the desired disk array WRTMT Perform a DESTRUCTIVE write read test on the desired disk ROMT Perform a read only test on the desired disk CREATELUN Adds a LUN to the selected disk array DELETELUN Removes a LUN from the selected disk array SHUTDOW Puts the selected disk array into the shutdown state SETOPTIONS View and modify SCSI mode parameters on the selected array controller RECOVER Reconstruct data mapping and array configuration Continue y n n 114 Chapter 5 Managing the HP SureStore E Disk Array 12H on MPE This chapter introduces the tools available for managing your HP SureStore E Disk Array 12H on MPE These tools provide complete control over all aspects of disk array operation The following information is included in this chapter e brief description of each management tool e Which tools can be used for each management task e Instructions for installing the disk array management software 115 Managing the HP SureStore E Disk Array 12H on MPE Disk Array Management Tools Disk Array Management Tools Two disk array management tools are available for managing the disk array an online tool and the disk array control panel AutoRAID Management ARM utilities Included with the disk array is a set of ARM utilities which provide advanced capabilities for managing the disk array You will need to use the
147. est of a disk type drivetest D slot r percent array id To perform a self test of a disk type drivetest D slot s array id percent is the percent 0 to 100 of the disk to be tested slot is the cabinet shelf containing the disk drive A1 through A6 or B1 through B6 Displaying Test Results After the disk testing is complete the test results can be displayed for analysis by using the dtest stat utility To display the results of a disk test type dteststat D slot array id 144 Using the ARM Command Line Utilities for MPE Performing Disk Array Maintenance Tasks If the D option is not specified results will be displayed for all disks in the array that have been tested Canceling a Disk Test If you do not want to wait for a disk test to complete you can cancel it using the dtest stat command and testing will stop immediately To cancel a disk test in progress type dteststat D slot c array id Printing ARMServer Log Contents The contents of the various log files maintained by ARMServer can be printed using the logprint command The contents of the logs may be useful in identifying any possible problems that may be occurring with the disk array For a detailed explanation of the log contents output see the Logprint man page To print the contents of the ARMServer logs type logprint d log directory name s start time e stop time t record type a array serial nu
148. et shelf containing the disk drive A1 through A6 or B1 through B6 Displaying Test Results After the disk testing is complete the test results can be displayed for analysis by using the dtest stat utility To display the results of a disk test type dteststat D slot array id 52 Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks If the D option is not specified results will be displayed for all disks in the array that have been tested Canceling a Disk Test If you do not want to wait for a disk test to complete you can cancel it using the dtest stat command and testing will stop immediately To cancel a disk test in progress type dteststat D slot c array id Printing ARMServer Log Contents The contents of the various log files maintained by ARMServer can be printed using the logprint command The contents of the logs may be useful in identifying any possible problems that may be occurring with the disk array For a detailed explanation of the log contents output see the Logprint man page To print the contents of the ARMServer logs type logprint d log directory name s start time e stop time t record type a array serial number log directory name identifies the location of the log files start timeand stop time limit the output to events between the specified times record type identifies the type of record s to print Records incl
149. etting Simplified Resiliency Setting Results Single Controller ENABLED Single Controller Dual Controller Warning Lock Write Cache On TRUE Not Supported Data Resiliency disabled Disable NVRAM on FALSE WCE False Disable NVRAM with TRUE One Ctrlr Disable NVRAM on FALSE UPS absent Force Unit Access 0 Response Disable Read Hits FALSE Resiliency Threshold 0 271 IN Windows NT Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Setting Data Resiliency The data map contents stored in the disk array controller NVRAM is copied to two disks on the array at regular intervals to protect against map loss The interval at which the disks are updated with recovery map information is controlled using the data resiliency setting Selecting a data resiliency setting involves making a tradeoff between data protection and performance The more frequently the recovery maps are updated the more impact it may have on performance To set the data resiliency mode type arraymgr J SingleController Secure Normal HighPerformance array id The data resiliency settings are described in Table 32 Table 32 Data Resiliency Settings Setting Description SingleController This setting should only be used if the disk array is operating with one controller This suppresses the single controller warning messages that are normally generated when the disk array is operat
150. ettings Resiliency Threshold 1 HighPerformance Mode Parameter Single Controller Warning Lock Write Cache On Disable on WCE False Disable NVRAM with One Ctrlr Disable on UPS absent Force Unit Access Response Disable Read Hits Resiliency Threshold Default Setting Simplified Resiliency Setting Results ENABLED Single Controller Dual Controller FALSE Not Supported Data map and write cache TRUE information will be flushed to the disk once every TRUE second FALSE 2 FALSE Default Setting Simplified Resiliency Setting Results Single Controller Dual Controller Not Supported Data Resiliency disabled FALSE TRUE FALSE FALSE 0 163 Chapter 7 Using the ARDIAG Offline Diagnostic on MPE This chapter defines the AutoRAID supported offline diagnostic commands for Series 800 MPE systems This functionality will be a combination of the support media offline diagnostic environment ODE commands and commands issued from the AutoRAID front panel Operating environment ARDIAG will only operate under ODE and thus it only operates in an offline environment When ARDIAG is executed ODE handles all the I O with the user and it handles several higher level commands like LOG HELP etc NOTE There should be no I O activity on the SCSI bus when using ARDIAG When operating in a
151. f d is specified clear the log for the disk installed in the cabinet slot identified by slot To display the contents of a hardware log type arraylog u e d slot array id To clear the hardware logs type arraylog C cl d slot array id Formatting the Disk Array Should it become necessary to do so you can format the entire disk array or a single LUN Formatting destroys all data on the array or LUN involved Formatting an array first requires that all LUNs be deleted To format the entire disk array type arrayfmt F h array id To format a LUN type arrayfmt L LUN lt array id gt 54 Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks Changing SCSI Settings The SCSI settings control the transfer of information over the SCSI channel connecting the host and the disk array The default SCSI settings listed in Table 9 have been chosen to work with all supported SCSI adapters and in most cases should not be changed NOTE Before changing any SCSI setting you should understand what the SCSI setting does and what effect changing it will have on disk array operation Be aware that using an incorrect SCSI setting may make it impossible for the host and disk array to communicate properly In this case it will be necessary to use the disk array control panel to return the SCSI setting to its original value to reestablish communication To ch
152. f manually include the disk as described in the next section Including a Disk NOTE In some situations the array will not include a disk automatically even if Auto Include is enabled This will happen if the new disk s status is not Normal See Auto Include in the HP SureStore E Disk Array 12H User s and Service Manual for more information about when this might occur 3 Depending on how the new disk will be used perform the appropriate next step e To use the disk to increase capacity create a LUN using all or a portion of the disk capacity For more information see Creating a LUN in this chapter e To use the disk to increase performance leave the disk capacity unallocated e To use the disk capacity for Active Spare enable the Active Spare feature if disabled 28 Using the HP UX System Administration Manager Adding a Disk Including a Disk A disk must be included in the disk array configuration before it can be used by the disk array There are two ways to include a disk e You can enable Auto Include which will automatically include a disk when it is inserted into the disk array enclosure e You can manually include each new disk For convenience Auto Include is enabled by default on a new disk array For information on disabling Auto Include see Changing Disk Array Configuration Settings in this chapter After including a disk you must decide how you want to use it For more information see the p
153. f printing this field is reserved for future functionality with UPS and is disabled in all resiliency modes 267 LN Windows NT Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Parameter Force Unit Access Response Disable Read Hits Resiliency Threshold 268 Default Setting 2 FALSE 4 Description Controls the array s response to the FUA bit The FUA bit is a command from the host when Write Cache is enabled It gives the host an opportunity to flush write cache on command If this field is 0 then the FUA bit is ignored If this field is 1 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed If this field is 2 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed along with the map journal before completing the write request If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored Controls the array s ability to satisfy read commands from rite cache contents during FUA processing If NVRAM use is disabled due to the chosen map resiliency mode then read data is never satisfied out of write cache Specifies the maximum time between delivery of the response to a write command to the host and initiation of associated wr
154. f these paths as the default hardware path and will use the default as the primary data path for access to the LUNs created on the disk array To improve disk array performance the paths to the LUNs on the disk array should be shared by both controllers Dividing the primary data paths across both controllers balances the I O load and optimizes disk array performance The primary hardware path to each LUN is established when adding the LUN to a volume group On HP UX 10 01 and 10 10 both controller hardware paths are displayed on the Disk Devices windows The hardware path for a LUN is established by selecting the appropriate path On HP UX 10 20 and later only the primary controller hardware path is displayed in the Disk amp File Systems window with an indication that there are two paths to the device This makes the selection of an alternate path to the LUN a bit more involved The following steps should clarify the process of defining an alternate path for a LUN To select an alternate hardware path 1 From the Disk amp File Systems window select the LUN you want to add Disks and File Systems hpbs1991 File List View Options Actions gt AutoRAID Array Controller gt AutoRAID AutoRAID gt AutoRAID Disk Array Controller Unused gt AutoRAID LUN LYM v gt AutoRAID LUN Unused Unused Ie LUN Unused 2577 z CSI Disk Array RAID 5 3 d v 2 a 2 a 2 2 a a ro ro ro 2 From the Act
155. fline Diagnostic on HP UX This chapter defines the AutoRAID supported offline diagnostic commands for Series 700 and Series 800 HP UX systems This functionality will be a combination of the support media offline diagnostic environment ODE commands and commands issued from the AutoRAID front panel Operating environment ARDIAG will only operate under ODE and thus it only operates in an offline environment When ARDIAG is executed ODE handles all the I O with the user and it handles several higher level commands like LOG HELP etc NOTE There should be no I O activity on the SCSI bus when using ARDIAG When operating in a multi initiator environment make sure none of the host systems are trying to access the disk array Bus activity make cause ARDIAG to not see the disk array Support Software ARDIAG requires all the software needed for ODE to run as ARDIAG relies on ODE to load and run Additionally ARDIAG needs the MAPPER SL and the DIODE SL from the protocol level down Minimum hardware For ARDIAG to run properly a working I O subsystem and at least one SureStore E AutoRAID Disk Array is needed ARDIAG must also meet ODE s hardware requirements for example a working console ARDIAG and ODE will use no more than 4 MB of main memory Minimum software ODE MAPPER SL and DIODE SL are needed to execute ARDIAG correctly 73 XN dH Using the ARDIAG Offline Diagnostic on HP UX Definition of Terms
156. formance you can use the disk to improve disk array performance by simply leaving it as unallocated capacity Enable Active Hot Spare you can use the additional capacity to enable Active Hot Spare if your disk array does not currently have the capacity to implement this feature NOTE The Auto Include feature will normally include the disk drive when it is installed However in some situations the array will not include a disk automatically even if Auto Include is enabled In this case you must manually include the disk For more information on Auto Include see the Concepts and Management chapter in the HP SureStore E Disk Array 12H User s and Service Manual To add a disk to increase capacity 1 Make sure the new disk has been installed in the array cabinet 2 Click the disk array from Detail pane 3 If Auto Include is on the disk is automatically added to the array and you can proceed to step 4 If Auto Include is off manually include the disk as described in Including a Disk Manually 4 Create one or more logical drives using all or a portion of the disk s capacity 5 Perform the necessary steps to configure the new logical drive into your host This is typically done using the Windows NT Disk Administrator Refer to Windows NT Help for information on performing this task To add a disk to improve array performance 1 Make sure the new disk has been installed in the array cabinet 2 Right click the disk
157. ftware installation Chapter 9 describes how to use the AutoRAID Manager for Windows to manage the disk array Chapter 10 describes how to manage the disk array using the Windows NT ARM command line utilities During installation of the AutoRAID Manager for Windows NT software an electronic copy of this book in Adobe Acrobat format is included in the Program Files VAutoRAID Doc directory default location Supporting Documentation The following documentation is included with the HP SureStore E Disk Array 12H and should be available for reference when installing and managing the disk array e HP SureStore E Disk Array 12H User s and Service Manual part number C5445 90901 For Windows NT users an electronic copy of this book in Adobe Acrobat format is included in the Program FilesNAutoRAID Doc directory Trademark Credits AutoRAID is a trademark of Hewlett Packard Company Microsoft Windows and Microsoft Windows are registered trademarks of Microsoft Corporation HP on the World Wide Web The latest information about your HP SureStore E Disk Array 12H is available on the HP web site at www hp com go support Check our web site for e Updated editions of product documentation e Firmware and software upgrades e Current supported system configurations e General information for optimizing the operation of your disk array Table of Contents CHAPTER 1 MANAGING THE HP SURESTORE E DISK ARRAY 12H ON HP UX
158. g AutoRAID Manager for Windows NT Performing Disk Array Maintenance Tasks 4 Under Array click Format If the Array control is not available it indicates that there are logical drives on the disk array The disk array cannot be formatted until all logical drives are deleted 5 Click OK in the confirmation box to begin the format Formatting a Logical Drive WARNING A format will destroy all the data on the logical drive Make sure all data is backed up before formatting the logical drive Each logical drive can be reformatted if necessary To format a logical drive Right click the disk array from the Detail pane Click Properties Click the Logical Drives tab Click the Logical Drive you want to format Sn eS YA Click Format Testing a Disk AutoRAID Manager includes diagnostics that test the operation and integrity of any disk in the array enclosure Testing a disk may impact the performance of the disk array so you may want to delay disk testing to times when the disk array is not being heavily accessed Two different types of testing can be performed Read verify a nondestructive test that does not alter any data on the disk being tested The disk is not downed when performing a read verify test e Write Read a destructive test that destroys data on the disk being tested The disk is downed before beginning the test If Auto Rebuild is enabled the disk array will immediately begin rebuilding the d
159. ges to AutoRAID status are immediately communicated to the client which updates the AutoRAID icon status indicators SureStore E Disk Arrays are managed from OpenView in the same manner as other network resources The following tips identify ways in which you can use OpenView to manage your disk array most efficiently AutoRAID Manager can be launched by double clicking an AutoRAID disk array icon on an OpenView submap The AutoRAID manager will open displaying the status window for the disk array icon that was clicked e The AutoRAID OpenView client requires that ARMServer and the AutoRAID SNMP agent both be running on the server e At OpenView initialization the client discovers all AutoRAID disk arrays connected to NT systems then adds the associated AutoRAID icons to the server submaps e The disk array submap layout is updated automatically once day to reflect the addition or deletion of devices The submap layout is also updated manually if a server rescan is performed e The SNMP configuration parameters used for each SNMP session opened by the OpenView Client for AutoRAID can be set from the OpenView console Disk Array Events Disk array events are logged to the NT Event Log by default If SNMP is installed on the server the events can be passed to network management applications such as HP OpenView using the AutoRAID Manager SNMP agent The AutoRAID SNMP agent is installed during setup if SNMP is detected on the serve
160. he array controller Synopsis SETOPTIONS Output ARDIAG gt 51 ETOPTIONS Using the ARDIAG Offline Diagnostic on HP UX Operational Command Descriptions Current information for the controller in slot X Subsystem Parameter Value Subsystem Parameter Value 1 Subsystem State Ready 15 Termination Power ON 2 Active Hot Spare OFF 16 Unit Attention ON 3 Volume Set Partitioning 17 Controller X Address 0 4 Rebuild in Progress OFF 18 Controller Y Address 1 5 Auto Rebuild 19 Enable Manual Override OFF 6 Auto Include 20 Manual Override Destination OFF 7 Balance in Progress OFF 21 Format Pattern Fill ON 8 Optimize in Progress 22 Disable Remote Reset OFF 9 Migrating Write Destination OFF 23 Language English 10 Log Full Warning 24 Capacity Depletion Threshold 0 11 Rebuild Priority OFF 25 Rebuild Progress 0 12 Parity O 26 Write Working Set Interval 0 13 SDTR 27 Subsystem Identifier 1 3 14 WDTR O Select the number of the parameter to modify Default quits 2 Allowed values for Active Hot Spare 0 OFF 1 ON Select th number of new parameter valu Default quits 1 99 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions 100 Modify the parameter with th Subsystem Parameter Subsystem State Active Hot Spare Volume Set Partitioning Rebuild in Progress Auto Rebuild
161. he arraydsp command check the following e Make sure all disk arrays are properly connected to the host This includes proper termination of the SCSI bus e Make sure all disk arrays are turned on and operating properly e Make sure the ARMServer process is running ARMServer must be running to execute this or any other ARM command You can easily check to see if the ARMServer process is running using the following command ps ef grep ARM e Rescan for disk arrays by typing arraydsp This will update the ARMServer information to reflect the current system configuration Changing Disk Array Configuration Settings A number of configuration settings control the operation of the disk array These settings are usually established during installation and once set should rarely need to be changed The default settings have been selected to provide the best operation for most systems However if you determine that any setting does not meet your needs you can easily change it Table 17 lists the various settings including factors you may want to consider when changing them 132 Using the ARM Command Line Utilities for MPE Changing Disk Array Configuration Settings Table 17 arraymgr Disk Array Configuration Settings Setting Default Command Comments and Considerations Option Active On h Active Hot Spare provides optimum Spare protection against disk failure Disabling Active Hot Spare will make additional capacity
162. he controller log is full When enabled the disk array will assert a Log Full Warning hen one of the disk array log pages fills or a log parameter reaches its maximum value When disabled no Log Full Warning will be indicated Language ENGLISH Volume Set DISABLED Indicates whether the disk array should boot when half or Partitioning more of the previously available drives are unavailable i e there is no drive quorum Enabled indicates that volume set partitioning should be allowed Disabled the system will remain in the No Quorum state when the required quorum is not available Format Pattern Fill DISABLED Indicates whether the disk array will fill in incomplete RAID blocks with a format pattern when performing new rites Some operating systems not HP UX and not NT expect that SCSI format commands completely reinitialize data to a non random pattern When using the array with of those operating systems pattern filling should be enabled Disk array Type ID 12 Uniquely identifies the disk array hardware configuration This field can be used to find the number of drives and SCSI channels supported within the disk array LUN Creation Limit 8 Controls the range of LUNs that may be created Maximum LUN 8 Specifies the maximum supported value for the LUN Creation Limit Creation Limit 264 Read Cache B Indicates whether the disk array should cache read data SCSI Parity Checking ENABLED I
163. he disk array from the Detail pane Click Properties Click the Diagnostics tab 2 3 4 Click on the disk you want to down 5 Click Down 6 From the Down Physical Disk Warning dialog select the desired option restricting the conditions under which the down can occur To return a downed disk to the array configuration you must manually include it See Including a Disk Manually for more information Switching Array Controllers In disk arrays with two controllers the array automatically switches to the secondary controller if the primary controller fails However you can manually switch controllers if necessary 237 LN SMOpUIM Windows NT Using AutoRAID Manager for Windows NT Performing Disk Array Maintenance Tasks To switch array controllers Right click the disk array from the Detail pane 2 Click Properties 3 Click the Diagnostics tab 4 Click Switch to Y or X as appropriate to switch to the secondary controller Displaying Hardware Logs Each controller and each disk drive in the disk array maintains its own internal hardware status log These logs contain information that may be useful in identifying or diagnosing problems with the array AutoRAID Manager regularly copies the contents of the controller logs to the log file on the host server You can read the contents of the hardware logs directly and also clear the logs To restart the hardware logs Right click the disk array from the Detai
164. hecking LUN Configuration When you are managing LUNs you may find it convenient to check the current LUN configuration and the available capacity To check the current LUN configuration and the available capacity type arraydsp 1 LUN lt array id gt Creating a LUN Only capacity assigned to LUNs is visible to the operating system When selecting the size for your LUNs consider the following factors e Any size limitations imposed by the operating system e Your backup strategy If you do unattended backup to a device such as a tape you may want to avoid creating a LUN that is larger than the capacity of the tape media This allows you to back up an entire LUN without changing tapes e Configuring the LUN for maximum performance as described in Table 15 NOTE Before creating a LUN check your operating system documentation for any additional information or steps that may be required to create a LUN To create a LUN type arraycfg L LUN a capacity array id LUN must be an unused value between 0 and 7 Capacity must be less than or equal to the currently available unallocated capacity 134 Using the ARM Command Line Utilities for Managing LUNs Logical Drives NOTE The following commands require LUN exclusive access This means nothing can access the LUN while the command is running Delete LUN Renumber LUN Renumbering a LUN NOTE Before renumbering a LUN check your operating syst
165. hecks the operation of the disk To perform a write read verify test of a disk type drivetest D slot w percent array id To perform a read verify test of a disk type drivetest D slot r percent array id To perform a self test of a disk type drivetest D slot s array id percent is the percent 0 to 100 of the disk to be tested slot is the cabinet shelf containing the disk drive A1 through A6 or B1 through B6 Displaying Test Results After the disk testing is complete the test results can be displayed for analysis by using the dtest stat utility To display the results of a disk test type dteststat D slot array id If the D option is not specified results will be displayed for all disks in the array that have been tested 256 Using the ARM Command Line Utilities for Windows NT Performing Disk Array Maintenance Tasks Canceling a Disk Test If you do not want to wait for a disk test to complete you can cancel it using the dtest stat command and testing will stop immediately To cancel a disk test in progress type dteststat D slot c array id Printing ARMServer Log Contents ARMServer maintains a number of configuration and status log files on the server The contents of the various log files maintained by ARMServer can be printed using the Logprint command The contents of the logs may be useful in identifying any possible problems that may be occurring with the disk
166. hpb10320 000000057022 no rights Normal hpb26440 000000057A5E 11872 GB og New Host View 2 Count 2 4 For Help press F1 215 IN Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT Setting Up AutoRAID Manager for Windows NT What is a view To more easily manage disk arrays connected to a host you can group sets of hosts into customized views For example you could group all arrays that are located in an accounting department and group them into a single view called Accounting When you first start the HP AutoRAID Manager a Local Host view and All Known Hosts view are automatically created these views are not customizable 216 Managing the HP SureStore E Disk Array 12H Windows NT Setting Up AutoRAID Manager for Windows NT How do add a customized view To add a customized view 1 2 Enter a for the view 3 Click OK 4 5 Click OK 6 Select the desired Hosts 7 Click OK Click 7 lt lt Add New Host View gt gt in the View pane Use the Columns Filter and Sort tabs to customize your view The view will now appear in the View pane All known hosts hpb10320 boi hp com hpb26440 boi hp com hpb10320 boi hp com hpb25440 boi hp com 217 Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT Setting Up AutoRAID Manager for Windows NT To select additional arrays you want
167. ic functionality via ISL support media for Series 800 MPE 10 04 systems Logical Interchange Format A standard disk format that may be used for interchange of files among various HP computer systems LIF volume contains a header identifying it as a LIF volume and a directory that defines the contents of the volume A boot device with a valid LIF is assumed to have bootable media Spectrum Object Module A SOM is the smallest unit that may be generated by a compiler and it may exist as a single entity or as part of a collection Using the ARDIAG Offline Diagnostic ARDIAG Operational Commands ARDIAG Operational Commands Table 23 lists the ARDIAG operational commands and their descriptions Table 23 ARDIAG Operational Commands Command CLRLOG CREATELUN DELETELUN DESCRIBE DOWNLOAD FORMAT INQUIRY READLOG RECOVER REQSENSE RESTART REVISION ROMT SETOPTIONS SHUTDOWN WRTMT Description Allows the user to clear the specified target s internal logs Allows the user to create a LUN via the specified controller path to the disk array Allows the user to delete a LUN via the specified controller path to the disk array Displays LUN configurations and warning states Allows the user to download firmware to the specified controller or internal disk Allows the user to format the specified target Issues an Inquiry to an array controller or an internal disk Allows the user
168. il pane Click Properties Click the Logical Drives tab Click the drive you want to renumber Click Renumber to display the Renumber Logical Drive window In the Change to field select the new number you want assigned to the logical drive Click OK wc OBS Deleting Logical Drive CAUTION Deleting a logical drive destroys all data on that logical drive Before deleting a logical drive make sure you backup any data you want to save When you delete a logical drive all data on that logical drive is destroyed and its capacity is returned to the pool of unallocated capacity on the array To delete a logical drive Right click the disk array from the Detail pane 2 Click Properties 3 Click the Logical Drives tab 4 Click the drive you want to delete 227 LN SMOpUIM Windows NT Using AutoRAID Manager for Windows NT Adding a Disk 5 Click Delete Adding a Disk At some time you will probably want to add another disk to your array Features such as hot plug disk modules and Auto Include simplify the process of adding a disk to the array even while the disk array is operating disk can be added to the array without disrupting client or host operation After you have added a new disk you have three options on how to use it e Increase capacity you can use the disk to increase the capacity available to the operating system by creating a new logical drive e Improve per
169. ill not let you down a disk if doing so would result in loss of data redundancy or data unavailability However you can override this protection by specifying the appropriate options The R option allows the disk to be downed even if a loss of redundancy would result but not data unavailability The Z option allows the disk to be downed even if data unavailability would occur complete description of the options is included in the on line help arraycfg description NOTE Two of the disks in the disk array are used to store recovery map information The disk array will not allow you to down either of these drives unless you use the R or Z option After testing a downed disk can be returned to the disk array configuration by manually including it For more information see Including a Disk Manually in this chapter 255 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Performing Disk Array Maintenance Tasks Testing a Disk Diagnostics allow you test the operation and integrity of a disk Three different types of testing can be performed e Write read verify a destructive test that will destroy data on the disk being tested The disk must be downed before beginning the test Read verify a nondestructive test that will not alter any data on the disk being tested It is not necessary to down the disk before performing a read verify test Self test a nondestructive internal test that c
170. in nein nnns 59 Firmware Download Procedure With 60 RECOVERING DATA MAPS 62 VIEWING THE DISK ARRAY GENERAL CONFIGURATION 5 5 2 00 64 Simplified Resiliency 70 CHAPTER 4 USING THE ARDIAG OFFLINE DIAGNOSTIC ON HP UX eeeceeeeee eren 73 OPERATING ENVIRONMENT ient dg teer teer p i Het e edet doe 73 SUPPOLt PL 73 Minimum hardware coner 73 DEFINITION OF TERMS PERCHE D 74 ARDIAG OPERATIONAL COMMANDS ccsscsccccecssssssececececsesseeeeceeeceesesseaeceeecseseseaeseeececsessaaeseeececeensaaseeeess 75 ARDIAG INTERFACE COMMAND S rena asset 76 OPERATIONAL COMMAND DESCRIPTIONS ccecessesececececeessnaececcececeeseneaececececsesaaeceeccecsenssaeceeeceeeensaaeeeeecs 77 CTE ROG eese ege etre de iani nisi TI DEDLETBEEUN imitata dan 78 DESCRIBE eere tiere ph te eee pe tr ido pete Pe ed eer eere ie 79 IDONINIEGOAI EC 81 FORMAT aie CO DOW gr ORARE 85 INOUIRY tiis emn ioi eere ie e p idt act 87 REBEADEQOQG rte Het MRSA RTS
171. in this chapter Checking Disk Array Status One of the most important management tasks is monitoring the status operation and configuration of the disk array It is important to know how well the disk array is operating and if any problems or failures have occurred Using the ARM arraydsp command you can easily check all aspects of disk array operation and configuration The arraydsp command options summarized in Table 27 allow you to display information about each disk array hardware component as well as information about the logical configuration of the disk array Table 27 arraydsp Options for Displaying Disk Array Status Option Status Information Displayed none General information about the disk array LUN Information for the specified LUN All information displayed for options l d c 5 v and h Array controller status Disk status Hardware status I Serial numbers for all disk arrays Display performance metrics Rescan Make performance recommendations 1 S Generate raw output used in combination with other options General configuration information For a complete description of all the configuration settings see Viewing the Disk Array General Configuration Settings at the end of this chapter V l Capacity information l 244 Using the ARM Command Line Utilities for Windows NT Displaying Disk Array Serial Numbers Displaying Disk Array Serial Numbers
172. ing with one controller This setting will affect I O performance This is the default setting for single controller mode Secure Continually updates the disks with any changes in the controller maps decreased I O performance Normal Updates the maps on the disks at regular intervals typically 4 seconds This option offers both data protection and good performance This is This is the highest level of data protection but it may result in the default setting for dual controller mode HighPerformance Updates the disk maps only during shutdown of the disk array This is the lowest level of data protection but it offers the highest level of performance 272 Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Creating a Disk Array Alias An alias can be created to identify the disk array The alias provides an alternative to the disk array serial number and raw device file name that can also be used to identify the array Aliasing can be used in a variety of ways to help identify disk arrays in large systems For example by assigning numbers to racks and to the shelf positions within the racks each disk array can be uniquely identified using an appropriate alias If a rack is assigned number 12 the disk array installed on shelf 3 of the rack could be identified using an alias of R12 S03 This technique simplifies locating the disk array should it need service To
173. intenance tasks HP UX using ARM 50 MPE using ARM 142 NT using ARM 254 NT using AutoRAID Manager for Windows 234 management tools HP UX 14 MPE 116 managing a remote host NT using AutoRAID Manager for Windows 219 managing logical drives NT using ARM 247 NT using AutoRAID Manager for Windows 226 managing LUNs HP UX using ARM 42 HP UX using SAM 25 MPE using ARM 134 manpages for ARM 33 126 manually including a disk HP UX using ARM 45 HP UX using SAM 29 MPE using ARM 137 NT using ARM 250 NT using AutoRAID Manager for Windows 230 maps recovering HP UX using ARM 62 MPE using ARM 153 missing disk arrays possible causes HP UX using ARM 40 277 Index Index MPE using ARM 132 NT using ARM 245 NT using AutoRAID Manager for Windows 219 MPE configuration tips 119 MPE ARM utilities 125 N new disk array configuring HP UX using ARM 37 HP UX using SAM 23 MPE using ARM 129 NT using ARM 243 NT using AutoRAID Manager for Windows 224 NT event log 209 OpenView integration with NT 209 P performance See also Analyzing Performance See also Analyzing Performance metrics on NT 233 thresholds on NT 233 working set metric 49 141 primary controller switching HP UX using ARM 56 HP UX using SAM 32 MPE using ARM 148 NT using ARM 260 NT using AutoRAID Manager for Windows 238 printing log contents HP UX using ARM 53 using ARM 145
174. ion on security see Controlling Access to the Disk Arrays Server hangs during boot process after installing the HP SureStore E Disk Array 12H This may occur if there is no logical drive created on the disk array At least one logical drive must be created on the disk array before the host server can access the array During the boot process the server must recognize a logical drive on the disk array to ensure the HBA device driver is loaded properly If there is no logical drive on the disk array the server may hang during the boot process A logical drive can be created using the disk array control panel as described in the HP SureStore E Disk Array 12H User s and Service Manual 221 IN Chapter 9 Using AutoRAID Manager for Windows NT AutoRAID Manager for Windows can be used to perform all the tasks involved in normal management of your HP SureStore E Disk Array 12H AutoRAID Manager can be used for e Configuring a New Disk Array e Checking Disk Array Status e Changing Disk Array Configuration Settings e Managing Logical Drives e Adding a Disk e Rebuilding the Disk Array e Analyzing Disk Array Performance e Performing Disk Array Maintenance Tasks This chapter describes the steps involved in performing each of these tasks Online Help AutoRAID Manager includes detailed online Help If you need assistance in performing any task the Help should provide the guidance you need Selecting a Disk Arra
175. ioning within ODE The user runs ODE at the ISL prompt ISL ODE CK Ck CK Ck KKK kk kk kk ox OFFLINE DIAGNOSTIC ENVIRONMENT eek kk kk KKK KKK aon x C copyright Hewlett Packard Co 1994 5 All Rights Reserved NOR E KKK KKK KKK KKK KKK KKK EASIER OK TC Version XX XX XX SysLib Version NOS det kk kk KKK KKK KKK KKK KKK KKK KKK KKK KK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KKK KKK KKK KEK KKK KK KKK KKK KKK KKK KK Type HELP for command information ODE gt help BASIC COMMANDS HELP Prints detailed information when help lt command gt or help lt variable gt is typed LS Lists modules available on the boot media MODULE NAME Load and initialize the module RUN Run modul after setting environment variables CONTROL Y CONTROL C Abort an ODE command pause a module RESUME Restart a paused module DISPLOG After running a module display the contents of the log XIT Return to the next higher level prompt El VIRONMENT VARIABLES XN dH 111 Using the ARDIAG Offline Diagnostic on HP UX ODE interface SHOWSTATE Display the values of the fo
176. ions menu select Add and then select Using The Logical Volume Manager 20 3 Using the HP UX System Administration Manager Selecting Hardware Paths on HP UX 10 20 The Create a Volume Group window indicates the default hardware path to the LUN Click the Use Hardware Path button to change the path Create a Volume Group hpbs1991 ume groups you can add up or crea a new one Selected Disk HP AutoRAID LUN Use Hardware Path Yolume Group Name cal Volume Groups Modify Default Options Cancel 21 Using the HP UX System Administration Manager Selecting Hardware Paths on HP UX 10 20 4 Fromthe Alternate Paths window select the alternate non default hardware path to the LUN Click OK Alternate Paths hpbs 1991 22 Using the HP UX System Administration Manager Configuring a New Disk Array Configuring a New Disk Array After installing a new disk array you can perform the initial configuration using the disk array utilities This establishes the desired operating environment for the disk array To configure a new disk array 1 Plan your capacity management strategy Decide how you want to use the capacity of the disk array Factors such as data redundancy and performance influence how you manage the capacity To support your strategy it may be necessary to disable Active Hot Spare Auto Include Auto Rebuild See Managing Disk Array Ca
177. iptions CLRLOG The CLRLOG command allows the user to clear the internal logs of the specified target Synopsis CLRLOG Output ARDIAG CLRLOG Clearing the contents of array controller log at Indx 0 Dk Ck ck ck Ck ck kk ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck Ck ck Ck ck ck ck ck ck ck ck ko ck kk ck Sk Sk ko Sk kv kx Sk ko kx ok Successful Completion 169 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions DELETELUN The DELETELUN command allows the user to delete a LUN from the specified disk array Synopsis DELETELUN Output ARDIAG DELETELUN Enter the number of the LUN to delete default 0 Kk ck Ck ck ck Ck ck ck Ck Ck C Sk Ck Ck ck kk ck kk ck kk Ck kk kk ck kk ck kk Ck ck ck kk Sk kk ck kk ck kk Sk kc KKK KK DATA LOSS WARNING Ck ck ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck Sk ck ck kk kk ck ko Sk kv kx kx kv ko ko ko ko ko Continuing with DELETELUN WILL DEFINITELY RESULT IN DATA LOSS 11 information stored on the LUN to be deleted will be lost and the physical storage previously used will be freed Ck ck ck 0k ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck kk ck kk ck ko ck ko Sk ck Sk ko ko ko Lun 2 selected Delete selected lun default n ly Lun 2 was successfully deleted
178. isk Array Management Software The AutoRAID Manager ARM disk array management software is distributed on the IPR Independent Product Release CD ROM B6191AA The IPR CD ROM should have been ordered and delivered with your disk array The IPR CD ROM includes an instruction sheet for installing the software Follow the instructions to install the ARM software NOTE The HP SureStore E Disk Array 12H disk array requires IPR CD ROM release IPR9810 or later Earlier releases of the IPR CD ROM do not contain the required software At the time of printing the following ARMServer patches were available PHCO 15699 for HP UX 10 X PHCO 15700 for HP UX 11 0 The latest patches are available from the HP Patch web site www hp com go support which will indicate if the above patches have been superceded Operating System Support The HP SureStore E Disk Array 12H is currently supported on the following HP UX releases 10 01 10 10 10 20 11 0 IPR Program To provide you with the latest software patches and drivers Hewlett Packard offers the Independent Product Release IPR program As a subscriber to this program you will receive a complete collection of all updated patches and drivers at regular intervals thus ensuring you always have the latest software Using the latest software upgrades will ensure optimal performance of the disk array and other hardware peripherals Contact your Hewlett Packard sales representative for more de
179. ite Working Set resulting in a Working Set value less than or equal to 1 A Working Set value greater than 1 indicates that the Write Working Set is larger than the available RAID 0 1 space and the disk array is servicing writes from RAID 5 space If the Working Set consistently exceeds 1 the amount of RAID 0 1 space available should be increased to improve performance This can be accomplished in several ways as described in the following section If the Working Set is consistently much less than 1 some of the RAID 0 1 capacity can be allocated to a new LUN without impacting performance The remaining RAID 0 1 space should be adequate to accommodate the Write Working Set 141 Using the ARM Command Line Utilities for Performing Disk Array Maintenance Tasks Performing Disk Array Maintenance Tasks There are several tasks that you may have to perform in the on going management of the disk array These maintenance tasks are typically performed infrequently and may involve taking the disk array off line Shutting Down the Disk Array CAUTION When an array shutdown is performed the disk array becomes unavailable to the host system An array that is shutdown appears to the operating system as if its power has been turned off As with any disk subsystem it is essential that file system access to the disk array be correctly removed before shutting down or powering off the disk array Before an array Shutdow
180. ites out of write cache and map journal to disks This time is specified in seconds This maximum time is submitted to the scheduler for implementation Depending on the priority of other events in the scheduler the desired time may not be implemented If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored The value of this field is not the same as disabling NVRAM use A value of Oh indicates that no maximum time will be enforced and that no recovery image will be available Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Simplified Resiliency Setting The simplified resiliency setting is derived from several other fields Collectively these settings control how data resiliency is managed Table 31 identifies the field settings for each of the resiliency modes There are four modes of data resiliency Normal SingleController Secure and HighPerformance The default setting is Normal for dual controller disk arrays If the disk array has been ordered with a single controller the default for this settings is SingleController For information on changing the data resiliency mode see Setting Data Resiliency in this chapter 269 IN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings 270 Table 31 Simplified Data Resiliency Settings Normal Mode Default P
181. k Sk kk ck kk Sk KKK KKK KKK About to update Unknown Product currently at firmware revision 0256 with new firmware revision 0284 Dk ck ck ck Ck ck kk ck Ck Ck ck Ck ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck kk ck ko ck ko ko ck ko kk kx kx ke k ko ko ko ko Notes for this firmware release 1G and 2G 5400 RPM Seagate Do you want to do the update Please be minutes WAR IF YOU INTI ING patient from SEA5400 y n y The firmware download process may take several ERRUPT THIS PROCESS THE DEVICE COULD BE RENDERED INOPERATIVE Shutting down the array Downloading the firmware image If this is the last firmware download you want to perform on this array you take effect wai can reset the array ting to do the reset after the last which causes firmware changes on the disks to If you have more downloads to do you will save time by one Note that a download to the array controller causes an automatic reset Do you want to perform a reset y1 n y Please wait while the array performs a reset Dk ck ck ck Ck ck Ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck oko ck kk ck kk kk ck ko ck ok kx kx ko ko ESSFULLY Firmware downloaded SUCC KKK Ck ck ck Ck ck ck Ck ck Ck Sk KKK KKK KKK ck kk Ck Sk ck kk ck kk ck kk kk ck ko kk kk ko kk
182. k ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck cko ck KK ck ck ck ko ck ko Sk Sk ko ko NOTICE Ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck ck Ck Ck Ck Sk Ck Ck Ck Ck ck ck ck ck ck ck ck Sk kk kk kk ko ko kock ok The following FORMAT will require approximately ONE HOUR to complete During this period you will be unable to perform any other diagnostic activities KC Ck ck ck ck ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck ck Ck kk Ck Ck Ck ck ck ck ck ck ck kk kk kk Do you want to continue y n y FORMATTING Dk ck ck ck Ck ck Ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck Ck ck ck ck ck Ck ck ck ck ck ck cko ck Ck ock ck ck ck ko ck kv kx Sk ko ko FORMAT Successfully Completed ck ck Ck ck ck kk Ck Sk ck Ck ck Ck ck ck kk ck kk Ck kk kk ck kk ck kk KKK Ck ck ck kk ck kk ck kk Sk kc k ko kc KKK KK ARDIAG gt Additional notes 178 Any problems detected by the FORMAT command will be displayed with an appropriate error message and a prefix of ERROR INVALID or FAILED Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions INQUIRY The INQUIRY command allows the user to view SCSI Inquiry data from the specified target Synopsis INQUIRY Output ARDIAG INQUIRY Performing Inquiry at TestLevel Indx 0 Ck ck
183. k ock ck ko ck ko ko ck Sk Sk Mk kx kv Mk ko ko Vendor ID HP Product ID 1 050 GB 3rd Page Code 0 Supported Log Sense Pages Page Code Page Code Error Counter Page Write Page Write Errors Error Counter Page Read Page Read Errors Page Code Error Counter Page Verify Page Verify Errors Page Cod Non Medium Error Page Non Medium Page Code 37 Vendor Unique or Non SCSI 2 Standard Vendor Unique Page Code 3e Vendor Unique or Non SCSI 2 Standard Vendor Unique Co Page Code 2 Write Errors Errors Corrected With Possible Delays otal Errors Corrected By Applying Retries otal Errors Corrected Total Times Correction Algorithm Processed Total Bytes Processed Total Uncorrected Errors 0 0000000020046200 Or Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Page Code 3 Read Errors Errors Corrected Without Substantial Delay Errors Corrected With Possible Delays otal Errors Corrected By Applying Retries otal Total Total Total Errors Corrected Times Correction Algorithm Processed Bytes Processed Uncorrected Errors Page Code 5 Verify Errors oop ROO 8 x000000000de82c00 Total Total Total Total Total Errors Corrected Without Substantial Delay Errors Corrected With Possible Delays Errors Correct
184. l pane 2 Click Properties 3 Click the Diagnostics button on the toolbar 4 Under Logs click Restart 238 Using AutoRAID Manager for Windows NT Downloading Controller Firmware Downloading Controller Firmware AutoRAID Manager includes a separate utility for downloading controller firmware This utility WinDownload simplifies the process of downloading new firmware to the controller in your disk array The WinDownload utility is used only for downloading firmware to the disk array controllers Firmware can also be downloaded to the disk modules using the ARM download command line utility described in Downloading Firmware to a Disk Module in the next chapter The download process shuts down the disk array while the firmware is being downloaded so the disk array will be inaccessible while the download is in progress CAUTION In multi host configurations other hosts must not access the disk array while the download is in progress Data can be lost if write requests are made to the disk array while a download is in progress Do not attempt to download controller firmware to a disk array that is serving as the NT boot device If you attempt to do so the operating system will crash To download controller firmware 1 Alert users that data on the disk array will be inaccessible during the download 2 From the Start menu select Programs AutoRAID Manager WinDownload 3 Select the disk array you want to upgrade with ne
185. lick Properties Click the Diagnostics tab Under Array click Shutdown Aus wn Click OK to confirm the Shutdown 234 Using AutoRAID Manager for Windows NT Performing Disk Array Maintenance Tasks Restarting the Disk Array The disk array must be restarted after being shutdown Restart returns the disk array to operation ready to once again service I Os from the server To restart the disk array Right click the disk array from the Detail pane 2 Click Properties 3 Click the Diagnostics tab 4 Under Array click Restart Resetting the Disk Array Resetting the disk array returns it to the power on state Access to the disk array may be interrupted during a reset To reset the disk array Right click the disk array from the Detail pane 2 Click Properties 3 Click the Diagnostics tab 4 Under Array click Reset Formatting the Disk Array WARNING A Format will destroy all the data on the disk array Make sure you backup all the data you want to keep before formatting the disk array The entire disk array can be formatted if necessary However the disk array can only be formatted if there are no logical drives on the array If there are logical drives on the disk array they must be deleted before the array can be formatted To format the disk array Right click the disk array from the Detail pane 2 Click Properties 3 Click the Diagnostics tab 235 LN SMOpUIM Windows NT Usin
186. lities Record the serial number for future reference 3 Ifthe planning in step 1 requires you to disable any of the configuration settings to implement your capacity management strategy do so now These settings include Active Hot Spare Auto Rebuild and Auto Include Change the configuration settings by typing arraymgr h array id Active Hot Spare arraymgr a 4 on off array id Auto Rebuild arraymgr i on off array id Auto Include NOTE Only one setting can be changed on each command line 4 Check the available unallocated capacity on the disk array by typing arraydsp SID The total unallocated capacity available for creating logical drives will be displayed Make sure there is adequate capacity to create the logical drive structure you need 5 Create each logical drive on the disk array by typing arraycfg L LUN a capacity array id Example arraycfg L 0 a 1000 00786B5C0000 243 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Checking Disk Array Status This command creates logical drive 0 with a capacity of 1000 Mbytes on the array identified by serial number 00786B5C0000 This step makes disk array capacity available to your operating system and it must be repeated for each logical drive to be created Make sure you observe any operating system limitations on logical drive size or number For more information see Creating a Logical Drive
187. llection Using the ARDIAG Offline Diagnostic HP UX ARDIAG Operational Commands ARDIAG Operational Commands Table 11 lists the ARDIAG operational commands and their descriptions Table 11 ARDIAG Operational Commands Command CLRLOG CREATELUN DELETELUN DESCRIBE DOWNLOAD FORMAT INQUIRY READLOG RECOVER REQSENSE RESTART REVISION ROMT SETOPTIONS SHUTDOWN WRTMT Description Allows the user to clear the specified target s internal logs Allows the user to create a LUN via the specified controller path to the disk array Allows the user to delete a LUN via the specified controller path to the disk array Displays LUN configurations and warning states Allows the user to download firmware to the specified controller or internal disk Allows the user to format the specified target Issues an Inquiry to an array controller or an internal disk Allows the user to read the specified target s internal logs Allows the user to restore controller map information in the specified SureStore E AutoRAID Disk Array Issues a Request Sense on the specified target Allows the user to bring the array out of the Shutdown state via the specified controller path Displays controller drives and software revisions Read Verify test to the specified internal mechanism Allows the user to view and set specific options for the array controller Allows the user to put the array into a Shutdo
188. llowing environment variables LOOP Run a test this many times ERRPRINT ON OFF Print low level error messages ERRNUM ON OFF Print one line numbered errors ERRPAUSE ON OFF Pause module on error ERRONLY ON OFF Print only error messages INFOPRINT ON OFF Print informational messages ISOPRINT ON OFF Print fault isolation messages ISOPAUSE ON OFF Pause module when isolation message is generated 106512 Set the size of the message log DEFAULT Reset environment variables to default state ODE 15 odules on this boot media are filename type siz created description ARDIAG XXXX XXX XXXX XXXXXX 112 Using the ARDIAG Offline Diagnostic HP UX ARDIAG interface to ODE ARDIAG interface to ODE ODE ARDIAG kk ck Ck ck ck Ck ck ck Ck KKK KKK KKK KKK KK KKK KKK ck kk KKK ck kk Ck kk kk ck KKK kk ck kk ko kk kk ck ko KKK KKK KKK KKK KKK KKK KKK KKK ARDIAG KKK KKK KKK KKK KKK KKK EARRAS C copyright Hewlett Packard Co 1996 ERES RA ERRER E All Rights Reserved EE kkk kx KKK kk KKK kk KKK KKK KKK kk KKK KKK SE Version XX XX XX EE KKK KKK KKK KKK KKK KKK KKK KKK KKK KK KKK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KKK KKK KKK KEK KKK KKK KK KKK KKK KKK KK Please wait while I scan the device busses Test Disk Path Product String Rev Size 0 56 40 1 0 HPC3586 AUTORAID disk array 2 1 0 Legend TestDisk Indx of the array liste
189. loff array id To change Rebuild Priority setting type arraymgr p high low array id 246 Using the ARM Command Line Utilities for Windows NT Managing Logical Drives Managing Logical Drives An important part of managing the disk array involves defining and maintaining the optimal logical drive structure for your system Your system requirements and limitations will influence the logical drive structure you choose Managing logical drives is a part of the overall task of managing disk array capacity For more information on managing disk array capacity to meet your system needs refer to the HP SureStore E Disk Array 12H User s and Service Manual Checking Logical Drive Configuration When you are managing logical drives you may find it convenient to check the current logical drive configuration and the available capacity To check the current logical drive configuration and the available capacity type arraydsp 1 LUN lt array id gt Creating a Logical Drive Only capacity assigned to logical drives is visible to the operating system When selecting the size for your logical drives consider the following factors e Any size limitations imposed by the operating system e Your backup strategy If you do unattended backup to a device such as a tape you may want to avoid creating a logical drive that is larger than the capacity of the tape media This allows you to back up an entire logical drive without
190. ltaneously CAUTION Do not attempt to download controller firmware to a disk array that is serving as the NT boot device If you attempt to do so the operating system will crash NOTE Downloading firmware to the disk array controllers is done most efficiently using the WinDownload utility described in Downloading Controller Firmware in the preceding chapter To download new firmware code to the disk array controllers 1 Gather the following information e The ID of the disk array The disk array ID can be displayed using the arraydsp i command The location and name of the file containing the new firmware code 2 Froma command prompt type in the following command download C codefilename array id codefilename identifies the file containing the firmware code 261 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Downloading Firmware Copying Firmware From the Primary Controller to the Secondary Controller For proper disk array operation the disk array controllers must have matching firmware If the firmware revisions do not match this procedure can be used to synchronize them Firmware can also be copied from the primary to the secondary controller using the disk array control panel CAUTION The firmware code is copied from the primary controller to the secondary controller so before performing this task make sure the controller that has the desirable code is designated the prim
191. lude a disk automatically even if Auto Include is enabled This will occur if the new disk s status is something other than Normal See Auto Include in the HP SureStore E Disk Array 12H User s and Service Manual for more information about when this might occur 3 Depending on how you intend to use the new disk perform the appropriate next step To use the disk to increase capacity create a logical drive using all or a portion of the disk capacity For more information see Creating a Logical Drive in this chapter e To use the disk to increase performance leave the disk capacity unallocated e To use the disk capacity for an Active Hot Spare enable the Active Hot Spare feature if not currently enabled For more information see Changing Disk Array Configuration Settings in this chapter 249 LN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Adding a Disk Including a Disk Manually A disk must be included in the disk array configuration before it can be used by the disk array There are two ways to include a disk e You can enable Auto Include which will automatically include a disk when it is inserted into the disk array enclosure e You can manually include each new disk For convenience Auto Include is enabled by default on a new disk array For information on disabling Auto Include see Changing Disk Array Configuration Settings in this chapter After including a disk y
192. main off line until a failure of the primary controller is detected at which point it becomes primary Only the primary controller will go on line Very Early Busy DISABLED Controls the behavior of the disk array with respect to SCSI bus selection during the early stages of the initialization sequence i e from about three seconds after reset until about fifteen seconds prior to initialization completion When disabled the disk array will ignore SCSI bus selection until the late stages of initialization When enabled the disk array will accept selection during early initialization and will return BUSY status until the late stages of initialization Queue Full Threshold 1952 Specifies the target queue depth beyond which the disk array will return QUEUE FULL status to subsequent host requests 67 XN dH Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Parameter Maximum Queue Full Threshold Simplified Resiliency Setting Single Controller Warning Lock Write Cache On Disable NVRAM on WCE False Disable NVRAM with One Disable NVRAM on UPS absent Default Setting 1952 Normal ENABLED TRUE FALSE TRUE FALSE Description Specifies the maximum supported value for the Queue Full Threshold parameter Attempts to set the QFT parameter to values higher than the MQFT will fail with CHECK CONDITION status and ILLEGAL RE
193. mand Line Utilities for Viewing the Disk Array General Configuration Settings Table 22 Simplified Data Resiliency Settings Normal Mode Default Parameter Setting Simplified Resiliency Setting Results Single Controller ENABLED Single Controller Dual Controller Warning Lock Write Cache On TRUE Not Supported Data map and write cache Disable NVRAM FALSE will be 21 WCE False to the disk every four Disable with TRUE seconds Disable FALSE UPS absent Force Unit Access 2 Response Disable Read Hits FALSE Resiliency Threshold 4 SingleController Mode Default Parameter Setting Simplified Resiliency Setting Results Single Controller DISABLED Single Controller Dual Controller Warning Lock Write Cache FALSE Data map and write cache Not Supported Disable NVRAM on TRUE Jinformation will be flushed WCE False to the disk once every Disable NVRAM with TRUE second One Disable NVRAM on FALSE UPS absent Force Unit Access 2 Response Disable Read Hits FALSE Resiliency Threshold 1 162 Secure Mode Parameter Single Controller Warning Lock Write Cache On Disable NVRAM on WCE False Disable NVRAM with One Disable NVRAM on UPS absent Force Unit Access Response Disable Read Hits Using the ARM Command Line Utilities for MPE Viewing the Disk Array General Configuration S
194. mation is controlled using the data resiliency setting Selecting a data resiliency setting involves making a tradeoff between data protection and performance The more frequently the recovery maps are updated the more impact it may have on performance 148 Using the ARM Command Line Utilities for MPE Performing Disk Array Maintenance Tasks To set the data resiliency mode type arraymgr J SingleController Secure Normal HighPerformance array id The data resiliency settings are described in Table 19 Table 19 Data Resiliency Settings Setting Description SingleController Secure HighPerformance This setting should only be used if the disk array is operating with one controller This suppresses the single controller warning messages that are normally generated when the disk array is operating with one controller This setting will affect I O performance This is the default setting for single controller mode Continually updates the disks with any changes in the controller maps This is the highest level of data protection but it may result in decreased I O performance Updates the maps on the disks at regular intervals typically 4 seconds This option offers both data protection and good performance This is the default setting for dual controller mode Updates the disk maps only during shutdown of the disk array This is the lowest level of data protection but it offers the highest level of performance
195. may take several ERRUPT THIS PROCESS THE DEVICE COULD BE RENDERED INOPERATIVE Shutting down the array Downloading the firmware image this is the last firmware download you want to perform on this array If you take effect wai can reset the array ting to do the reset after the last which causes firmware changes on the disks to If you have more downloads to do you will save time by one Note that a download to the array controller causes an automatic reset Do you want to perform a reset y1 n y Please wait while the array performs a reset Dk ck ck ck Ck ck Ck ck Ck ck ck Ck ck Ck ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck kk ko ko ck ko Sk ko kx ko ko ko ESSFULLY Firmware downloaded SUCC kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk ck Ck Sk ck Ck ck ck Ck Sk ck kk Ck Sk ck kk ck kk ck kk Ck kk kk Sk kk ck kk ck kk Sk kc k ko ko KK KKK ARDIAG gt Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions FORMAT The FORMAT command allows the user to format the specified target Synopsis FORMAT Output Example 1 Controller ARDIAG TESTLEVEL 0 ARDIAG FORMAT kk ck Ck ck ck Ck ck ck Ck Ck Ck Sk ck Ck KKK KKK KKK KKK ck kk ck kk ck kk Ck kk kk Sk kk Sk kk ck kk ko kc k ko KKK KKK p Array Subsystem FORMAT ck ck ck ck ck ck ck ck Ck Ck Ck
196. mber log directory name identifies the location of the log files start time and stop time limit the output to events between the specified times record type identifies the type of record s to print Records include system usage log usage disk error log disk controller error log ct x1r system change log change and performance log perf array serial number limits the output to only those entries associated with the specified disk array Displaying Hardware Logs In addition to the logs maintained by ARMServer hardware logs are also stored on the disk array The arraylog command provides access to the controller and disk logs maintained by the disk array These logs contain information useful for diagnosing and troubleshooting the disk array The logs can also be cleared using arraylog The arraylog options for accessing the disk array hardware logs are listed in Table 18 145 Using the ARM Command Line Utilities for Performing Disk Array Maintenance Tasks Table 18 arraylog Options for Displaying Log Contents Option Description Display the contents of the disk array controller usage log Display the contents of the disk array controller event log d slot Display the contents of the log for the disk installed in the cabinet slot identified by slot Slot numbers must be of the form An or Bn where A or B correspond to a cabinet column and n corresponds to a shelf position 1 6
197. mpossible for the host and disk array to communicate properly In this case it will be necessary to use the disk array control panel to return the SCSI setting to its original value to reestablish communication To change SCSI settings type arraymgr W T m r onloff array id Changing the Controller SCSI ID Changing the SCSI ID directly impacts the operating system s ability to access the disk array Before performing this task check your operating system documentation for information on how to change the SCSI ID of a disk subsystem To change the controller SCSI ID type arraymgr C X Y addr array id addr is the new address 0 15 for the indicated controller 147 Using the ARM Command Line Utilities for Performing Disk Array Maintenance Tasks Switching Primary Controllers In dual controller disk array configurations the array automatically switches to the secondary controller if the primary controller fails However you can switch controllers manually if necessary This will cause the secondary controller to assume the role of primary controller To switch primary controllers type arraymgr c X Y array id Setting Data Resiliency The data map contents stored in the disk array controller NVRAM is copied to two disks on the array at regular intervals to protect against map loss The interval at which the disks are updated with recovery map infor
198. mprove performance use the disk to improve the disk array performance by simply leaving it as unallocated capacity Enable Active Spare use the additional capacity to enable Active Hot Spare if the disk array does not currently have the capacity to support this feature This also improves performance as the spare space is used as RAID 0 1 space until it is needed To add a disk to the array 4 Make sure the new disk has been physically inserted into the array 5 If Auto Include is on the disk is automatically added to the array and you can skip to the next step If Auto Include is off manually include the disk as described in the next section Including a Disk NOTE In some situations the array will not include a disk automatically even if Auto Include is enabled This will occur if the new disk s status is something other than Normal See Auto Include in the HP SureStore E Disk Array 12H User s and Service Manual for more information about when this might occur 6 Depending on how you intend to use the new disk perform the appropriate next step e To use the disk to increase capacity create a LUN using all or a portion of the disk capacity For more information see Creating LUN in this chapter To use the disk to increase performance leave the disk capacity unallocated e To use the disk capacity for an Active Hot Spare enable the Active Hot Spare feature if not currently enabled For more information se
199. multi initiator environment make sure none of the host systems are trying to access the disk array Bus activity make cause ARDIAG to not see the disk array Support Software ARDIAG requires all the software needed for ODE to run as ARDIAG relies on ODE to load and run Additionally ARDIAG needs the MAPPER SL and the DIODE SL from the protocol level down Minimum hardware For ARDIAG to run properly a working I O subsystem and at least one SureStore E Disk Array is needed ARDIAG must also meet ODE s hardware requirements for example a working console ARDIAG and ODE will use no more than 4 MB of main memory Minimum software ODE MAPPER SL and DIODE SL are needed to execute ARDIAG correctly 165 Using ARDIAG Offline Diagnostic MPE Definition of Terms Definition of Terms The following terms are used in this chapter MPE SIO SERIES 800 ISL ODE LIF SOM 166 Hewlett Packard s version of the MPEG Operating System Server I O The driver environment for the current Series 800 MPE system Proprietary non reentrant port server drivers Communication is via message passing MPE PA RISC multi user systems These systems utilize the 5 0 subsystem Initial System Loader This is the first piece of software loaded from outside the SPU and executed during the boot process Offline Diagnostic Environment The software architecture consisting of several modules to provide diagnost
200. n hich to perform a rebuild process Active Hot Spare Auto Include Desired Indicates the action to be taken when a drive is physically inserted into the array Auto Rebuild ENABLED Indicates the action to be taken when a drive becomes unusable Indicates the priority the rebuild process is given with respect to host I O 64 Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Capacity Depletion 0 Indicates the amount of space below which the disk array Threshold should signal a Capacity Depletion warning For example if this field is set to 99 then when the disk array reaches 99 capacity a Capacity Depletion warning will be indicated 0 means that Capacity Depletion warnings ill not be issued Write Working Set 8640 ndicates the period ten second intervals over which Interval seconds write performance measurements should be gathered For example 8640 X 10 86400 seconds 24 hours Language ENGLISH Indicates the language used when displaying information lon the front panel Log Full Warning DISABLED Indicates whether the disk array should assert warning hen some portion of the controller log is full When enabled the disk array will assert a Log Full Warning hen one of the disk array log pages fills or a log parameter reaches its maximum value When disabled no Log Full War
201. n is performed all mounted file systems mapped to the disk array must be unmounted The umount operation synchronizes data in the MPE internal buffers with the data stored on the disk array The disk array must be Shutdown prior to performing any maintenance The Shutdown process copies vital data mapping information from the controller NVRAM to the disks This protects the data mapping information should the contents of the NVRAM be lost or corrupted due to battery failure Shutdown then takes the disk array off line making all data unavailable to the host The disk array can still be managed and tested but all data is inaccessible while the disk array is Shutdown Shutdown is initiated automatically each time the disk array is turned off using the power switch so it is usually not necessary to initiate a Shutdown using the ARM utility To Shutdown the disk array type arraymgr s shut lt array id gt Restarting the Disk Array Following Shutdown the disk array can be brought back on line by performing a restart This makes the data on the disk array available to the host once again To restart the disk array type arraymgr s start lt array id gt 142 Using the ARM Command Line Utilities for MPE Performing Disk Array Maintenance Tasks After the array completes initialization file systems mapped to the disk array must be mounted for MPE to access the disk array Resetting the Disk Array The disk array can be reset if ther
202. n the array To change configuration settings 1 Select the desired disk array controller from the list on the screen Qv ue 090 09 Select Actions on the menu bar Select Disk Array Maintenance Select Modify Array Configuration Click setting boxes to make changes Click to effect the change or Cancel to ignore Managing LUNs Logical Drives An important part of managing the disk array involves defining and maintaining the optimal LUN structure for the disk array Your system requirements will influence the LUN structure you choose Managing LUNs is a part of the overall task of managing the disk array capacity For more information on managing array capacity to meet your system needs see Managing Disk Array Capacity in the HP SureStore E Disk Array 12H User s and Service Manual 25 XN dH Using the HP UX System Administration Manager Managing LUNs Logical Drives Checking LUN Configuration Anytime you are managing LUNs you may find it convenient to check the current LUN configuration and the available capacity The current LUN definitions can be seen on the standard Disk Devices screen The available capacity is displayed as Unallocated on the View Array Status Information screen See Checking Disk Array Status in this chapter Creating a LUN Only capacity assigned to LUNs is visible to the HP UX operating system When creating LUN
203. n the array Rebuild High Rebuild priority determines how quickly a rebuild operation will Priority complete 225 IN Windows NT Using AutoRAID Manager for Windows NT Managing Logical Drives To change disk array configuration settings Right click the disk array from the Detail pane 2 Click Properties 3 Click the Configuration tab 4 Select the appropriate value for each setting Managing Logical Drives Establishing and managing the logical drive structure of your disk array is an important management task Your system requirements and file directory structure influence the logical drive strategy you choose Managing logical drives is a part of the overall task of managing disk array capacity For more information on managing array capacity to meet your system needs see the Concepts and Management chapter in the HP SureStore E Disk Array 12H User s and Service Manual Creating a Logical Drive Creating a logical drive makes array capacity available to the operating system Each logical drive appears to the host as a physical disk device Logical drives are created during initial installation of the disk array and typically after installing a new disk module To create a logical drive Right click the disk array from the Detail pane 2 Click Properties 3 Click Logical Drives 4 Click Create to display the Create Logical Drive window If there is no unallocated capacity on the disk array
204. name Intended Product ID Rev Size Z1CS HPC3586A disk array Z1CS 1048576 SEA5400 SEAGATEST31230N disk drive 0284 261632 5267200 Generic Disk 1208 261632 5467200 ST15150W disk drive 1207 261632 Legend File name name of the firmware file Intended Product ID firmware file s intended product name Rev firmware Revision of the firmware file Size exact byte size of the firmware image Enter the firmware file name displays the files Default quits sea5400 WARNING Firmware file SEA5400 was made for a SEAGATEST31230N disk drive TESTDISK selects a Unknown Product STOP CONTINUING MAY PERMANENTLY DESTROY ALL DATA ON DISK Do you still want to continue y n Y 83 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions 84 kk ck Ck ck ck Ck ck ck Ck ck Ck Sk ck Ck ck ck Ck ck ck kk ck kk Ck ck ck kk ck kk ck kk ko kk kk Sk kk ko kk ck kk Sk kc k ko kc KKK KK About to update Unknown Product currently at firmware revision 0256 with new firmware revision 0284 Dk ck ck ck Ck ck Ck ck Ck ck ck 0k ck CC ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ok ck ck ck ko ck ko ko ck ko Sk ko kx kx kv ko ko ko koko Notes for this firmware release 1G and 2G 5400 RPM Seagate Do you want to do the update Please be minutes WAR IF YOU INTI ING patient from SEA5400 y n y The firmware download process
205. nd the disk array to communicate properly To view the general configuration settings type arraydsp s array id A complete list of the settings will be displayed Table 21 General Configuration Settings Default Description Setting READY Specifies the current state of the array ENABLED Indicates whether the array should reserve space within hich to perform a rebuild process Overall State of Array Auto Include Active Hot Spare Desired Indicates the action to be taken when a drive is physically inserted into the array Auto Rebuild ENABLED Indicates the action to be taken when a drive becomes unusable Indicates the priority the rebuild process is given with respect to host I O 155 Using the ARM Command Line Utilities for Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Capacity Depletion 0 Indicates the amount of space below which the disk array Threshold should signal a Capacity Depletion warning For example if this field is set to 99 then when the disk array reaches 99 capacity a Capacity Depletion warning will be indicated 0 means that Capacity Depletion warnings ill not be issued Write Working Set 8640 ndicates the period ten second intervals over which Interval seconds write performance measurements should be gathered For example 8640 X 10 86400 seconds 24 hours Indica
206. ndicates whether the disk array should check SCSI bus Terminator Power ENABLED Indicates whether the disk array should provide power for Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Array SCSI N A configuration Controller X SCSI Indicates the SCSI bus address to be used by controller Address X Changes in this field will take affect only after controller X is reset Controller Y SCSI Indicates the SCSI bus address to be used by controller Address Y Changes in this field will take affect only after controller Y is reset Write Cache ENABLED Indicates whether the disk array should cache write data Generally write cache is used even though a host system automatically disables this field Note This field does not represent what is actually happening The disk array uses read cache algorithms This field is put in place to facilitate some third party operating systems parity Disabled means that bus parity checking is disabled Enabled indicates that bus parity checking is enabled SDTR ENABLED Indicates whether the disk array should initiate SDTR Synchronous Data Transfer Request Disabled indicates the disk array will initiate SDTR WDTR ENABLED Indicates whether the disk array should initiate Wide Data Transfer Request Disabled indicates the disk array will not initiate WDTR Enabled indica
207. nfigure a new disk array 1 Plan your capacity management strategy and LUN configuration Decide how you want to use the disk array capacity Factors such as data redundancy and performance influence how you manage the capacity See Managing the Disk Array Capacity in the HP SureStore E Disk Array 12H User s and Service Manual for help in planning your strategy Also see Tips for Configuring the Disk Array On HP UX in Chapter 1 of this guide Display the serial number of the disk array by typing arraydsp i The serial number provides a way to identify disk arrays when using the ARM utilities Record the serial number for future reference or set a shell variable to hold this value for example export ID 00786B5C0000 Or export ID dev rdsk c2t0d0 If the planning in step 1 requires you to disable any of the configuration settings to implement your capacity management strategy do so now These settings include Active Hot Spare Auto Rebuild and Auto Include Change the configuration settings by typing arraymgr h array id Active Spare arraymgr a array id Auto Rebuild arraymgr i array id Auto Include NOTE Only one setting can be changed on each command line 37 XN dH Using the ARM Command Line Utilities for HP UX Configuring a New Disk Array 4 Check the available unallocated capacity on the disk array by typing arraydsp SID The
208. ng assigned to the array using the arraymgr command For example assume a disk array has a serial number of 00786B5C0000 special raw device file dev rdsk c2t0d0 is defined for this array and it has an alias of autoraid4 To check the available unallocated capacity on this particular disk array you could use any of the following commands arraydsp 00786B5C0000 arraydsp dev rdsk c2t0d0 arraydsp autoraid4 The serial number of all arrays in the system can be obtained using the command arraydsp i For more information see Displaying Disk Array Serial Numbers in this chapter A list of all special device filenames assigned to the disk array can be displayed using the following command ioscan fn C disk Many commands affect the operation of the entire disk array regardless of whether the array id field contains the array serial number or a special device file name Commands that involve only a specific LUN on the array will include an option L LUN for identifying the LUN involved For example to format LUN 3 on a disk array that is referenced by dev rdsk c2t0d0 you would use the following command arrayfmt L 3 dev rdsk c2t0d0 36 Using the ARM Command Line Utilities for HP UX Configuring a New Disk Array Configuring a New Disk Array After installing a new disk array you can perform the initial configuration using the ARM utilities This establishes the operating environment for the disk array To co
209. ning will be indicated Volume Set DISABLED Indicates whether the disk array should boot when half or Partitioning more of the previously available drives are unavailable i e there is no drive quorum Enabled indicates that volume set partitioning should be allowed Disabled the system will remain in the No Quorum state when the required quorum is not available Format Pattern Fill DISABLED Indicates whether the disk array will fill in incomplete RAID blocks with a format pattern when performing new writes Some operating systems not HP UX and not NT expect that SCSI format commands completely reinitialize data to a non random pattern When using the array with of those operating systems pattern filling should be enabled Disk array Type ID 12 Uniquely identifies the disk array hardware configuration This field can be used to find the number of drives and SCSI channels supported within the disk array LUN Creation Limit 8 Controls the range of LUNs that may be created Maximum LUN 8 Specifies the maximum supported value for the LUN Creation Limit Creation Limit 65 XN dH Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Parameter Default Description EE Array SCSI NA configuration Controller X SCSI Indicates the SCSI bus address to be used by controller Address X Changes in this field will take affect only aft
210. ningful analysis and recommendations For example if the highest load on your disk array occurs between the hours of 8 AM and 5 PM restrict the analysis to this time period If periods of lower activity are included the analysis may yield different results 232 Using AutoRAID Manager for Windows NT Analyzing Disk Array Performance and consequently different recommendations This occurs because activity is averaged over the entire analysis period and periods of lower activity will offset the effects of busier periods For the most accurate analysis select a period of time that represents normal system operation Avoid any unusual events such as a rebuild or changes made to array capacity If you select a time period that includes an event which may distort the analysis the utility will alert you and will not provide any recommendations To analyze disk array performance Right click the disk array from the Detail pane 2 Click Properties 3 Click the Performance tab 4 Position the interval marker over the desired time period in the Performance Events window If the time period you want is not displayed use the horizontal scroll bar to display the desired time 5 Click Recommend to display the Performance Recommendations window Displaying Performance Metric Thresholds Several of the performance metrics maintained by the disk array can be displayed These are the same metrics used by AutoRAID Manager to produce performanc
211. nterval 0 13 SDTR 27 Subsystem Identifier 13 14 WDTR O Select the number of the parameter to modify Default quits 2 Allowed values for Active Hot Spare 0 OFF 1 ON Select th number of new parameter valu Default quits 1 191 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions 192 Modify the parameter with th Subsystem Parameter ES Subsystem State Active Hot Spare Volume Set Partitioning Rebuild in Progress Auto Rebuild Auto Include Balance in Progress Optimize in Progress Migra ting Write Destination Log Full Warning Rebuild Priority Pari SDTR WDTR ty selected value y n y Value Subsystem Parameter Value Ready 15 Termination Power ON 16 Unit Attention ON 17 Controller X Address 0 OFF 18 Controller Y Address 1 19 Enable Manual Override OFF 20 Manual Override Destination OFF OFF 21 Format Pattern Fill ON 22 Disable Remote Reset OFF OFF 23 Language English 24 Capacity Depletion Threshold 0 OFF 25 Rebuild Progress 0 O 26 Write Working Set Interval 0 27 Subsystem Identifier 1 3 Additional notes SCSI Specification Ready Warning Not Enough Drives No Quorum No Map Shutting Down Shut Down Starting Up No Code RESERVED Shutdown Warning RESERVED Mismatched Code Controller Mismatch Using the ARDIAG Offline Diagnostic on MP
212. ntroller The internal disks are identified using the REVISION command which will display the disks within the array along with their corresponding TESTLEVEL identifier The TESTLEVEL command must be set to a value from one to twelve to indicate the appropriate disk selection 173 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions Output Example 1 Controller 174 ARDIAG testlevel 0 TESTLEVEL 0 ARDIAG download T KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK Array Controller DOWNLOAD id KKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKKK he following options are available for DOWNLOAD to the array controller 1 Download a firmware image file to the primary controller with an automatic update of same firmware to the secondary controller 2 Update the secondary controller with the same firmware as the primary controller Choose 1 2 The primary controller is at XXXX Do you want to do the update y n y Please be patient The firmware download process may take several minutes WARNING IF YOU INTERRUPT THIS PROCESS THE DEVICE COULD BE RENDERED INOPERATIVE Shutting down the array Mirroring the microcode image Please wait while the array performs a reset Dk ck ck ck 0k ck Ck ck Ck Ck ck 0k ck Ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck
213. ntroller The primary controller is identified using the INQUIRY command If the secondary controller is identified you must change the controller selection using the TESTDISK command The TESTLEVEL command must be set to zero to indicate a controller selection A presently installed secondary controller is automatically updated with the new firmware image after a successful DOWNLOAD to the primary controller Thus there is no interaction needed to update the presently installed secondary controller A newly installed secondary controller can be interactively updated to the firmware image that is presently in the primary controller using the DOWNLOAD command After successful DOWNLOAD the REVISION command can be used to verify firmware revision Reset ARDIAG to update DISPMAP revision information DOWNLOAD to an internal disk The DOWNLOAD process to an internal disk can be performed through either the primary or secondary controller The internal disks are identified using the REVISION command which will display the disks within the array along with their corresponding TESTLEVEL identifier The TESTLEVEL command must be set to a value from one to twelve to indicate the appropriate disk selection 81 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Output Example 1 Controller 82 ARDIAG testlevel 0 TESTLEVEL 0 ARDIAG download T KKKKKKKKKKKKK
214. o use the capacity of the disk array Factors such as data redundancy and performance influence how you manage the capacity For example you can increase the amount of available capacity by disabling Active Hot Spare but this may mean you sacrifice data redundancy in the event of a disk failure See the Concepts and Management chapter in HP SureStore E Disk Array 12H User s and Service Manual for help in planning your strategy 3 Create logical drives on the disk array This required step makes disk array capacity available to your operating system Each logical drive appears to the host as a physical disk device Make sure you observe any limitations imposed by your system regarding disk size or number For more information see Creating a Logical Drive Checking Disk Array Status One of your most important management tasks is monitoring the status operation and configuration of your disk array Because it is a vital piece of your system it is important to know how well the disk array is operating and if any problems or failures have occurred AutoRAID Manager continually monitors the operation of the disk array and updates status information regularly This provides you with the latest information on the operation of your disk array 224 Using AutoRAID Manager for Windows NT Changing Disk Array Configuration Settings To check disk array status Right click the disk array from the Detail pane 2 Click Properties 3 Click the
215. ocated disk array capacity LVM configuration Performance Unallocated capacity is used as additional RAID 0 1 space The amount of RAID 0 1 space required to maintain optimal performance is determined by the write working set parameter Capacity To create the maximum amount of storage capacity allocate all available capacity to LUNs Performance To improve performance configure LVM as follows e Stripe logical volumes across LUNs on different disk arrays Including LUNs on different disk arrays will improve throughput by spreading I Os across arrays e Divide LUN access between both disk array controllers Mapping every LUN to the same controller does not utilize the two data paths available on the disk array The default path is through the primary disk array controller but this can be changed to the secondary controller when adding the physical volume to an LVM volume group The process for changing the path differs for each version of HP UX HP UX 10 1 From the Disk Devices list select the disk hardware path through the secondary controller before adding the disk to a volume group HP UX 10 2 When adding the disk to a volume group in the Creating A Volume Group dialog change the hardware path to the secondary controller Capacity These LVM configurations have no impact on disk array capacity configuration Chapter 2 Using the HP UX System Administration Manager Most disk array management can be done
216. opsis REVISION Output ARDIAG REVISION Test Level FRU Slot Vendor ID Product ID Rev 0 CNTRL Y HP Arrays R Us YPR1 1 ECH A1 HP 1 050 GB 3rd 0256 2 B1 HP 2 13 GB 2nd 0256 3 ECH A2 HP 1 050 GB 3rd 0256 4 2 2 13 GB 2nd 0256 5 ECH A3 HP 1 050 GB 3rd 0256 6 B3 HP 2 13 GB 2nd 0256 7 ECH A4 HP 1 050 GB 3rd 0256 8 ECH B4 HP 2 13 GB 2nd 0256 ARDIAG gt 188 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions ROMT The ROMT command allows the user to perform a read only media test on the specified target Synopsis ROMT Output ARDIAG romt Dk ck ck ck Ck ko ko ok WARNING Dk ck ck ck Ck ck Ck ck Ck Ck ck Ck ck Ck ck Ck ck ck ck ck Ck ck ck Ck ck ck ck ck ck ck ck ck ck ck ck ck Ck ck ck ck ck ck ck ck oco ck kk ck ko ck kk ck ko ko ko Sk Sk A ko ko ko This test cannot continue without a loss of redundancy on the AutoRAID Array being tested Redundancy will be restored when the test completes However failure of another disk during the test will result in data unavailability until the test completes KC Ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck ck Ck Ck Ck Ck Ck Ck Ck Ck C Ck C C C C CC CC CC CC C C CK C C C C C C C Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Ck Sk Do you want to continue y n y
217. ou can also view arraydsp man page for a complete explanation of all the performance command options To analyze disk array performance type arraydsp r stime etime 4 m stime etime int array id Command Examples The following command displays performance recommendations for disk array serial number 00786B5C0000 Performance is analyzed for the time period starting at 8 00 AM and ending at 5 00 PM 1700 on March 15 The format of the st ime and et ime arguments is mnmddhhmm yy arraydsp r 03150800 03151700 00786B5C0000 The following command displays the performance metrics for disk array serial number 00786B5C0000 Metrics are displayed for the time period starting at 11 00 AM and ending at 6 00 PM on April 6 A display interval of 30 minutes is specified arraydsp m 04061100 04061800 2 00786B5C0000 140 Using the ARM Command Line Utilities for Analyzing Disk Array Performance Selecting a Time Period for Analysis When analyzing performance you must define the time period over which the analysis will be performed A starting time st ime and ending time et ime establish the analysis period For the best results select a time period when performance may be a concern This will produce the most meaningful analysis and recommendations For example if the heaviest load on the disk array occurs between the hours of 8 00 AM and 5 00 PM restrict the analysis to this time period If you include pe
218. ou must decide how you want to use it For more information see the preceding section Adding a Disk To manually include a disk type arraycfg D slot lt array id gt slot is the cabinet shelf containing the disk drive A1 through A6 or B1 through B6 250 Using the ARM Command Line Utilities for Windows NT Rebuilding the Disk Array Rebuilding the Disk Array To maintain data redundancy in the event of a disk failure it is important to rebuild the disk array as soon as possible Auto Rebuild does this automatically and it is enabled by default If you want more control over the rebuild process you can disable Auto Rebuild This will allow you to manually start a rebuild at the time you choose A Rebuild impacts disk array performance while it is in progress so before starting a rebuild make sure the appropriate Rebuild Priority is set See Setting Rebuild Priority in this chapter For convenience and maximum protection in the event of a disk failure Auto Rebuild is enabled by default on a new disk array For information on disabling Auto Rebuild see Changing Disk Array Configuration Settings in this chapter Rebuilding the Disk Array Manually If you have chosen to disable Auto Rebuild you will have to start the rebuild manually The rebuild will begin immediately and continue to completion If no rebuild is necessary the command will be ignored To start a rebuild manually type arrayrbld r arr
219. pacity in the HP SureStore E Disk Array 12H User s and Service Manual for help in planning your strategy Also see Tips for Configuring the Disk Array On HP UX in Chapter 1 of this guide If necessary change disk array configuration settings to implement your capacity management strategy If the planning in step 1 requires you to change any of the default configuration settings do so now These include Active Hot Spare Auto Rebuild and Auto Include For more information see Changing Disk Array Configuration Settings in this chapter Check the available unallocated capacity on the disk array See Checking Disk Array Status in this chapter Create each LUN on the disk array This required step makes disk array capacity available to your operating system and it must be repeated for each LUN you are creating For more information see Creating a LUN in this chapter 23 XN dH Using the HP UX System Administration Manager Checking Disk Array Status Checking Disk Array Status One of the most important management tasks is monitoring the operation and status of the disk array Because it is a vital piece of your system it is important to know how well the disk array is operating and if any problems or failures have occurred To view disk array component status 1 Select the desired disk array controller from the list displayed on the screen 2 Select Actions on the menu bar 3 Select View More
220. psis TESTDISK lt Indx gt Output ARDIAG TESTDISK 2 201 Using the ARDIAG Offline Diagnostic MPE Interface command descriptions TESTLEVEL The TESTLEVEL environment variable allows the user to reference a physical mechanism within an SureStore E AutoRAID Disk Array for test Synopsis TESTLEVEL Indx Output ARDIAG TESTLEVEL 2 202 Using the ARDIAG Offline Diagnostic ODE interface ODE interface The following examples illustrate ARDIAG functioning within ODE The user runs ODE at the ISL prompt ISL ODE CK CK CK Ck Ck Kk kk CK Sk kk Sk Kk Ck Kk kk CK Sk Ck Sk Ck Ck Kk Ck CK Sk CK Sk Kk Sk Sk ke kk x Gk kx Kk kk ko KKK KK KKK kk kkk kk ox OFFLINE DIAGNOSTIC ENVIRONMENT eek kk kk KKK KKK aon x C copyright Hewlett Packard Co 1994 5 All Rights Reserved NOR E KKK KKK KKK KKK KKK KKK EASIER OK TC Version XX XX XX SysLib Version NOS det kk kk KKK KKK KKK KKK KKK KKK KKK KKK KK KK KKK KKK KKK KKK KK KKK KKK KKK KKK KKK KKK KKK KEK KKK KK KKK KKK KKK KKK KK Type HELP for command information ODE gt help BASIC COMMANDS HELP Prints detailed information when help lt command gt or help lt variable gt is typed LS Lists modules available on the boot media MODULE NAME Load and initiali
221. r disk array events are listed in the online Help for AutoRAID Manager In addition an ASCII text file of all the events message strings is included in the Program Files AutoRAID Doc folder default location This file may be useful in creating filters and templates for network management applications 209 LN SMOpUIA Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT System Requirements System Requirements AutoRAID Manager requires the following system environment e Microsoft Windows NT 4 0 Advanced Server or NT Server Enterprise Edition with Service Pack 3 host server or later e Microsoft Windows 4 0 Workstation or Windows 95 client e TCP IP services installed e SNMP Services installed required for installation of the AutoRAID SNMP agent Checking the System Hardware Configuration Before installing and using the AutoRAID Manager software the system hardware should be checked to make sure it meets the necessary requirements for proper operation on Windows NT Check the following system components and make sure they are configured as indicated SCSI Hardware CAUTION The disk array uses a differential SCSI bus Use only the HP A5252A SCSI HBA and C2905A differential terminator when connecting the disk array Using single ended SCSI hardware may damage the disk array Check the following SCSI bus hardware e The A5252A SCSI HBA is installed properly in the host e S
222. rameter Default Setting Single Controller ENABLED Indicates whether the disk array should assert a warning Warning Lock Write Cache On Disable NVRAM on WCE False One Disable NVRAM on UPS absent TRUE FALSE Disable NVRAM with TRUE FALSE Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Description when only one controller is present When enabled the disk array will assert a Single Controller Warning when there are not two controllers present in the array disk array When disabled no Single Controller Warning will be indicated Controls the modification of the Write Cache setting When LCWO is set to true then the host cannot change the Write Cache setting When LWCO is false the host ill be able to change Write Cache setting Enables the disk array response to the Write Cache setting If false NVRAM use does not depend on the Write Cache setting and will be enabled if no other condition inhibits it If true then NVRAM is disabled when Write Cache is disabled Couples the use of NVRAM to the presence of an operational dual controller If true and no operational dual controller is present then NVRAM use is disabled Otherwise NVRAM use does not depend on the presence of the second controller and will be enabled if no other condition inhibits it Couples use of NVRAM to the presence of an operational UPS or BBU At the time o
223. re includes several components The AutoRAID Manager Server and the AutoRAID Manager Client are required components that work together to manage the HP SureStore E Disk Array 12H The exact combination of components you install will be determined by your system and network configuration 208 AutoRAID Manager Clients these required components provide the user interface for managing the HP SureStore E Disk Array 12H Two clients are provided the AutoRAID Manager for Windows client and the ARM command line utilities The Windows client can be used to manage disk arrays remotely The command line utilities can only be used to manage disk arrays from the host server OpenView Client for AutoRAID this optional component runs under HP OpenView and automatically discovers and displays HP AutoRAID devices on the HP OpenView map The HP OpenView client communicates with the SNMP agent The OpenView Client is installed if OpenView Network Node Manager is installed on the client AutoRAID Manager Server this required component is the core of the AutoRAID Manager The AutoRAID Manager Server ARMServer is installed as a service on the host server to which the disk array is connected The HP AutoRAID Manager service must be running to manage disk arrays using AutoRAID Manager Installation of the AutoRAID Manager Server must be done from the host server console by a user with Administrator access Remote Management SNMP Agent This optional component is
224. reating Logical Drive ae iae etri ep IEEE D tr rE EEEE 247 Renumbetning Logical Dr Ve degit ote yates Suchen b eae P I et pee eei 248 Deleting logical Drive teu ette ne eter iei e Ie 248 DISKS Eu E S E MEE EIE E 249 Including Disk Manually rette ceo Dn ree pO e ertet Rig 250 REBUILDING THE DISK ARRAY erre eere ie ee et i rer rei eres 251 Rebuilding the Disk Array Manually 251 Setting Rebuild Priority eee ee pet ete eee 251 Checking the Progress of 252 Canceling Ru 252 ANALYZING DISK ARRAY PERFORMANCE 253 Command 2 2 beri eie RERUM EPOD Biel 253 PERFORMING DISK ARRAY MAINTENANCE 5 8 254 Shutting Down the Disks Array identities rhe eb eer tpe te ehe poene 254 Restarting the Disk Array eite tern e E 254 Resetting the Disk Array iine reni Er ehe Vete Eo ea tue eei e eret ER ee n 255 Downing Excluding DISK niet petere he eat tree ree eerte eere is 255 Testina DISK esi etu ep eR One Ote de ente et EU Rips 256 Displaying Test Results ict ee tete De he Ree ete deci e decre eo subs se ete eed 256 Canceling a Disk Test zug PSOE ee ibn idee 257 Printing 5
225. receding section Adding a Disk To manually include a disk 1 Select the desired disk array controller from the list on the screen 2 Select Actions on the menu bar 3 Select Disk Array Maintenance 4 Select Include Disk Disks not currently included will be highlighted on the display dh Select one of the highlighted disks to include 6 Click OK to effect the change 29 XN dH Using the HP UX System Administration Manager Adding a Disk Downing Excluding a Disk Downing or excluding a disk is typically done in preparation for testing the disk After the disk is downed testing can be done without impacting disk array operation If testing reveals that the disk is good the disk can be included back in the array configuration Downing a disk has the same effect as if the disk failed or was physically removed from the cabinet If Auto Rebuild is enabled the disk array will immediately begin a Rebuild when a disk is downed The down operation can be either destructive or nondestructive The type of down performed determines whether the disk array will assume there is any valid data on the disk if it is returned to the array configuration If a destructive down is performed the disk array will assume no data on the disk is valid If the down is nondestructive the disk array will assume any data on the disk that was not updated in the disk s absence is valid To protect data availability th
226. required to delete a logical drive To delete a logical drive type arraycfg L LUN d array id LUN is the logical drive to be deleted 248 Using the ARM Command Line Utilities for Windows NT Adding a Disk Adding a Disk At some time you may want to add another disk to your array Features such as hot pluggable disks and Auto Include simplify the process of adding a disk to the array even while it is operating A disk can be added to the array without disrupting current I O operations After you have added a new disk you have three options on how to use it e Increase capacity use the disk to increase the capacity available to the operating system by creating a new logical drive e Improve performance use the disk to improve the disk array performance by simply leaving it as unallocated capacity Enable Active Hot Spare use the additional capacity to enable Active Hot Spare if the disk array does not currently have the capacity to support this feature This also improves performance as the spare space is used as RAID 0 1 space until it is needed To add a disk to the array 1 Make sure the new disk has been physically inserted into the array 2 If Auto Include is on the disk is automatically added to the array and you can skip to the next step If Auto Include is off manually include the disk as described in the next section Including a Disk Manually NOTE some situations the array will not inc
227. res 16 bit wide data transfer Synchronous data transfer Tagged Command Queuing This device responds to a RESET condition with a SOFT RESET alternative Vendor Identification HP Product Identification C3586A Product Revision Level ZPRF Requested Lun is Supported Product Serial Number is ffffffffffffffffffffffffffffffff Controller Pair Serial Number is TITfTtft amp tftfttffftfffftffffttftttftfttftfifftffftfftf Backplane Serial Number is 007870ca0000 Requested Lun is Supported Manufacturing Product Code is ManProdCod Firmware Revision is 5D03159518 Number of Supported Logical Units 32 Lun Capacity Logical Blocks 0 2097152 1 4194304 2 6291456 87 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions READLOG The READLOG command allows the user to read the internal logs of the specified target Synopsis Output READLOG The following are examples of the READLOG command ARDIAG READLOG Example 1 Controller 88 ARDIAG readlog Contents of array controller log at TestLevel Indx 0 Dk kk ck Ck ck Ck ck Ck Ck ck 0k ck Ck ck Ck ck ck ck ck Ck ck ck ck ck koc ck ck ck ck ck ok kk ck ko ck ko ck kv kx kv Mk ko ko ko ko ko Vendor ID HP Product ID C3586A Usage Log ECC Error Count 0 0x0 Select number to indicate display option for Event Log Information
228. retrieving status from the disk array at regular intervals arraymond can alert the system operator to any array problems The arraymond process is included with the ARM software and is installed with the other software components Like ARMServer the arraymond process is launched automatically when the system is booted Routing arraymond Error Messages By default the status messages generated by arraymond are sent to the system console and emailed to root The destination for the error messages can be rerouted by editing the contents of the destination configuration file etc hpC2400 arraymon dest This file identifies the destinations for all error messages The first line in the file identifies a computer screen dev console or none dev nu11 if no screen messages are desired The rest of the file identifies electronic mail addresses The following example of an arraymon dest file directs the arraymond process to send its error messages to the system console and to also mail the messages to three people on other systems dev console root bobGhpdml69 alex hpdmm99 35 XN dH Using the ARM Command Line Utilities for HP UX Selecting a Disk Array to Manage Selecting a Disk Array to Manage When using the ARM utilities the array id field is used to identify the disk array The array id field can contain the disk array serial number the raw device file name of any LUN on the array or an alias text stri
229. riods of less activity the analysis may yield different results and consequently different recommendations This occurs because activity is averaged over the entire analysis period and periods of less activity will offset the effects of busier periods Typically you should select a period of time that represents normal system operation Avoid any unusual events such as a Rebuild or changes made to array capacity If you select a time period that includes an event that may distort the analysis the utility will alert you and will not provide any recommendations You can also control the display interval using the int option This allows you to control how much detail you get when displaying the metrics The display interval is the number of 15 minute increments Checking the Working Set Metric A key factor in monitoring and maintaining optimal performance of the disk array is the Working Set metric To ensure that disk array performance is maintained you should access the performance metrics regularly and check the Working Set value The Working Set performance metric is derived from the Write Working Set parameter It indicates the ratio of the Write Working Set size to the amount of RAID 0 1 space available For a detailed explanation of the Write Working Set and its impact on performance refer to the HP SureStore E Disk Array 12H User s and Service Manual To maintain performance the amount of RAID 0 1 space should equal or exceed the Wr
230. rting the Disk Array onere reete Rare kr Me ERE tre Erie Eee 235 Resetting the Disk Array RE e 235 10 Formatting the ee toe pe ip pee 235 Formatting a Logical 236 Testing a Disk C Pts 236 Down ga See t dm I TH RO OE Ads 237 Switching Array Controllers tee tete suds e ee de 237 Displaying Hardware 238 DOWNLOADING CONTROLLER FIRMWARE 239 CHAPTER 10 USING THE ARM COMMAND LINE UTILITIES FOR WINDOWS NT 241 ONEINEHEEP e NER 241 COMMAND SYNTAX CONVENTIONS isi secs reti tret rhet CHI P RE HD sues REX DRE ENS 242 SELECTING A DISK ARRAY TO MANAGE 55 RU 242 CONFIGURING A NEW DISK ARRAY 4 44 1 00 ener 243 CHECKING DISK ARRAY STATUS anonimo Oe eere ei epa eU ERES 244 DISPLAYING DISK ARRAY SERIAL NUMBERS eese nene en en enne nenne 245 CHANGING DISK ARRAY CONFIGURATION 5 5 2 246 MANAGING LOGICAL DRIVES Errea ESSEEN EE ESE SE ero oai 247 Checking Logical Drive Configuration ene enne ener 247 C
231. running a Windows 95 client A Windows 95 client must be rebooted following installation of the AutoRAID Manager software before remote severs can be accessed Reboot the client and remote servers should be accessible from AutoRAID Manager e Can tlocate Help for ARM NT command line utilities The Help for the command line utilities is accessed by typing armhelp from the NT command prompt Help can also be accessed from the NT desktop Start menu by selecting ProgramslAutoRAID ManagerlAutoRAID Manager Command Line Help e Can tsee the disk array in the NT Disk Administrator At least one logical drive must be created on the disk array before it will be visible in the NT Disk Administrator e Disk array events are not being communicated to network management application This is typically caused by the AutoRAID SNMP agent not being installed If the NT SNMP service is 220 Managing the HP SureStore E Disk Array 12H on Windows NT Solving Common Installation Problems not installed on the server when running Setup the AutoRAID SNMP agent is not installed To load the AutoRAID SNMP agent install the NT SNMP service and reinstall the AutoRAID Manager software Security Violation error when attempting to access the disk array using AutoRAID Manager from remote client This is typically a caused by the AutoRAID Manager Administrator folder default Program Files AutoRAID ARMadmin not being shared on the server for access For more informat
232. s consider the following factors e size limitations imposed by the operating system for example LVM e Your backup strategy If you do unattended backup to a device such as tape you may want to avoid creating a LUN that is larger than the capacity of the media This allows you to backup an entire LUN without changing media e Configuring the LUN into LVM for maximum performance as described in Table 2 NOTE Before creating a LUN check your operating system documentation for any additional information or steps that may be required to create a LUN To create a LUN 1 Select the desired disk array controller from the list on the screen Select Actions on the menu bar Select Disk Array Maintenance Select Bind LUN Set the LUN size not greater than unallocated capacity Select OK DX dO Note the new LUN definition in the list of disks and arrays 26 Using the HP UX System Administration Manager Managing LUNs Logical Drives Deleting a LUN CAUTION AII data on a LUN is lost when it is deleted Make sure you backup any important data on the LUN before deleting it When a LUN is deleted its capacity is returned to the pool of unallocated capacity Deleting a LUN is a good way of freeing up capacity for the Active Hot Spare or for RAID 0 1 space to improve disk array performance NOTE Before deleting a LUN check your operating system documentation for any additional
233. s menu The Recover command is only supported in controller firmware versions later than HP40 and any patch delivered after IPR9808 release 150 Using the ARM Command Line Utilities for MPE Downloading Firmware Downloading Firmware The download command copies new firmware code to the controller s or individual disk mechanisms in the disk array Firmware also can be copied from a primary array controller to a secondary controller The ARMServer process continues to run while the download is in progress However this process may not be able to access the array during the download because the download operation shuts down the disk array This may cause warning messages such as the following to be displayed access error Unable to get status from disk array on raw disk file name at machine name These warning messages can be ignored CAUTION The firmware code is copied from the primary controller to the secondary controller so before performing this task make sure the controller that has the desirable code is designated the primary controller See Switching Primary Controllers for instructions on changing the state of the controller To download firmware to a disk type download D slot codefilename array id slot identifies the cabinet shelf containing the disk drive codefilename identifies the file containing the firmware code To download firmware to the disk array controllers type downlo
234. s also improves performance as the spare space is used as RAID 0 1 space until it is needed To add a disk to the array 1 Make sure the new disk has been physically inserted into the array 2 If Auto Include is on the disk is automatically added to the array and you can skip to the next step If Auto Include is off manually include the disk as described in the next section Including a Disk NOTE In some situations the array will not include a disk automatically even if Auto Include is enabled This will occur if the new disk s status is something other than Normal See Auto Include in the HP SureStore E Disk Array 12H User s and Service Manual for more information about when this might occur 3 Depending on how you intend to use the new disk perform the appropriate next step e To use the disk to increase capacity create a LUN using all or a portion of the disk capacity For more information see Creating a LUN in this chapter e To use the disk to increase performance leave the disk capacity unallocated 44 Using the ARM Command Line Utilities for HP UX Adding a Disk e To use the disk capacity for an Active Hot Spare enable the Active Hot Spare feature if not currently enabled For more information see Changing Disk Array Configuration Settings in this chapter Including a Disk A disk must be included in the disk array configuration before it can be used by the disk array There are two ways to includ
235. s performance recommendations for disk array serial number 00786B5C0000 Performance is analyzed for the time period starting at 8 00 AM and ending at 5 00 PM 1700 on March 15 The format of the stime and et ime arguments is mmddhhmm yy arraydsp r 03150800 03151700 00786B5C0000 The following command displays the performance metrics for disk array serial number 00786B5C0000 Metrics are displayed for the time period starting at 11 00 AM and ending at 6 00 PM on April 6 A display interval of 30 minutes is specified arraydsp m 04061100 04061800 2 00786B5C0000 48 Using the ARM Command Line Utilities for HP UX Analyzing Disk Array Performance Selecting a Time Period for Analysis When analyzing performance you must define the time period over which the analysis will be performed A starting time st ime and ending time et ime establish the analysis period For the best results select a time period when performance may be a concern This will produce the most meaningful analysis and recommendations For example if the heaviest load on the disk array occurs between the hours of 8 00 AM and 5 00 PM restrict the analysis to this time period If you include periods of less activity the analysis may yield different results and consequently different recommendations This occurs because activity is averaged over the entire analysis period and periods of less activity will offset the effects of busier periods Typically you should
236. scriptions RANGE The RANGE environment variable allows the user to set the desired range of blocks for testing Synopsis RANGE start end start end Output ARDIAG gt RANGE 300 500 1000 2000 Range 300 500 1000 2000 RDIAG gt RANGE ange 300 500 1000 2000 RDIAG gt RANGE 10 20 30 50 ange 10 20 30 50 Arn Dp ARDIAG gt 107 XN dH Using ARDIAG Offline Diagnostic on HP UX Interface command descriptions SHOWENV The SHOWENV command allows the user to view the current values of the environment variables TESTDISK TESTLEVEL and RANGE Synopsis SHOWENV Output ARDIAG SHOWENV TESTDISK 0 TESTLEVEL 3 RANGE 200 600 108 Using the ARDIAG Offline Diagnostic HP UX Interface command descriptions TESTDISK The TESTDISK environment variable allows the user to reference a particular disk array for test Synopsis TESTDISK lt Indx gt Output ARDIAG TESTDISK 2 109 XN dH Using the ARDIAG Offline Diagnostic on HP UX Interface command descriptions TESTLEVEL The TESTLEVEL environment variable allows the user to reference a physical mechanism within a SureStore E AutoRAID Disk Array for test Synopsis TESTLEVEL Indx Output ARDIAG TESTLEVEL 2 110 Using the ARDIAG Offline Diagnostic HP UX ODE interface ODE interface The following examples illustrate ARDIAG funct
237. sensaeeecececeenesesesececeeseaseaecececeesescaecececeesentaaeeeeees 37 CHECKING DISK ARRAY STATUS cccccesssssscecececeessnaececcecececseseaececececseseaeeeeececsesaaeceeeceesensaaeeeeececeeneaaeeeeecs 39 DISPLAYING DISK ARRAY SERIAL NUMBERG ssssssssssececeessssececececceesesececececsenessececeecceeseaeaesececeesensaaeeeeees 40 CHANGING DISK ARRAY CONFIGURATION 5 68 2 2 012 2240000 00001000000000000000000000 40 MANAGING LUNS LOGICAL DRIVES 102 0 2225202000 610600000000000000005050 42 Checking LUN Configuration nest enin nest 42 Creating LUN pep rap OPER REPAS US 42 Renumbering a LUN zu itae pert tein ien Re 43 Deleting LUN 5 1 tette ed ect tee e Bs teet ele i 43 INDDING A ey ee toe Ie 44 Including DISK cene the pen RE e Dritte 45 REBUIEDING THE DISK ARRAY ds ertet metn re t er i Here e t e ere 46 Rebuilding the Disk Array Manually 46 Setting Rebuild Priority ce ee Een Pe RO 46 Checking the Progress of a Rebuild 47 Canceling a Rebuild nce REC eta 47 ANALYZING DISK ARRAY 1 4 0000000 48 Command Examples de eos Be
238. stalled Check the README file for important updates about the AutoRAID Manager software Do not install the AutoRAID Manager software into the root directory of the local drive Installing the software into the root directory will cause the AutoRAID Manager to be inoperative Make sure you share the AutoRAID Manager Administrator folder Program Files AutoRAID ARMadmin the default location on the server if you want to manage the disk arrays from a remote client If this folder is not shared you will get a security access violation when you try to manage the disk array using AutoRAID Manager Windows NT allows the static assignment of drive letters on volumes partitions and CD ROM drives This means that a drive letter can be permanently assigned to a specific hard disk partition or volume and CD ROM drive When a new hard disk is added to an existing computer system it will not affect statically assigned drive letters Installing the failover driver can cause remapping of your disk drives if they have not been assigned as static To prevent re mapping use the Disk Administrator StartlIProgramslAdministrative Tools and reassign the current disk drive letters to each drive Reassigning the letters will make the assignments static and prevent remapping To install the AutoRAID Manager software 1 2 212 Logon to the network as a user with administrator privileges Load the HP AutoRAID on Windows NT CD into the CD ROM
239. sted RANGE Select range of blocks to be tested by ROMT and WRTMT SHOWENV Display the current settings for TESTDISK and RANGE READLOG Read the internal logs of the desired disk array controller or internal disk CLRLOG Clear the internal logs of the desired disk array controller or internal disk REQSENSE View SCSI Sense data of the desired disk array controller or internal disk INQUIRY View SCSI Inquiry data of the desired disk array controller or internal disk DESCRIBE View Configuration data of the desired disk array REVISIO View revision information of the desired disk array WRTMT Perform a DESTRUCTIVE write read test on the desired disk ROMT Perform a read only test on the desired disk CREATELUN Adds a LUN to the selected disk array DELETELUN Removes a LUN from the selected disk array SHUTDOW Puts the selected disk array into the shutdown state SETOPTIONS View and modify SCSI mode parameters on the selected array controller RECOVER Reconstruct data mapping and array configuration Continue y n n 206 Chapter 8 Managing the HP SureStore E Disk Array 12H on Windows NT This chapter introduces AutoRAID Manager the complete software package for managing your HP SureStore E Disk Array 12H on Microsoft Windows NT AutoRAID Manager is a management tool that allows you to monitor and manage your HP SureStore E Disk Array 12H from a network Windows computer either locally or remo
240. t was started manually A Rebuild that was started by Auto Rebuild cannot be canceled If a Rebuild is canceled it must be started over again and any progress made during the first Rebuild will be lost When canceling a Rebuild the Rebuild may not stop immediately This occurs if the disk array is busy servicing higher priority I O requests from the host The Rebuild will be canceled when the disk array has serviced all higher priority commands To cancel a manual Rebuild type arrayrbld c array id 139 Using the ARM Command Line Utilities for Analyzing Disk Array Performance Analyzing Disk Array Performance The disk array monitors and stores a number of metrics that reflect how the disk array is performing The disk array management software periodically retrieves these performance metrics and stores them for your viewing The ARM software can also analyze the performance metrics to identify any potential performance problems Based on this analysis recommendations are made on how to improve disk array performance Checking the metrics regularly is a quick and easy way for you to monitor the performance of the disk array and identify any problems that may be developing You may choose to display only the recommendations or you may want to view the metrics for further analysis For a complete description of each performance metric refer to the HP SureStore E Disk Array 12H User s and Service Manual Y
241. tails on subscribing to the IPR program 16 Managing the HP SureStore E Disk Array 12H on Tips for Configuring the Disk Array On HP UX Tips for Configuring the Disk Array On HP UX Before installing a new disk array you should determine what is more important for your operating environment performance or capacity This decision will influence how you configure the disk array hardware and LVM There are several configuration options that impact the performance of the disk array Table 2 identifies the various configuration options their impact on disk array performance and any considerations regarding disk array capacity NOTE OvVverall system performance is a complex issue influenced by many factors The configuration options described here will increase the potential performance of the disk array However the actual performance of the disk array will be determined largely by host demand AutoRAID technology is particularly suited to l O intensive application environments such as OLTP and NFS It is in these environments that the performance benefits offered by AutoRAID will be fully realized Table 2 Configuration Options Configuration Impact Performance and Capacity Option Number of LUNs Performance Increase the number of LUNs per disk array to improve per disk array performance More LUNs increases the size of the I O command queue allocated by HP UX which increases throughput The recommended
242. tely Using AutoRAID Manager you can easily and quickly perform all the tasks involved in the daily management and operation of your disk arrays AutoRAID Manager makes it easy for you to disk capacity to the disk array e Monitor array performance e Change disk array operating parameters e Identify and diagnose problems with the disk array Most disk array management can be done using the AutoRAID Manager for Windows The AutoRAID Manager software also includes the HP AutoRAID Manager ARM command line utilities for Windows NT These commands are typically used for performing advanced tasks such as reading logs and downloading firmware Using the disk array control panel An alternative to using the AutoRAID Manager to manage your disk array is the disk array control panel Although you can use the disk array control panel to perform many of the same management tasks the added convenience and functionality provided by AutoRAID Manager makes it a more convenient tool for managing the disk array Find the Latest Information on the World Wide Web For the latest information about operating your HP SureStore E Disk Array 12H on Windows NT visit our web site at http www hp com go support You ll find the latest software firmware and operating tips 207 1 SMOpUIA Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT AutoRAID Manager Components AutoRAID Manager Components AutoRAID Manager softwa
243. tes the indicates the disk array will not initiate SDTR Enabled disk array will initiate WDTR the SCSI bus terminators Disabled indicates the disk array will not provide termination power Enabled indicates the disk array will provide termination power Unit Attention ENABLED Indicates whether the disk array should signal a Unit Attention condition immediately following power on or reset Disabled indicates the disk array will not signal unit attention Enabled indicates the disk array will signal unit attention 265 IN SMOpUIM Windows NT Using the ARM Command Line Utilities for Windows NT Viewing the Disk Array General Configuration Settings Parameter Default Description Setting Disable Remote Reset ENABLED Controls the bus reset behavior of the second controller hen a SCSI reset reset signal or Reset Disk array command is received in the first controller When this bit is disabled the second controller will assert the bus reset signal to indicate that all outstanding requests were cleared in response to the reset When this bit is enabled the second controller will not assert the SCSI reset signal to indicate the commands were cleared Hosts that cannot tolerate target bus resets should enable this setting Secondary Controller DISABLED Controls the behavior of the secondary controller with Offline respect to bus selection When disabled any secondary controller present will respond to host s
244. tes the language used when displaying information the front panel Log Full Warning DISABLED Indicates whether the disk array should assert a warning hen some portion of the controller log is full When enabled the disk array will assert a Log Full Warning hen one of the disk array log pages fills or a log parameter reaches its maximum value When disabled no Log Full Warning will be indicated Language ENGLISH Volume Set DISABLED Indicates whether the disk array should boot when half or Partitioning more of the previously available drives are unavailable i e there is no drive quorum Enabled indicates that volume set partitioning should be allowed Disabled the system will remain in the No Quorum state when the required quorum is not available Format Pattern Fill DISABLED Indicates whether the disk array will fill in incomplete RAID blocks with a format pattern when performing new rites Some operating systems not HP UX MPE or NT expect that SCSI format commands completely reinitialize data to a non random pattern When using the array with one of those operating systems pattern filling should be enabled Disk array Type ID 12 Uniquely identifies the disk array hardware configuration This field can be used to find the number of drives and SCSI channels supported within the disk array LUN Creation Limit 8 Controls the range of LUNs that may be created Maximum LUN Specifies the maximum supported
245. then read data is never satisfied out of write cache Specifies the maximum time between delivery of the response to a write command to the host and initiation of associated writes out of write cache and map journal to disks This time is specified in seconds This maximum time is submitted to the scheduler for implementation Depending on the priority of other events in the scheduler the desired time may not be implemented If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored The value of this field is not the same as disabling NVRAM use A value of Oh indicates that no maximum time will be enforced and that no recovery image will be available 160 Using the ARM Command Line Utilities for MPE Viewing the Disk Array General Configuration Settings Simplified Resiliency Setting The simplified resiliency setting is derived from several other fields Collectively these settings control how data resiliency is managed Table 22 identifies the field settings for each of the resiliency modes There are four modes of data resiliency Normal SingleController Secure and HighPerformance The default setting is Normal for dual controller disk arrays If the disk array has been ordered with a single controller the default for this settings is SingleController For information on changing the data resiliency mode see Setting Data Resiliency in this chapter 161 Using the ARM Com
246. tices which if not observed could result in personal injury About this Book This book describes the tasks and tools involved in managing the HP SureStore E Disk Array 12H on HP UX MPE and Windows NT This material is intended for system administrators and others involved in the installation operation and management of network storage The content of this book is organized as follows The first section describes the tasks involved in managing the HP SureStore E Disk Array 12H on HP UX Chapter 1 provides an overview of the management tools system requirements and software installation Chapter 2 describes how to use the HP UX System Administration Manager SAM to manage the disk array Chapter 3 describes how to manage the disk array using the HP UX ARM command line utilities Chapter 4 explains how to use the ARDIAG Offline Diagnostic to isolate and solve disk array problems The next section describes the tasks involved in managing the HP SureStore E Disk Array 12H on MPE Chapter 5 provides an overview of the management tools system requirements and software installation Chapter 6 describes how to manage the disk array using the MPE ARM command line utilities Chapter 7 explains how to use the ARDIAG Offline Diagnostic to isolate and solve disk array problems The last section describes management of the HP SureStore E Disk Array 12H on Windows NT Chapter 8 provides an overview of the system requirements and so
247. time you choose A Rebuild impacts disk array performance while it is in progress so before starting a Rebuild make sure the appropriate Rebuild Priority is set See Setting Rebuild Priority in this chapter For convenience and maximum protection in the event of a disk failure Auto Rebuild is enabled by default on a new disk array For information on disabling Auto Rebuild see Changing Disk Array Configuration Settings in this chapter Rebuilding the Disk Array Manually If you have chosen to disable Auto Rebuild you will have to start the Rebuild manually The Rebuild will begin immediately and continue to completion If no Rebuild is necessary the command will be ignored To start a Rebuild manually type arrayrbld r array id Setting Rebuild Priority The rebuild priority determines how quickly a Rebuild completes It allows you to balance the servicing of host I Os with the rebuilding of the disk array The same rebuild priority is used for both Auto Rebuilds and manual Rebuilds To set the rebuild priority type arrayrbld P high low array id 138 Using the ARM Command Line Utilities for Rebuilding the Disk Array Checking the Progress of a Rebuild You can easily check the progress of a Rebuild This allows you to determine approximately when a Rebuild will complete To check the progress of a Rebuild type arrayrbld p array id Canceling a Rebuild A Rebuild can be canceled only if i
248. tional Command Descriptions 6 MECH B3 Included ECH A4 Included 8 MECH B4 Included Lun Capacity GB GB GB GB GB MB oun 172 Using the ARDIAG Offline Diagnostic MPE Operational Command Descriptions DOWNLOAD The DOWNLOAD command allows the user to download firmware to the specified target s Synopsis DOWNLOAD Additional information DOWNLOAD to the array controller The DOWNLOAD process must be performed using the primary controller The primary controller is identified using the INQUIRY command If the secondary controller is identified you must change the controller selection using the TESTDISK command The TESTLEVEL command must be set to zero to indicate a controller selection A presently installed secondary controller is automatically updated with the new firmware image after a successful DOWNLOAD to the primary controller Thus there is no interaction needed to update the presently installed secondary controller A newly installed secondary controller can be interactively updated to the firmware image that is presently in the primary controller using the DOWNLOAD command After successful DOWNLOAD the REVISION command can be used to verify firmware revision Reset ARDIAG to update DISPMAP revision information DOWNLOAD to an internal disk The DOWNLOAD process to an internal disk can be performed through either the primary or secondary co
249. to Include simplifies the task of adding Include a new disk to your array Disabling it will require you manually to include each disk you install in the array Rebuild High p Rebuild priority determines how quickly a Priority Rebuild operation will complete To change Active Spare Auto Rebuild or Auto Include settings type arraymgr h a i onloff array id To change Rebuild Priority setting type arraymgr high low array id 41 XN dH Using the ARM Command Line Utilities for HP UX Managing LUNs Logical Drives Managing LUNs Logical Drives An important part of managing the disk array involves defining and maintaining the optimal LUN structure for your system Your system requirements and limitations will influence the LUN structure you choose Managing LUNs is a part of the overall task of managing disk array capacity For more information on managing disk array capacity to meet your system needs refer to the HP SureStore E Disk Array 12H User s and Service Manual Checking LUN Configuration When you are managing LUNs you may find it convenient to check the current LUN configuration and the available capacity To check the current LUN configuration and the available capacity type arraydsp 1 LUN lt array id gt Creating a LUN Only capacity assigned to LUNs is visible to the operating system When selecting the size for your LUNs consider the following f
250. to manage 1 the View menu click Select Hosts 2 Selecta disk array you want to manage from the Known Hosts You can select multiple arrays at once 3 Click the Add gt gt button to move the disk array to the Selected Hosts list Click OK the disk arrays you added are now displayed in the Detail pane There may be a delay as the status of each disk array is retrieved Until the status is retrieved each disk array is assigned a status of UNKNOWN You are now ready to begin managing your disk arrays using the AutoRAID Manager The next chapter provides step by step instructions for performing the tasks involved in managing the disk array NOTE Disk arrays connected to an NT cluster will appear twice in the Discovered Arrays once for each server in the cluster 218 Managing the HP SureStore E Disk Array 12H on Windows NT DiscoveringDisk Arrays Missing Disk Arrays If you know there are disk arrays connected to your network but they are not displayed in the Discovered Arrays check the following e Make sure the host the disk array is connected to is operational and logged on to the network e sure the disk array is turned on and operating properly e Make sure the disk array is properly connected to the host This includes proper termination of the SCSI bus e sure there is at least one logical drive created on the disk array A logical drive must be created on the disk array before the array can
251. to run automatic discovery you can do the following 219 IN Windows NT Managing the HP SureStore E Disk Array 12H on Windows NT Editing the Disk Array List To add a remote host and its disk arrays 1 the HP AutoRAID Manager s View menu click Select Hosts 2 Click Discover 3 Click Specified Host and enter the IP address or the domain name of the host server and click OK 4 The AutoRAID Manager will locate the remote host and add each HP SureStore E Disk Array 12H connected to it to the Discovered Arrays 5 Select the disk array s and click Add gt gt to add them to the Selected Hosts list if the view is not configured for All Known Hosts 6 Click OK and the disk arrays will be included in the Detail pane Editing the Disk Array List Over time the disk arrays you are responsible for managing may change You can use the Selected Hosts list dialog box to add or remove disk arrays from the Detail pane Only disk arrays in the Selected Hosts list window can be managed using AutoRAID Manager Solving Common Installation Problems AutoRAID SNMP agent was not installed during Setup If the NT SNMP service is not installed on the server when running Setup the AutoRAID SNMP agent is not installed To load the AutoRAID SNMP agent install the NT SNMP service and Service Pack 3 or later then reinstall the AutoRAID Manager software access remote servers from AutoRAID Manager
252. total unallocated capacity available for creating LUNS will be displayed Make sure there is adequate capacity to create the LUN structure you need 5 Create each LUN on the disk array by typing arraycfg L LUN a capacity array id Example arraycfg L 0 1000 00786 5 0000 This command creates LUN 0 with a capacity of 1000 Mbytes on the array identified by serial number 00786B5C0000 This step makes disk array capacity available to your operating system and it must be repeated for each LUN to be created Make sure you observe any operating system limitations on LUN size or number For more information see Creating a LUN in this chapter 38 Using the ARM Command Line Utilities for HP UX Checking Disk Array Status Checking Disk Array Status One of the most important management tasks is monitoring the status operation and configuration of the disk array It is important to know how well the disk array is operating and if any problems or failures have occurred Using the ARM arraydsp command you can easily check all aspects of disk array operation and configuration The arraydsp command options summarized in Table 4 allow you to display information about each disk array hardware component as well as information about the logical configuration of the disk array Table 4 arraydsp Options for Displaying Disk Array Status Status Information Displayed 1 LUN Information for the specified LU
253. u are now prepared to perform all rebuilds manually To rebuild the disk array manually 1 Right click the disk array from Detail menu Click Properties Click the Configuration tab Under Rebuild click Manual Select the Priority you want the disk array to use as it rebuilds the disk array Qu pss ss Click Start in the Rebuild controls The rebuild will begin and the progress of the rebuild will be displayed NOTE If there is no Active Hot Spare or not enough unallocated capacity available you must increase the amount of unallocated capacity so there is enough to perform the rebuild You can increase the amount of unallocated capacity by adding another disk or deleting an existing logical drive Checking the Progress of a Rebuild The progress of a rebuild is displayed in the Configuration window Access this window to monitor the progress of a rebuild To monitor the progress of a rebuild Right click the disk array currently undergoing a rebuild from Detail pane 231 IN SMOpUIM Windows NT Using AutoRAID Manager for Windows NT Analyzing Disk Array Performance 2 Click Properties 3 Click the Configuration tab 4 Under Rebuild check the rebuild progress bar for the current status of the rebuild Canceling a Rebuild It is possible to cancel a rebuild in progress before it completes However you should be aware that when you restart a stopped rebuild it starts at the beginning not at th
254. u bar Select Disk Array Maintenance Select Modify Array Configuration Select desired rebuild priority high or low d MES CP E Select OK to effect the change 31 XN dH Using the HP UX System Administration Manager Switching Primary Controllers Switching Primary Controllers In dual controller disk array configurations the array automatically switches to the secondary controller if the primary controller fails However you can switch controllers manually if necessary This will cause the secondary controller to assume the role of primary controller To switch primary controllers 1 Ow 32 Select the desired disk array controller from the list Select Actions on the menu bar Select Disk Array Maintenance Select Modify Array Configuration Select desired primary controller Select OK to effect the change Chapter 3 Using the ARM Command Line Utilities for HP UX The AutoRAID software includes a set of AutoRAID Manager ARM command line utilities These commands provide the ability to manage the disk array from the HP UX command prompt rather than from SAM The ARM commands can be used to perform a number of tasks that cannot be performed using SAM These tasks are intended for advanced users and involve procedures such as diagnostics performance monitoring setting SCSI values and disk array maintenance CAUTION Many of the tasks availa
255. ude system usage log usage disk error log disk controller error log ct x1r system change log change and performance log perf array serial number limits the output to only those entries associated with the specified disk array Displaying Hardware Logs In addition to the logs maintained by ARMServer hardware logs are also stored on the disk array The arraylog command provides access to the controller and disk logs maintained by the disk array These logs contain information useful for diagnosing and troubleshooting the disk array The logs can also be cleared using arraylog The arraylog options for accessing the disk array hardware logs are listed in Table 6 53 XN dH Using the ARM Command Line Utilities for HP UX Performing Disk Array Maintenance Tasks Table 6 arraylog Options for Displaying Log Contents Option Description Display the contents of the disk array controller usage log Display the contents of the disk array controller event log d slot Display the contents of the log for the disk installed in the cabinet slot identified by slot Slot numbers must be of the form An or Bn where A or B correspond to a cabinet column and n corresponds to a shelf position 1 6 c d slot Clear the specified logs If c is specified clear the disk array controller usage and event logs Both logs will be cleared when using this option It is not possible to clear just one of the logs I
256. ue Page Code 3e Vendor Unique or Non SCSI 2 Standard Vendor Unique Page Code 2 Write Errors Errors Corrected With Possible Delays otal Errors Corrected By Applying Retries otal Errors Corrected Total Times Correction Algorithm Processed Total Bytes Processed Total Uncorrected Errors 0 0000000020046200 Or Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Page Code 3 Read Errors Errors Corrected Without Substantial Delay Errors Corrected With Possible Delays otal Errors Corrected By Applying Retries otal Errors Corrected Total Times Correction Algorithm Processed Total Bytes Processed Total Uncorrected Errors 0x000000000de82c00 ll OoOoB 5oo0W1 Page Code 5 Verify Errors Errors Corrected Without Substantial Delay Errors Corrected With Possible Delays Total Errors Corrected By Applying Retries Total Errors Corrected Total Times Correction Algorithm Processed Total Bytes Processed Total Uncorrected Errors 0 0000000000000000 ll Page Code 6 NonMedium Errors 334 Errors Corrected Without Substantial Delay Page Code 37 Addr Hexadecimal ASCII 0 0x37 0x00 8 0 01 Ox3e x00 Oxle 0x00 0x00 0x80 0x02 ae 16 0x10 0x00 0x02 x00 0x01 0x80 0x02 Oxff Oxff DPI x80 0x02 0x00 0x00 0x00 Ox03
257. ues and disk array maintenance LUNs and Logical Drives To maintain consistency with MPE terminology the term LUN is used to refer to a disk array logical drive The two terms are used interchangeably and refer to the same logical entity on the disk array Setting Up MPE POSIX Shell To set up the MPE POSIX shell to execute the ARM command line utilities refer to the following example 1 SETCATALOG HPPXUDC APPEND SYSTEM 2 3 cd opt hparray bin 125 Using the ARM Command Line Utilities for Information in man pages Information in man pages The procedures in this chapter summarize the use of the ARM utilities Detailed information about the ARM command line utilities and their proper usage is included in the MPE operating system man pages A man page also exists for ARMServer the server portion of the disk array management software To access MPE man pages information type man lt commandname gt Substitute one of the following ARM utility names for commandname ARMServer arraycfg arraydsp arrayfmt arraylog arraymgr arrayrbld arrayrecover download drivetest dteststat logprint Command Syntax Conventions The following symbols are used in the command descriptions and examples Symbol Meaning lt gt Integer value whose units are not defined Exclusive OR Exactly one of the parameters displayed will be used Items enclosed are optional
258. ups deactivated in step 2 eb pax Remount the LVM logical volumes unmounted in step 1 For example if two LVM logical volumes 17011 and 1vo12 are defined on the disk array and both logical volumes are contained in the volume group vg02 the steps to download the firmware 1 Umount the logical volumes umount dev vg02 lvoll umount dev vg02 1vol2 2 Deactivate the volume group vgchange a n dev vg02 60 Using the ARM Command Line Utilities for HP UX Downloading Firmware 3 Load the controller firmware download C firmware file 000000123456 4 Reactivate the volume group vgchange a y dev vg02 5 Mount the logical volumes mountall If the disk array has LUNs that are being used as raw devices for example a database system doing raw I O the array should be disconnected from all systems except the one doing the download An alternative to disconnecting the systems is to use the HP UX shutdown 1m command to halt the other systems until the download is complete 61 XN dH Using the ARM Command Line Utilities for HP UX Recovering Data Maps Recovering Data Maps If the data maps stored in the disk array controller NVRAM are lost the arrayrecover command can be used to reconstruct the maps The data maps are reconstructed using the latest recovery maps stored on two of the disk drives The recovery operation involves performing a parity scan on the contents
259. ure functionality with UPS and is disabled in all resiliency modes 68 Parameter Default Setting Force Unit Access 2 Response Disable Read Hits FALSE Resiliency Threshold 4 Using the ARM Command Line Utilities for HP UX Viewing the Disk Array General Configuration Settings Description Controls the array s response to the FUA bit The FUA bit is a command from the host when Write Cache is enabled It gives the host an opportunity to flush write cache on command If this field is 0 then the FUA bit is ignored If this field is 1 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed If this field is 2 then the FUA bit in a write command disables immediate report on the write and all write cache data for the involved LUN will be flushed along with the map journal before completing the write request If NVRAM use is disabled due to the chosen map resiliency mode then this field is ignored Controls the array s ability to satisfy read commands from rite cache contents during FUA processing If NVRAM use is disabled due to the chosen map resiliency mode then read data is never satisfied out of write cache Specifies the maximum time between delivery of the response to a write command to the host and initiation of associated writes out of write cache and map journal to disks This time is specified in seconds This maximum tim
260. used improperly could destroy data or disrupt access to the disk array For this reason it is recommended that you restrict user access to AutoRAID Manager When sharing the AutoRAID Manager Administration folder make sure you restrict access to the folder The default permissions when sharing a folder give everyone access so you will need to change the permissions to restrict access to the desired users only The AutoRAID Administrator folder must be shared to allow remote clients to use AutoRAID Manager to manage the disk array Disk array management is controlled by the permissions applied to the AutoRAID Administrator folder which is located by default at Program Files AutoRAID ARMadmin If you selected another location for the Administration folder during setup the specified folder will control access to the disk array Standard Windows NT security procedures are used to define the permissions for the Administrator folder Any user account with read write access to the Administrator folder will be able to view and manage the disk array The folder will inherit the share access rights currently defined for the system If the default permissions are not suitable add or remove users from the access list as necessary to impose the desired security on the folder Also make sure the NTFS file permissions are also set properly to control access HP UX Server Access If you will be using AutoRAID Manager to manage a disk array connected to an HP
261. using ARM 143 NT using ARM 255 NT using AutoRAID Manager for Windows 237 downloading firmware HP UX using ARM 59 MPE using ARM 151 NT using ARM 261 NT using WinDownload 240 E error messages NT installation 220 events on NT 209 excluding a disk See downing a disk failover driver Windows NT 208 firmware downloading HP UX using ARM 59 MPE using ARM 151 NT using ARM 261 formatting the disk array HP UX using ARM 54 MPE using ARM 146 NT using ARM 258 NT using AutoRAID Manager for Windows 236 H hardware logs HP UX using ARM 53 using ARM 145 NT using ARM 257 NT using AutoRAID Manager for Windows 239 HBA SCSI 210 help command line for HP UX 33 command line for MPE 126 command line for NT 241 HP OpenView integration with NT 209 HP UX configuration tips 17 System Administration Manager 19 HP UX ARM utilities 33 HP UX servers access for AutoRAID Manager 213 I including a disk manually HP UX using ARM 45 HP UX using SAM 29 MPE using ARM 137 NT using ARM 250 NT using AutoRAID Manager for Windows 230 installing software on HP UX 16 on 118 IPR program 16 Index L log contents printing HP UX using ARM 53 MPE using ARM 145 NT using ARM 257 log location Windows NT 214 logical drives managing NT using ARM 247 NT using AutoRAID Manager for Windows 226 LUNs managing HP UX using ARM 42 HP UX using SAM 25 MPE using ARM 134 M ma
262. vides an alternative to the disk array serial number and raw device file name that can also be used to identify the array Aliasing can be used in a variety of ways to help identify disk arrays in large systems For example by assigning numbers to racks and to the shelf positions within the racks each disk array can be uniquely identified using an appropriate alias If a rack is assigned number 12 the disk array installed on shelf 3 of the rack could be identified using an alias of R12 S03 This technique simplifies locating the disk array should it need service To create a disk array alias type arraymgr D alias name array id alias name can be up to 12 characters in length and can include upper case letters numbers spaces number sign underscore and period Recover If the disk array is not shutdown properly it is possible that the data maps in NVRAM memory will be lost For this reason the disk array allows the data maps to be periodically written to the disk drives If the maps are lost an error code such as Address Table will appear on the display If your disk array is a boot device you may have to recover the maps by using the front panel command called Recover under the Cntrl Changes menu The Recover command is only supported in controller firmware versions later than HP40 and any patch delivered after IPR9808 release 58 Using the ARM Command Line Utilities for HP UX Downloading Firmw
263. w firmware 4 Click Select Firmware File 5 Select the firmware file to be downloaded to the disk array The latest version of firmware is installed in the AutoRAID FW folder during setup If the desired firmware file is in another location locate and select it 6 Click Download The download process will begin The process can take up to 5 minutes to complete When the download is complete the disk array will be brought back on line ready to process I Os from the host 239 LN SMOpUIM Chapter 10 Using the ARM Command Line Utilities for Windows NT Included with the AutoRAID Manager ARM software are the ARM command line utilities for Windows NT These commands provide the capability of managing the disk array from the NT command prompt The command utilities are emulated by the AutoRAID Manager Windows GUI so the GUI should be the primary tool used to manage the disk array NOTE The command line utilities can only be used on the host server to manage local disk arrays The command line utilities cannot be used to manage disk arrays remotely The command line utilities do provide some additional functionality not available from the GUI However these are specialized tasks not typically performed in the day to day management of the disk array e The download command downloads new firmware to the disk array controller and disk drives e The logprint command accesses the log files maintained by AutoRAID Manager e The arr
264. wn state via the specified controller path Destructive write read and verify test to the specified internal mechanism 75 XN dH Using the ARDIAG Offline Diagnostic on HP UX ARDIAG Interface Commands ARDIAG Interface Commands Table 12 lists the ARDIAG test module interface commands and their descriptions Table 12 ARDIAG Interface Commands Command Description DISPMAP Displays a list of AutoRAID controllers p DISPMECH Displays the disk mechanisms within the selected disk array RANGE Environment variable that represents a range of blocks to test SHOWENV Displays environment variables TESTDISK Points to the disk array on which to execute commands TESTLEVEL Points to a physical mechanism within the disk array on which to execute commands 76 Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions Operational Command Descriptions CLRLOG The CLRLOG command allows the user to clear the internal logs of the specified target Synopsis CLRLOG Output ARDIAG CLRLOG Clearing the contents of array controller log at Indx 0 Dk Ck ck ck Ck ck kk ck Ck ck ck Ck ck Ck Ck ck Ck ck ck ck ck Ck ck ck ck ck ck ck ck ck ck ck Ck ck Ck ck ck ck ck ck ck ck ko ck kk ck Sk Sk ko Sk kv kx Sk ko kx ok Successful Completion 77 XN dH Using the ARDIAG Offline Diagnostic HP UX Operational Command Descriptions
265. y Before you can manage a disk array using AutoRAID Manager you must select the array from the Detail pane See Setting Up AutoRAID Manager in the preceding chapter for information on adding disk arrays to the Array List window To select a disk array 1 In the Detail pane click the disk array you want to manage If it is not visible in the Detail pane you may need to select another view 2 Right click the array and use the property pages tabs for the operation s you want to perform 223 LN SMOpUIM Windows NT Using AutoRAID Manager for Windows NT Configuring a New Disk Array Configuring a New Disk Array After installing a new disk array you can perform the initial configuration using AutoRAID Manager This establishes the desired operating environment for the disk array and makes the capacity of the disk array available to the host server The following steps guide you through the typical process of configuring a new disk array NOTE Atleast one logical drive must be created on the disk array before the host can access the array For more information see Logical Drive Configuration in the preceding chapter To configure a new disk array 1 Check the available unallocated capacity on the disk array The total unallocated capacity available for creating logical drives is displayed in the Capacity screen 2 Plan your capacity management strategy and logical drive configuration Determine how you want t
266. y on this particular disk array you could use any of the following commands arraydsp 00786B5C0000 arraydsp autoraid4 The serial number of all arrays in the system can be obtained using the command arraydsp i For more information see Displaying Disk Array Serial Numbers in this chapter 128 Using the ARM Command Line Utilities for MPE Configuring a New Disk Array Configuring a New Disk Array After installing a new disk array you can perform the initial configuration using the ARM utilities This establishes the operating environment for the disk array To configure a new disk array 1 Plan your capacity management strategy and LUN configuration Decide how you want to use the disk array capacity Factors such as data redundancy and performance influence how you manage the capacity See Managing the Disk Array Capacity in the HP SureStore E Disk Array 12H User s and Service Manual for help in planning your strategy Also see Tips for Configuring the Disk Array On MPE of this guide 2 Display the serial number of the disk array by typing arraydsp i The serial number provides a way to identify disk arrays when using the ARM utilities Record the serial number for future reference or set a shell variable to hold this value for example export ID 00786B5C0000 3 Ifthe planning in step 1 requires you to disable any of the configuration settings to implement your capacity management strategy do so now
267. ze the module RUN Run modul after setting environment variables CONTROL Y CONTROL C Abort an ODE command pause a module RESUME Restart a paused module DISPLOG After running a module display the contents of the log XIT Return to the next higher level prompt El VIRONMENT VARIABLES 203 Using the ARDIAG Offline Diagnostic MPE ODE interface SHOWSTATE Display the values of the following environment variables LOOP Run a test this many times ERRPRINT ON OFF Print low level error messages ERRNUM ON OFF Print one line numbered errors ERRPAUSE ON OFF Pause module on error ERRONLY ON OFF Print only error messages INFOPRINT ON OFF Print informational messages ISOPRINT ON OFF Print fault isolation messages ISOPAUSE ON OFF Pause module when isolation message is generated 106512 Set the size of the message log DEFAULT Reset environment variables to default state ODE 15 odules on this boot media are filename type siz created description ARDIAG XXXX XXX XXXX XXXXXX 204 Using the ARDIAG Offline Diagnostic ARDIAG interface to ODE ARDIAG interface to ODE ODE ARDIAG kk ck Ck ck ck Ck ck ck Ck ck Ck Ck ck Ck ck ck Ck KKK KKK KKK KKK ck kk ck kk ck kk Ck kk kk ck kk ck kk ck kk ck kk Sk kc k ko kc KKK KK KKK KKK KKK KKK

HP SureStore E Disk Array 12H System Administrator`s

Contents

Download Pdf Manuals

Related Search

Related Contents