Home

HP SureStore E Disk Array 12H User`s and Service Manual

1. The controller continually monitors various points along the data path to determine how efficiently data is moving These measurements form the metrics used to identify performance problems and make recommendations for their solution 106 Performance Metrics Concepts and Management Optimizing Performance As shown in Table 4 the disk array uses an internal set of metrics to monitor its own performance These metrics reflect internal events and conditions that are continually measured and stored by the disk array A host disk array management utility can access the metrics and use them to evaluate performance and even make recommendations on how to improve performance For more information on accessing the performance metrics refer to your operating system documentation Table 4 Disk T Performance Metrics isk Array Performance Metric Typical Limit Disk Diff 0 1 gt 1 Relocate Blocks 0 008 gt 0 07 Working Set 1 gt 1 1 Cache Utilization 0 1 0 2 Definition the internal SCSI channels If the disks are not installed properly one of the SCSI channels may be andling more than its share of activity reducing erformance Redistributing the disks will solve this roblem NEN HR if the disks are properly distributed across For more information see SCSI Bus Topology RAID 0 1 and 5 If too much data is being transferred between RAID 0 1 and RAID 5 erformance declines Indicates th
2. Troubleshooting Replacing Modules CAUTION To maintain proper cooling within the disk array a failed fan module must remain in the fan slot until a replacement fan module can be installed A missing fan module will cause a power off pending error since eventually the loss of cooling air could cause the disk modules to overheat The power off pending means that the disk array will power off automatically within ten minutes Figure 29 Removing and Replacing a Fan Module 117 Sunoouso qno Troubleshooting Troubleshooting Replacing Modules CAUTION Before replacing a controller module you should always perform a successful Shutdown on the array using the control panel to avoid loss of data Figure 30 Removing and Replacing a Controller Module 118 Troubleshooting Troubleshooting Procedures Troubleshooting Procedures This section provides troubleshooting procedures that can be used to identify components within the disk array that have failed Fan Module The disk array enclosure is designed to ensure proper airflow with two fan modules operational regardless of the number of disk modules installed No filler panels are required for empty disk module slots three fans must be installed for proper airflow If a fan module fails an amber fan status light indicates that the fan speed is too slow or that the fan has stopped blowing air
3. 221 Sjouiqe Cabinets Appendix C Cabinet Configurations Deskside Cabinet Cabling Deskside Cabinet Cabling To attach the power and SCSI cables to the deskside cabinet perform the following steps 1 Ifpresent remove the rear door 7 from the deskside cabinet as shown in Figure 40 2 Connect the SCSI cables 1 and route them through the hook and loop strap 5 on the left rear of the cabinet 3 Connect the short SCSI cable 2 if necessary Refer to Connecting SCSI Cabling in the Chapter Product Description for more details about SCSI cabling 4 Connect the power cords 6 to the disk array and route them through the hook and loop strap on the right rear of the cabinet 5 Reinstall the rear door onto the deskside cabinet 6 Attach the front door of the disk array as shown in Figure 45 7 keep the deskside cabinet from rolling unscrew the four leveling feet on the bottom four corners of the deskside cabinet until they touch the floor Figure 40 Deskside Cabinet Cabling rN wW 7 p 4 c p u PY 228 Appendix C Cabinet Configurations Rackmount Cabinets Rackmount Cabinets Up to four disk array enclosures can be installed into a 1 6 meter cabinet and up to six disk array enclosures can be installed into a 2 0 meter cabinet Each disk array requires a cabinet space equal to 13 EIA units Figure 41 shows the
4. WER _ 8120 6903 OUTLET PDU LINE IN Appendix D Back to Back Racking Installation Procedure Figure 53 Power Cabling Diagram OS o ooo G 9 alg 251 Sup ewxy Racking Appendix D Back to Back Racking Troubleshooting Troubleshooting A computer instrument rack which has been field upgraded for back to back racking has improved cooling and power features In addition the upgrade the exhaust fan assembly provides circuitry which can report fan failures and provide an indication of power failures Fan and power failures are reported via indicator LEDs on the front of the fan assembly If a failure occurs the rack will continue to operate with its remaining non failed components To isolate a possible power or fan failure use one of the following procedures based on the status of the indicator LEDs NOTE The two exhaust fan assembly power cables need to be connected into two separate PDUS such as PDU 1 and PDU 2 The new PDU
5. Assumptions Use this procedure to download firmware if you are unable to use the host computer or the AutoRAID Manager ARMServer This procedure is used to update firmware using an array controller as the source of the new firmware 1 Obtaina third controller with the desired version of firmware 2 Reset disconnect and then re connect both batteries on third controller at the same time to erase NVRAM 3 Quiet quiesce or stop all I Os deactivate volume set unmount file systems stop traffic to the array 4 front panel display should say Ready Check the System State on the front panel display It should say System State Ready Also check to see which controller is set to primary The primary controller is KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready Primary _ 46 5 Shutdown the array using the front panel KEY CANCEL CANCEL MENU ENTER ENTER HP SureStore E Disk Array 12H Controller Firmware Download Procedure DISPLAY Ready Language View Settings Shutdown Confirm Shutdown Complete After shutdown has completed push the on off button to the off position completing the shutdown This will turn off power to all modules Disconnect all disk drive modules from the back plane You do not need to remove them completely Just pull them out so that the cam handle is fully extended Remove
6. C3595 40043 Battery Cover Not Shown 1 C3595 60018 Shipping Package Enclosure Not Shown 1 C3595 80049 Power Module Not Shown C3595 60055 Shipping Package Fan Module Not Shown 1 3595 60053 Shipping Package Controller Module Not Shown C2905 60150 Fast Wide SCSI Terminator Not Shown C3595 40007 Module Cam Handle Not Shown A3701 60100 Rear Door Assembly Deskside Cabinet Only 5064 2408 C3595 60055 Shipping Package Disk Module and 1 SCSI Cable Right Angle Deskside Cabinet Only C3595 40030 Module Cam Handle Controller Not Shown 1 A3701 60001 Cabinet Assembly Deskside Cabinet Only 5064 2413 Hook and Loop Straps Deskside Cabinet Only 208 Removal and Replacement Replaceable Part Numbers Item HP Replacement No Part Number Description Qty z s 873 V Class SCSI Cable 2 m 3 m Y Cable 68 pin Inline REF Term Cable 871 V Class SCSI Cable 2 m 5 m Y Cable 68 pin Inline REF Term Cable 906 8120 2104 Power Cord SEV Type 12 Power Cord Swiss REF 3912 8120 2956 Power Cord Denmark REF 917 8120 4211 Power Cord CA ASSY 18 AWG South Africa REF 918 8120 4753 Power Cord Japan 100 REF 209 Appendix A Product Specifications This appendix contains the following information Enclosure Physical Specifications Operating Characteristics and Requirements Environmental Requirements Module Speci
7. If necessary remove disk array enclosures see step 4 below The disk array enclosures are heavy Prior to removing the enclosure the individual disk modules should be removed Also when re installing the disk modules they should be installed back into the same enclosure from which they were removed but not necessarily the same slot Thus when removing the disk modules identify on the module from which array enclosure it was removed Remove the top cap from the rack Remove any fans these will be replace by high volume exhaust fans 3 Install the exhaust fan assembly Install this assembly as described in the Exhaust Fan Assembly Installation Guide E7687 90001 4 Install rail kits and or reposition rails as needed Install new rail kits as needed to hold the number of disk enclosures to be racked Ensure that the rails are installed as indicated in Table 21 for either the 1 6 meter or the 2 meter rack Rail installation 15 described in Appendix C of this document Refer to Figure 50 Rack Array Component Positioning for a visual representation of the placement of the arrays components Also install the clip nuts for the array enclosure flange as indicated in Table 21 Remember if the rack has side access you can install disk enclosures in the high density configuration if side access is not an option one enclosure space must remain vacant see Figure 49 243 Sup exX Racking Appendix D Back to Back Racking Ins
8. Secondary Array Controller can also improve the performance throughput of the disk array host dependent No RAID knowledge is required to use the disk array since the logistics of RAID are all managed by the Disk Array Controller Self test is an internal diagnostic test sequence that is performed whenever the power is switched on The results of the Self test pass or fail are displayed by a status light on the front panel of each device Shutdown is a coordinated process of taking the disk array offline During a Shutdown vital configuration information is copied from the array controller NVRAM to the disks This provides more permanent storage for this information In the shut down state the disk array can still execute some SCSI commands from the host but the host cannot access any data on the array The disk array allows a total of three 32 Megabyte SIMMs on the Disk Array Controller The maximum capacity is therefore 96 Megabytes The SIMMs provide Error Correction Code ECC The Status Light is a light on the front panel of each disk module power module and fan The status light is used to indicate normal operating conditions and fault conditions of a particular device The Stripe Depth used by the disk array in RAID 5 mode is 64K blocks Normally RAID 5 mode incurs a performance penalty since in order to determine the proper parity previously written data must be read This is called the read modify write penalty
9. Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Number Logged DEC hex Predictive Maintenance Implication 59 0x3b Error Internal SCSI N More than two per hot plug event Suspected Description Components Disk Drive A disk drive responded Controller incorrectly The result was an Internal internal SCSI bus reset and a SCSI Bus hot plug recovery sequence to regenerate the bus state 60 0x3c Internal Data Y Single Controller A data transfer failed due to a Transfer Error Occurrence Disk Drive parity error This typically Internal SCSI indicates a hardware problem Bus 61 0x3d Host SCSI Bus Y Ignore Host Host SCSI The target timed out on a Timeout Reported Error Bus Host SCSI transaction Or Controller Compatibility 62 0x3e Reassign Failed Y Single Occurrence Disk Drive The controller s attempts to reassign a bad block failed due to a disk drive error Reassign Out Of Y Resources 63 0x3f Single Occurrence Disk Drive The controller was unable to complete a Reassign Blocks command because the sparing buffer is too small There are too many bad blocks in the sparing buffer indicating a problem with the disk drive Optimal Space Y Error In RAID 5 64 0x40 Ignore None The controller detected that space allocated for an optimal only migrating write i e rebuild policy was no longer optimal after the write had completed This
10. e Gigabytes available capacity This is the capacity used for creating logical drives LUNs It represents the total amount of capacity available to the host 101 sjdo uo Concepts Concepts and Management Managing the Disk Array Capacity Some General Rules Regarding Capacity Allocation Building on the information provided in the preceding example there are some general rules you can follow to achieve the most efficient use of your disks capacity Using disks of all the same size in the array produces the most efficient use of capacity In this case there is only one segment and only one parity block When adding a new disk the array will be able to make almost the entire capacity of the disk available to the host assuming the disk is not larger than the other disks This is because the Active Hot Spare and parity requirements are already satisfied For example adding another 4 3 Gigabyte disk to the array would yield almost 4 3 Gigabytes of additional capacity A small amount would be required for RAID 0 1 and controller overhead Adding one or two large disks to an array of smaller disks will not yield the capacity you might expect The upper segment created by the new larger disks will be used for Active Hot Spare and parity and not for available capacity For this reason it is much better to have twelve 4 3 Gigabyte disks than six 9 1 Gigabyte disks When installing a 36 Gigabyte disk drive module into an HP Su
11. Concepts and Management Disk Array Hardware NVRAM Each controller contains NVRAM non volatile RAM used to store vital operating parameters and mapping information used in managing the array Because data is spread across the disks the array controller must have some means of determining where each block of data is located It accomplishes this using a logical to physical data map stored in controller NVRAM This map is vital to managing data without it all data on the array is unrecoverable The contents of the NVRAM is maintained by dual battery backup if power to the disk array is off Each controller module has its own dual set of batteries If the batteries fail or lose their charge while power is off the NVRAM contents will be lost To protect against this the Shutdown operation copies the contents of the NVRAM to each disk when power to the array is turned off by the power switch This allows reconstruction of the data in NVRAM memory if the original map in NVRAM is lost or corrupted The process of copying the contents of NVRAM to the disks makes the disk set independent of its controller Because all the necessary mapping information is on the disks it is possible to install a new controller or move the entire disk set to another controller The new controller will determine that it has a new disk set and the controller will attach itself to those disks Data Maps and Array Recovery Two of the disks in the disk array are u
12. The Disk Changes menu enables you to change disk parameters for the array Include Disk Include Disk Includes a new disk in the array The array cannot use a newly installed disk until it has been included Once included you can use the disk to 1 increase capacity by creating a logical drive using part of or all of the disk capacity or 2 protect against disk failure by enabling Active Hot Spare if not already enabled Before a logical drive is created the disk array can however have increased performance capability once the new disk has been included Auto Include NOTE disk used by another array will be marked Previously Used and will not be automatically included Use Include Disk to include a Previously Used disk Auto Include should be disabled if you want to run a diagnostic on a new disk before the array begins to use it Auto Include enables default or disables Auto Include when a new disk is installed in the array 79 pued 10002 Control Panel Operating the Control Panel Using Control Panel Menus Auto Rebuild CAUTION Until a rebuild is complete data may be lost if a second disk fails or is removed If you stop a rebuild process before it completes the array is operating in an unprotected mode Use Start Rebuild to complete the rebuild process Auto Rebuild enables default or disables automatically rebuilding the array with the redundant data in the event of a disk failure If Auto Rebuild
13. g Q 211 P ckaging Dimensions s u a cao ns dep e eee eie eR ERR Iber EE Gu u eee 212 OPERATING CHARACTERISTICS AND 598 2 00 22222 00 0000000000000000000000000000000000000 212 Power Requirements ini o eso ha ER RE Ee u E Re u Q a 212 A Lr LOWS pace eo aet RO RUD I e OR DE E UD era 212 Heat Dissip tionis a e de e eer CU RUE Og S ee ERR E D ERR AR Fee RR 212 Electromagnetic EMISSIONS seed UTR EE REGI LEG ERU DR UN Vrbe s 213 Acoustic EMISSIONS 213 ENVIRONMENTAL REQUIREMENTS a 213 T mp erature Ranges ioi Cr as ERR 213 Humidity R ng s iico eue pop er Er e EE I rr Eee ha 213 Altitude Ranges oed atate De eee aene eee eve pest 214 214 SHOCK dcr 214 Line Input Requirements eaae 214 Electromagnetic Susceptibility Operating Range eese eene nennen rennen 215 ENCLOSURE tete espe mu bred Eee Ee Pepe eoe E Peer xe eee Reo 215 MODULESPECIFICATIONS etie tete Re eene eee eese eee et 216 Disk Modules uaa Q u n tin e eie n T u anu au ra E 216 Gontroller Module niis
14. other configurations have lower emission levels Gerauschemission For Germany Only 220 LpA 45 0 dB suchend Am fiktiven Arbeitsplatz nach DIN 45635 T 19 Die Daten sind die Ergebnisse von Typpr fungen an Geratekonfigurationen mit den h chsten Gerduschemissionen 12 Plattenlaufwerke Alle andere Konfigurationen haben geringere Ger uschpegel F r weitere Angaben siehe unter Umgebungsbedingungen Appendix Regulatory Safety Statements Electromagnetic Compatibility Electromagnetic Compatibility For U S A designed to meet 47 CFR Part 15 of the FCC rules for Class A digital devices For Europe complies with the requirements of CISPR 22 Class A Low Voltage Directive 73 23 EEC and the EMC directive 89 336 EEC For Canada complies with CSA EMC Class A requirements For Japan complies with VCCI Class A requirements For Korea complies with MIC Class A requirements For Taiwan complies with BCIQ Class A requirements Safety Specifications This product meets the following safety specifications UL 1950 3 Edition CSA 22 2 No 950 95 950 1991 A1 A2 EN 60950 1992 Al A2 IEC 825 1 1993 EN 60825 1 1994 Class 1 Laser LED EMOKO TSE 74 SEC 207 94 TUV EN60950 1992 2 Edition 221 amp 8 Regulatory Safety Appendix B Regulatory Safety Statements Declaration of Conformity Declaration of Conformity DECLARATION OF CONFOR
15. 1 2 3 4 Select the Cntrl Changes from the main menu Select Unit Attn Term Power Parity WDTR SDTR or VEB Select Enable or Disable Select Confirm Switching Controllers 1 2 3 4 Select the Cntrl Changes from the main menu Select Switch Cntrl Select Reset Select Confirm Testing a Controller 1 2 3 4 5 Select the Cntrl Changes from the main menu Select Reset Cntrl Select a controller Cntrl X or Cntrl Y Select Reset Select Confirm 73 pued 10002 Control Panel Operating the Control Panel Using the Control Panel Buttons Formatting the Array CAUTION Once Logical Drives LUNs have been created on your disk array you will not be permitted to Format the disk array again You must delete all of the existing Logical Drives before a disk array can be formatted Deleting Logical Drives LUNs and formatting the disk array destroys all data on the disk array Before formatting a disk array make backup copies of the entire disk array or you will lose all data 1 2 3 4 Follow the earlier procedure called Deleting a Logical Drive to delete all LUNSs Select the Cntrl Changes from the main menu Select Format Array Select Confirm Shutting Down the Array NOTE Shutting down the array will take it offline 1 Select the Shutdown from the main menu 2 Select Confirm Setting the Language 1 Select the Language from the main me
16. 46 SIMM 0 Ist half test 48 SIMM 0 2nd half test 32 MB SIMMs only 50 remote SIMM 0 Ist half test 52 remote SIMM 0 2nd half test 32 MB SIMMs only 54 remote SIMM 0 2nd half test retry 32 MB SIMMs only 60 initialize remote board NVRAM 62 test local NVRAM checksum 64 test remote NVRAM checksum part 1 66 test remote NVRAM checksum part 2 68 NVRAM consistency check 70 sync communication areas 72 determine if download is required 74 test for nocode on primary or secondary 76 no download required check for personality mismatch 78 chip initialization 80 initialize NVRAM for clients 82 jump to OS 90 drive spinup and configuration 95 map and cache upload 99 final initialization 122 Troubleshooting Control Panel Error Messages Control Panel Error Messages The control panel is capable of displaying error messages which indicate possible disk array failures Different types of errors may be displayed such as System States Disk States and other errors Table 10 shows all possible errors regarding System States Table 11 shows all possible errors regarding Disk States Table 12 shows the remaining control panel error messages listed alphabetically Each table shows the control panel error message an explanation of the error and the action that should be taken to resolve the problem NOTE The letter x or y that is displayed on the right side of the display panel usually indicates which array con
17. C 12X 192A p to 5 redundant disk enclosures D 25 400A p to 5 redundant disk enclosures or up to 8 non redundant disk enclosures Z 4X 64A Do not use with BTBR K 17X 272A p to 5 redundant disk enclosures or 8 non redundant disk enclosures In Europe therefore site preparation must include 16 amp circuit breakers preferably with the D or K tripping characteristic code CAUTION Intermittent power losses will occur if improper circuit breakers are used 242 Appendix D Back to Back Racking Installation Procedure 2 Take disk arrays offline power down rack and remove existing components Of the items that are removed some are removed to be replace by new components Some may need to be removed such as the disk array enclosures to allow the enclosure rails to be repositioned since back to back racking in the 2 meter rack utilizes all EIA units including the bottom unit For additional clarification refer to the detailed installation steps following this step a b 4 f g h j Unmount all file systems associated with the devices in the rack Bring devices in the rack offline Power down the devices in the rack Unplug the rack s PDU cables from the power source Remove the rack s rear door and side panels Remove the rear door hinges and latch catch These will not be used again Remove the PDUs 56 inch and brackets from the rack These will be replaced with new PDUs
18. Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 159 0x9f Controller Data Y More than 1 Controller The controller experienced a Bus in 6 months parity error during Microprocessor microprocessor access to Access Parity PRAM or host SRAM Error Typically occurs during self test 160 0xa0 Internal Data Y More than 1 Controller The controller experienced a Bus Error in 6 months SIMM parity error during a DMA operation Typically occurs during self test 161 0xal Mirrored Y Ignore see None A communication problem has Memory Down accompanying developed that prevents the errors two controllers from communicating properly In this state X will be on line and controller Y will be offline The secondary controller will remain offline until a power cycle is performed controller 162 0xa2 Mirrored Y Ignore if Controller A communication problem has Memory recovered by SIMM developed that prevents the Blocked power cycle Backplane two controllers from properly maintaining their memory mirror image The secondary controller will be offline in this state 163 0xa3 Disk Set Y Ignore None The controller has detached Detached Operator Host itself from the disk set This activity occurs when a shutdown has completed successfully Note The component identifier in the log record is the disk set number for this system change event 166 Troubleshooting
19. View Settings System State Warnings Primary Cntrl SCSIID x SCSIID y Firmware SCSI Values Auto Include Auto Rebuild ActivSpare Number of Disks Disk States Rebuild SIMMs Installed MBytes of RAM L Drv States 82 Displays the system state Displays the warnings Displays the primary array controller x y Displays the SCSI ID for controller x Displays the SCSI ID for controller y Displays the current version of firmware for the primary array controller Displays the SCSI values for Unit Attn Term Power Parity WDTR SDTR DRR and VEB Displays the status of the Auto Include Enabled Disabled Displays the status of the Auto Rebuild Enabled Disabled Displays the status of Active Hot Spare Enabled Disabled Displays the current number of disks configured in the array Displays the current states of all disks 1 and B1 B6 in the array Displays the current status of the rebuild operation Displays the number of SIMMs currently installed in the controller indicated Displays the Megabytes of RAM currently installed in the controller indicated Displays the size in Megabytes of all logical drives O to 7 in the array Chapter 4 Concepts and Management This chapter describes in greater detail the concepts and management of the disk array including features and hardware design of the disk array This information will give you a better understanding of how the disk array operates which should help you mak
20. HP SureStore E Disk Array A3700A Location 4 HP SureStore E Disk Array A3700A Location A 1 6 Meter Rack 245 SUDpPey Racking Appendix D Back to Back Racking Installation Procedure 5 Install Replacement PDU brackets Install the PDU angle brackets supplied in the A4915A HP Power Upgrade Kit as described below see Figure 51 a Install the PDU brackets at the rear of the cabinet as described in the A4915A Upgrade Kit Quick Installation Guide However when installing the brackets two on the right column and two on the left column install them in the positions as indicated in the table below not as described in the quick installation guide Rack Accessory Holes as counted from the top of the rack down 1 6 Meter Rack 12and 33 2 meter Rack I and37 NOTE Do not install the PDUs until after the disk array enclosures have been installed 6 Install the disk enclosures Replace any array enclosures you removed from the rack and install any new arrays Refer to appendix C of this document for array enclosure installation Remember when installing the disk modules ensure that they are installed back into the same enclosure from which they were removed NOTE For optimum cooling efficiency array enclosures should be racked from top to bottom in the front of the rack and from bottom to top in the back of the rack If side access to the rack is not goi
21. Make sure the correct down type is specified when performing the command 37 0x25 Recovered With Ignore See None Retries were used successful RAID 5 Retries accompanying completion of a read or write errors to RAID 5 space 38 0x26 Recovered With Ignore See None Redundancy was used in the RAID 5 accompanying successful completion of a Redundancy errors read or write to RAID 5 space 39 0x27 Unavailable See None An attempt to read or write Data Detected Accompanying failed because the array is in By RAID 5 Errors a data unavailable state This may be a result of removing disk drives Reinstall any disks that may have been removed 40 0x28 Send Diagnostic Single Controller The current system state and Self Test Failure Occurrence warnings indicate that the self test should return bad status Check system state and warning states for more information 41 0x29 Invalid Pass NA None The host attempted an execute Through pass through command without a valid Set Pass Through Mode command preceding 143 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 42 0x2a Pass Through Y Ignore None A command to the disk drive Reset Failed failed Drive status should be returned and can be examined to determine the condition of the disk drive 43 0x2b Recovered With Y
22. This procedure assumes that the Model 12 and HP SureStore E Disk Array 12H controllers have been updated to at least HP25 and HP24 firmware revisions respectively If they have not been updated to these versions of firmware then refer to the service notes indicated below to update the array before continuing e 515 03 A3516A 03 and A3516Z 01 HP AutoRAID Disk Array for the Model 12 e A3700A 01 A3700 AD 01 and 3700 7 01 HP AutoRAID Disk Array for the HP SureStore E Disk Array 12H NOTE It is required that a backup of the array be performed before downloading the firmware CAUTION Follow all the procedures exactly as described If you fail to follow the procedures exactly you could lose data CAUTION upgrading to 40 or later firmware for the purpose of using 36 Gigabyte disk drives it is important that the controller firmware be upgraded before adding the 36 Gigabyte disk drives as any data written on those disk drives using any controller firmware previous to version 40 will be lost 39 J9 0 9u0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Controller 40 Figure 15 Downloading Controller Firmware Flowchart Downloading Controller Firmware NOTE Downloading firmware is an offline process In most cases the LUNs on the array need to be either un mounted or deactivated from the volume group Perform the Yes Downl
23. USING THE CONTROL PANEL ETC RE EATE ins 69 CONTROL PANEL STATUS LIGHT OPERATION cssssscccccecsesssececececescseececececsenenseaesececeeseneaeeeeecsesenseaeeeeecs 70 Selecting a Meny ltem ier eo eet ET Re ea iSS EARTE 70 Returning to the Main Menu a 70 USING THE CONTROL PANEL BUTTONS 22 20 66600 0000000000000000000000000000000000000 71 Viewing Array Settings sauna a aaa upas payaa 71 Creating a asan et erbe eee Yee ERU Pee eere etae 71 Renumbering Logical Drive eese esee enne then e SEE nne 71 Deleting a Logical Driyey sessi ERG 72 In luding a Disk ite He rre ORE RR e 72 Enabling Disabling Auto Include aene siriani 72 Enabling Disabling Auto Rebuild cii esie betreiben teretes deeds 72 Starting Stopping a Rebuild sss seisi eee eiae teet nee tret tie iHe ii trend 72 Setting the Controller SCSI ID ectetur red er Herr i SESSE 73 Changing Other Controller SCSI Settings eese eee 73 Switching Controllers e nee eee reir EGER Hr e er SEESE 73 Testing a Controller ide eie etie p a p rei ar 73 Formatting
24. index number that the array s is on If you select the default range index numbers of all busses found this process could take 5 minutes After checking the array busses ARDIAG will search for firmware files on the tape and report those to you as well The firmware files located on the tape will present you a table that looks like the following File name Intended Product ID Revision Size C3586AHPxx HPC3586A disk array HPxx 1048576 C5447 AHPxx HPC5447A disk array HPxx 2097152 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Now to set the environment variables 10 11 12 13 14 Run the DISPMAP command This command allows you to display all available AutoRAID disk arrays controllers on the system Note the Test Disk of the controller you ve identified as the primary by its path in ioscan Type TESTDISK where _ is a space and is the Test Disk from DISPMAP Your returned message will read Array state is ready Run the DISPMECH command This command allows you to display physical mechanisms within a selected AutoRAID disk array Note the Test Level of the FRU you want to download the firmware to i e Cntrl X TESTLEVEL where _ is a space and is the Test Level from DISPMECH Your returned message will read TESTLEVEL where is the number you selected At this point you can run the INQUIRY command to verify that you have selected the primary cont
25. 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 new 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 Aut oRAIDNFW 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 43 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Download A Downloading Disk Array Firmware from a File Using ARMServer AutoRAID Manager CAUTION Controller firmware HP4x and any later versions have a different data map format in NVRAM memory Once you HP48 or later you can not downgrade to any firmware version prior to HP4x without first completely backing up all d
26. 264 In either RAID 0 1 or RAID 5 mode the data is stored with Data Redundancy so that at any time a single disk failure will cause no loss of user data and no interruption of data transfer While Logical Drive space is defined and thus fixed the available amount of free space can vary because Dynamic Data Migration is able to switch user data from RAID 0 1 to RAID 5 mode as the Disk Modules become full Similarly installation of additional Disk Modules allows more data to be stored in RAID 0 1 mode thus increasing the disk array performance The box or set of boxes containing one or more SCSI devices It may provide the power cooling mechanical support and external electronic interfaces for those devices Erasable Programmable Read Only Memory A device that uses Fibre Channel technology Fast Wide SCSI is a 68 pin bus implementation that utilizes some of the data lines of the bus for address lines One of three existing Fibre Channel FC topologies in which two to 126 devices are interconnected serially in a single loop circuit The arbitrated loop topology supports all classes of service and guarantees in order of delivery of frames when the source and destination are on the same loop The process of reinitializing the disk media A format destroys all data on the portion of the disk media being formatted Two types of formats are available a hard format that overwrites all data on the disk media and a soft format that s
27. 60001 A3714 69001 18 2 Gigabyte Disk Module 10 000 rpm 5292 A5292 60001 A5292 69001 36 4 Gigabyte Disk Module 10 000 rpm A3706A A3706 60004 A3706 69004 96 Megabyte Controller Module N A C5445 60005 N A Backplane Assembly A3708A A3708 60002 A3708 69002 Power Module A3709B 206 A3709 60002 N A Fan Module Removal and Replacement Replaceable Part Numbers Table 17 Disk Array Enclosure Replaceable Parts HP Replacement Part Number Description Not Available Enclosure Chassis C5445 60005 Backplane Assembly 5064 2404 AC Power Adapter Kit 1 Power Receptacle 2 T 15 Screws with Captive Lock Washers C5445 60042 Power Supply Lockout Kit 1 Standoff 1 Screw C5445 60043 Backplane Assembly Hardware Kit 11 M4 Nuts 4 Standoffs 4 Star Washers 3 Flat Washers 2 T 15 Screw Display Module 2 MA Captive Screws Rear Cover Assembly 2 6 Star Washers Rear Cover Assembly 13 14 15 Not Available Rear Cover Assembly 207 Removal Removal and Replacement Replaceable Part Numbers Item HP Replacement No Part Number Description Qty 16 C5445 00024 Hinge Bracket 1 17 C2786 00002 Rack Rail Rackmount Only 2 8120 6514 Power Cord CA ASSY IEC 320 Rackmount Only 3 5064 2407 Module Cam Tooth Kit 1 Module Cam Tooth 1 1 Nut 1 Washer Not Shown 1420 0532 Battery Pack Not Shown 2 C3595 60141 Battery Box Cover Not Shown 1
28. 63016 Option 843 2 0 m V Cable 68 pin HD VHDCI 68 pin high density part number A3639 63017 Option 844 2 0 m V Cable 68 pin HD VHDCI I L Term 68 pin high density part number A3639 63018 Battery Pack part number 1420 0532 ac power cords options for different countries Strapping Kit for repackaging the disk array part number C5445 80019 Unpacking and Repackaging the Disk Array When you unpack your disk array be careful to save all packaging items including the strapping material The strapping material will be needed if you ever have to ship or return your disk array Figure 1 shows the packing items and strapping material shipped with the disk array 14 Product Description Unpacking and Repackaging the Disk Array Figure 1 Repackaging the Disk Array for Shipment 15 uondrnbsoq Description Product Description Product Features Product Features Figure 2 shows the front of the disk array Figure 3 shows the rear of the disk array Figure 4 shows the controls and indicators on the disk array Figure 5 shows the slot numbers for the disk modules Figure 2 Disk Array Front Panel 1 Power Standby Switch 2 Control Panel Display 3 Control Panel Status Light 16 Figure 3 Disk Array Rear Panel Controller X SCSI Connector 1 2 C
29. A failed fan module has no airflow If two fans fail power will automatically shut down to avoid overheating Power Module Each power module contains a power supply with an internal fan for cooling Three power modules redundant configuration use active current sharing to distribute the power load If one power module fails or is removed or if one of the ac power cords is accidentally removed the other two power modules take over the entire power load for the enclosure without interruption CAUTION The following conditions will cause a power module to shut down automatically e two failed fan modules apower module internal fan failure or other internal failure e anover current condition caused by a faulty disk module or by using an unsupported disk type with an excessive current rating 119 Sunoouso qno Troubleshooting Troubleshooting Troubleshooting Procedures Power Module General Troubleshooting Procedure Follow these steps for any power module failure except if the disk array contains three power modules see Three Power Modules Troubleshooting Procedure in this chapter 1 Remove all disk modules Install the disk modules one at a time until a power module fault occurs Remove all disk modules except the last one installed If a power module fault occurs again replace the disk module ON ME Jt If a power module fault occurs after replacing the disk module replace the power module Three P
30. FAX 49 7031 143143 222 Appendix C Cabinet Configurations Cabinet Configurations Available This chapter shows the different disk array cabinet configurations The disk array can be configured in an EIA cabinet a 1 1 meter cabinet a 1 6 meter cabinet a 2 meter cabinet or can be enclosed in a deskside cabinet Deskside Cabinet The deskside cabinet is a single disk array enclosure in a cabinet on wheels The deskside disk array cabinet is shown in Figure 37 Figure 37 Deskside Cabinet 223 Sjouiqe Cabinets Appendix C Cabinet Configurations Unpacking the Deskside Cabinet Unpacking the Deskside Cabinet To unpack the deskside cabinet perform the following steps 1 Cutthe strapping bands and slide the fully packaged product off of the pallet as shown in Figure 38 View 1 2 Remove the four plastic clips and lift the carton off of the cabinet Also remove the accessory tray and then remove the top front and rear foam cushioning as shown in Figure 38 View 2 3 Liftone edge of the cabinet and then remove the bottom cushion from the opposite side Repeat this step for the other side of the cabinet See Figure 38 View 3 4 Rollthe cabinet off of the bottom tray as shown in Figure 38 View 4 224 Appendix C Cabinet Configurations Unpacking the Deskside Cabinet Figure 38 Unpacking the Deskside Cabinet 225 Sjouiqe Cabinets Appendix C Cabinet Configurations Installing t
31. However when the data to be written exceeds the size of the stripe depth or 64K no read modify write penalty is incurred because all of the data including parity already exists in RAM 269 JUSSO Glossary Glossary Target Terminator Block Time Stamp Unallocated Capacity Uninterruptable Power Supply VEB 270 A SCSI device usually the peripheral that responds to an operation requested by a SCSI initiator usually the host system SCSI peripherals are targets but for some commands for example a COPY command the peripheral may need to act temporarily as an initiator An electrical connection at each end of the SCSI bus composed of a set of resistors or possibly other components Its function is to provide a pull up for open collector drivers on the bus and also impedance matching to prevent signal reflections at the ends of the cable The SCSI bus requires termination at both ends of the bus One end of the SCSI bus is terminated by the adapter s internal termination The other end should have a terminator placed on the 68 pin high density SCSI connector on the last SCSI peripheral If this device is not terminated data errors may occur data blocks stored on the disk array are written along with a Time Stamp The Time Stamp allows the Disk Array Controller to determine how frequently blocks of data have been accessed Array capacity that has not been assigned to a Logical Drive and therefore
32. I O Retry the front panel operation until you are successful 129 Sunoouso qno Troubleshooting Troubleshooting Control Panel Error Messages Error Message Explanation Action physically removed from the disk array enclosure this disrupts the flow of cooling air to the other modules in the enclosure Not Enough There are not enough fewer than three disk Install more disk modules Disks modules in a logical drive set for redundancy NVRAM The NVRAM images miscompared array If possible Mismatch doesn t know which image to use This error can 1 remove both controllers occur if you remove two controllers from two 2 reinstall them in their original arrays separate arrays neither of which were Shutdown and install them into a new array 3 run Shutdown 4 return them to the array where the error Caution The data memory maps stored in occurred or NVRAM are necessary to read the user data 1 remove one controller 2 power on the array and wait for it to become Ready 3 Shutdown the array 4 reinstall the controller you removed 5 power on the disk array NVRAM x y The NVRAM in the controller module in slot X Replace the controller Failed or Y failed Offline SCSI ID One or more controller in this disk array has not Set the desired SCSI ID for the controller s been assigned a SCSI ID Optimizing The array is migrating data to RAID 0 1 to None Information only optimi
33. Ignore See None Redundancy was used by the RAID 1 Accompanying controller in the successful Redundancy Errors completion of a read or write 44 0x2c Unavailable Y Ignore See None The controller detected that Data Detected accompanying the disk drives necessary for a By RAID 1 errors read or write were not available for access before the access was attempted This can apply to an initial attempt to read or write or to a retry 45 0x2d Shutdown N NA None The controller could not Failed shutdown the disk array because all the disk drives were missing Reinstall the missing disks to shutdown the array 47 0x2f Disk Drive N NA None A disk drive was removed Removed from the disk array while an During Add Add Physical Drive command Physical Drive was in progress against the drive 144 Event Number DEC hex 48 0x30 Event Name Cancel Rebuild With Auto Rebuild Enabled Event Logged Predictive N NA Maintenance Implication Suspected Components None Troubleshooting Event Code Descriptions Description An attempt was made to cancel a rebuild operation while Auto Rebuild was enabled Because Auto Rebuild is enabled the rebuild would only start up again after the cancel rendering it ineffective Therefore the controller always rejects a cancel rebuild command while Auto Rebuild is enabled 49 0x31 Disk Drive Stamp Write Failed Y Single Occurrence See accompanying
34. Modules susan ter i RR OR ERE RED UR EE REDE THREE ERES 95 Power Modes o adc oim CUR RR EU E REP UR DR RE La ate 98 Fan Modules ii RR REP ERR 98 SCST Bus tite ea eo aee e Pe edic ge Ue Pe e et Sl a 99 MANAGING THE DISK ARRAY CAPACITY nennen rennen inneren eren inneren nennen 100 Sample Capacity Allocation u a 101 Capacity Management Strategies 103 UPGRADING THE CAPACITY OF A FULLY LOADED DISK 104 OPTIMIZING PERFORMANCE yess xen tet eer tede e e HIR Seer Ee e p E Pt Rs 105 Controller Data Path ie eite eta te itin in ente eee Ene 106 Performance Metrics ise ie tiene te eee ied e ed eee re Pe i Renee altered tet 107 Rebuild Impact on Performance 2 109 Write Working Set eet UR ete te maq Rr Cree e Pe 109 Increasing the Amount of RAID 0 1 Space 20 110 Why Deleting Files Won t Increase RAID 0 1 Space nennen 110 Deleting a Logical Drive to Increase RAID 0 1 Space essen eene eee 111 CHAPTER 5 TROUBLESHOOTING i desse ees heres sue aos heo ee deo rabo sabes ee ie ees soie riae orare ike 113 MODULE SLOT NUMBERS 5 z quia 1
35. Panel Error Messages Error Message Table 11 Control Panel Error Messages Disk States Explanation Action 126 Disk States Log information is being logged on the None Information only Current Log Disk specified log disk Disk States The disk module is in a Not Included state The user requested the array to enter this state Downed Use the Include Disk command if you want to include the disk module Disk States The disk module is in a failed state Replace the disk module Failed Disk States The disk module failed during initialization Replace the disk module Failed Setup Disk States The log disk has failed Return the log disk to the factory for a Log Disk Failed replacement Disk States The log disk is full and ready to be evaluated Return the log disk to the factory for Log Disk Full evaluation of the log Disk States The disk module is not part of a logical drive You must manually include the disk module to Not Included Auto Include is OFF make it become part of the logical drive set Disk States The disk module has been used in a different None Information only Previously Used array CAUTION The disk module may be used in the present array but the original data will be lost when Include Disk is executed Disk States The disk module is in a ready state None Information only Ready Disk States The disk module is in an unknown
36. Remove the controller with the bad batteries and install new batteries The controller batteries are Ni Cad Nickel Cadmium type batteries When replaced these batteries must be recycled or disposed of properly Replace batteries only with the same type Press down on the battery cover latch 2 Remove the battery cover latch and RFI shield if necessary you can use a flat bladed screwdriver to remove the RFI shield Disconnect the cable from the controller PCA to ONE of the battery packs Replace the first battery pack with a new battery pack and reconnect the cable Disconnect the cable from the controller PCA to the other battery pack Replace the second battery pack with a new battery pack and reconnect the cable B n Ro Reinstall the RFI shield and the battery cover latch 36 HP SureStore E Disk Array 12H Controller Downloading Array Controller Firmware Downloading Array Controller Firmware When two controllers are installed in the disk array enclosure dual controllers both controllers must have the same version of firmware If you update or download new firmware the new firmware is copied to both controllers When a controller is replaced however the replacement controller may have a later version of firmware than the currently installed controller If the installed controllers have different versions of firmware you will get an error message on the display panel that says Firmware Needed In t
37. SCSI bus reset switches see Figure 33 are located on the top left corner of the backplane assembly To locate the switch remove the rear cover to expose the switch Figure 33 shows the switch segments on the SCSI bus reset switch and Table 15 shows the switch functions The SCSI bus reset switches are preset at the factory for proper operation A SCSI bus reset is a very high priority interrupt The SCSI bus reset option is used when the host controller or software requires a SCSI bus reset to be sent as notification of the removal of a controller or of a power shutdown The default setting for the SCSI bus reset switch is shown in Figure 33 There are two switches for each bus one switch for power detect and the other is for controller detect When enabled the SCSI bus is reset when the following conditions occurs Power detect conditions e power switch is cycled e When power is first applied power on e When all power is removed power off Controller detect conditions e controller module is removed e Acontroller module is inserted The following actions will not cause a power or controller detect reset condition e Removing replacing or adding a single power module if there are two additional functioning power modules in the enclosure A single power module is not sufficient to operate a fully loaded disk array e Removing a fan module NOTE Removing a single fan module will power off the disk array
38. X whenever power is applied or removed Controller Detect Switch is enabled default the disk array controller generates a SCSI Bus Reset on Bus X whenever disk array controller X is inserted or removed When the Bus X Controller Detect Switch is disabled the disk array controller will not generate a SCSI Bus Reset on Bus X whenever disk array controller X is inserted or removed 198 Removal and Replacement Backplane Assembly HP qualified personnel only Figure 33 Replacing the Backplane 199 Removal Removal and Replacement Power Button and Push Rod HP qualified personnel only Power Button and Push Rod HP qualified personnel only CAUTION The following procedures require that you perform a shutdown turn off the disk array and make it inaccessible Check with the system administrator before proceeding To replace the power button and push rod see Figure 36 1 CUR 200 Remove the Backplane Assembly Grasp the power button with pliers and pull out to remove the power button Locate and carefully remove the spring from the power button housing on the front of the enclosure Slide the push rod to the rear Reassemble in the reverse order holding the push rod while sliding on the spring and then snapping the power button onto the push rod Note the following e The rear of the push rod is angled downward to fit the switch on the backplane a
39. a controller error although the disk array may still support some I Os without adding disks 137 0 89 Rebuild Started Y Ignore See None A rebuild has been started accompanying either automatically or by errors command 138 0x8a Rebuild Y Ignore None A rebuild has completed Complete successfully 139 0x8b Insufficient Y Ignore See None A rebuild was not started due Space To Start accompanying to insufficient disk space This Rebuild errors event is most likely to occur when Active Hot Spare is not enabled To create the necessary disk space add a disk or delete a LUN 140 0x8c Rebuild Failed Y Ignore See None Rebuild has failed to complete accompanying due to multiple disk failures or errors a controller failure Previous messages in the log should indicate whether or not there have been multiple disk failures 159 sunoous qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex Event Name Event Logged Predictive Suspected Maintenance Components Implication Description 141 0x8d Cancel Rebuild Y Ignore None The cancellation of a rebuild Complete Operator or has completed Host Activity 142 0x8e Disk Drive Y Ignore None The controller is capable of Table Overflow Operator retaining information about 16 activity drives Disk drives that are removed from the array and enter the missing drive list are included in this count If the
40. an array power supply P1 P2 and P3 to a different PDU PDU 1 PDU 2 and PDU 3 as shown in Figure 52 Power Cable Schematic To help ensure that individual array supplies get power from a different PDU divide the cabinet into vertical thirds looking from the back All power supplies on the right side go to PDU 1 mounted on the upper right side all power supplies in the middle of the arrays go to PDU 2 the horizontal PDU and all power supplies to the left go to PDU 3 on the upper left side Figure 52 illustrates the cabling for a 2 meter rack with 6 units However this diagram applies to any configuration a 2 meter with 5 units or 1 6 meter with 3 or 4 units Simply follow the cabling for the number of arrays you have installed Route the cables from the back of the disk array out and around the outside of the rack column and into the PDU as shown in Figure 53 Do not connect more than 8 power cables into one PDU 248 Appendix D Back to Back Racking Installation Procedure 9 Install the SCSI cables Use 0 5 meter flexible SCSI cables to make connections between Bay A and Bay B of each storage enclosure and between storage enclosures Install the SCSI cables to conform to the system requirements Refer to Connecting SCSI Cabling in the Chapter Product Description for more details about SCSI cabling 10 Complete the installation a Install filler panels to completely fill any unused space in the front and back of
41. assumption that in most situations 90 of the data on the disk remains essentially unchanged That is it is only read or not accessed at all and can therefore be stored in RAID 5 without impacting performance Thus 1046 of the capacity should be large enough to service the remaining data that is frequently updated and performance can be maintained However if your applications make heavy write demands on the disk array the write working set may exceed the 1076 RAID 0 1 minimum In this case the array must begin servicing write I Os from RAID 5 and the performance of the array begins to degrade Large data base applications that frequently update many records may have a write working set large enough to exceed the 1076 RAID 0 1 space 109 sjdo uo Concepts Concepts and Management Optimizing Performance Increasing the Amount of RAID 0 1 Space Available CAUTION Deleting a logical drive LUN destroys all data on the logical drive Before deleting a logical drive make backup copies of the files you want to save or move the files to another logical drive If the write working set is exceeding the amount of available RAID 0 1 space you can restore performance by increasing the amount of RAID 0 1 space You can do this in one of the following ways e Adda disk and leave its capacity unallocated This is an effective way of permanently increasing the amount of RAID 0 1 space available for the write working set Delete an unne
42. command dd ifZARDhHif of dev rmt tapedrive bs 2k 2 Onthe machine you want to download and while the operating system is running get the following information Perform an ioscan fnCdisk and ioscan fnCtape to locate the arrays and the tape device and note their location Tape device hardware address Tape device location 1 dev rmt Disk array s bus number s Firmware file name needed to download This file should be in the file Primary controller 53 J9 0 ju0 Controller HP SureStore E Disk Array 12H Controller Firmware Download Procedure If you are unsure which controller is the primary verify from the front panel Use the information below to discover which controller is the primary Pu ON SA roe 54 KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready Primary Once this is complete you can use this tape as a boot tape to run ARDIAG on any PA RISC system using an AutoRAID product as a boot device Shutdown HP UX system and reboot the system from the tape At Interact with IPL y or n answer y At ISL type ODE ARDIAG Enter the password at this point ARDIAG will scan the array busses and report those on the screen each bus will be referenced by an index number At the Please enter a range of the disk array busses you want to check select only the range by
43. contents of the rackmount kit shipped with the disk array enclosure WARNING The disk array may weigh up to 38 2 kg 84 Ib To avoid personal injury remove all modules from the disk array before installing it into the EIA cabinet Also to ensure cabinet stability and to avoid personal injury install disk arrays into the bottom slots of the EIA cabinet first Figure 41 Rackmount Accessory Kit 1 Support Rails 2 5 Door Snap 1 2 Clip Nut 8 6 Door 1 3 T 25 Screw 8 7 Retainer Clip 2 4 T 15 Screw 1 8 Screw 2 229 Sjouiqe Cabinets Appendix C Cabinet Configurations Rackmount Cabinets Installing the Disk Array into a Rack 1 Install one clip nut Figure 42 Item 1 on each of the four rail standards on hole number 3 Secure the rails 2 with four T 25 screws 3 See Table 18 for rail standard clip nut spacing for multiple disk arrays This spacing begins installation by installing an array in the lowest cabinet EIA position 2 Install two clip nuts on the front left and two on the front right ear standards one clip nut Figure 42 Item 4 on hole number 7 and one clip nut 5 on hole number 33 See Table 18 for ear standard clip nut spacing for multiple disk arrays 3 Slide the enclosure into the cabinet see Figure 43 Secure the enclosure with four T 25 screws 4 Slip the Retainer Clip Figure 41 Item 7 up against the back of the array enclosure and secure t
44. controller is NOT new reset disconnect and then re connect both batteries on the replacement controller at the same time to erase NVRAM 3 Verify which controller has failed by examining the logs using the Logprint command This is to make sure you do not remove the functioning good controller 4 Perform a shutdown of the operating system Insure that there is no I O traffic presented to the array 5 Remove the failed controller After Initialization the display panel should say Ready CAUTION Itis very important that the array be in a Ready state The array should not present any warning other than a single controller warning statements at this point 63 J9 0 9u0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 6 Shutdown the array using the front panel KEY CANCEL CANCEL MENU ENTER ENTER DISPLAY Ready Language View Settings Shutdown Confirm Shutdown Complete After shutdown has completed push the on off button to the off position completing the shutdown This will turn off power to all modules 7 Disconnect all disk drive modules from the back plane You do not need to remove them completely Just pull them out so that the cam handle is fully extended 8 Pullthe existing good controller and reset disconnect and then re connect both batteries at the same time This will mean that the only copy of valid maps reside on the disk
45. controller was discovered to be bad during power on Specifically the secondary controller appears to have failed due to the inability for the communication area of NVRAM to be cleared of ECC errors A controller was discovered to be bad during power on Specifically the secondary controller appears to have failed due to the mirror not coming to an up state within 1 second of power on 181 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Number Logged DEC hex Secondary Y Controller Failed With Communication Timeout 218 0xda Member Disk Y Drive Added Back Into Disk Set 219 0xdb Controller Y Initiated Host SCSI Reset 220 0xdc 182 Predictive Maintenance Implication See Controller accompanying Backplane errors Ignore None Operator or host activity Ignore None Suspected Components Description A controller was discovered to be bad during power on Specifically the secondary controller appears to have failed due to its failure to respond within the message timeout period The controller has detected that a member of the disk set that was either down failed or missing has been returned to the ready state This can occur as a result of a Add Physical Drive command or a hot plug installation of the drive The disk drive ID number assigned to the drive added appears in the component identifier field
46. disk array logical drive as an individual disk subdividing it again if necessary into partitions The disk array can be divided into up to eight logical drives To maintain data redundancy and improve performance the data in each logical drive is distributed across all disks in the array Consequently there is no correlation between a logical drive and a specific physical disk each logical drive encompasses all disks This is particularly important to remember when adding a disk and creating a logical drive to use the new capacity The new logical drive you create is not confined to the new disk like all the other logical drives it is distributed across all the disks in the array Figure 21 Logical Drives KRE a LOGICAL DRIVEL a LOGICAL lt DRIVE2 DISK 1 DISK 2 DISK 3 DISK 4 88 Concepts Management Disk Array Features Auto Include The Auto Include feature of the disk array simplifies the process of adding new disks to your array New disk capacity can be added to the disk array while it is running eliminating the need to bring the host computer system down You avoid the time consuming backup restore process required by some disk arrays when adding new capacity Auto Include automatically adds new disks to the disk array configuration as soon as they are installed in the array enclosure Once included the array immediately begins using the new disk to increase performance by a
47. is disabled you must use Start Rebuild to manually rebuild data see Start Rebuild Auto Rebuild will start after a disk failure if Active Hot Spare is enabled or if you install a new disk of equal or greater capacity to replace the failed disk Active Hot Spare Active Hot Spare enables default or disables the allocation of an Active Hot Spare space If Active Hot Spare is enabled and Auto Rebuild is enabled and if there is enough unallocated capacity Active Hot Spare will rebuild data if a disk fails If you enable Active Hot Spare before creating any logical drives LUNS Active Hot Spare will guarantee that there is sufficient unallocated space to perform an Auto Rebuild If there is not enough unallocated capacity available you must add a disk to the array or delete a logical drive to create sufficient unallocated space When Active Hot Spare is disabled the array may not be able to rebuild the data until another disk is added to the array Start Rebuild Start Rebuild begins the rebuild process to recover redundant information that was on a failed disk The array reconstructs the data from the failed disk using redundant data from the remaining disks Stop Rebuild CAUTION Until a rebuild is complete data may be lost if a second disk fails or is removed If you stop a rebuild process before it completes the array is operating in an unprotected mode Use Start Rebuild to complete the rebuild process Stop Rebuild ends th
48. is not available to the server The disk array uses all Unallocated Capacity to improve array performance by using it for RAID 0 1 storage An Uninterruptable Power Supply is a power supply that is capable of maintaining power even if the input ac mains supply loses its source of power VEB Very Early Busy is a SCSI parameter VEB instructs the disk array to return a BUSY response during its power on sequence Following a power on it can take the disk array up to three minutes to complete its internal self test sequence During this time the disk array will not respond to host commands if VEB is disabled This may cause some host systems to time out while waiting for the disk array to respond If VEB is enabled the disk array will return a BUSY status to the host rather than simply ignore any commands during this interval This will alert the host system to the presence of the disk array and thus avoid the possibility of the host system timing out before the disk array is ready WDTR Write Working Set Glossary WDTR Wide Data Transfer Request controls whether or not the additional eight bits on a wide bus will be utilized in most data phases of a SCSI command WDTR also determines the negotiation protocol of the host If WDTR is enabled default the disk array will initiate for negotiation for wide transfer If WDTR is disabled the host will initiate negotiation protocol In either case the disk will always respond to any r
49. left to right top to bottom Overloading one or two of the internal SCSI busses can impact throughput and decrease performance Figure 24 SCSI Bus Details SE SCSI CONTROLLER X CONTROLLER Y 99 sjdo uo Concepts Concepts and Management Managing the Disk Array Capacity Managing the Disk Array Capacity This section explains how to manage your disk array efficiently It concentrates on two primary management tasks managing capacity and optimizing performance The overall capacity of the disk array can be configured in a variety of ways Depending on your system needs you can configure the disk array capacity to provide maximum available capacity improved performance or optimum protection from disk failure The entire capacity of the disk array is divided into several categories each playing a different role in the operation of the array To manage array capacity efficiently you should understand what each of these categories do e Logical Drive LUN capacity assigned to the logical drives LUNs created on the array This is the capacity visible to the operating system You must create the desired logical drive configuration on your disk array observing any specific limitations and requirements imposed by your operating system Each logical drive appears to the operating system as a separate disk drive The capacity for each logical drive is distributed across all the disks in th
50. may need to be removed to be repositioned see step 4 below Install the exhaust fan assembly Install rail kits as needed Install replacement 19 PDU angle brackets Install re install disk enclosures Install replacement 19 PDUs Install the power cables Install the SCSI cables Complete the installation Some of the installation procedures identify the installation of kits The detailed installation procedures for these kits are included in documentation supplied with the kit such as the exhaust fan assembly kit rail kit and power PDU upgrade kit 241 Sur e Racking Appendix D Back to Back Racking Installation Procedure Installation Procedure 1 Confirm power requirements Ensure that the installation site meets the following requirements e Input voltage 200 230 240 V AC single phase e Input frequency 50 Hz 60 Hz e Input current 16 A maximum operating PDU US installations Standard 20A Square D circuit breakers are adequate for all authorized configurations of back to back racking European circuit breakers In Europe some sites may require upgraded circuit breakers This is because inrush current for a single disk enclosure may be has high as 57 16 amps peak for up to three milliseconds Note the figures in the following comparison chart Tripping Current Total trip current Back to Back Rack usability characteristic code multiplier B 6X 96A Do not use with BTBR
51. right mounting flange 2 Secure the door snap with a T 15 screw 193 IEAo ur q Removal Removal and Replacement Display Module HP qualified personnel only Display Module HP qualified personnel only CAUTION The following procedures require that you perform a shutdown turn off the disk array and make it inaccessible Check with the system administrator before proceeding To replace the display module see Figure 32 1 Shutdown the disk array turn off the disk array disconnect the ac power cords and SCSI interface cables 2 Remove the two captive screws from the rear cover Tilt back the top of the rear cover and remove it completely 3 Remove the two 15 4 Screws Figure 32 Item 3 that secure the display module to the backplane assembly 4 Remove both controller modules from the enclosure 5 Remove the display module from the front of the disk array 194 Removal Replacement Display Module HP qualified personnel only Figure 32 Replacing the Display Module 195 Removal Removal and Replacement Backplane Assembly HP qualified personnel only Backplane Assembly HP qualified personnel only CAUTION The following procedures require that you perform a shutdown turn off the disk array and make it inaccessible Check with the system administrator before proceeding To replace the backplane assembly see Figure 33 Item 2 1 196 Shut do
52. share the power load equally If one power module fails the other two power modules are capable of powering the disk array If two power modules fail the controller will shutdown the disk array Fan Modules Three cooling fan modules are included in the standard disk array enclosure The fan modules provide the air flow necessary to maintain the proper operating temperature for the disk array 98 Concepts and Management Disk Array Hardware SCSI Bus Topology The array controller communicates externally with the host over a fast wide differential SCSI bus Each array controller is addressed by the host as a separate SCSI target therefore they must each have different addresses if they are connected to the same SCSI bus The array controller SCSI address and all other SCSI operating parameters are set programmatically from the control panel on the disk array enclosure or using the host software utility There are no hardware jumpers on the array controller Internally the array controller communicates with the disks over four separate single ended SE SCSI buses Disk SCSI addresses and all other operating parameters are set internally by the disk array controller There are no hardware jumpers on the disk modules To ensure optimum performance the disk modules should be distributed equally across the internal SCSI busses This is accomplished by simply making sure the disk modules are installed in the enclosure in the correct order
53. shipped with each new controller module In the disk array shipping carton the batteries are located in a cardboard sleeve next to the array controller If you are replacing a failed controller module first remove and use the battery packs from the failed controller module in the new controller module To install the controller battery packs do the following see Figure 14 1 Ifthe battery cover latch is not already removed press down on the battery cover latch 2 Ifthe battery cover latch is not already removed remove the battery cover latch and then remove the RFI shield if necessary you can use a flat bladed screwdriver to remove the RFI shield 3 Install both battery packs and connect the cables as shown 4 Reinstall the RFI shield and the battery cover latch CAUTION Make sure you reinstall the battery cover latch carefully so each of the locking tabs see Figure 14 View 5 Items 1 and 2 slide beneath the battery compartment If the battery cover latch is not secured properly it could prevent the removal of the controller module once it is installed in the disk array enclosure After new battery packs are installed and the array controllers are installed and powered on about one minute is required before the Battery Discharged message on the control panel disappears About twelve hours are required to fully charge new battery packs During this time a Battery Charging message will be displayed on the control p
54. that future other errors that occur during the same I O are reported back to the host A SMART event does not affect the I O except that it does count as one try of the I O 128 0x80 Access Time Y Ignore None The frequency distribution Count Error table used to locate recently modified blocks was found to be inconsistent This problem is self correcting 129 0x81 Disk Drive Y Single Disk Drive The controller failed a disk Placed In Failed Occurrence Controller drive due to its accumulated State Internal SCSI 1101 history Replace the Bus failed disk drive 130 0x82 Image Y Ignore None This event would occur if the Out Of Date Operator disk drives were moved from activity one disk array to another and then reinstalled in the original array The result is that the controller will detect a different stamp on the disks 157 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex Event Name Event Logged Predictive Maintenance Implication Suspected Components Description 131 0x83 No Quorum For Y Ignore None During startup fewer than half Disk Set Access Operator of the original disk drives in activity or the disk set were present To accompanying recover the data reinstall the errors missing disk drives The disk array can be reformatted with the remaining disk drives but this will result in all data being lost 132
55. the disk array busy queue An excessive number may reflect address conflicts in host I Os caused by repeated access to the same block s of data host queue problem here typically reflects another ottleneck further along in the controller data path For example if the cache queue is full commands will stack up in the host queue waiting for the cache queue to empty The counter does not show until 32 Host Queue 0 532 Indicates the number of host commands waiting in the commands are in the queue eing transferred over the SCSI channel This metric eflects channel activity and does not indicate any Throughput Not Used Indicates the average number of bytes per second average number of bytes per second roblems with the disk array I O Rate Not Used Indicates the average number of I Os per second being serviced by the disk array This metric reflects channel activity it does not indicate problems with the disk Cache Queue 0 05 220 32 Indicates the number of commands in the cache I array EH 108 Concepts Management Optimizing Performance Rebuild Impact on Performance Because the rebuild process uses internal array resources it may decrease array performance while a rebuild is in progress This is particularly true if the host is issuing many I Os to the disk array The disk array uses a rebuild priority that allows you to determine which is more important system performance or data redundancy A high rebuil
56. the disk array uses the Active Hot Spare space as RAID 0 1 capacity which improves array performance An Active Hot Spare does not sit idle it is used to increase the disk array performance until it is needed A printed circuit assembly that translates data between the host processor s internal bus and a different bus such as SCSI See Fibre Channel Arbitrated Loop Most disk arrays require disk drives to be of matched capacities performance and often manufacturers A disk array is not restricted to use disks of the same capacity or even performance to function in the array A process of Auto Configuration permits the Disk Array Controller to recognize and include disks of varying capacity and performance This means that future disk drives of higher capacity and better performance may be supported for use in the disk array When two controllers are present one serves as the Primary Array Controller and the other becomes the Secondary Array Controller either controller fails the other controller has the ability to assume the role of Primary Array Controller thus allowing uninterrupted access to all user data with no downtime Meanwhile the failed controller can be removed then replaced by a new array controller without any downtime loss of data or interruption to the host computer system AutoRAID is a trademark of Hewlett Packard Company for the disk array product HP AutoRAID implements RAID technology automatical
57. the rack this is important to maintain cooling b Re install the side panels the rear rack door is not required for this installation it should not be installed and can be set aside c Connect the PDUs to the power source For redundant configurations the PDU cables must be connected to separate circuits or UPS for high availability power d Bring the array enclosures on line and mount the file systems 249 Sur e Racking Appendix D Back to Back Racking Installation Procedure NI 3NI1 069 0218 Figure 52 Power Cable Schematic 8120 6514 HP SureStore E Disk Array A3700AZ Location 4 1 P2 P3 FAN UNIT E7687A FAN 1 LINE IN LINE IN NI INN 8069 0218 8120 6514 8120 6514 8120 6514 HP SureStore E Disk Array A3700AZ Location 1 P1 P2 P3 20 E x 4 zi HP SureStore E Disk Array A3700AZ Location 5 8120 6514 8120 6514 8120 6514 8120 6514 8120 6514 8120 6514 HP SureStore E Disk Array A3700AZ Location 2 8120 6514 HP SureStore E Disk Array A3700AZ Location 6 P1 P2 P3 8120 6514 8120 6514 LINN NOLLOgIH LSIO H3MOd 8120 6514 7199 0218 8120 6514 HP SureStore E Disk Array A3700AZ Location 3 0199 0218 8120 6514 250 POWER DISTRIBUTION UNIT E7676A PDU 2
58. those users who desire to gain a better understanding of how the disk array hardware operates Disk Array Controller Module The disk array controller contains the intelligence and functionality required to manage the operation of the entire disk array The major responsibilities of the disk array controller include Implementing Hewlett Packard AutoRAID to ensure optimum performance and cost efficient data storage Managing all data transfers between the host and the disks Maintaining data integrity by automatically correcting any data errors that occur Rebuilding the array in the event of a disk failure Monitoring the operation of all hardware components including itself e Alerting the host in the event of a problem with the disk array The disk array is operated with both a primary array controller and a secondary array controller for controller hardware redundancy The use of two controllers allows the disk array to have controller redundancy which protects against a single controller failure In dual controller configurations both controllers may be active thus allowing a possible increase in disk array performance while simultaneously providing redundancy The contents of controller NVRAM is updated on both controllers simultaneously By maintaining a mirror image of all vital operating information the secondary array controller can take over immediately should a controller fail 93 s3do uo Concepts
59. time Install this previously removed original controller in the now empty controller slot Y Push the on off button to the on position and allow the array to complete its initialization process After the initialization has completed the front panel should say Not Enough Disks Check the system state using the key strokes below The system state should be Not Enough Disks with a Firmware Needed warning KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed 49 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 20 Verify if the primary controller currently installed has the desired firmware using the front panel display The following table shows the proper front panel commands to view which controller is the primary array controller and what version of firmware is on the primary array controller KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Warning Warning Firmware Needed Primary Cntrl X SCSI ID x Offline SCSI ID y Offline Firmware HP 21 Copy the firmware from the primary controller to the secondary controller using the front panel KEY CANCEL CANCEL MENU ENTER ENTER ENTER DISPLAY Not Enough Disks Language Cntrl Chang
60. when files are deleted because none of the data stored in RAID 5 has been deleted This situation represents a trade off between disk array performance and operating system file management Most operating systems have the ability to recover undelete or salvage files that have been deleted They can do this because the data remains on the disk media If the disk array really deleted files to regain RAID 0 1 space you would no longer be able to use these operating system recovery features 110 Concepts Management Optimizing Performance Deleting a Logical Drive to Increase RAID 0 1 Space Unlike deleting files from the operating system level deleting a logical drive does free up capacity for use as RAID 0 1 space To understand how file space is allocated you must differentiate between file system free space and free space or unallocated capacity within the disk array The following example should help make this distinction clear If you create a three Gigabyte logical drive on the disk array and then write three Gigabytes of data to that logical drive you have used three Gigabytes of space on the disk array If you then erase the data by deleting all the files you have written you now have no user data on the disk array but the logical drive is still occupying three Gigabytes of space Therefore even if you delete files from the file system updating the structures within the file system to indicate deletion i
61. 0 0xaa Internal SCSI Y See disk drive Disk Drive The controller successfully Reassign vendor s completed a sparing operation Completed reassign rate The reported FRU is the slot spec number 171 0xab Internal SCSI Y Ignore None A sparing operation was Reassign canceled typically due to a Canceled hog plug event on the bus The sparing operation will not be retried until another the bad block is accessed again The reported FRU is the slot number 172 0xac Error On Y Ignore unless Disk Drive The controller has determined Disk Media medium error that media for which a disk After Medium recurs drive had reported a Error MEDIUM ERROR is actually OK The data stored at the affected block may or may not have been recovered The reported FRU is the slot number 173 0xad Incompatible Y Ignore None The NVRAM image stored on RAM Image Operator Error the disk set when it was Revision shutdown is not compatible with the current version of controller firmware This is typically caused when upgrading controller firmware Returning to the previous firmware revision will correct this problem without data loss 169 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex 174 0xae 175 0xaf 176 0xb0 177 0xb1 Event Name Other Controller Interrupts Stuck On No Interrupts From Other Controller Internal SCSI Disconnect Disk Drive Downed Event Logged Pred
62. 0 Disk Drive Data Transfer Failed Maintenance Components Description The I O completed with status indicating the data transfer failed retries are exhausted or the I O is not retryable 33 0x21 Disk Drive Miscompare on Write And Verify Internal SCSI The read verify following a write indicated a Miscompare error status retries are exhausted or the I O is not retryable 34 0x22 Disk Drive State Conflict A disk drive was in the wrong state for read write or pass through For a read or write the wrong states are down missing or initialization failed For pass through the wrong state is missing Check the state of the disk drive and try to return it to ready by adding it if it was downed or reinstalled it if it is missing 35 0 23 Disk Drive Data Out Of Sync After Access accompanying Data being read or written to a disk drive was out of sync after the access was complete The redundant copy of the data stored on the disk drive is no longer valid A rebuild will ensure the data is valid 142 Event Event Name Number DEC hex Illegal Drive State For Down Command 36 0x24 Predictive Logged Maintenance Implication NA Suspected Components None Troubleshooting Event Code Descriptions Description An attempt to down a disk failed Downing the disk would have resulted in loss of redundancy or data unavailability
63. 0 m V Cable VHDCI VHDCT 68 pin high density Option 842 2 0 m V Cable VHDCI VHDCI Term 68 pin high density Option 843 2 0 m V Cable 68 pin HD VHDCI 68 pin high density Option 844 2 0 m V Cable 68 pin HD VHDCI I L Term 68 pin high density 11 uondrnsoq Description Product Description Options 12 Option 851 V Class 10 0 m SCSI Cable 68 pin high density Inline Term Cable P N A4801 63002 Option 871 V Class 2 0 m 5 0 m SCSI Y Cable V Inline Term Cable 68 pin P N A4801 63012 Option 873 V Class 2 0 m 3 0 m SCSI Y Cable V Inline Term Cable 68 pin P N A4801 63010 Option 875 V Class 5 0 m SCSI Cable 68 pin high density Inline Term Cable P N A4801 63004 Option B25 4 0 m V Cable VHDCI VHDCT 68 pin high density Option B26 4 0 m V Cable VHDCI VHDCI I L Term 68 pin high density Option B27 4 0 m V Cable 68 pin HD VHDCI 68 pin high density Option B28 4 0 m V Cable 68 pin HD VHDCI I L Term 68 pin high density Option ASJ Disk Array 12H Software Interface Kit for NT Product Description Available Accessories Available Accessories The following accessories are available A3701A Deskside Cabinet Upgrade Kit A3702A Single 4 3 Gigabyte Disk Module 7200 rpm A3703A Single 9 1 Gigabyte Disk Module 7200 rpm A3710A Single 18 2 Gigabyte Disk Module 7200 rpm A5289A Single 36 4 Gigabyte Disk Module 7200 rpm A3713A Single 9 1 Gigabyte Disk Module 10 000 rpm A3714A Single 18 2 Gigabyte Disk Module
64. 0x84 Quorum For Y Ignore None During startup fewer than half Disk Set Operator of the original disk drives in Attachment activity or the disk set were present To accompanying recover the data reinstall the errors missing disk drives The disk array can be reformatted with the remaining disk drives but this will result in all data being lost 133 0x85 No Quorum Due Y Ignore See None During startup with disk set To Broadcast accompanying related NVRAM invalid the Failure errors controller was unable to invalidate the NVRAM disk image on more than half of the members of the disk set The disk have disappeared or are not working 134 0x86 No Address Y Ignore None During startup the controller Table Operator was unable to find a disk set activity or with a valid shutdown image accompanying on any of the members errors present The data mapping 158 information has been lost and it will be necessary to reformat the disk array Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 135 0x87 Disk Drives Y Ignore None During startup no disk drives Found During Operator were present A new disk set Initialization activity was created with no members 136 0x88 Insufficient Y Ignore See None There is insufficient disk Space During accompanying space for migrating data Rebuild errors during a rebuild This is
65. 10 000 rpm A5292A Single 36 4 Gigabyte Disk Module 10 000 rpm A3702AM Four 4 3 Gigabyte Disk Modules 7200 rpm A3703AM Four 9 1 Gigabyte Disk Modules 7200 rpm A3710AM Four 18 2 Gigabyte Disk Modules 7200 rpm A5289AM Four 36 4 Gigabyte Disk Modules 7200 rpm A3713AM Four 9 1 Gigabyte Disk Modules 10 000 rpm A3714AM Four 18 2 Gigabyte Disk Modules 10 000 rpm A5292AM Four 36 4 Gigabyte Disk Modules 10 000 rpm A3706A Single 96 Megabyte Controller Module A3708A Power Module A3709B Fan Module A5329A Single Front Door Assembly Quartz Gray Color Optional A5329AM Ten Front Door Assemblies Quartz Gray Color Optional A5253A Disk Array 12H Software Interface Kit for NT C2905A Fast Wide SCSI terminator Ultra flexible 0 5 m SCSI Cable part number 5064 2408 Option 851 V Class 10 0 m SCSI Cable 68 pin high density Inline Term Cable part number A4801 63002 13 uondrbsoq Description Product Description Unpacking and Repackaging the Disk Array Option 875 V Class 5 0 m SCSI Cable 68 pin high density Inline Term Cable part number A4801 63004 Option 873 V Class 2 0 m 3 0 m SCSI Y Cable V Inline Term Cable 68 pin part number A4801 63010 Option 871 V Class 2 0 m 5 0 m SCSI Y Cable V Inline Term Cable 68 pin part number A4801 63012 Option 841 2 0 m V Cable VHDCI VHDCI 68 pin high density part number A3639 63015 Option 842 2 0 m V Cable VHDCI VHDCI I L Term 68 pin high density part number A3639
66. 119 proper installation of 105 replacement 115 slot numbers 113 status lights 113 Not In Use 100 NVRAM 94 options available 10 packaging 14 Parity 71 73 77 82 part numbers replaceable 205 PDU 234 performance 105 107 power off 32 on 32 power cords 20 234 power module 98 119 power on sequence 121 Previously Used 79 89 primary array controller 76 priority 90 R rackmount EIA 230 rackmount installation 229 RAID 83 RAID 0 1 83 RAID 5 83 RAM 94 rebuild 73 80 auto 90 impact on performance 109 interrupted 90 priority 90 recovery 94 removal ac power receptacle 201 backplane assembly 196 display module 194 front door assembly 192 module cam lever 203 power button 200 push rod 200 rear cover assembly 194 Renumber L Drv 81 replaceable parts 205 replaceable parts enclosure 207 replaceable parts modular 206 replacing modules 115 Reset Cntrl 76 reset switches Index SCSI bus 197 S safety specifications 221 SCSI bus 99 120 SCSI bus reset switches 197 SCSI cable 13 SCSI cabling 25 SCSI ID 73 76 SCSI settings 73 SDTR 71 73 77 82 secondary array controller 76 self test 70 121 sequence test 121 settings 82 shutdown 74 92 Shutdown 82 SIMMs 94 single controller 22 93 slot numbers 113 specifications 211 stamps 97 Start Rebuild 80 status light 70 status lights 70 113 Stop Rebuild 80 Switch Cntrl 76 switch cover 19
67. 13 MODUDESTATUS EIGHT ott te teda bI D EO P RIEN E a ee dee te 113 REPEACING MODYUDES 5 115 TROUBLESHOOTING PROC DURE O S aan aussi 119 Fan S Ill sl A e Aa ITEE essa nasa 119 119 Disk MOGUL Ecc AAA ect 120 ID NON EDUC 120 POWER ON SEQUENCE TEST siiis feret tetto ees prete e epe Ee pr ever PEE e en 121 CONTROL PANEL ERROR 55 8 2 2 2 0 0 01 1 00000000000000000000 0 123 133 EE Nee 134 Term Definitions tete tereti sae eae Wg kbs m Am naha that 12949 134 Predictive Dat Tesal uu n usu a SS Qu cA i 135 CHAPTER 6 PREVENTIVE MAINTENANCE 189 TOOLS RE OUR aa ANES RRt 189 REQUIRED PREVENTIVE 1200 01111000000000000000000 000000000 sees ette 189 LH M a 189 Gontroller Batteries c E
68. 15 soodg Specs Appendix A Product Specifications Module Specifications Module Specifications Disk Module e Size 1 6 inch 3 5 inch form factor e 1 6 Disk Dimensions 54 x 120 x 240 mm 2 1 x 4 7 x 9 5 in e Weight 1 6 kg 3 5 Ib e Type 4 3 Gigabyte 9 1 Gigabyte 18 2 Gigabyte or 36 4 Gigabyte SEW SCSI 2 e SCSI ID set by slot Controller Module Dimensions 290 x 58 x 240 mm 11 4 x 2 3 x 9 5 in e Weight 2 2 kg 4 9 Ib e AutoRAID with 96 Megabytes SCSIID 0 15 set by control panel Power Module Dimensions 120 x 100 x 230 mm 4 7 x 3 9 x 9 1 in e Weight 2 7 kg 6 0 Ib e Type autoranging power factor correcting active current sharing internal axial fan Fan Module e Dimensions 62 x 120 x 240 mm 2 4 x 4 7 x 9 5 in e Weight 0 45 kg 1 0 Ib radial blower 216 Appendix Regulatory Safety Statements This appendix contains the following information e FCC Statement For U S A Only e Statement Worldwide CSA Statement For Canada Only e VCCIStatement For Japan Only e BCIQ Class A Warning Statement For Taiwan Only e Statement For Korea Only e Sp cification ATI Classe A France seulement Product Noise Declaration For Germany Only e Ger uschemission For Germany Only e Electromagnetic Compatibility e Safety Specifications Declaration of Conformity FCC Statement For U S A Only The Federal Commu
69. 197 Removal Removal and Replacement Backplane Assembly HP qualified personnel only Table 15 SCSI Bus Reset Switch Functions Switch Number 1 2 3 4 SCSI Bus Bus Y Bus Y Bus X Bus X Switch Name Power Detect Controller Detect Power Detect Controller Detect SCSI Bus Default OFF ON OFF ON Switch Setting Disabled Enabled Disabled Enabled Switch Function When the Bus Y When the Bus Y IWhen the Bus X When the Bus X When Enabled Switch Function When Disabled Power Detect Switch is enabled the disk array controller generates a SCSI Bus Reset on Bus Y whenever power is applied or removed Power Detect Switch is disabled default the disk array controller will not generate a SCSI Bus Reset on Bus Y whenever power is applied or removed When the Bus Y Controller Detect Switch is enabled default the disk array controller Reset on Bus Y whenever disk array controller Y is inserted or removed When the Bus Y Controller Detect Switch is disabled the disk array controller will not generate a SCSI Bus Reset on Bus Y whenever disk array controller Y is inserted or removed Power Detect Switch Is enabled the disk array controller generates a SCSI Bus generates a SCSI BusReset on Bus X whenever power is applied or removed IWhen the Bus X Power Detect Switch 15 disabled default the disk array controller will not generate a SCSI Bus Reset on Bus
70. 2 WDTR 73 77 control panel display 69 control panel menu 75 controller dual 93 failure 191 single 93 controller module 22 93 Copy Firmware 79 Create L Drv 81 D data recovery map 79 94 data redundancy 100 Delete L Drv 81 deskside cabinet 223 dimensions 211 Disable Remote Reset 78 273 JUSSO Index Index disk cache 95 module 120 stamps 97 disk module 216 disks 95 door 192 door enclosure 233 downloading firmware 37 DRR 71 78 dual controllers 93 dynamic data migration 84 E EIA rackmount 230 enclosure door 233 environmental requirements 213 error messages 123 errors disk states 123 system states 123 Failed 89 failover 91 fan module 98 119 features 16 215 fibre channel 32 filler panels 233 Firmware Needed 37 format 74 Format 77 front door 192 front door snap 193 H hardware 93 hot spare See Active Hot Spare 89 I Include Disk 79 including a disk 72 Initialization Failed 89 274 installation batteries 34 controller modules 22 installing modules 20 interrupted rebuild 90 language 74 76 localization 74 76 Log Disk 79 logical drives LUNs 81 88 100 Logical Drv 71 M management 103 maps 94 Mark Log Disk 79 memory 94 menus 75 migration 84 mirroring 83 models available 9 module controller 93 disk 95 120 fan 98 119 fibre channel 32 installation 20 MUX 32 power 98
71. 3 sunoous qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 149 0x95 Logical Unit Y Ignore None The controller has Deleted operator or successfully deleted a LUN host activity The number of the deleted LUN appears in the last 8 bytes of the component identifier field of this log entry 150 0x96 Logical Unit Y Ignore host None The controller has Renumbered activity successfully renumbered a LUN The component identifier field of this log entry contains the original number of the LUN first 8 bytes and the new number of the LUN last 8 bytes 151 0x97 Disk Set Y Ignore None The controller has Attached operator successfully attached to a new activity disk set This occurs after every format disk array and whenever a power on or reset occurs while the controller is not attached to a disk set In some cases the currently attached disk set may be empty 152 0x98 Unable To Y Ignore See None The system experienced a Complete accompanying memory error and was unable Internal Data errors to resolve the error because Transfer DMAs did not complete The disk drive must be reset 164 Event Number DEC hex Event Name Event Predictive Logged Maintenance Implication Suspected Components Troubleshooting Event Code Descriptions Descripti
72. 3 switches SCSI bus reset 197 switching controllers 73 Synchronous Data Transfer Request 77 System Warning 23 T Term Power 71 73 78 82 terminator SCSI 13 tools 191 troubleshooting 119 U Unallocated 100 Unit Attn 71 73 78 82 Unsupported 89 275 xopu Index Index V VEB Very Early Busy 78 View Settings 71 82 276 W WDTR 71 73 77 82 weight 211 Wide Data Transfer Request 77 write working set 109
73. 870 VA 7 3 A 120 Vac 3 8 A 230 Vac e Voltage Range 88 140 Vac 180 269 Vac e Nominal Frequencies 60 Hz 120 Vac 50 Hz 230 Vac e Inclusive Frequency Range 47 to 66 Hz e Transparent Line Surge 293 Vac for 0 5 seconds e Recoverable Line Surge 298 Vac for 0 5 seconds e Line voltage sag 80 5 Vac for 0 5 seconds e Power Line Dropout Must not exceed 20 ms 214 Appendix A Product Specifications Enclosure Features Electromagnetic Susceptibility Operating Range Radiated 3V m 14 kHz to 1000 MHz Conducted 3 Vrms 30 kHz to 50 kHz 1 Vrms 50 kHz to 400 MHz Magnetic 47 5 Hz to 198 Hz at 4 Gauss external field Electrostatic Discharge 1 10K V operating 1 25 Nonoperating Magnetic interference Nonoperating 5 25 mGauss at 4 6 m 2 mGauss at 7 0 ft Magnetic Field Interference operating 5 Gauss Peak to Peak at the Surface of the product Enclosure Features Disk modules 4 3 Gigabyte 9 1 Gigabyte 18 2 Gigabyte or 36 4 Gigabyte SEW SCSI 2 Controller modules 1 or 2 second controller provides redundancy Power modules 2 or 3 third power module provides redundancy Fan modules 3 one redundant but all three must be installed for proper airflow Control panel 2 x 20 back lit LCD display with function keys Protected power standby switch Rackmountable enclosure mounts in EIA cabinets Airflow is from front bottom to top rear Hot pluggable modules with activity fault LEDs Front access 2
74. ALLATION ET DE MISE EN EXPLOITATION d un mat riel de traitement de l information ATI class en fonction des niveaux de perturbations radio lectriques mis d finis dans la norme europ enne EN 55022 concernant la Compatibilit Electromagn tique Cher Client Conform ment la R glementation Francaise en vigueur l installation ou le transfert d installation et l exploitation de cet appareil de classe A doivent faire l objet d une d claration en deux exemplaires simultan ment aupr s des services suivants Comit de Coordination des T l communications 20 avenue de S gur 75700 PARIS e Pr fecture du d partement du lieu d exploitation Le formulaire utiliser est disponible aupr s des pr fectures La d claration doit tre faite dans les 30 jours suivant la mise en exploitation Le non respect de cette obligation peut tre sanctionn par les peines pr vues au code des Postes et T l communications et celles indiqu es dans la loi du 31 mai 1993 susvis e Arr t du 27 Mars 1993 publi au J O du 28 Mars ATI 219 Ayayes A1078 NSIY Regulatory Safety Appendix B Regulatory Safety Statements Product Noise Declaration For Germany Only Product Noise Declaration For Germany Only Acoustic Noise Emissions LpA 45 0 dB seeking At bystander position per ISO 7779 data are the results from type tests of the product configuration having the highest acoustic emissions 12 disk modules
75. By Related Hot Plug Disk Drive Reset Itself Event Logged Predictive Maintenance Implication See information provided by drive vendor See information provided by disk drive vendor See information provided by drive vendor NA NA Suspected Components Disk Drive Disk Drive Disk Drive None None Description Data was recovered be the disk drive using its own internal recovery process The sense information provided by the disk drive should provide more information on the nature of the failure The disk drive returned a status indicating a media error condition Retries have been exhausted This event will only occur in degraded mode The disk drive returned a status indicating a hardware error condition The controller retries are exhausted or the I O is not retryable This event will only occur in degraded mode A disk drive sharing the same internal bus as the target drive was either installed or removed causing a bus reset The controller could not retry the I O either because it is not retryable RAID 5 or retries are exhausted The disk drive returned a status indicating it had been reset The controller could not retry this I O The host should retry the I O 140 Event Number DEC hex 27 0x1b Event Name Disk Drive Reset by Pass Through Event Logged Predictive Maintenance Implication Ignore None Suspected Components Troub
76. CSI 68 pin to 68 pin Cable HBA to Y2 3 SCSI Passive Differential Terminator X2 2 SCSI 68 pin to 68 pin Cable Y1 to X1 27 Product Description Connecting SCSI Cabling Figure 10 SCSI Cabling for 1 Disk Array 2 Controllers 2 SCSI Buses 1 SCSI 68 pin to 68 pin Cable HBAI to X2 3 SCSI 68 pin to 68 pin Cable HBA2 to Y2 2 SCSI Passive Differential Terminator X1 4 SCSI Passive Differential Terminator Y1 28 Product Description Connecting SCSI Cabling Figure 11 SCSI Cabling for 2 Disk Arrays 2 Controllers 1 SCSI Bus 1 SCSI 68 pin to 68 pin Cable HBA to X2 3 SCSI Passive Differential Terminator X1 2 SCSI 68 pin to 68 pin Cable X1 to X2 29 Product Description Connecting SCSI Cabling Figure 12 SCSI Cabling for 2 Disk Arrays 4 Controllers 1 SCSI Bus 1 SCSI 68 pin to 68 pin Cable HBA to Y2 4 SCSI 68 pin to 68 pin Cable Y1 to X1 2 SCSI 68 pin to 68 pin Cable to 5 SCSI Passive Differential Terminator X2 3 SCSI 68 pin to 68 pin Cable X2 to Y2 30 Product Description Connecting SCSI Cabling Figure 13 SCSI Cabling for 2 Disk Arrays 4 Controllers 2 SCSI Buses uondrnsoq SCSI 68 pin to 68 pin Cable HBAI to X2 4 SCSI Passive Differential Terminator X1 2 SCSI 68 pin to 68 pin Cable HBA2 to Y2 5 SCSI 68 pin to 68 pin Cable Y1 to Y2 3 SCS
77. Control Panel Buttons Deleting a Logical Drive CAUTION Deleting a logical drive LUN destroys all data on the logical drive Before deleting a logical drive make backup copies of the files you want to save or move the files to another logical drive 1 Select the Logical Drv Changes from the main menu 2 Select Delete L Drv 3 Select a drive number 0 to 7 4 Select Confirm Including a Disk Select the Disk Changes from the main menu 2 Select Include Disk 3 Selecta disk number A1 to A6 or B1 to B6 4 Select Confirm Enabling Disabling Auto Include Select the Disk Changes from the main menu 2 Select Auto Include 3 Select Enable or Disable 4 Select Confirm Enabling Disabling Auto Rebuild 1 Select the Disk Changes from the main menu 2 Select Auto Rebuild 3 Select Enable or Disable 4 Select Confirm Starting Stopping a Rebuild 1 Select the Disk Changes from the main menu 2 Select Start Rebuild or Stop Rebuild 72 3 Operating the Control Panel Using the Control Panel Buttons Select Confirm 4 Use the View Settings menu to check the rebuild progress Setting the Controller SCSI ID 1 Qv COME 19 Select the Cntrl Changes from the main menu Select SCSI ID Select a controller Cntrl X or Cntrl Y Select SCSI ID 0 to 15 Offline takes disk array offline Select Reset Select Confirm Changing Other Controller SCSI Settings
78. E er ee 169 CHAPTER 7 REMOVAL AND REPLACEMENITAI 191 IREOUIRED2 255 n et etus belit tentiae 191 REPLACING A FAILED CONTROLLER MODULE HP QUALIFIED PERSONNEL ONLY eene 191 FRONT DOOR ASSEMBLY HP QUALIFIED PERSONNEL ONLY 192 DISPLAY MODULE HP QUALIFIED PERSONNEL 2 212 2 1 2220 04 0000000000000000000000000 194 BACKPLANE ASSEMBLY HP QUALIFIED PERSONNEL ONLY enne rn nnn 196 SCSI Bus Reset Switches HP qualified personnel only eerte renes 197 POWER BUTTON AND PUSH ROD HP QUALIFIED PERSONNEL ONLY 200 AC POWER RECEPTACLE HP QUALIFIED PERSONNEL ONLY 201 MODULE CAM LEVER HP QUALIFIED PERSONNEL ONLY eene eene 203 REPLACEABLE PART NUMBERS uz aa nawanpas Eze ER EE ee eee E ee Pro 205 APPENDIX A PRODUCT SPECIFICATIONS 211 ENCLOSURE PHYSICAL SPECIFICATIONS cccccccccsesssaececececsessaueeecececeesenseaececececsenseaececeesesessaaeeeeeceeseneaees 211 Weights Ure ete d dw 211 Produet Dimension s usu ha EI OE a EE
79. EL MENU ENTER ENTER ENTER DISPLAY Not Enough Disks Language Cntrl Changes SCSI ID Copy Firmware Confirm Copying Firmware then Firmware Loading then array will initialize then Offline SCSI ID 12 The front panel display should say Not Enough Disks Check the System State on the front panel display It should say Not Enough Disks with no other warnings 48 KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Primary Cntrl X HP SureStore E Disk Array 12H Controller Firmware Download Procedure 13 Verify that both controllers have the desired firmware using the front panel The following table shows the proper front panel commands to view the firmware revision installed 14 15 16 17 18 19 KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Primary Cntrl X SCSIIDx SCSIIDy Firmware HP Push on off button to off position and wait for the array to turn off Remove the third controller that was obtained is step 1 It should be in controller slot X Move the controller that is in slot Y into controller slot X Take the previously removed original controller and reset disconnect and then re connect its batteries both at the same
80. Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 164 0xa4 Microprocessor Y More than 1 Controller A problem occurred which Software Fault in 1 month caused the microprocessor to vector to the software fault interrupt service routine The controller always resets following this log entry This error is caused by illegal instruction fetches divide by Zero or other illegal processor activity 165 0xa5 Microprocessor Y More than 1 Controller A problem occurred which Hardware Fault in 1 month caused the microprocessor to vector to the hardware fault interrupt service routine The controller always resets following this log entry This error occurs when the microprocessor attempts an illegal interrupt vector 166 0xa6 Internal SCSI Y Single Disk Drive This is an error detected by the Event Occurrence not Controller interface driver during a bus with other Internal SCSI scan or an error that cannot be entries Bus linked to a particular command It is normal for this event to accompany hot plug events If it occurs independent of any hot plug events it may indicate a hardware failure 167 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Number DEC hex Event Logged Predictive Maintenance Implication Suspected Components Description 167 0xa7 Insuff
81. HP SureStore E Disk Array 12H User s and Service Manual LA ooa with AutoRAID Technology HP Part Number C5445 90901 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 Company 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 Company Print History Edition 1 May 1997 Edition 2 April 1998 Edition 3 September 1998 Edition 4 February 1999 Edition 5 June 1999 Edition 6 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 practices which if not observed could result in personal injury Trade
82. I 68 pin to 68 pin Cable X1 to X2 6 SCSI Passive Differential Terminator Y 1 31 Description Product Description Fibre Channel FC SCSI Multiplexer MUX Connection Fibre Channel FC SCSI Multiplexer MUX Connection Fibre channel host connection is available through a Fibre Channel FC SCSI Multiplexer MUX Configurations of the FC are detailed in the HP 9000 Enterprise Servers Configuration Guide For information regarding the Fibre Channel technology refer to the Fibre Channel SCSI Multiplexer Service and User Manual part number A3308 90005 Switching Power On To turn on the disk array power 1 Lift the power standby switch door up or open the front door see Figure 2 Item 1 2 Press the power standby switch to the power in position see Figure 4 Item 4 3 During and after all the disk modules have spun up and passed self test the control panel will display power on sequence test numbers Once the power on sequence is finished the control panel will display the word Ready NOTE f the controller batteries are not fully charged you may need to wait a few minutes for the Ready message to be displayed Switching Power Off WARNING When the power standby switch is in the standby position the dc power output circuits are off within the power supplies but the ac power input circuits are still active To switch off the disk array power 1 Lift the power standby switch door up or open th
83. In most cases the LUNs on the array need to be either un mounted or deactivated from the volume group Does this disk array have a single controller No Dual Controller Does the replacement controller have the same Yes Single Controller 5 Perform procedure Replace D Replacing a Single Controller replace single Yes firmware that is in the disk array No Is the disk array a Replacing controller and firmware Perform procedure Replace A matches replace matching Yes boot device No Y Perform procedure Replace B Replacing controller and firmware does NOT match replace non boot Perform procedure Replace C Replacing controller on boot device and firmware does NOT match replace boot 41 J9 0 ju0 Controller HP SureStore E Disk Array 12H Controller Firmware Download Procedure Links to Downloading Firmware Download Windows NT Download A Downloading Disk Array Firmware from a File Using ARMServer AutoRAID Manager Download B Downloading Disk Array Firmware from a Third Controller Using the Front Panel Download C Downloading Disk Array Firmware from a File Using Off Line Diagnostics ARDIAG Download D Downloading Disk Array Firmware from a File on a V Class system using dfdutil Utility Links to Replacing an Array Controller Replace A Replacing an Array Control
84. Loss backed RAM loss completed complete with at least partial success Maps were recovered Multiple failures may or may not have occurred Occurrence of multiple failures are reported in the log between the RECOV STARTED and RECOV DONE events 185 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex 227 0xe3 186 Event Name Redundancy corrected Event Logged Predictive Maintenance Implication Suspect data if followed by error code 93 Ox5d Suspected Components Disk Drive Back End SCSI Bus Controller Description This error code indicates that the parity scan which executes during recovery from RAM loss found an instance of incorrect redundant data Data which was being updated when RAM was lost may produce this error in single failure scenarios This error can also result from multiple failure conditions Firmware differentiates the two scenarios by placing a limit on the number of these conditions which can occur before recovery is terminated due to multiple component failures If this error is followed by error 93 0 54 then the contents of data blocks reported in this error should be considered suspect The block address of the error appears in the first 4 bytes of the controller serial number field of this log record The LUN appears in the fifth byte of the controller serial number and the length of the affected area i
85. MENU Language View Settings ENTER System State Ready Primary Cntrl X Assumption 2 Since there is only one controller in the array it is very likely that when the controller failed the NVRAM memory maps were destroyed as well If the maps were destroyed or the procedure failed above there are only two other options 1 Tryto restore maps using arrayrecover command See the man page for details on command usage NOTE The arrayrecover utility is only available on ARMServer patch released IPR 9810 or newer 15699 for 10 x and 15700 for 11 0 or newer Also the array controller firmware must be running version HP4x or newer 2 Restore array configuration LUN and or LVM plus data from backup storage medium 67 J9 0 ju0 Chapter 3 Operating the Control Panel This chapter explains how to operate the control panel on the front of your disk array The control panel enables you to perform many array configuration and maintenance tasks The control panel display can be seen through the front door but the door must be open in order to access the control panel function keys Using the Control Panel Figure 17 shows the control panel display and function keys Figure 17 Control Panel Display and Function Keys CANCEL MENU 1 Available Choice 7 Select Menu Items 2 Selection History 8 Scroll Text Right 3 S
86. MITY according to ISO IEC Guide 22 and EN 45014 Manufacturer Name Manufacturer Address Declares that the product Product Name Product Number s Product Options Hewlett Packard Company Enterprise Storage Solutions Division P O Box 15 Boise Idaho U S A 83707 HP SureStore E Disk Array 12H A3700A AY A370XA AY 71 X 0 9 Y A Z conforms to the following Product Specifications Safety EMC Supplementary Information 950 1991 A1 A2 A3 EN 60950 1992 Al A2 IEC 825 1 1993 EN 60825 1 1994 Class 1 Laser LED CISPR 22 1993 EN 55022 1994 Class A EN 50082 1 1992 Generic Immunity including IEC 801 2 1991 4 CD 8 AD IEC 801 3 1984 3 V m IEC 801 4 1988 1 kV Power Lines 0 5 kV Data Lines IEC 1000 3 2 1995 EN 61000 3 2 1995 Harmonic Current IEC 1000 3 3 1994 EN 61000 3 3 1995 Voltage Fluctuation and Flicker STSB GB9254 1988 China The product herewith complies with the requirements of the Low Voltage Directive 73 23 EEC and the EMC Directive 89 336 EEC and carries the CE Marking accordingly 1 The product was tested with a Hewlett Packard NetServer host computer system fia er ys t BOISE IDAHO U S A 09 30 98 Dan T Michaud QA Manager European Contact Your local Hewlett Packard Sales and Service Office or Hewlett Packard GmbH Department Herrenberger Stra e 130 71034 B blingen
87. NG THE DISK ARRAY INTO THE DESKSIDE CABINET n nennen 226 DESKSIDE CABINET CABLING ee reed bia Pret ee ue ecu 228 RACKMOUNT GABINETS eee eee e e ee petet n 229 Installing the Disk Array into Rack 230 APPENDIX D BACK TO BACK RACKING 235 PRODUCT DESCRIPTION rr REP VERRE VE PERENNE CERERI 235 uiti ee it aap desta gu espe eye tege tee E PER CH EO eA E e ie Eee pee Re d 236 238 238 Power R dunddhty a tu o e ase RR REI ua ERR RYAN NES E 238 NS TAEDATION2 eset tumet sauku mtis Ie ded 239 Installation Strategy i iiie tue aiu iie tert iet oe ret epe tr I C tegere 239 Brief Summary of Installation Steps sees esee eee ener nennen rene enne 241 INSTALLATION PROCEDURB a ss us usaspa si Su ase E nasse Th Baad 242 ae u za RECS ERE d nus 244 TROUBLESHOOTING 3 toot piste DDR pite PE vp nus 252 Indic tor LEDs uos pi Rt et te iem 253 Indicator EED Stale S n ececet ee pee use ee eee ee ine ye bee e bee eve e
88. SI ID Copy Firmware Confirm Copying Firmware then Firmware Loading then array will initialize then Not Enough Disks 13 The front panel display should say Not Enough Disks Check the System State on the front panel display It should say Not Enough Disks with no other warnings KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Primary Cntrl X 65 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 14 15 16 17 18 19 20 21 66 Verify that both controllers have the desired firmware using the front panel The following table shows the proper front panel commands to view the firmware revision installed KEY DISPLAY CANCEL CANCEL Not Enough Disks MENU Language View Settings ENTER System State Not Enough Disks Primary Cntrl X SCSI ID x SCSIID y Firmware HP Push on off button to off position and wait for the power to turn off Remove both controllers from slot X and Y and reset disconnect and then re connect both batteries on each controller at the same time Re install both controllers It doesn t matter which controller goes into which slot at this time Plug in all drive modules Push on off button to the on position and wait for the initialization to co
89. Sense N NA None No sense data available See the SCSI Command Specification for more information 1 0x01 Event Log Full N NA None The controller event log is full The log can be cleared using the SCSI Access Log command 2 0x02 Controller Log N NA None The controller log is full The Full log can be cleared using the SCSI Access Log command 3 0x03 Parameter List N NA None A SCSI parameter list length Length Error error occurred See the SCSI Command Specification for more information 4 0x04 Invalid Field In N NA None A SCSI invalid field in Parameter List parameter list error occurred See the SCSI Command Specification for more information 5 0x05 No Physical N NA None A command tried to address a Device Present disk drive that doesn t exist Check the physical configuration of the disk array to see what disk drives are installed 136 Event Event Name Number DEC hex 6 0x06 Invalid Opcode Logged Maintenance Components Troubleshooting Event Code Descriptions Description A SCSI invalid opcode error occurred See the SCSI Command Specification for more information NOTE In certain disk array states this event may occur even if the opcode is valid These are typically states that indicate a larger problem with the disk array 7 0x07 Invalid Bit In CDB Invalid bit in CDB See the SCSI Command Specification for more information 8 0x08 Logical Block Address Out Of Rang
90. States A fan in slot F1 F2 or F3 failed Check the fan status lights to determine which Fan Failed fan module failed then replace the fan module or CAUTION Do not remove a fan module Fan Missing until a replacement fan is available Also replace only one fan at a time Removing two fans at one time shuts down the power module System States The controller event fault logs are full Use a SCSI command to clear the logs Log Full You may want to read the logs first System States A disk that is required is no longer present Find the missing disk to recover data Missing Disk System States NVRAM errors have occurred Check the control panel display for more NVRAM Failed information regarding what to replace System States A power module has failed Replace the indicated power module Power Supply Failed System States The RAM is not being reflected in both Determine what the other error condition is by RAM controllers due to some other error condition the front panel Unmirrored System States The rebuild process terminated abnormally Restart the rebuild or add more capacity Rebuild Failed System States A Hot Pluggable operation has occurred None Information only Verifying Disks System States warning condition exists on the disk array Check the front panel for appropriate action to Warning identify and correct the problem 125 Sunoouso qno Troubleshooting Troubleshooting Control
91. ULTIPLEXER MUX 22 20000 0000 32 SWIICHING POWER ON triviis e D RR TR GREEN RED TREO 32 SWITCHING POWER OFF ssssccccccecsssssccecececesesssececececsessnseaecececsensseaesececeeseseaesecscseseaaaecececeesensaaeeeeeceeees 32 CHAPTER 2 HP SURESTORE E DISK ARRAY 12H CONTROLLER 33 0 OPERE EE AEAESEU UE 33 INSTALLING CONTROLLER BATTERIES c cccccccccsssssscecececsessuuecececececsesaaeceeececsesnsaeceeececsensaaeceeeceeeessaaseeeess 34 REPLACING CONTROLLER BATTERIES 0202 2 222 0 608010000000000000200000000000000000000 nass seen setate nnns nen 36 DOWNLOADING ARRAY CONTROLLER 2 2 42 2 00000 0000000000000000000 0 37 Before Beginning Firmware Downlodd 37 Other Considerations About the Firmware Download Process eee een 38 FIRMWARE DOWNLOAD PROCEDURE 2 0 22 2 20 2 2601000000000000000000000002200000050000000000000 39 Links to Downloading Firmware iae ese eret eei tee ers ee Pa Ee eias 42 Links to Replacing an Array Controller eese nennen eene trennen 42 CHAPTER 3 OPERATING THE CONTROL PANELA 69
92. ailed the I O The Retry recommendatio controller retried the I O and n the disk drive returned a status indicating the I O was successful 152 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 86 0x56 Not Ready N Ignore None The disk array is not ready Manual and the action required to Intervention solve the problem will Required require user intervention Check system state information for more details 87 0x57 Data loss Y NA None An otherwise successful read detected op failed due to data loss recorded in at least 1 of the disk blocks read 88 0 58 Format N NA None The subsystem could not be Precluded due formatted because logical to LUN units were created and still in Existence existence Delete all LUNs prior to issuing the format subsystem command 89 0x59 Can t N NA None Download is not possible even download though the target controller is because NV is in NO CODE because the NVRAM image is not valid on the disks i e the subsystem is not shutdown Make the secondary controller the primary controller have it shutdown and then do the download not shutdown 153 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 90 0 Host R
93. ample the disk image may support 8 GB disk drives but the current configuration does not To use the existing disk image the current controller map configuration must be changed to match the disk map configuration 190 0xbe Internal SCSI Y Single Disk Drive A disk drive sent received less Underrun Occurrence not Controller data than expected Event accompanied Internal SCSI by other errors Bus 173 Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 191 0xbf Internal SCSI Y Single Disk Drive A disk drive sent received Overrun Event Occurrence not Controller more data than expected accompanied Internal SCSI by other errors Bus 192 0xc0 Internal SCSI Y More than 2 Disk Drive A disk drive didn t respond Timeout Event Controller within a specified time limit per hot plug Internal SCSI The disk drive may be failing event Bus or it may be busy performing data retries 193 0xc1 Controller ECC Y Single SIMM The system experienced a Miscorrection Occurrence Controller correctable error which was most likely miscorrected 194 0xc2 Controller Data Y Single Controller The controller experienced a Bus Occurrence microprocessor access error Microprocessor without the needed other Access Error qualifiers Microprocessor access errors should only occur in conjunction
94. an module until a replacement fan is available Also replace only one fan at a time Removing two fans at one time shuts down the power module Firmware Loading EPROMs are being reprogrammed None Information only Firmware Needed Format Failed The controllers installed in the disk array have different firmware revisions new controller may have been installed that has a later version of firmware The firmware revisions must be the same on both array controllers SCSI command to format failed Perform ONE of the following procedures 1 Copy firmware from the primary to the secondary array controller or 2 Update firmware on both controllers For both of these procedures refer to the Downloading Firmware section in this document and also to the ARM Download utility procedure in the Array Utility Manual The array is malfunctioning Call the support line Incompatible The two controllers installed have different Do not use controllers from different vendors CNTLR personality profiles Unique controller in the same disk array enclosure Remove the personalities are created for different vendors incompatible controller and replace it with a These personalities are incompatible with one controller from the correct vendor Call the another support line Incompatible The controller s installed in the disk array are Remove the incompatible controller s and Format the wr
95. anel 34 HP SureStore E Disk Array 12H Controller Installing Controller Batteries Figure 14 Installing the Controller Batteries 35 J9 0 ju0 HP SureStore E Disk Array 12H Controller Replacing Controller Batteries Replacing Controller Batteries This procedure is different than installing batteries the first time the disk array is used since replacing batteries assumes that you have valuable data stored on your disk array Many factors affect battery life including not only time of battery use but also time of battery storage Controller batteries should be replaced every three years or sooner if the display module indicates a constant Battery Discharged message To replace controller batteries do the following see Figure 14 CAUTION If the disk array power is on you should always perform a successful Shutdown using the control panel to avoid loss of data before replacing a controller battery If a message appears on the control panel display indicating that the Shutdown did not complete successfully the disk array WILL LOSE DATA if you remove the batteries In this case you should perform the following steps 1 Install a second functional controller 2 Wait about two minutes for the data to be posted from the controller with the bad batteries to the second controller The Ready message appears at this time 3
96. are exhausted or the I O is not retryable 151 Troubleshooting Troubleshooting Event Code Descriptions Event Name Event Number DEC hex Event Logged Predictive Maintenance Implication Suspected Components Description 80 0 50 Disk Drive Y Single Disk Drive The disk drive returned bad Failed Start Unit Occurrence status for a start unit Command command The controller s retries are exhausted 81 0 51 Unexpected Y Single Disk Drive The disk drive returned a Disk Drive Occurrence Controller status inconsistent with the Status possible Internal SCSI current operation The compatibility Bus controller s retries are problem exhausted or the I O is not retryable 82 0x52 Target Storage Y Ignore None The requested compare and Unchanged By swap command was valid but Compare And did not change the contents of Swap target storage 83 0x53 Internal Data Y Single Controller A transfer failed due to a Transfer Occurrence Disk Drive timeout Typically results Timeout unless with Internal SCSI because the disk drive took other errors Bus too long to complete the command 84 0x54 Internal SCSI N Single Controller A data transfer failed This Bus Data Occurrence Disk Drive error is typically accompanied Transfer Error unless with Internal SCSI other errors that overwrite other errors Bus it 85 0 55 Recovered Y See drive Disk Drive On a previous attempt s the Disk Drive On vendor disk drive f
97. ata on the array including any meta data e g LUN and LVM configuration After you have downgraded the firmware and formatted the array all meta data and real data will need to be restored from backup Assumptions Use this procedure if the desired version of firmware is available from the host system and you want to update the firmware on the installed array controllers 1 Get array id using ARMServer arraydsp i or front panel View Settings Array S N You may also use the raw device file that is connected to the array 2 Obtain the desired version of firmware and determine its lt codefilename gt filename and location path 3 Quiet quiesce or stop all I Os deactivate volume set unmount file systems stop traffic to the array 4 Shutdown the array using ARMServer arraymgr s shut array id or front panel Shutdown The following table shows the proper front panel commands to shutdown the array KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings Shutdown ENTER Confirm ENTER Shutting Down then Shutdown Complete 5 Download the firmware download C codefilename array id 44 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 6 Verify updated firmware revisions on both controllers using ARMServer arraydsp c array id gt or front panel The following table shows the proper front panel commands to verify the firmware on the arr
98. ate negotiation protocol if disabled the host will initiate negotiation protocol In either case the disk array will always respond to any requests made by the host WDTR NOTE This command requires a SCSI Reset or power cycle to take effect Wide Data Transfer Request WDTR controls whether or not the additional eight bits on a wide bus will be utilized in most data phases of a SCSI command WDTR also determines the negotiation protocol of the host If WDTR is enabled default the disk array will initiate for negotiation for wide transfer If WDTR is disabled the host will initiate negotiation protocol In either case the disk will always respond to any requests made by the host 77 pued 10002 Control Panel Operating the Control Panel Using Control Panel Menus Term Power NOTE This command requires a SCSI Reset or power cycle to take effect Termination Power Enables default or disables SCSI Term Power SCSI Terminator Power If Term Power is enabled the disk array will supply power to the SCSI terminator connected to the array and to the SCSI termination on the HBA Unit Attn NOTE This command requires a SCSI Reset or power cycle to take effect Unit Attention Enables default or disables SCSI Unit Attn Unit Attention for the primary array controller If Unit Attn is enabled and a command failed with Unit Attention status the controller reports to the host that a Reset has occurred since the last
99. ay KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready Primary _ SCSIIDx SCSIIDy _ Firmware HP 7 If ARMServer issues a switch controller command you may have to repeat step 5 download NOTE some older versions of ARMServer the download command had a problem recognizing the primary array controller This happens when the primary array controller SCSI ID is a higher number than the secondary array controller SCSI ID If this problem occurs the download command will issue a switch controller command and then stop When this happens issuing the download command again causes the download to complete properly 8 The disk array will display the Ready state when the download command has completed 45 J9 0 9u0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Download B Downloading Disk Array Firmware from a Third Controller Using the Front Panel This is not the preferred method Use this procedure only if the disk array firmware is NOT available from the host CAUTION Controller firmware HP4x and any later versions have a different data map format in NVRAM memory Once you have HP4x or later you can not use this procedure to downgrade to any firmware version prior to HP4x NOTE This procedure is not recommended if you have a single controller installed in your array
100. be on an installed controller on a replacement controller or from a file Make sure you know which controller contains the desired version of firmware If the desired version of firmware is not on any controller but is in a file make sure you know the location path to the file and its filename 37 J9 0 ju0 HP SureStore E Disk Array 12H Controller Downloading Array Controller Firmware Other Considerations About the Firmware Download Process NOTE For Windows NT Systems Since the Download utility does not support long directory names it is better to change to the directory where the firmware is located first and then the codefilename is simply the filename of the firmware Otherwise a path with a directory named Program Files will have to be entered as its MS DOS directory name equivalent for example download C c Progra 1 AutoRAID Fw Model_12H_HP26 fw 000000057D22 e Ifthe array does not show up using ARMServer arraydsp i you may have to re connect to see the array using the ARMServer arraydsp R command which scans the bus for all arrays This command can take a few seconds to execute Once re connected you will see the array again 38 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Firmware Download Procedure See the flowcharts Figure 15 and Figure 16 to determine which download procedure to follow Pick the appropriate procedure NOTE
101. bles and terminators are installed correctly 146 Event Number DEC hex Event Name Predictive Maintenance Implication Suspected Components Troubleshooting Event Code Descriptions Description 53 0x35 Host Ignore Host None Commands were attempted to Overlapped Protocol be overlapped from the host Commands Incompatibilit This is usually an incorrect Attempted y initiator connection error and usually reflects a SCSI protocol violation on the part of the host 54 0x36 Unsupported Ignore Host None An illegal unsupported SCSI Host SCSI Protocol message was received This Message Incompatibility usually reflects a SCSI protocol violation on the part of the host 55 0x37 Host SCSI Ignore See None The initiator detected an error Initiator host error condition usually parity All Detected Error report retries have been exhausted Check host error reports for more information 56 0 38 Host Invalid Ignore Host None An invalid SCSI message was Message compatibility received This usually reflects problem SCSI protocol violation on the part of the host 57 0x39 Host Single Host SCSI The initiator did not respond Reselection Occurrence Bus to reselection attempts Failure Ignore if Host retries have been exhausted Controller timeouts 58 0x3a Host Data Single Host SCSI The initiator timed out in a Phase Error Occurrence Bus Host data phase Controller 147 Sunoouso qno
102. byte disks Active Hot Spare is enabled for maximum data protection This explanation describes how information is logically distributed Physically the information is spread evenly throughout all the disks Figure 25 Capacity Allocation 2 GBYTE DISKS din Z UPPER 1 GBYTE PARITY 1 GBYTE DISKS SEGMENT ACTIVE 1 GB i 2 XX UE HOT s LowER SPARE 0 5 GB 1 GBYTE 2 GB 1GB PARITY SEGMENT 2 98 OVERND 1GB Starting with 12 Gigabytes of disk storage the capacity is distributed as follows e Gigabytes for Active Hot Spare When using Active Hot Spare the disk array always reserves enough capacity to rebuild the largest disk which in this example is 2 Gigabytes e Gigabytes for parity This is the capacity used to support the data protection provided by RAID 5 When using disks of different sizes the disk array divides the capacity into segments To ensure adequate protection for all the data parity must be maintained for each segment In this example 1 Gigabyte of parity is required for the upper segment and 1 Gigabyte for the lower segment e Gigabytes for RAID 0 1 and controller overhead This is the capacity used by the disk array to support dynamic data migration and maintain performance This calculation is approximate and will vary as more disks are added or removed
103. cated capacity available If the disk is reinstalled before the rebuild is complete the array controller will recognize the missing disk and stop the rebuild It is not necessary for the rebuild to continue because any data on the disk that has not changed in its absence is still valid However data that gets written while the disk is missing must be rebuilt so the rebuild operation might not stop The rebuild must continue if there were any write I Os while the disk was missing from the array 90 Concepts and Management Disk Array Features Auto Failover To provide maximum hardware redundancy a second array controller protects the disk array against a single array controller failure If the first array controller fails data continues to be available by the use of the second array controller This capability is called Auto Failover NOTE To have uninterrupted access to data during Auto Failover your host disk array driver must support failover capability multiple paths to the same device The contents of the controller NVRAM is updated on both controllers simultaneously By maintaining a mirror image of all vital operating parameters the secondary array controller can take over immediately in the event of a failure with no disruption to the host should the primary array controller fail Failure of the first array controller can be detected by the host operating system or by the second array controller which continually monitor
104. channel A multiplexer enables SCSI devices to interface directly with hosts that have Fibre Channel technology The MUX contains one or two FC ports and up to four fast wide SCSI differential ports Taking a SCSI adapter offline means terminating inputs outputs and suspending all transactions going from the MUX to the specified SCSI devices The SCSI adapter is no longer active or available for access Taking the MUX offline indicates that all SCSI and FC adapters in the MUX are offline For a SCSI adapter online indicates the SCSI adapter is active and available for access and input output processing For the MUX online indicates that at least one adapter in the MUX is active and available for access A disk that has been used in another disk array The Disk Stamp indicates that this disk was once part of another array Including it will destroy any data on the disk In disk arrays with two controllers the Primary Array Controller is the one that manages the disk array hardware devices such as the fan modules and the control panel display module RAID 0 1 RAID 5 Read Verify Rebuild Rebuild Priority Recommended Operating Range Redundancy Glossary RAID stands for Redundant Array of Independent Disks The disk array implements this technology to connect several disk drives to one Disk Array Controller Several different forms of RAID implementations have been defined The RAID implementations sup
105. ck 2786 has a total height of 32 EIA units and the 2 meter rack C2787A has a total height of 41 EIA units One EIA unit is equal to 1 75 inches or 44 45 mm The disk array enclosure has a height of 13 EIA units 235 SuD e Racking Appendix D Back to Back Racking Product Description In addition to the power upgrade and exhaust fan kits mentioned above additional rail kits power cords and SCSI cabling should be identified and ordered as required see Table 19 Components The following kits and components are used for back to back racking To more accurately determine the components you need for your installation review the Installation Procedure below Table 19 Upgrade Kits and Components Item Part Number Exhaust Fan Assembly includes 3 fans LED fault circuit and 1 E7687A EIA unit air scoop 240 V AC Power Distribution Units PDUs 10 IEC 320 receptacles 4915 Unterminated line cord See Table 20 for specific option Computer rail kits as needed C2788A SCSI cables as needed See Table 17 Replaceable Parts Power cables as needed Disk array to PDU 8120 6514 PDU to AC power supplied with 4915A Upgrade Kit Opt ABA E7803 60001 Opt ABB 8120 6895 Opt 024 8120 6961 Rail Kits C2788A Filler panels as needed 1 EIA space 40101A 2 EIA spaces 40102A 3 EIA spaces 40103A 4 spaces 40104A 5 EIA spaces 40105A 6 EIA spaces 40106A 7 EIA spaces 40107A 236 A
106. command was executed DRR NOTE This command requires a SCSI Reset or power cycle to take effect Disable Remote Reset When DRR is ON or enabled default it prevents a host SCSI reset from resetting both disk array SCSI buses With DRR set to ON disk array controllers are not allowed to reset their own SCSI bus even if the host resets that controller If DRR is set to OFF a host SCSI reset will also allow that controller to reset both disk array SCSI buses This parameter controls the SCSI bus reset behavior or the remote other disk array controller when a SCSI reset results in both disk array controllers processing a reset request These include an ordinary reset signal a bus device reset or a reset subsystem command When DRR is set to ON the remote controller will not assert the SCSI reset signal to indicate that the commands were cleared Hosts that cannot tolerate target bus resets should have DRR set to ON When DRR is OFF the remote controller will assert the bus reset signal to indicate that all outstanding requests were cleared in response to the reset VEB NOTE This command requires a SCSI Reset or power cycle to take effect Very Early Busy VEB instructs the disk array to return a BUSY response during its power on sequence If VEB is enabled the disk array will return a BUSY status to the host rather than simply ignore any commands during the power on sequence This will alert the host system to the presence of th
107. compromised LED Indicators Front panel indicator LEDs show normal operation and fan failures Power failures are indicated by no LEDs being lit Power Redundancy The disk arrays in the rack can be configured for power redundancy In redundant mode if power to one of the rack s PDUs fails the other two PDUs will continue to power all disk array units To provide power redundancy each PDU must be connected to a separate dependable or UPS power source 238 Appendix D Back to Back Racking Installation Installation WARNING Back to back rack mounting of the HP AutoRAID Disk Array should be performed by factory trained personnel only Customers should not attempt to perform this procedure Installation Strategy The following general information and strategy applies to both the 1 6 meter and 2 0 meter racks To determine detailed components required for installation refer to the Installation Procedures section later in this appendix 1 Upgrade the rack s power The HP 4915 SureStore E Disk Array 12H Power Upgrade kit is supported only for 240 VAC power systems This upgrade provides three 10 receptacle 240 VAC PDUs PDUs for non US installations must be ordered with the appropriate plugs refer to the Table 20 below Table 20 A4915A Power Upgrade Kit Power Options PDU Description Use unterminated A4915A 200 240 VAC North American PDU US installations PDUs Opt ABA 10 IEC 320 receptacles line cord i
108. croll Text Left 9 Control Panel Status Light 4 Cancel Selection 10 Controller Indicator x or y 5 Display Main Menu 11 Menu Item Scroll Indicator 6 Scroll Menus Items 69 pued 10002 Control Panel Operating the Control Panel Control Panel Status Light Operation Control Panel Status Light Operation When first powered on the disk array controller performs an extensive self test At first a memory test is performed During the memory test the control panel status light will remain off for about one minute Then a build process occurs During the build process the control panel status light is amber for about twenty seconds At this time the power on self test is finished and now the control panel status light flashes green with input output I O activity If the power on self test passed successfully the indicator will flash green with I O activity If any part of the disk array does not pass self test or if a problem occurs the control panel status light will be amber solid or flashing Table 1 Control Panel Status Light Indications Color Indication power disk array not available or initializing R Selecting a Menu Item When using the control panel follow these steps 1 Press MENU to select the main menu Press or to scroll through to a menu Press ENTER to select a menu Press or to scroll through to a menu item Press ENTER to select a menu item B
109. d Note 1 English Language menu shown Note 2 Only if two controllers installed Note 3 Only if Manufacturer Log Disk is installed L Drv States 75 pued 10002 Control Panel Operating the Control Panel Using Control Panel Menus Language The Language menu enables you to set the control panel language to one of five languages After the language is set the control panel will display all messages in the selected language Deutsch Sets the control panel language to German English Sets the control panel language to English Espanol Sets the control panel language to Spanish Francais Sets the control panel language to French Italiano Sets the control panel language to Italian Changes The Cntrl Changes menu enables you to change SCSI controller parameters for the array SCSI ID NOTE If two controllers are installed and are connected to the same SCSI bus they must each have unique SCSI IDs Changes the SCSI ID SCSI address of controller x or controller y to any address from 0 to 15 or offline and resets the controller Switch Cntrl For arrays with two controllers Switch Cntrl manually makes the secondary array controller become primary If the primary array controller fails during operation of the array the array automatically enables the secondary array controller to become primary Reset Cntrl NOTE During a controller reset all disk module lights turn amber for th
110. d priority ensures the rebuild will be completed as quickly as possible but at the expense of disk array performance A low rebuild priority maintains disk array performance but delays completion of the rebuild leaving your data vulnerable for a longer period of time Write Working Set A key factor in maintaining disk array performance is the balance between the RAID 0 1 space and the write working set The write working set is the number of unique data blocks written to the disk array over a period of time The disk array monitors the number of write I Os to create the write working set size parameter As long as the amount of RAID 0 1 space available is equal to or greater than the write working set size maximum performance is maintained On newly installed disk arrays there is enough RAID 0 1 space for the write working set and performance is not a problem However as more data is stored on the disk array and the amount of RAID 5 space increases the RAID 0 1 space available for the write working set decreases If a point is reached where the write working set exceeds the size of the available RAID 0 1 space some writes will have to be serviced from RAID 5 and then there may be a performance degradation To maintain performance the disk array reserves at least 1096 of the array capacity as RAID 0 1 for the write working set This value was chosen after careful analysis of many diverse operating environments and application loads It makes the
111. d to the Scrub policy by a device read It is logged to provide information about the cause of the error Additional information about the error from the device should precede this entry 187 Chapter 6 Preventive Maintenance This chapter provides instructions for preventive maintenance of the disk array Tools Required No tools are required for preventive maintenance of the disk array Required Preventive Maintenance Airways Monthly or more frequently if operating in a dusty environment clean any dust accumulation from the front door and module grills Clean any dust accumulation from the rear power module fan grill Controller Batteries Controller batteries should be replaced every three years or sooner if the display module indicates See Replacing Controller Batteries 189 Chapter 7 Removal and Replacement This chapter provides detailed instructions for enclosure parts removal and replacement This chapter also describes and identifies part numbers WARNING The procedures in this chapter should only be performed by a qualified service representative If you are not a qualified service representative performing these procedures may result in personal injury or loss of data CAUTION Check with system administrator before powering down the system Tools Required e T 15 TORX driver e T 20 TORX driver e T 25 TORX driver e Standard flat bladed screwdriver e Smal
112. d with an optimum cache The disk array contains SIMMs with Error Correction Code ECC which are capable of both error detection and correction on the Disk Array Controller Most RAID systems use SIMMs without ECC A Channel refers to a SCSI bus on the Disk Array Controller Each Disk Array Controller is connected to one channel See Auto Configuration and Self Configuring Data Redundancy protects the disk array from lost data if a disk fails With Data Redundancy the array can reconstruct the data that was on the failed disk If a disk fails Data Redundancy is lost until the array rebuilds the data that was on the failed disk Data Redundancy Capacity Data Transfer Rate Disk Array Controller Disk Module Disks Not in Use Disk Stamp Downing A Disk DRR Dynamic Data Migration Glossary Data Redundancy Capacity is the capacity required to support the RAID 0 1 and RAID 5 storage techniques used by the disk array for Data Redundancy This capacity is managed by the Disk Array Controller and cannot be altered or reduced The Data Transfer Rate is the speed at which data is transferred between a host computer system and a peripheral The Disk Array Controller implements the HP AutoRAID Array Technology and Balancing of the data The Disk Array Controller manages all data transfers to and from the host computer and also to and from each Disk Module The disk array enclosure holds up to twelve Disk Modules Each Dis
113. dding its capacity to the RAID 0 1 space You can continue to use the new disk to increase performance or you can make the capacity of the new disk available to the operating system by creating a new logical drive which is also done online without disrupting array operation Auto Include is performed only for new disks which are operating normally and are ready to be used Such disks typically are assigned a state of Normal when they are installed If the disk array determines that the new disk has one of the following states the disk will not be included automatically e Previously Used to protect any data that may be on the disk you must manually add a disk with this state e Unsupported the disk has not been tested and certified for operation in the disk array Only supported disks are guaranteed to work properly in the disk array You can include an unsupported disk manually but there is no assurance that it will operate properly in the disk array e Failed the disk is not functional and cannot be included even manually e Initialization Failed the disk failed the initialization test performed by the controller and cannot be included even manually Active Hot Spare The Active Hot Spare feature of the disk array offers increased protection against disk failure Active Hot Spare used in conjunction with Auto Rebuild ensures that the disk array can restore data redundancy and performance as quickly as possible following a dis
114. drive modules 9 Install both controllers Put the controller with the desired firmware into controller slot X 10 Push the on off button to the on position and allow the array to complete its initialization process After the initialization has completed the front panel should say Not Enough Disks Check the system state using the key strokes below The System state should be Not Enough Disks with a Firmware Needed warning KEY CANCEL CANCEL MENU ENTER 64 DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed HP SureStore E Disk Array 12H Controller Firmware Download Procedure 11 Verify if the primary controller currently installed has the desired firmware using the front panel display The following table shows the proper front panel commands to view which controller is the primary array controller and what version of firmware is on the primary array controller KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed Primary Cntrl X SCSI ID x _ SCSI ID y __ Firmware HP__ 12 Copy the firmware from the primary controller to the secondary controller using the front panel KEY CANCEL CANCEL MENU ENTER ENTER ENTER DISPLAY Not Enough Disks Language Cntrl Changes SC
115. e The specified logical block address was out of range See the SCSI Command Specification for more information 9 0x09 Not Ready Not Ready Init command required 10 0x0a Invalid Identify Received Invalid Identify Received See the SCSI Command Specification for more information 11 0x0b Power On Unit Attention Power on unit attention See the SCSI Command Specification for more information 137 Sunoouso qno Troubleshooting Event Code Descriptions Event Event Name Event Number DEC hex Predictive Suspected Logged Maintenance Components Implication Description Troubleshooting 12 0x0c Hardware Error N NA None Power on or self test failure Unit Attention Check system state and warning states for more information 13 0 0 Commands N NA None Commands cleared by another Cleared Unit initiator See the SCSI Attention Command Specification for more information 14 0x0e Mode N NA None Mode parameters changed by Parameters other initiator See the SCSI Changed Unit Command Specification for Attention more information 15 0x0f Unconfigured N NA None A command tried to address a LUN LUN that hasn t been created 16 0x10 Invalid Target N NA None A command tried to perform LUN an illegal operation on the specified LUN For example trying to create a LUN that already exists 17 0x11 Invalid Source N NA None LUN A command tried to renumber a LUN that d
116. e array Data Redundancy capacity required to support the RAID 0 1 and RAID 5 storage techniques used by the disk array for data redundancy This capacity is managed by the array controller and cannot be altered or reduced This is the only portion of the array capacity that you cannot control e Active Hot Spare capacity reserved to perform a rebuild if a disk fails The disk array creates an Active Hot Spare large enough to rebuild the largest disk in the array Until needed the array uses the Active Hot Spare capacity as RAID 0 1 space to improve performance The capacity for the Active Hot Spare is distributed across all the disks in the array Unallocated capacity that has not been allocated to a logical drive or the Active Hot Spare The array uses unallocated capacity as RAID 0 1 storage to improve system performance Should you need to increase capacity you can use the unallocated capacity to create a new logical drive e Disks Not In Use capacity contained on any disk that is installed in the array enclosure but is not included in the array configuration Such disks are not being currently used by the disk array to store any user data 100 Concepts and Management Managing the Disk Array Capacity Sample Capacity Allocation The following example illustrates how the disk array uses capacity in a typical configuration In this example there are eight disks installed in the array four 2 Gigabyte disks and four 1 Giga
117. e disk array and thus avoid the possibility of the host system timing out before the disk array is ready 78 Operating the Control Panel Using Control Panel Menus Mark Log Disk Mark Log Disk Marks the log disk with information to record the starting time and stopping time of a benchmark test The log disk is a specially formatted disk that records disk activity IOs with the host it cannot store user data Copy Firmware Copy Firmware will copy controller firmware from the primary array controller to the secondary array controller This command will take the disk array offline Copy Firmware requires that the SCSI bus be quiescent quieted Be sure to unmount all LUNs deactivate any Volume Groups and make certain that any data based activity that uses LUNs on the disk array has achieved quiescence before using this command 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 No 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 Disk Changes
118. e e ee iare dece 254 PROBLEM RESOLUTION 255 NO EEDS liE E E 255 One b nk of LEDS not lit eee ee Pe ge et redes 256 One amber LED liie etae etie S pce 256 Two mber LEDS lit a a e eher Nee 256 REMOVAL AND 20 6 svbseraves tedeuesvetebcatssvesstervantevessesedeedies 257 SPECIFICATIONS 5 E A T A EEr ATIVE 258 Electrical dui an ede a u edicere eee esee ediscere ci e cbe erede 258 Exhaust fan assembly eed tnde na Eee eee eee Un detis deeds ene 258 A COHST Deed aae 258 FREQUENTLY ASKED QUESTIONS cscssssscecececeesesscaececcececsessnaecececeeseseaecececeesensaaeceeceeesessaaecesececeeneaees 259 GLOSSARY qe 261 INDEX ER 273 Chapter 1 Product Description This chapter describes the disk array models and configurations available This chapter also shows how to install disk modules controller modules power modules and fan modules into your disk array In addition this chapter shows how to connect the SCSI cabling to your disk array Models and Configurations Available The disk array is an AutoRAID storage device that holds up to twelve disk modules in a single enclos
119. e following e Push on off button to the off position Remove both controllers from slot X and Y and reset disconnect and then re connect both batteries at the same time on each controller Re install both controllers It doesn t matter which controller goes into which slot at this time e Plug in all drive modules e Push on off button to the on position and wait for the initialization to complete 24 The front panel display should say Ready Check the System State on the front panel display It should say Ready KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready 25 If the array does not show up using ARMServer arraydsp i you may have to re connect to see the array using the ARMServer arraydsp R command which scans the bus for all arrays This command can take a few seconds to execute Once re connected you will see the array again 51 J9 0 ju0 Controller HP SureStore E Disk Array 12H Controller Firmware Download Procedure Download C Downloading Disk Array Firmware from a File Using Off Line Diagnostics ARDIAG Assumptions Use this procedure if the desired version of firmware is available on a tape or other such media that can boot with the offline diagnostic environment ODE Do not use this procedure if the system is on a V class computer CAUTION Controller firmware HP4x and any later versions have a different data map format
120. e front door Figure 2 Item 1 2 Press the power standby switch to the standby out position Figure 4 Item 4 CAUTION When the disk array power is turned off the control panel will display a Shutting Down message Shutdown is not complete until ALL lights are OFF and the display panel shows the Shutdown Complete message 32 Chapter 2 HP SureStore E Disk Array 12H Controller This chapter explains how to install batteries download firmware into a disk array controller and how to replace an array controller About Battery Packs If your disk array goes without power the controller batteries are capable of retaining data in NVRAM for a minimum of one week Always perform a successful Shutdown before replacing the batteries For further protection battery packs should be replaced one at a time so that there is always at least one battery providing power for memory backup If the disk array power is to be turned off for an extended period of time more than five days then the NVRAM memory maps should be saved to disk by successful Shutdown and both battery packs should be disconnected to prevent damage to the battery packs and the controller 33 J9 0 ju0 Controller HP SureStore E Disk Array 12H Controller Installing Controller Batteries Installing Controller Batteries Two battery packs part number 1420 0532 must be installed in each controller module before the module can be used Two battery packs are
121. e incompatible with one controller from the correct vendor Call the another support line Mismatched The two controllers installed are from different Do not use controllers from different product CNTLR product families families different vendors in the same disk array enclosure Call the support line No Address No data maps are available for the controller If your disk array is a boot device you may Table module data was lost due to missing disk have to recover the maps by using the front modules or no NVRAM image to match disk panel command called Recover under the modules Cntrl Changes menu The arrayrecover utility is described in detail in the System Administrator s Guide which is shipped with each disk array No Code The firmware download was unsuccessful there Attempt downloading firmware upgrade again is insufficient processor firmware to operate the Call the support line controller No L Drv The SCSI command to Create Logical Drive The array diagnostics may have been running Created failed when someone accessed the control panel or the array is malfunctioning Call the support line No Quorum Less than half of the disk modules in the disk set are present Install missing disk modules or format the array Caution Formatting the array requires deleting logical drives LUNs and deleting LUNS erases all data No Resources A front panel request is in contention with user
122. e module lever to lock the module in place Figure 6 Installing a Disk Module 21 uondrnbsoq Description Product Description Controller Module Controller Module NOTE If two or more controller modules are connected daisy chained to the same SCSI bus whether in the same disk array enclosure or across two or more disk array enclosures each controller module on the same SCSI bus must have its own unique SCSI address which is set by the control panel on the front of the disk array refer to Operating the Control Panel To install a controller module CAUTION Before installing a controller module ensure all of the pins on the controller connector are straight Bent pins will cause the controller module to fail and the array will not function properly Also do not attempt to install the controller module upside down Pullthe controller module levers all the way out see Figure 7 View 1 2 Insert the controller module into the disk array while pushing on both sides of the controller module see Figure 7 View 2 3 When the enclosure engages the levers press the center of the levers with the palm of your hand until they lock securely into place You should hear an audible click when the levers lock see Figure 7 View 3 NOTE When using a single controller the controller should be installed in slot x Although either slot can be assigned to be the primary array controller slot
123. e number assigned by the controller to each disk drive installed in the array This number is used by the controller for its internal management of the disks and is in no way related to the slot position of the disk drive or to the disk mechanism s serial number The disk drive ID number is frequently used in the component identifier field of disk system change events to indicate the disk drive involved Internal SCSI Bus This refers to all of the components that comprise the SCSI channel between the disk array controller and the disk drive This includes the disk array controller the enclosure backplane and the disk module In general the term back end refers to the communication path between the disk array controller and the disk drives Host SCSI Bus This refers to all of the components that comprise the SCSI channel between the disk array and the host computer This includes the disk array controller the enclosure backplane external SCSI bus cabling and components internal to the host computer for example the host SCSI adapter In general the term front end refers to the communication path between the disk array controller and the host SIMM A controller NVRAM memory module Troubleshooting Event Code Descriptions Predictive Data Terms This column indicates if the event code is useful to predictive diagnostics If it is then this column contains the occurrence rate which should trigger predictive maintenance The foll
124. e phase e Current 16 A maximum per PDU Exhaust fan assembly e Number of fans 3 e Fan capacity 233 per fan Acoustic e Sound pressure 7 2 Bels in idle mode at 25 C ambient 258 Appendix D Back to Back Racking Frequently asked Questions Frequently asked Questions What site preparation is required Power There must be a source of 240 VAC power for each of the rack s two PDUs Each PDU draws a maximum of 16A European locations may require upgraded circuit breakers as described in the previous section Redundant power configurations require redundant power sources that is power to the rack s two PDUs must come from separate circuits Location The rack location must allow sufficient clearance for access to array enclosures and access to service Normally about 2 feet 0 61 meters clearance between the rack s back and any wall is sufficient Does the installation change the height of the rack Yes Installing the Exhaust Fan Assembly increases the height of both the 1 6 meter rack and the 2 0 meter rack by 1 EIA unit 1 75 or 44 45 mm Do I have to use filler panels Yes If the rack is not fully populated with array enclosures you must install enough filler panels to close the remaining open spaces in the front and back of the rack This is to optimize cooling efficiency What if have an uninterruptible power supply UPS at my site The indicator LEDs are unlikely to report power outage conditions They w
125. e ratio of the write working set to the amount of RAID 0 1 space available To maintain erformance the amount of RAID 0 1 space should equal or exceed the write working set a value of 1 or less A value greater than 1 indicates that the working set is larger than the available RAID 0 1 space In this case the disk array must service writes from RAID 5 space which degrades performance For more information see Write Working Set mr how much data is being moved between Indicates how efficiently the write cache on the disk efficiently the write cache on the disk array controller is being utilized If the write cache is consistently full too many commands begin stacking up in the cache queue resulting in decreased erformance 107 sjdo uo Concepts Concepts and Management Optimizing Performance Disk Array Values Performance Metric Typical Limit Definition Disk Queue 2 4 gt 10 per disk Indicates the average number of transactions on all disk drives waiting in the internal disk SCSI queues many commands in the queues decrease erformance Concurrency gt drive count Indicates the number of internal simultaneous disk lt drive count the disk array is making The disk array achieves optimum performance when it is accessing multiple disks at the same time so some concurrency is desirable Too low a value indicates that the host is not requesting enough concurrent I Os to keep
126. e rebuild process before it completes 80 Operating the Control Panel Using Control Panel Menus Logical Drv Changes The Logical Drv Changes menu enables you to change logical drive parameters for the array A logical drive is a portion of the array capacity that appears to the operating system as a physical disk Each logical drive corresponds to a SCSI Logical Unit LUN zero through seven However the data contained in each logical drive is distributed across all disks in the array so there is no correlation between a logical drive and a single disk in the array The entire array capacity can be divided into up to eight logical drives Only capacity that has been assigned to a logical drive is available to the operating system Unassigned capacity is used to increase the array performance or may be used during a rebuild operation Create L Drv Creates a logical drive SCSI LUN number 0 to 7 with a capacity size of 100 Megabytes to a maximum of the total remaining capacity available in the array Logical drives may be created any time disk space is available or becomes available Creating a logical drive makes the specified disk capacity available to the operating system After creating a logical drive you must configure the logical drive into your operating system to use the capacity CAUTION Deleting a logical drive LUN destroys all data on the logical drive Before deleting a logical drive make backup copies of the files you
127. e some of the decisions involved in managing the disk array Included is a very brief description of some of the basic concepts of disk array technology for those unfamiliar with disk arrays If you would like to know more about disk arrays many good articles have been published on the subject Disk Arrays in Brief A disk array uses a group or array of disk drives connected to an array controller which distributes or stripes data across the disks in the array Disk array technology offers the potential to provide three important benefits high availability high performance and cost efficient data storage Perhaps the most important feature of disk arrays is their ability to provide high availability or protection against disk failure This protection is achieved through the use of redundant information which is used to reconstruct data that is lost when a disk fails Disk arrays typically offer a variety of techniques for managing redundant information These techniques are referred to as RAID levels The RAID level used has a direct impact on the remaining two factors performance and cost efficient data storage Although there are a number of different RAID levels the only two we are concerned with here are RAID 0 1 and RAID 5 e RAID 0 1 gives you data redundancy and good performance but the performance is achieved by using a less efficient technique of storing redundant data called mirroring Mirroring maintains a bac
128. e voltage and current These power cords have two conductors and a ground Failure to use the proper power cord may result in a shock or fire hazard Installing Modules This chapter shows how to install modules into the disk array NOTE To improve enclosure cooling and balance the load on the internal SCSI buses the disk modules should be installed from left to right and from top to bottom Disk module SCSI addresses are automatically preset by the backplane inside the disk array enclosure upon insertion 20 Product Description Installing Modules Disk Power and Fan Modules Use the same installation procedure to install power modules and fan modules See Figure 5 to determine where to install the module Figure 6 shows how to install a disk module To install a disk power or fan module open the front door and follow these steps CAUTION The module lever must be pulled all the way out or the module will jam in the slot upon insertion Also always handle the disk modules carefully when they are out of the disk array enclosure since they are susceptible to shock and vibration When removing a disk module pull it only part of the way out of the enclosure first then wait 30 seconds for the disk spindle to spin down before fully removing the disk module This will avoid damaging gyroscopic effects between the disk surfaces and heads Pullout the module lever 2 Insert the module into the disk array 3 Pushin th
129. ead Y Ignore see None A host write was flushed from Collision With accompanyin cache but ended with a non Failed Cache g errors SUCCESS status this caused Post cache to mark the write cache entry as RETRY which means it is stuck in cache A new host read partially hits the same address which normally causes the write to flush but the write is stuck in cache The response is to fail the read 91 0x5b Ambiguous Y Ignore None A volume set ID referenced in Volume Set a recover command was Reference ambiguous If a null wild card volume set was used in a recover command then this error means that the controller found drives multiple volume sets present If a non null volume set was used then this error means that the controller is already attached to a volume set other than the one referenced in the recover command 92 0x5c Recovery Y Single Disk Drive This error indicates that the From RAM Occurrence Back End controller was unable to Loss Failed SCSI Bus recover an NVRAM image troll any disk in the volume set 2 This could be because of the disks present have images or the disks that have images were removed or failed before the could be fully uploaded into the controller 154 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 93 0xsd Multiple Y Ignore see None Conditions detected during Failu
130. ed every three years or sooner if the display module indicates a constant Battery Discharged message Typically a Shutdown is performed when the disk array is turned off which copies the contents of the NVRAM to each of the physical disks in the array If power to the array is lost before a proper Shutdown can be performed the contents of the NVRAM must be maintained by the battery backup system until power is restored When fully charged the batteries are capable of sustaining the contents of the NVRAM for a minimum of one week Disk Modules The disk modules also referred to as disks provide the storage medium for the disk array A minimum of four disk modules is required to operate the disk array Different capacity disk modules can be installed in the same array A new disk can be added at any time even while the disk array is operating The array controller will recognize that a new disk has been added and will include the disk in the array configuration automatically if the Auto Include feature is enabled To make the additional capacity available to the host a new logical drive must be created and the resulting drive must be configured into the operating system 95 sjdo uo Concepts Concepts and Management Disk Array Hardware Using Disks of Different Capacities Disks of different capacities can be installed in the same array providing maximum flexibility and expandability for your disk array This
131. eded existing logical drive and leave its capacity unallocated This too will permanently increase the amount of RAID 0 1 space available for the write working set Move the data to another device Add a disk and create a new logical drive with its capacity This is a temporary way of increasing the amount of available RAID 0 1 space As the new logical drive begins to fill up with data it will be converted to RAID 5 space and you may again find that the available 10 RAID 0 1 minimum is too small to accommodate the write working set Delete a logical drive and recreate it This has the same temporary effect as adding a new disk and creating a logical drive Why Deleting Files Won t Increase RAID 0 1 Space It might appear that simply deleting some unneeded or archived files from the disk array would increase the amount of RAID 0 1 space available for the write working set However this is not the case due to the way the disk array and most other disk systems handles deleted files When you delete files from the disk array the operating system updates its file system to record the deletion From the operating system s perspective there is now more disk capacity available However the disk array itself is unaware of the deletion and keeps the data files intact From the disk array s perspective the amount of stored data has not changed Consequently a disk array that has reached the 10 RAID 0 1 minimum cannot increase the RAID 0 1 space
132. eee Qu eara ee W Q n te ee qu eee coe a che EP E ge a SS 216 Power tiim IEEE 216 Fan Module niece aeta ee aan u eios eise 216 APPENDIX B REGULATORY SAFETY STATEMENITS 217 FCC STATEMENT FOR U S A 217 SPECIFICATION ATI CLASSE FRANCE SEULEMENT 219 PRODUCT NOISE DECLARATION FOR GERMANY ONLY entere nne innen 220 GERAUSCHEMISSION FOR GERMANY 1 222 12 220 ELECTROMAGNETIC COMPATIBILITY nennen enne t nnnnr sienne te 221 SAFETY SPECIFICATIONS nete nnns sess estesa E s estesa a aS 221 DECLARATION OF CONFORMITY eren ennt nnn asse enne tre setate aane sese setate anas nans seen 222 APPENDIX C CABINET CONFIGURATION S 223 CABINET CONFIGURATIONS ette than nente tena annees 223 DESKSIDE GABINET rs rene DU SR E E be SEINE EI ERE E ES PEE Ge EE EIE EY AERE cae duas 223 UNPACKING THE DESKSIDE CABINET ar enn rnnn asses tette rn nass seen 224 INSTALLI
133. en lost for any part of the disk set The disk array must be reformatted in this case 179 0xb3 Insufficient Y Ignore See None There is insufficient disk Optimal Space accompanying space for rebuild to execute For Rebuild errors This event is most likely to occur when using disk drives of different capacities To create the necessary disk capacity add another disk equal in capacity to the largest disk in the array or delete a LUN 180 0xb4 Rebuild Y Ignore None A rebuild has been Restarted automatically restarted due to a disk failure or a disk removal 171 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 183 0xb7 Cancel Rebuild Y Ignore None A rebuild has been canceled Started Operator or either automatically or by Host activity command The disk array automatically cancels a rebuild if it needs to restart rebuild due to a disk addition a LUN deletion or a LUN format 184 0xb8 Rebuild IO Y Ignore Host None The host has changed the Priority activity rebuild priority Changed 185 0xb9 Non Member Y Ignore None The controller detected that a Drive Removed Operator disk drive was removed This activity event occurs for drives that are not disk set members when they are removed The device number field is not used 186 0xba Drive Missing Y I
134. ents that must be considered before connecting SCSI cables to the disk array 1 The external bus connected to the disk array must be a SCSI 2 wide differential device cable with 68 pin high density external SCSI connectors 2 Each SCSI bus must be terminated with a passive differential SCSI terminator on the last device Therefore for dual controller disk array configurations you will need two passive differential terminators one for each bus 3 Each disk array contains an equivalent of 0 5 meters of internal SCSI cabling per bus 4 total combined length of internal SCSI cabling plus external SCSI cabling must not exceed 25 meters Connecting SCSI Cabling NOTE Before connecting SCSI cables press the power standby switch to the standby out position It may take a moment for the disk array to fully shut down SCSI cables are available in standard lengths Each controller must have a different SCSI ID even when using dual controller configurations To change controller SCSI IDs refer to the chapter titled Operating the Control Panel 25 uondrosoq Product Description Connecting SCSI Cabling Figure 8 SCSI Cabling for 1 Disk Array 1 Controller 1 SCSI Bus 1 SCSI 68 pin to 68 pin Cable HBA to X2 2 SCSI Passive Differential Terminator X1 26 Product Description Connecting SCSI Cabling Figure 9 SCSI Cabling for 1 Disk Array 2 Controllers 1 SCSI Bus 1 S
135. equests made by the host The Write Working Set is the number of unique data blocks written to the disk array over a period of time To maintain performance the write working set should not consistently exceed the amount of RAID 0 1 space available 271 JUSSO Index A ac power cord 20 ac power receptacle 201 accessories available 13 power cords 14 active hot spare 89 Active Hot Spare 80 100 105 airflow 212 alias 75 array controller primary 76 secondary 76 arrayrecover 79 94 auto failover 91 auto include 72 89 Auto Include 79 auto rebuild 72 90 Auto Rebuild 80 AutoRAID Capacity Planning Tool 102 available options 10 B backplane assembly 196 batteries 34 95 189 cabinet deskside 223 rackmount 229 cabinet filler panels 233 cabling SCSI 25 cache 95 capacity 100 104 capacity allocation 101 capacity management 103 Capacity Planning Tool 102 configuration 223 control panel Active Hot Spare 80 Auto Include 72 79 Auto Rebuild 72 80 Copy Firmware 79 Create L Drv 81 Delete L Drv 81 DRR 78 Format 74 77 Include Disk 72 79 Language 74 Logical Drv 71 Mark Log Disk 79 Parity 77 Recover 79 Renumber L Drv 71 81 Reset Cntrl 76 SCSI ID 73 76 SDTR 73 77 Select Parity 73 Shutdown 74 82 Start Rebuild 73 80 Stop Rebuild 73 80 Switch Cntrl 73 76 Term Power 73 78 Unit Attn 73 78 VEB 78 View Settings 71 8
136. errors Disk Drive Controller Internal SCSI Bus In order to make a disk drive a member of a disk set it must have a stamp written onto it by the controller This error indicates that the writing or updating of the stamp failed during an Add Physical Drive command causing the command to fail This is typically caused by disk drive errors 145 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Components Number DEC hex 50 0x32 Disk Drive State Changed During Add Physical Drive Logged Maintenance Description During an Add Physical Drive command the disk drive is reconfigured for proper operation with the controller If the reconfiguration fails the first attempt but succeeds the second the controller simulates a hot plug event on the disk drive to determine the true status of the disk drive The add physical drive command fails if this condition is encountered The disk drive state should be reviewed by the host and the add command reattempted if appropriate 51 0x33 Host SCSI Controller Parity Error A SCSI DATA IN phase received a parity error which indicates a problem on the controller not the SCSI bus retries have been exhausted 52 0x34 Host SCSI Parity Error A parity error was detected on the Host SCSI bus retries have been exhausted This may be caused by an improperly configures SCSI bus Make sure all ca
137. es SCSI ID Copy Firmware Confirm Copying Firmware then Firmware Loading then array will initialize then Offline SCSI ID 22 The front panel display should say Not Enough Disks Check the System State on the front panel display It should say Not Enough Disks with no other warnings 50 KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Primary Cntrl X HP SureStore E Disk Array 12H Controller Firmware Download Procedure 23 You are now at a point in which the drives need to be re installed The drives were disconnected earlier to further protect the NVRAM The NVRAM is very important NVRAM stores all configuration and meta data that point or map drive location information to real data If this procedure has been successful up to this point the correct NVRAM memory maps are located in both controllers and also in the disk drive modules If the NVRAM is located in both the controllers and the disk drive module correctly there were no errors encountered during this procedure Then do the following e Push on off button to the off position e Plug in all drive modules e Push on off button to the on position and wait for the initialization to complete If you are unsure that the NVRAM is correct in both controllers there were errors especially errors like No Address Map or NVRAM errors Then do th
138. ets all requirements of the Canadian Interference Causing Equipment Regulations Cet appareil num rique de la classe A respecte toutes les exigences du R glement VCCI Statement For Japan Only COKE ERUREEGEXEEBIGSeBMe VCcCI Oo s 2 SA ATHE CT CORE CERI 2 C OBSI RAE AUR GERE SEIBRENSCEMBVES This equipment is in the Class A category information technology equipment based on the rules of Voluntary Control Council For Interference by Information Technology Equipment VCCI When used in a residential area radio interference may be caused In this case user may be required to take appropriate corrective actions Class A Warning Statement For Taiwan Only a BR AC HUI TER GER Emir HI AX PARTE gt ARBOR PRE aia SE 218 Appendix Regulatory Safety Statements Sp cification ATI Classe A France Seulement MIC Statement For Korea Only 2121 amp AML Hondas ve J 210I L BOA SE ASAE WAS 2 0 PY S9 PAUSAS Hg SHS 560 uso dle Li Ct User Guide Class A Please note that this equipment has been approved for business purposes with regard to electromagnetic interference If purchased in error for use in a residential area you may wish to exchange the equipment where you purchased it Sp cification ATI Classe A France Seulement DECLARATION D INST
139. event is logged for debugging purposes only and requires no action 148 Event Number DEC hex 65 0x41 Event Name Optimal Space Error In RAID 1 Event Logged Predictive Maintenance Implication Ignore Suspected Components None Troubleshooting Event Code Descriptions Description The controller detected that space allocated for an optimal only migrating write i e rebuild policy was no longer optimal after the write had completed This event is logged for debugging purposes only and requires no action 66 0x42 No Shutdown Image Posted NA None The controller could not shutdown the disk array because all the disk drives were missing Reinstall the missing disks to shutdown the array 67 0x43 Bad Disk Drive N NA None An Add Physical Drive State During command failed with the disk Add Physical returning a status of Drive Initialization Failed The drive is failing initialization so information on the nature of the failure should appear in the event log in a separate entry 68 0x44 Internal SCSI N NA None disk drive sent received less Underrun data than expected 69 0x45 Internal SCSI N NA None A disk drive sent received Overrun more data than expected 70 0x46 Internal SCSI N Morethan2 Disk Drive A disk drive didnt respond Timeout occurrences Controller within a specified time limit per hot plug Internal SCSI The disk drive may be failing event bus or i
140. f INo power self test or unterminated SCSI bus Solid Amber fault Replace controller module 114 Troubleshooting Replacing Modules Replacing Modules If any module has a fault you should acquire a replacement module before removing the failed module and install a replacement module Since the modules are hot pluggable they can be replaced even when the array is online with the host When replacing a fan module do not leave a fan slot empty for more than ten minutes or a power off pending error will occur When a fan slot is empty the disk array will not be cooled properly Without adequate cooling disk module failures may occur The control panel display module or backplane assembly should be replaced only by a qualified Field Service Engineer To maintain availability while replacing a power module always make sure that there are three power modules installed to provide redundancy then remove the failed power module Never remove a power module from a disk array with only two power modules while the disk array is functioning modules are replaced in the same way see Figure 27 through Figure 30 Pullout the module lever 2 Remove the module from the enclosure 3 Install the new module 115 Sunoouso qno Troubleshooting Troubleshooting Replacing Modules Figure 27 Removing and Replacing a Power Module Figure 28 Removing and Replacing a Disk Module 116
141. f the new disk modules available to the disk array you must install at least two of the higher capacity disk modules The disk array cannot make full use of the added capacity if only one higher capacity disk module is installed e Because the Rebuild process can impact disk array performance you may want to perform this operation during periods of low activity This will minimize the impact on users who may be accessing the disk array 104 Concepts Management Optimizing Performance Optimizing Performance Disk array performance is influenced by many factors some host related some related to the configuration of the disk array itself Host related disk array performance issues can typically be identified and corrected using operating system utilities or third party applications designed for this purpose The following list identifies some of the factors that directly influence disk array performance Other factors such as a rebuild can temporarily affect array performance Amount of available RAID 0 1 In most cases the more RAID 0 1 space available the better the disk array performance With enough RAID 0 1 space the disk array can service all write I Os from RAID 0 1 thus allowing an optimum write working set RAID 0 1 space can be increased by increasing the amount of unallocated capacity on the disk array Enabling Active Hot Spare With Active Hot Spare enabled the disk array can recover quickly from a disk failu
142. ferent data map format in NVRAM memory Once you have HP4x or later you not downgrade to firmware version prior to HP4x without first completely backing up all data on the array including any meta data e g LUN and LVM configuration After you have downgraded the firmware and formatted the array all meta data and real data will need to be restored from backup Note For information on dfdutil refer to the man page 1 Obtain the desired version of DFDUTIL VHPxx LIF xx the controller firmware revision and copy it the file spp data DFDUTIL LIF on the teststation You can get the LIF file from the web On the World Wide Web http www hp com go support click next on Enterprise Storage click next on HP Enterprise Disk Arrays click next on SureStore E Disk Arrays e on Disk Array 12H e click GO on Software Firmware beneath Software amp Drivers 56 HP SureStore E Disk Array 12H Controller Firmware Download Procedure or inside the HP firewall HP qualified personnel only http essd boi hp com click on Products click on Disk Arrays click on Disk Array 12H click on AutoRAID firmware left click on DFDUTIL VHPxx LIF xx the controller firmware revision you want to download save the file using save as to your system 2 Once you have the DFDUTIL_VHPxx LIF copy it to a tape
143. fications Enclosure Features Enclosure Physical Specifications Weight Weight Disk Array Maximum Configuration Net 77 2 kg 170 Ib Maximum Configuration Shipping 92 0 kg 202 5 Ib Maximum configuration 3 fans 3 power modules 12 disk modules 2 controllers Product Dimensions Dimensions Disk Array Height 578 mm 22 8 in 10 EIA Width 425 mm 16 7 in Depth 310 mm 12 2 in 211 soodg Specs Appendix A Product Specifications Operating Characteristics and Requirements Packaging Dimensions Packaging Disk Array Height 706 mm 27 8 in Width 610 mm 24 0 in Depth 508 mm 20 0 in Operating Characteristics and Requirements Power Requirements NOTE Power requirements are specified for a unit during spin up with a maximum configuration of three fan modules three power modules twelve disk modules and two controller modules e 870W 7 3 A at 120 Vac 3 8 A at 230 Vac Airflow Space e Minimum 102 mm 4 in front and rear Heat Dissipation e Minimum Configuration 135 watts 460 Btu hr 115 kcals hr Maximum Configuration 750 watts 2560 Btu hr 645 kcals hr e Typical Configuration with twelve 9 1 Gigabyte 7200 rpm disk modules in the enclosure 420 watts 1433 Btu hr 361 kcals hr 212 Appendix A Product Specifications Environmental Requirements Electromagnetic Emissions e Radiated and conducted interference For USA this equipmen
144. fter The test takes about 36 hours to complete Replace both batteries 175 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 198 0xc6 Fan Failed Y Single Fan A fan has failed and should be Occurrence Controller replaced Backplane 199 0xc7 Power Supply Y Single Power Supply A power supply has failed and Failed Occurrence Controller should be replaced Backplane 200 0 8 Power Down Y Single Fans The controller has detected Due To Fan Occurrence Controller that not enough fans are Failure Backplane functioning to cool the disk array properly As a result the controller is shutting off the power supply to avoid overheating Note The NVRAM is not posted to disk so it is critical to repair the system quickly 201 0xc9 Missing At Y Ignore See None During startup the controller Initialization accompanying detected that a fan had been errors removed from the enclosure The total length of time a fan has been missing will be checked periodically and when it exceeds approximately 10 minutes the controller will shut off the power supply 202 0 Power Down Y Single Fan The controller is shutting off Due To Missing Occurrence Controller the power supply because a Fan Backplane fan has been missing from the enclosure for too long approximate
145. g power on self test and then blinks with bus activity The primary array controller x or y appears on the control panel display when the array is ready to service a read write request and is ready for all normal I O activity Depending on the number of disk modules in the array and the spin up times for the disks there can be a delay of 2 to 90 seconds between the time the status light goes on and the time the controller is online Table 5 Fan Module Status Light Indications Color Indication Off IN o power or fan module not operating Green ormal operation Amber Fan speed too low Replace the fan module 113 Sunooquso qno Troubleshooting Troubleshooting Module Status Lights Table 6 Power Module Status Light Indications Color Indication Off IN o power or power module not operating Green Normal operation Amber Power supply fault or excessive current draw from power supply Replace power module Table 7 Disk Module Status Light Indications Color Indication Off power no data activity or self test completed Random IRead write data activity Flashing Green Solid Green Heavy data activity or waiting in an I O state Solid Amber Power on self test or disk module fault Table 8 Controller Module Status Light Indications Color Indication SolidGreen Ready State with no data activity Random Read write data activity Normal operation Flashing Green Of
146. gnore unless Disk Drive Indicates that a disk drive was At Power On no Operator Controller discovered to be missing activity Internal SCS during a power on or reset Bus This event occurs for disk set members only The disk drive ID number assigned to the missing drive appears in the component identifier field of this log entry A disk drive ID number consisting of the controller serial number followed by a 0 indicates that the disk drive ID number has not yet been assigned 172 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components zi DEC hex Implication 5 187 0xbb RAM Image Y Single Disk Drives The controller was unable to Upload Failed Occurrence Controller upload a usable disk copy of Internal SCSI during startup This ee Bus may be a result of missing or failing disk drives 188 0xbc Cache Y Ignore See None Write data found in the Configuration errors uploaded NVRAM image Used associated with blocked a requested transition shutdown to a new cache configuration The previous configuration is being used 189 0xbd Incompatible Y Ignore None The current RAM RAM Image Operator error configuration is not Content compatible with the RAM configuration of the NVRAM disk image This is caused by a different mapping of NVRAM in the disk image that cannot be supported by the current configuration For ex
147. he Disk Array into the Deskside Cabinet Installing the Disk Array into the Deskside Cabinet If the disk array is already installed into the deskside cabinet skip this set of steps To install the disk array into 1 226 the deskside cabinet see Figure 39 and perform the following steps Attach the edge protector 5 along the lower front of the cabinet as shown in Figure 39 This will protect the cabinet from being scratched when installing the disk array enclosure Install the four provided clip nuts 6 onto the front of the deskside cabinet as shown in Figure 39 Lift the disk array enclosure into the cabinet and secure it with the four screws 4 as shown in Figure 39 Remove the edge protector Loosen the two screws on the clamp 7 at the rear of the deskside cabinet and slide the clamp forward until the tabs of the clamp engage in the slots in the lower rear corners of the disk array as shown in Figure 39 Loosen the two screws on the clamp at the rear of the deskside cabinet and slide the clamp forward until the tabs of the clamp engage in the slots in the lower rear corners of the disk array as shown in Figure 39 Retighten the two screws Install the disk array door 3 as shown in Figure 45 Proceed to the next section to install the cabling Appendix C Cabinet Configurations Installing the Disk Array into the Deskside Cabinet Figure 39 Installing the Disk Array into the Deskside Cabinet
148. he clip with retainer clip screw 5 Install the door snap see Figure 44 Item 3 on the latch bracket 1 using the alignment holes 2 as a guide Secure the door snap with one T 15 screw 4 6 Install the enclosure door see Figure 45 Item 1 Drop the lower corner mounting hole onto the lower hinge bracket pin 2 then swing the upper corner mounting slot into the upper hinge bracket pin 3 Gently press on the upper left corner of the door and snap the upper corner onto the hinge bracket 7 Repeat steps 1 through to install additional disk arrays 8 Install two filler panels in the top two EIA slots see Figure 46 9 Connect the disk array power cords see Figure 47 to the cabinet power distribution units PDUs 10 Press the cabinet power switch see Figure 48 Item 1 to switch on the cabinet power The cabinet power light 2 should be green NOTE For greater power redundancy you can connect each power module into a separate PDU This will require ordering the Power Upgrade Kit A4915A to obtain additional PDUs refer to Appendix D for additional information CAUTION To prevent the current rating of the PDU from being exceeded observe the maximum current rating of the PDU Do not connect more enclosures than the PDU rating You may have to upgrade your PDU to provide enough total current for a full back to back rack configuration 230 Appendix C Cabinet Configurations Rackmount Cabinets Figure 42 Install
149. he green LED on the rack front that was not lit should now light One amber LED lit One or two of the three fans is not operating probably due to a physical failure A power failure is ruled out as that would cause the complete bank of LEDs to not light 1 Which fan is not spinning The exhaust fan assembly is located at the back top of the Inspect the assembly visually to see which fan is not spinning Replace the exhaust fan assembly Two amber LEDs lit Two or three fans have failed This condition is unlikely to occur power failure is ruled out as that would cause both banks of LEDs to not light 1 Have all fans failed The exhaust fan assembly is located at the back of the rack at the top Inspect the assembly visually to confirm that no fans are spinning Replace the exhaust fan assembly CAUTION This condition is critical as the fans are not cooling the array enclosures in the rack and disk drive reliability may be compromised The exhaust fan assembly should be replaced within 48 hours 256 Appendix D Back to Back Racking Removal and Replacement Removal and Replacement The exhaust fan assembly can be removed and replaced without powering down the rack The replacement part number for the exhaust fan assembly is E7687 63001 WARNING Avoid contact with any electrical parts inside the exhaust fan assembly while power is applied as there is an electrical shock hazard present 1 Remove the old exhau
150. he module cam lever see Figure 35 1 Observe the module cam lever is guided by locking bracket that fits into a track Notice the track has a notch observable when the lever is fully out Figure 35 Remove the lever by squeezing the tabs of the pivot and lifting the lever away from the module Position the lever locking bracket over the track notch observed in step 1 and lift it from the track To replace insert the cam lever locking bracket into the track at the notched area then snap the cam lever onto the pivot point 203 Removal 204 Removal and Replacement Module Cam Lever HP qualified personnel only Figure 35 Replacing the Module Cam Lever Removal and Replacement Replaceable Part Numbers Replaceable Part Numbers Figure 36 Replaceable Parts Removal 205 Removal Removal and Replacement Replaceable Part Numbers Table 16 Modular Replaceable Parts HP Product HP Replacement HP Exchange Number Part Number Part Number Description A3702A A3702 60003 A3702 69003 4 3 Gigabyte Disk Module 7200 rpm A3703A A3703 60002 A3703 69002 9 1 Gigabyte Disk Module 7200 rpm A3710A A3710 60001 A3710 69001 18 2 Gigabyte Disk Module 7200 rpm A5289A A5289 60001 A5289 69001 36 4 Gigabyte Disk Module 7200 rpm A3713A A3713 60001 A3713 69001 9 1 Gigabyte Disk Module 10 000 rpm A3714A A3714
151. his case you must copy the desired version of firmware from one controller to the other Copy Firmware always copies from the PRIMARY to the SECONDARY controller Firmware be copied using ARMServer download m or the front panel Copy Firmware You can also download the desired version of firmware from a file to both array controllers simultaneously using the ARMServer AutoRAID Manager NOTE The Download utility is described in further detail in the System Administrator s Guide which is shipped with each disk array Before Beginning Firmware Download Before you begin any of the following download procedures have the following ready e Get lt array id gt using ARMServer arraydsp or front panel View Settings Array S N You may also use the raw device file that is connected to the array NOTE Alphabetic characters in the Array Serial Number are CASE SENSITIVE Once you have the Array S N it may be easier to define the long serial number string as an alias To create an alias for the Array S N of 000000057D22 for example On HP UX or MPE systems export lt id gt 000000057D22 use id in place of lt array id gt On Windows NT systems C gt set id 000000057D22 use id in place of array id e Obtain the desired version of firmware and determine its lt codefilename gt filename and location path If you are downloading firmware you must know the source of your firmware The firmware could
152. icient Y Ignore None This errors occurs if a disk set RAM For Image Operator error has been moved from a Upload or controller with more RAM to accompanying a controller with less RAM errors Consequently the current RAM configuration is not large enough to hold the NVRAM image stored on disk More RAM can be added to the controller to correct this problem 168 0 8 Internal SCSI Y Immediate Disk Drive This event indicates that the Queuing Performance controller couldn t set the Disabled Problem SCSI configuration settings Unsupported for the disk drive correctly Disk This could be caused by a failing disk drive or the use of an unsupported disk drive 169 0 9 Internal SCSI Y Immediate Disk Drive The controller could not Disconnects Performance configure the disk drive to Disabled Problem disable disconnects during Unsupported data transfer phase This will Disk cause the interface driver to disable disconnects on the channel thus causing degraded drive performance This event may also indicate that the controller could not enable queuing on this drive so the interface driver will disable disconnects on the channel This is typically caused by using an unsupported disk drive that the controller cannot configure properly 168 Event Number DEC hex Event Name Event Logged Predictive Maintenance Implication Suspected Components Troubleshooting Event Code Descriptions Description 17
153. ictive Maintenance Implication Single Occurrence Single Occurrence Ignore Drive compatibility problem Ignore Operator or Host activity Suspected Components Controller Backplane Controller Backplane Disk Drive None Description The reporting controller has detected that the other controller s interrupt seems to be stuck on This error is logged only once per power on The reporting controller has detected that the other controller s interrupt is not working properly This error is logged only once per power on A drive disconnected during a data transfer This results in failure of the transfer because there is no method for ensuring correct generation of parity data This is typically accompanied by other errors A disk drive was successfully downed in response to user request 170 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 178 0xb2 Y Ignore See None The data unavailable condition Unavailable accompanying was triggered This is caused errors by multiple failed downed or missing disk drives The condition persists as a warning until fewer than 2 drives are down failed or missing If this condition occurred as a result of removing disk drives reinstall them to correct the problem This event also occurs if data and redundancy have be
154. ild Priority is able to balance the speed of the Rebuild with host system performance The range of temperature humidity vibration shock and altitude recommended for extended periods of operation of a disk array See Maximum Environmental Limit See Data Redundancy 267 JUSSO Glossary Glossary Reload Relocated Blocks Restart SCSI SCSI HBA SCSI ID SDTR 268 Traditionally RAID systems required the entire user data set to be rewritten in order to add more capacity or adjust the RAID mode The disk array does not require a Reload to make use of additional disk space because of the automatic Balancing and Dynamic Data Migration inherent in HP AutoRAID Array Technology Relocated Blocks are data blocks that are relocated from RAID 0 1 to RAID 5 or vice versa when performing a write The process of bringing the array back on line Typically done following a Shutdown a restart copies vital configuration information from the disks to NVRAM on the array controller SCSI stands for Small Computer System Interface This is an industry standard interface that defines the mechanical electrical and functional requirements for connections and communication between small computers and disk drives and other peripherals The SCSI ID identification or address uniquely identifies the device on the SCSI channel The SCSI HBA Host Bus Adapter is the card installed in the server that provides the con
155. ill still report power outage conditions if power from the UPS to the rack PDUs has failed They will still report fan failure conditions How do l rack less than the maximum number of array enclosures For optimum cooling efficiency array enclosures should be racked from top to bottom in the front of the rack and from bottom to top in the back of the rack If side access to the rack is not going to be possible install the bottom and then the top array enclosure leaving the array position in the middle or top open middle for a 2 meter rack or top for a 1 6 meter rack What replacement parts are there The only replacement part is the exhaust fan assembly part number E7687 63001 A failed exhaust fan assembly must be replaced by the Hewlett Packard service provider How rapidly must replace a failed exhaust fan assembly One failed fan in an exhaust fan assembly will not cause a critical cooling condition immediately However it s best to replace the exhaust fan assembly as soon as possible For racks fully populated with disk storage units replace the exhaust fan assembly within 48 hours 259 Sur ey Glossary Active Hot Spare Adapter Arbitrated Loop Auto Configuration Auto Failover AutoRAID Array Technology An Active Hot Spare is a portion of the disk array capacity reserved to perform a rebuild An Active Hot Spare ensures that the disk array can maintain Data Redundancy if a disk fails Until it is needed
156. imply rewrites the headers The or Host Bus Adapter is an internal card that is located in the host computer Each can connect to several SCSI devices Devices are Hot Pluggable if they can be removed while the host computer system is running When the disk array is configured with redundant controllers and power supplies all devices are Hot Pluggable Operation Including A Disk JBOD Load Balancing L DRV Logical Drive Logical Drive Number Loop Address LUN Glossary An Input Output I O Operation is an operation initiated by a host computer system during which data is either written to or read from a peripheral The process of adding a Disk Module to the array configuration is called Including A Disk The disk array cannot use a Disk Module until it has been included Once included the disk array begins using the Disk Module for data storage even if you have not created a Logical Drive with the newly added disk capacity As opposed to disk enclosures that offer Data Redundancy an enclosure without a Disk Array Controller or another type of RAID is called Just a Bunch Of Disks or JBOD When a JBOD enclosure has a disk failure data is lost See Balancing See Logical Drive A Logical Drive is a portion of the array capacity that appears to the operating system as a physical disk The entire array capacity can be divided into up to eight Logical Drives Only capacity that has been a
157. in NVRAM memory Once you have HP4x or later you can not downgrade to any firmware version prior to HP4x without first completely backing up all data on the array including any meta data e g LUN and LVM configuration After you have downgraded the firmware and formatted the array all meta data and real data will need to be restored from backup NOTE HP qualified personnel only For more information on ARDIAG refer to the web address http essd boi hp com products DiskArrays autoraid ARMdoclist htm click on Off Line Diagnostic Command Specification 1 Obtain a bootable LIF file You can get the LIF file from the web On the World Wide Web e http www hp com go support click next on Enterprise Storage click next on HP Enterprise Disk Arrays click next on SureStore E Disk Arrays click next on Disk Array 12H e click GO on Software Firmware beneath Software amp Drivers 52 HP SureStore E Disk Array 12H Controller Firmware Download Procedure or inside the HP firewall HP qualified personnel only e http essd boi hp com e click on Products e click on Disk Arrays e click on Disk Array 12H e click on AutoRAID firmware e left click on ARDIif HPxx xx the controller firmware revision you want to download e save the file using save as to your system Once you have the ARDIif file copy it to a tape using the following
158. ing Clip Nuts and Rails o Table 18 Multiple Disk Array Clip Nut Hole Spacing Enclosure Number Rail Standard Spacing Ear Standard Spacing 1 3 7 33 2 42 46 72 3 81 85 111 Note 1 order of installation from bottom to top Note 2 Hole numbers from bottom to top Note 3 Hole numbers from bottom to top two holes per bracket 231 Sjouiqe Cabinets Appendix C Cabinet Configurations Rackmount Cabinets Figure 43 Sliding the Enclosure into the Rackmount Cabinet Figure 44 Installing the Door Snap 232 Appendix C Cabinet Configurations Rackmount Cabinets Figure 45 Installing the Enclosure Door 233 sjouiqe Appendix C Cabinet Configurations Rackmount Cabinets Figure 47 Connecting Enclosure Power Cords to the PDU 234 Appendix D Back to Back Racking This appendix explains how to install disk arrays back to back in either a 1 6 meter rack 2786 or a 2 0 meter rack 2787 The following information is included e Product description Installation e Troubleshooting e Removal and replacement e Specifications Product Description Customers may obtain maximum disk storage capacity with a minimum use of floor space by filling their Hewlett Packard computer instrument racks completely with the maximum number of a
159. isk array in single controller mode or install a second array controller 131 Sunoouso qno Troubleshooting Troubleshooting Control Panel Error Messages Error Message System Warning Explanation An array warning condition exists warning designators follow This error can occur if you are using a single controller Action Use the control panel View Settings Warnings menu to see what error occurred If you are using a single controller you may have to run the arraymgr utility to suppress these errors The arraymgr utility is described in detail in the System Administrator s Guide which is shipped with each disk array 132 Unused Log Disk The specified log disk is not being used None Information only Troubleshooting FRU Codes FRU Codes Each replaceable hardware assembly in the disk array is assigned a FRU Field Replaceable Unit code These FRU codes are used to identify suspect hardware in the controller event logs Table 13 lists the FRU codes and the hardware assembly associated with each Table 13 Disk Array FRU Codes FRU Code Hardware Assembly 0 Disk in slot A1 1 Disk in slot 2 Disk in slot A2 3 Disk in slot B2 4 Disk in slot A3 5 Disk in slot B3 6 Disk in slot A4 7 Disk in slot B4 8 Disk in slot A5 9 Disk in slot B5 10 Disk in slot A6 11 Disk in slot B6 129 Reporting Controller or No FRU 130 First installable SIMM on X con
160. izes performance and storage efficiency by changing the storage techniques used as the amount of data stored on the disk array increases The following examples illustrate how Dynamic Data Migration manages the distribution of data during three different stages of capacity 84 Concepts and Management Disk Array Features Stage 1 Low Data Storage A newly installed disk array will typically have only a small amount of data stored on it In this case the disk array has room to keep all data in RAID 0 1 I Os can be serviced from RAID 0 1 so disk array performance is optimized Figure 18 Low Data Storage RAID 0 1 RAID 0 17 RAID 0 1 Qj DISK 1 DISK 2 DISK 3 DISK 4 85 sjdo uo Concepts Concepts and Management Disk Array Features Stage 2 Moderate Data Storage As more data is stored a point is reached where it becomes necessary for the disk array to begin moving or migrating data from RAID 0 1 to RAID 5 Because RAID 5 is more storage efficient this migration frees up capacity for more data The more full the array gets the larger the percentage of data stored in RAID 5 becomes The disk array examines all data and moves it to RAID 5 only if it has not been updated recently Write I Os are serviced faster from RAID 0 1 so this technique helps maintain disk array performance by keeping data that has the highest probability of changing in RAID 0 1 Read I Os are serviced just as quickly from RAID 5 a
161. k Module contains a single hard disk assembly A Disk Not in Use is any disk that is installed in the array enclosure but is not included in the array configuration A unique identifying code written to each disk The Disk Stamp identifies the disk as belonging to the set of disks installed in the array The stamp is written to the disk when it is included in the array The process of logically removing a disk from the array configuration Typically done before testing a disk downing a disk has much the same effect as physically removing it from the array enclosure DRR or Disable Remote Reset When DRR is ON or enabled default it prevents a host SCSI reset from resetting both disk array SCSI buses With DRR set to ON disk array controllers are not allowed to reset their own SCSI bus even if the host resets that controller If DRR 15 set to OFF a host SCSI reset will also allow that controller to reset both disk array SCSI buses Dynamic Data Migration is the process of moving data that is accessed frequently according to both the Disk Stamp and the Time Stamp which is updated when the data is written into RAID 0 1 space which is optimized for performance and also moving data that is accessed less often into RAID 5 space which is optimized for its higher storage density 263 JUSSO Glossary Glossary Enclosure EPROM FC Device Fast Wide Fibre Channel Arbitrated Loop Format HBA Hot Pluggable
162. k failure Active Hot Spare reserves capacity to perform a rebuild in the event of a disk failure If you are using disks of different capacities in your array Active Hot Spare reserves enough space to rebuild the largest disk drive Like logical drives the capacity reserved for the Active Hot Spare is distributed across all the disks in the array no one physical disk contains the Active Hot Spare Active Hot Spare is similar to the dedicated hot spare disks used in some conventional disk arrays However unlike conventional disk arrays that let the hot spare remain idle until it is needed the disk array uses the Active Hot Spare for RAID 0 1 storage until the spare is needed This provides the added benefit of enhancing performance while also protecting against disk failure 89 sjdo uo Concepts Concepts and Management Disk Array Features Auto Rebuild Auto Rebuild works in conjunction with Active Hot Spare to provide maximum protection against disk failure Auto Rebuild allows the disk array to automatically begin rebuilding a failed disk drive thus restoring data redundancy as quickly as possible A rebuild priority setting allows you to balance array performance with rebuild speed If a disk fails with Auto Rebuild enabled the disk array immediately begins rebuilding the contents of the failed disk on the Active Hot Spare or other available unallocated capacity The disk array uses redundant information stored on the re
163. kup copy of all data so half of the disk space is consumed by redundant data This results in a higher cost to store your data e RAID 5 provides data redundancy and improves cost efficiency by using a more efficient method of storing redundant data However the storage method extracts a performance penalty for each write operation This can impact system performance if your applications frequently update large amounts of data 83 sjdo uo Concepts Concepts and Management Disk Array Features Disk Array Features This section describes some of the key features of the disk array Understanding how these features work will allow you to use them to best advantage when managing your disk array Dynamic Data Migration Unlike most conventional disk arrays the disk array automatically manages its disks for the optimum balance of performance and storage efficiency It achieves this through the use of an innovative technique called Dynamic Data Migration which uses the best RAID level for storing data Most disk arrays operate using a single RAID level However Dynamic Data Migration uses two levels RAID 0 1 which optimizes performance and RAID 5 which optimizes storage efficiency Dynamic Data Migration combines the benefits of RAID 0 1 and RAID 5 while reducing the disadvantages of each Dynamic Data Migration uses RAID 0 1 and RAID 5 exclusively therefore all data is fully protected redundant Dynamic Data Migration optim
164. l flat bladed screwdriver less than 4 inches long and less than 0 25 inch width e 3 8 inch nut driver e 19 deep socket 5 mm nut driver Needle nosed pliers Replacing a Failed Controller Module HP qualified personnel only Replacement controller modules are shipped without batteries Therefore if you are installing a replacement controller you should remove and use the good battery packs from the controller module being replaced 191 Removal Removal and Replacement Front Door Assembly HP qualified personnel only Front Door Assembly HP qualified personnel only To replace the front door See Figure 31 1 Open the door to a position 90 degrees from the closed position Center the top hinge pin in the top slot of the door Push the door top outward so the hinge pin exits through the perpendicular slot Pull the door up and free it from the bottom hinge pin Replace the door by dropping the lower door mounting hole onto the lower hinge pin Qv Ew d Ue d Swing the upper corner mounting slot into the upper hinge bracket pin until it snaps into place Figure 31 Replacing the Front Door 192 Removal and Replacement Front Door Assembly HP qualified personnel only Switch Cover To replace the switch cover Snap the cover in place from the front of the door Front Door Snap To replace the front door snap see Figure 36 Item 14 1 Align the door snap with the alignment pins on the
165. ler Version of Firmware on the Replacement Array Controller Matches Version of Firmware on the Installed Array Controllers Replace B Replacing an Array Controller In an Array that IS NOT used as a boot device and the Version of Firmware on the Replacement Array Controller DOES NOT Match Version of Firmware on the Installed Array Controllers Replace C Replacing an Array Controller in an Array that is Used as a Boot Device Version of Firmware on the Replacement Array Controller DOES NOT Match Version of Firmware on the Installed Array Controllers Replace D Replacing an Array Controller in an Array with one controller installed 42 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Download Windows NT 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 the System Administrator s Guide which is shipped with each disk array 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 multi host configurations other hosts must not access the disk array while
166. leshooting Event Code Descriptions Description A Pass Through Reset command has reset the disk drive causing I Os to fail The controller could not retry I Os This typically occurs when performing diagnostic testing on the disk drive 28 0x1c Disk IO Failed Due To Drive Removal NA None A disk drive was removed causing I Os to fail Any I Os in the controller queue are flushed with this status 29 0x1d 30 0 1 31 0x1f Disk Drive Failed Initialization Disk Failed Due To Related Hot Plug Unexpected Disk Drive Error Y NA None NA None Single Disk Drive Occurrence Unless Excluded For Drive Type Following a device reset the controller could not successfully initialize the disk drive This is typically caused by a disk drive s failure to spin up The controller flushes all non pass through I Os from the queue when this event occurs A disk drive was in unknown state during command execution The I O was not retryable or the all retries have been exhausted The host should retry the I O This event will occur only if the disk array is operating in degraded mode The disk drive returned a status that could not be handled by the controller s error recovery algorithms retries are exhausted or the I O is not retryable 141 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Number DEC hex 32 0x2
167. ller X becomes the primary controller and controller Y is offline This may be recovered by power cycle 214 0xd6 Secondary Y Ignore if Controller A controller was discovered to Controller corrected by Backplane be bad during power on Failed Entered reset or power Specifically the secondary Separation cycle controller appears to have Mode failed due to an event which caused separation mode to be entered In separation mode the NVRAM contents is not mirrored between controllers This may be recovered by reset or power cycle 180 Event Number DEC hex 215 0xd7 216 0xd8 217 0xd9 Event Name Secondary Controller Failed With Memory Error Secondary Controller Failed Communication ECC Errors Secondary Controller Failed To Enable RAM Mirroring Predictive Implication Ignore if SIMM SIMM Controller configurations Backplane do not match See None accompanying errors See None accompanying errors Suspected Maintenance Components Troubleshooting Event Code Descriptions Description A controller was discovered to be bad during power on Specifically the secondary controller appears to have failed due to some memory error or violation If the memory contents on each board does not match in number this error can occur Additionally if a memory component fails it will cause this error This will occur if a different number of SIMMs are used on each controller A
168. lly loaded with disk modules By simply replacing the original disk modules with higher capacity modules you can add more logical drives or you can improve disk array performance by creating more unallocated capacity Replacing the disk modules is done while the disk array is operating thus avoiding any downtime There are some important points to remember when replacing disk modules in a fully loaded disk array NOTE Anytime a rebuild is in progress the disk array is vulnerable to a disk failure If another disk fails before the rebuild is complete data loss will occur Although this is unlikely to occur it is something to consider before making the decision to increase the capacity of the disk array by replacing the disk modules If another disk fails you can minimize data loss by replacing the failed disk with the original disk removed to perform the upgrade As added protection it is a good idea to backup all your data before upgrading the capacity of the disk array disk modules must be replaced one at a time and the disk array must be allowed to rebuild the data from the disk module before a another disk module is removed This ensures that no data is lost as a result of two disk modules being removed from the disk array simultaneously Once the Rebuild is complete the capacity of the new disk module is added to the disk array configuration and another disk module can then be replaced e Tomake the additional capacity o
169. ly without requiring you to know all the complexities of determining and setting up different RAID modes 261 JUSSO Glossary Glossary Auto Rebuild Availability Balancing Block Mirroring Cache Channel Configuration Data Redundancy 262 Auto Rebuild begins immediately if a disk failure occurs as long as enough space is available to perform the rebuild No operator intervention is required to perform an Auto Rebuild Once the Auto Rebuild has completed the disk array is once again Fault tolerant since all user data is once again redundant Availability is achieved by using redundant data to prevent the loss of use of a storage system in the event of a disk failure Balancing is the process of automatically spreading data equally across all disks to increase performance When a fixed amount of data is either written or read from multiple disks instead of to or from just one disk the throughput or speed of the process increases greatly Block Mirroring is a technique in which duplicate copies of blocks of data are stored on an array of disks An on board cache or memory buffer greatly enhances the speed of data transfers to and from disk devices since the next block of data required by the host computer is often already available in high speed cache memory Caching does require controller overhead however so unlike traditional RAID systems that often utilize huge cache space the disk array is tune
170. ly 10 minutes 176 Event Number DEC hex 203 0xcb 204 0xcc 205 0xcd 206 0xce 207 0xcf Event Name Replace Batteries Cache Shrink Attempted After Shutdown Warning Controller Failed SIMM Failed Extended Drive Installation Event Predictive Maintenance Implication Single Occurrence Ignore see errors during shutdown Ignore if corrected by reset or power cycle Single Occurrence Ignore unless no operator activity Suspected Components Batteries Controller None Controller SIMM Controller Disk Drive Controller Internal SCSI Bus Troubleshooting Event Code Descriptions Description The batteries have failed a discharge test or have dropped below an acceptable voltage level Both batteries should be replaced The amount of cache on the controller has shrunk due to SIMM failure or removal This happened with valid writes still in cache and all the writes no longer fit in the reduced amount of cache During power on a bad controller was detected This may result if the controller has difficulty establishing communication Ignore if recovered by power cycle or reset SIMM was discovered to be bad during power on An internal SCSI channel is being held reset longer than expected Typically caused by a partially installed drive module a failing drive module a failing controller or a bent pin on a connector 177 Sun
171. maining disks to reconstruct the data that was on the failed disk Until the rebuild is complete the array is vulnerable to another disk failure and performance will be impacted If you want more control over the rebuild process you can disable Auto Rebuild and start a rebuild manually This allows you to control when the rebuild is performed but it can leave your data vulnerable to a second disk failure until the rebuild is complete To perform a rebuild automatic or manual the disk array must have adequate available capacity The best way to make this capacity available is by enabling the Active Hot Spare This ensures that the disk array always has enough capacity to rebuild even the largest disk in the array Rebuild Priority During a rebuild the disk array is trying to do two things at once perform the rebuild and service I O requests from the host A rebuild priority allows you to define which of these operations is more important High rebuild priority allows the disk array to complete the rebuild as quickly as possible but system performance may suffer because host I Os are delayed Low rebuild priority instructs the array to give precedence to server I Os thus maintaining system performance but delaying the completion of the rebuild Interrupted Rebuild Removing a disk from the array with Auto Rebuild enabled will cause the array controller to immediately begin a rebuild assuming there is an Active Hot Spare or enough unallo
172. makes it easy to increase your array s data capacity as new higher capacity disks become available There is an important point to remember when using disks of different capacities in the same array You should add at least two high capacity disks to the array to use their entire capacity If you add a single high capacity disk the array will not be able to use its full capacity In this situation the array can only use capacity equal to that of the largest disk already installed in the array For example if your disk array has only 4 3 Gigabyte disks and you add a single 9 1 Gigabyte disk the array can only use 4 3 Gigabytes of the high capacity disk at least until another 9 1 Gigabyte disk is added Figure 22 Using Disks of Different Capacities T GB USED 1 GB USED 1 GB DISK 1 GB DISK 2 GB DISK This is a result of the way in which redundant data is stored using RAID technology Because there is no corresponding space on another disk for redundant data the array cannot use the upper portion of the high capacity disk Thus adding a single high capacity disk will not yield all the additional capacity you had expected To avoid this situation always install at least two high capacity disks in the array 96 Concepts and Management Disk Array Hardware Figure 23 Installing a Second High Capacity Disk 1GBDISK 1GBDISK 2GBDISK 268 DISK Disk Stamps Because data is distributed across all disks the array contr
173. mand which scans the bus for all arrays This command can take a few seconds to execute Once re connected you will see the array again HP SureStore E Disk Array 12H Controller Firmware Download Procedure Replace C Replacing an Array Controller in an Array that is Used as a Boot Device Version of Firmware on the Replacement Array Controller DOES NOT Match Version of Firmware on the Installed Array Controllers CAUTION Controller firmware HP4x and any later versions have a different data map format in NVRAM memory Once you have HP4x or later you can not downgrade to firmware version prior to HP4x without first completely backing up all data on the array including any meta data e g LUN and LVM configuration After you have downgraded the firmware and formatted the array all meta data and real data will need to be restored from backup NOTE This is an offline procedure Because the array is a boot device the operating system needs to be shut down Assumptions Use this procedure if an existing array controller has failed on a boot device and the replacement array controller DOES NOT have the same version of firmware as the installed array controllers NOTE The offline diagnostic is another way to install firmware when the array is a boot device Offline diagnostics are not covered in this paper 1 Ifreplacement controller is new install the controller batteries supplied 2 Ifreplacement
174. mark Credits AutoRAID is a trademark of Hewlett Packard Company TORX hardware is used in this product TORX hardware requires the use of special drivers In this manual any reference to TORX hardware will be accompanied by the required driver size for example 15 or 25 TORX is a product of the Camcar Division of Textron Inc Table of Contents CHAPTER 1 PRODUCT DESCRIPTION 9 MODELS AND CONFIGURATIONS AVAILABLE eneenennnr ness entra nnns seen entere nasse s entere nnns 9 OPTIONS 5 ayasa 10 ANWAILABEE ACCESSORIES hee Hr POE E Ere pe 13 UNPACKING AND REPACKAGING THE DISK 14 PRODUCPEBATURBS TN ER 16 CONNECTING 5 eeclesie prre Nr E PERO e pee ke Pu Eee Cerrar Nn 20 INSTAEEING MODU LES dee c pee e eb reed ee tede eed 20 Disk Power and Fan Modules eerte tette rette eye cr ED E REPRE TERN 21 CONTROLEER MODULE e e 22 Suppressing Single Controller Warning Messages a 00 23 SCSLCABEING REQUIREMENTS ege ng eed pede e eut cede 25 CONNECTING SCSI CABLING 3a eere rer eee eie yas ERE vus 25 FIBRE CHANNEL FC SCSI M
175. memory was added other situations this event represents a true memory error 211 0xd3 Controller Y Ignore if SIMM A controller was discovered to Failed Local corrected by Controller be bad during power on Address Decode reset or power Specifically the secondary Test cycle controller appears to have failed the request to perform a local memory decode test or was unable to establish communication with the other controller This may be recovered by power cycle or reset 212 0xd4 Controller Y Ignore if SIMM A controller was discovered to Failed Remote corrected by Controller be bad during power on Address Decode reset or power Backplane Specifically the secondary Test cycle controller appears to have failed the request to perform a remote memory decode test or was unable to establish communication with the other controller This may be recovered by power cycle or reset 179 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex Event Name Event Logged Predictive Maintenance Implication Suspected Components Description 213 0xd5 Secondary Y Ignore if Controller A controller was discovered to Controller corrected by Backplane be bad during power on Failed Entered power cycle Specifically the secondary Isolation Mode controller appears to have failed due to an event which caused isolation mode to be entered In isolation mode contro
176. mpanyin g errors None Retries were used by the R5 module in the successful completion of a read 97 0x61 Write Ignore See None Retries were used by the Recovered accompanyin R5 module in the With RAID 5 g errors successful completion of a Retries write 98 0x62 Read Ignore See None Redundancy was used by Recovered accompanyin the 5 module in the With RAID 5 g errors successful completion of a Redundancy read 99 0x63 Write Ignore See None Redundancy was used by Recovered accompanyin the 5 module in the With RAID 5 g errors successful completion of a Redundancy write 100 0x64 Read Ignore See None Redundancy was used by Recovered accompanyin the R1 module in the With RAID 1 g errors successful completion of a Redundancy read 101 0x65 Write Ignore See None Redundancy was used by Recovered accompanyin the R1 module in the With RAID 1 g errors successful completion of a Redundancy write 156 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 102 0x66 Disk drive Y Signals Disk Drive The disk drive reported a reported a internal drive SMART event The event is SMART event errors logged and the common sense threshold data is updated for all regular exceeded read write and passthru babl di k commands The precedence 19 a N a for SMART event data in the failure in common area is low so
177. mplete The disk array should display the Ready state Check the system state There should be no other warnings KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready Primary Cntrl X If the array does not show up using ARMServer arraydsp i you may have to re connect to see the array using the ARMServer arraydsp R command which scans the bus for all arrays This command can take a few seconds to execute Once re connected you will see the array again HP SureStore E Disk Array 12H Controller Firmware Download Procedure Replace D Replacing an Array Controller in an Array with one controller installed Assumption 1 If a successful shutdown has occurred in which the memory maps were copied to the disk drive modules then you can replace the controller and restore the maps using the following procedure 1 Ifreplacement controller is new install the controller batteries supplied 2 Ifreplacement controller is NOT new reset disconnect and then re connect both batteries on the replacement controller at the same time to erase NVRAM 3 Replace the failed controller with the new one in controller slot X 4 Push the on off button to the on position and wait for the initialization to complete The disk array should display the Ready state Check the system state There should be no other warnings KEY DISPLAY CANCEL CANCEL Ready
178. n 512 byte blocks appears in the last 3 bytes The FRU field reports the FRU of the drive corrected Troubleshooting Event Code Descriptions Event Name Event Predictive Event Suspected Description Number Logged Maintenance Components zi DEC hex Implication 5 228 0xe4 Local Controller Y Ignore if DRAM SIMM This error code indicates that Failed DRAM recovered by Controller the DRAM SIMM on the local z Address Decode reset or power controller failed the DRAM ee Test on address decode test during the us 229 0xe5 230 0xe6 231 0xe7 Remote Controller Failed DRAM Address Decode Test Controller Failed Background ROM checksum test Scrub policy detected error Ignore if recovered by reset or power on DRAM SIMM Controller Ignore if recovered by reset or power on ROM Controller Ignore should Disk drive be preceded by Drive errors power on process Specifically the DRAM SIMM on the primary controller failed the decode test This error code indicates that the DRAM SIMM on the remote controller failed the DRAM address decode test during the power on process Specifically the DRAM SIMM on the secondary controller failed the decode test This error code indicates that the ROM on the local controller failed the ROM checksum test done in the background during idle time The firmware stored in the ROM is corrupted This error code indicates that an error was returne
179. nection for the disk array The SCSI HBA manages the transfer of data between the server and the disk array Multiple SCSI devices be connected to each HBA The SCSI ID is a unique number assigned to each device connected to a SCSI bus This number is used by the HBA to address each device on the bus Each controller in the disk array is assigned its own SCSI ID SDTR stands for Synchronous Data Transfer Request SDTR controls the data rate on the SCSI bus used by the host and the array SDTR also determines the negotiation protocol of the host If SDTR is enabled the array will initiate negotiation protocol if disabled the host will initiate negotiation protocol In either case the disk array will always respond to any requests made by the host Secondary Array Controller Self Configuring Self test Shutdown SIMMs Status Light Stripe Depth Glossary In disk arrays with two controllers one controller is called the Primary Array Controller and the other is called the Secondary Array Controller The Secondary Array Controller offers redundant controller operation should the primary array controller ever fail The Secondary Array Controller offers redundant controller operation full access to all data and it also monitors the status of the Primary Array Controller the status of the Primary Array Controller is ever questionable the Secondary Array Controller is able to become the Primary Array Controller
180. ng to be possible install the bottom and then the top array enclosure leaving the array position in the middle or top open middle for a 2 meter rack or top for a 1 6 meter rack 246 Figure 51 PDU Installation Appendix D Back to Back Racking Installation Procedure PDU 10 Plug 7676 63001 3ea 6 Power Cord 2 Bracket A4915 00001 4 ea Opt ABA 7803 60001 3 ea 3 Screw 5 8 TORX 2680 0323 4 ea Opt ABB 8120 6895 3 ea 4 Nut 10 32 2740 0003 6 ea Opt 024 8120 6961 3 ea 5 Screw gt Pozidriv 2680 0055 6 ea 247 Sup exX Racking Appendix D Back to Back Racking Installation Procedure 7 Install Replacement PDUs a Attach two PDUS vertically to the brackets installed in the previous step Position the PDU with its large power receptacle toward the bottom and install as described in the A4915A Power Upgrade Kit Quick Installation Guide b Install the horizontal PDU with the large power receptacle to the right facing the rear of the rack Install the PDU in the holes indicated below for the specific cabinet as described in the power upgrade kit quick installation guide Rack EIA Holes as counted from the bottom of the rack 1 6 Meter Rack 89 middle hole of EIA unit 30 2 meter Rack 41 middle hole of EIA unit 14 8 Install the power cables Connect each power cable from
181. nications Commission in 47 CFR 15 105 has specified that the following notice be brought to the attention of the users of this product This equipment has been tested and found to comply with the limits for a Class A digital device pursuant to Part 15 of the FCC Rules These limits are designed to provide reasonable protection against harmful interference when the equipment is operated in a commercial environment This equipment generates uses and can radiate radio frequency energy and if not installed and used in accordance with the Instruction manual may cause harmful interference to radio communications Operation of this equipment in a residential area is likely to cause harmful interference in which case the user will be required to correct the interference at his own expense The end user of this product should be aware that any changes or modifications made to this equipment without the approval of the manufacturer could result in the product not meeting the Class A limits in which case the FCC could void the user s authority to operate the equipment 217 K39jeg A10je nS9 s Regulatory Safety Appendix B Regulatory Safety Statements FCC Statement For U S A Only IEC Statement Worldwide This is a CISPR 22 Class A product In a domestic environment this product may cause radio interference in which case the user may be required to take adequate measures CSA Statement For Canada Only This Class A digital apparatus me
182. nu 2 Selectalanguage 3 74 Select Confirm Using Control Panel Menus Table 2 shows all of the control panel main menus except the Shutdown menu which has no menu items Note that some words have been abbreviated to fit the display window such as the word Cntrl for Controller L Drv for Logical Drive and ActivSpare for Active Hot Spare Operating the Control Panel Using Control Panel Menus When the display panel shows that the disk array is Ready it may also display an alias name for the disk array if one has been defined An alias must be defined with the Array Management ARM Utility as described in the System Administrator s Guide which is shipped with each disk array Table 2 Control Panel Main Menus Language Cntrl Chanqes Disk Chanqes L Drv Changes View Settings Svstem State Recover SIMMs Installed MBytes of RAM Deutsch SCSI ID Include Disk Create L Drv English 2 Switch Cntrl Auto Include Delete L Drv Warnings Espanol Reset Cntrl Auto Rebuild Renumber L Drv Primary Cntrl Francais Format Array Activspare SCSIIDx Italiano Parity Start Rebuild SCSI ID y SDTR Stop Rebuild Firmware WDTR AmaySN Term Power SCSI Values Unit Attn Auto Include DRR Auto Rebuild VEB ActivSpare Mark Log Disk Number of Disks Post Log Disk Disk States Copy Firmware Rebuil
183. numbers are displayed on the front panel The numbers step higher as each initialization step is completed The changing sequence numbers assure you that the disk array is performing the various power on tests When the tests have all completed the front panel will display the word ready If the word ready is not eventually displayed the sequence number at which the power on stops will help diagnose what might have went wrong Table 9 Power On Sequence Test Sequence Code Description of Power On Sequence Test Performed 1 host SRAM test parity SRAM test 4 chip register test data path tests 6 ROM test copyperfcodetoSRAM 10 delay for mirror to come up start mirrored communications 12 check backplane 16 start early spinup 20 local NVRAM pretest 22 remote NVRAM pretest 24 local NVRAM decode test 26 setup remote NVRAM decode test 28 remote NVRAM decode test 30 SIMM 1 Ist half test 32 SIMM 1 2nd half test 32 MB SIMMs only 34 SIMM 2 1st half test 36 SIMM2 2ndhalftet 32 MBSIMMsonl 38 remote SIMM 1 Ist half test 40 remote SIMM 1 2nd half test 32 MB SIMMs only 42 remote SIMM 2 Ist half test 121 Sunoouso qno Troubleshooting Troubleshooting Power On Sequence Test Sequence Code Description of Power On Sequence Test Performed 44 remote SIMM 2 2nd half test 32 MB SIMMs only
184. o the disk drive 150 Troubleshooting Event Code Descriptions Event Name Event Predictive Event Suspected Description Number Logged Maintenance Components zi DEC hex Implication 75 0x4b Internal Data N NA None A data transfer with the disk Transfer drive failed due to a hot plug Aborted By Hot event on the channel When ee Plug performing RAID 1 mirrored um transfers a hot plug event may disrupt the transfers on both SCSI channels in use 76 0x4c Internal SCSI N Single Controller A disk drive responded Bus State Occurrence Internal SCSI incorrectly The result was an Error Bus internal SCSI bus reset and a Disk Drive hot plug recovery sequence to regenerate the bus state The control panel display may indicate verifying drives as a result of the internal bus reset 77 0x4d Write Sequence N Ignore None A part of a write cache stream In Cache Broken was aborted by the host The By Host Abort failed write was forced because the operation could not be completed This is a normal part of an abort operation The host should retry the operation 78 0x4e Silent Disk N NA None On a previous attempt s the Drive Error disk drive failed the I O The Recovery controller retried the I O and the drive returned good status 79 0xAf Check Y Single Disk Drive The disk drive returned a Condition From Occurrence check condition with no sense Disk Drive With data The controller s retries No Sense
185. oad Windows NT Procedure Operating System Windows NT No Perform procedure Download D Yes Downloading firmware via the dfdutil utility download boot V Class Is controller firmware available on the host System Is the disk array a boot device Is the system a V Class No Tc Perform procedure Download A Perform procedure Download C No Downloading firmware from Downloading firmware via off line a file using ARMServer diagnostic ARDIAG download non boot download boot not V Class Perform procedure Download B Downloading firmware from a third controller download from third controller Do you have a third controller with the desired firmware Y In order to download firmware you either need the firmware file on the system or in a controller For more information on downloading firmware refer to either the User s and Service Manual or the System Administrator s Guide Both documents can be found on ESBU s Intranet web page at http essd boi hp com on the World Wide Web at http www hp com go support HP SureStore E Disk Array 12H Controller Firmware Download Procedure Figure 16 Replacing an Array Controller Flowchart Replacing an Array Controller NOTE In some cases you may need to download firmware when replacing a controller Downloading firmware is an offline process
186. oesn t exist 138 Event Event Name Event Predictive Suspected Number Logged Maintenance Components DEC hex Implication 18 0x12 Not Enough Y Ignore None Space 19 0x13 Command N NA None Illegal In Current State 20 0x14 Becoming N NA None Ready 21 0x15 Recovered With Y Ignore Disk Drive Low Level Controller Retries Internal SCSI Bus Troubleshooting Event Code Descriptions Description A write command had to move data in order to find enough space to complete Errors were encountered during this process which prevented the controller from providing enough space for the write This would typically result from multiple failing or missing drives If disk drives have been removed reinstall them in the array The requested operation is not allowed due to the current state of the disk array Certain operations require the disk array to be in a specific state for execution For example the disk array must be shutdown to perform a download The disk array has not finished its initialization After failing the initial I O data was recovered using retries It was necessary to use redundancy to recover the data 139 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Number DEC hex 22 0x16 23 0x17 24 0x18 25 0x19 26 0 1 Event Name Recovered By Disk Drive Disk Drive Media Error Disk Drive Hardware Error Disk Drive Reset
187. of this log entry The controller is going to initiate a Host channel SCSI reset This will occur if the controller detects that the channel has not already seen a reset and DRR Disable Remote Reset indicates that a SCSI reset should be generated Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 221 0xdd Cache Version Y See errors None A new version of controller Mismatch In associated with firmware has been download RAM Image shutdown with writes stuck in cache Normally there will be no writes stuck in cache and a firmware download will complete without any problems However if there are writes stuck in cache the upload will fail with this event code To correct this it may be necessary to revert to the old version of firmware and solve whatever problem is causing writes to be stuck in cache probably one or more disks have failed 222 0xde Version Y NA None The upload routine was unable Mismatch to upload part of the disk NVRAM image because the current firmware does not support that use of RAM Most likely an older version of firmware is trying to upload a disk image posted by some other firmware version 183 Sunoouso qno Troubleshooting Event Code Descriptions Troubleshooting Event Event Name Event Predictive Suspected Description Number Logged Maintenance C
188. oller could be either Cntrl Y Disabled and controller X has become primary and primary or secondary array controller at fault controller Y is silent Also could be a backplane failure Replace backplane System States Data may have become inaccessible Replace the faulty disk module s Data Warning System States Data redundancy has been lost If Auto Rebuild is disabled perform a manual Degraded rebuild If Auto Rebuild is enabled wait for the Warning array to restore data redundancy data rebuild is in progress You may need to add more disk modules to complete the Rebuild System States DRAM errors have occurred Replace the controller module or the DRAM DRAM Failed System States A SIMM was added without first performing a Shutdown the array and then power it on DRAM Unused Shutdown 124 Error Message Explanation Troubleshooting Control Panel Error Messages Action System States An invalid configuration has been detected The following action should be performed for Disk Warning caused by one of the following cases each of the cases listed Case 1 a failed disk module in the array Case 1 Replace the disk module and perform a Rebuild Case 2 a Previously Used disk module has Case 2 Use Include Disk on the Previously been added Used disk module or remove the Previously Used disk module Case 3 a down disk drive system Case 3 Fix or replace the disk drive System
189. oller as primary e Time offline is approximately 2 minutes 91 s3do uo Concepts Concepts and Management Disk Array Features Shutdown The disk array uses a coordinated process called Shutdown to take the disk array offline The primary function of Shutdown is to copy the contents of the controller NVRAM to the disks This protects critical data stored in the NVRAM against loss should battery backup for the NVRAM fail in the absence of ac power In the Shutdown state the disk array can still execute some SCSI commands from the host but the host cannot access any of the data on the disk array Shutdown is performed automatically when the disk array is turned off using the power switch on the enclosure The action of the power switch delays shutting off power long enough for the array controller to copy the NVRAM contents to each disk NOTE If power to the disk array is lost by means other than turning off the switch the array will not have time to perform a successful Shutdown In this case the battery can sustain the NVRAM contents for a minimum of one week Shutdown can be initiated manually using the disk array control panel or using a host software utility However a Shutdown makes all data on the disk array unavailable to the host 92 Concepts and Management Disk Array Hardware Disk Array Hardware This section describes the operation and design of the disk array hardware This information is provided for
190. oller must keep track of which disks are installed in the array To do this the array controller writes a unique stamp on each disk The disk stamp includes a unique number assigned to the disk typically its serial number and the numbers of the other disks in the disk set This information allows the array controller to determine if any disk is missing from the disk set The disk stamp identifies each disk s logical position in the array configuration relative to the other disks When managing data the array controller deals in terms of logical position rather than physical location in the array enclosure Thus disks can be moved to different enclosure slots without any impact on array operation long as the correct disks are installed the array controller is satisfied When another disk is installed in the array enclosure the array controller checks the disk stamp to determine if the disk is new or if it was previously installed in another array In the latter case the array controller assumes there may be some valuable data on the disk and does not include it even automatically until you tell it explicitly to do so Once included the disk is given a new stamp indicating it now belongs to the array s disk set 97 sjdo uo Concepts Concepts and Management Disk Array Hardware Power Modules The fully redundant disk array includes three power modules which provide power to all components in the enclosure The power modules
191. omponents DEC hex Implication 223 0xdf Disk Format NA None The NVRAM format of the Version Mismatch 224 0xe0 Shutdown Due To Power Supply Failure 225 0xel Recovery from battery backed RAM Loss Started 184 Single Power Occurrence Supplies Controller Backplane Single Controller Occurrence disks does not match the format used by the controller This event is logged when an icicle controller attempts to use disks previously shutdown by an Ice controller or visa versa The controller cannot use an NVRAM image in the incorrect format This error code indicates that the Background Manager BGM has discovered that there are not enough good power supplies to run the system The BGM is shutting down the subsystem to minimize system operation with inadequate power Note The NVRAM is posted to disk This error code indicates that recovery from a battery backed RAM loss has been initiated The contents of the NVRAM that is battery backed up has been lost and the arrayrecover command has been initiated The contents of the NVRAM has started to be rebuilt This is a feature only in controller firmware revisions HP54 and later Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 226 0xe2 Recovery from Y Ignore None This error code indicates that Battery Backed recovery from a battery RAM
192. on 153 0x99 Unable To Y Ignore None The controller experienced a Recreate memory error but was unable Memory Error to recreate the failed memory accesses The disk array will reset in response to this condition A reset will rescan all memory and attempt to locate the problem 154 0x9a Memory Error Y More than 1 SIMM The controller experienced an Not Recovered in 6 mo Controller unrecoverable error that could With Backplane not be recovered using Redundancy redundant memory This may cause data loss if the error occurs in non redundant NVRAM memory 155 0x9b Memory Error Y Ignore See None The controller experienced an Recovered With accompanying unrecoverable error that was Redundancy errors recovered using redundant memory or a controller reset 156 0x9c Uncorrectable Y More than 1 SIMM The controller experienced an Memory Error in 6 months Controller uncorrectable memory error per controller Backplane ECC error 157 0x9d Correctable Y See correctable SIMM The controller experienced a Memory Error rate associated Controller correctable error with usage Backplane page 158 0x9e Memory Y Ignore See SIMM The system experienced a Compare Error accompanying Controller compare error between the errors Backplane redundant memories The mirrored image maintained by the two controllers does not match but the exact cause is not apparent 165 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions
193. ong type Different HP disk array products install the correct controller s in the disk use different controllers Although the controller array Controller family can be determined by a may fit mechanically into the enclosure the disk small label on the controller handle Call the formats used by different product families are support line incompatible Incompatible The firmware on the primary array controller is Copy the image number displayed on the Image not compatible with the NVRAM image control panel display and contact the support line for assistance in determining if the image is compatible with your firmware Initializing The array is running the initialization process None Information only 128 Error Message Explanation Troubleshooting Control Panel Error Messages Action Log Disk A requested mark on the log disk has failed Retry If error continues return log disk for a Not Marked replacement L Drv Not The SCSI command to Delete Volume failed Call the support line Deleted L Drv Not The SCSI command to Renumber Logical Drive Call the support line Renumbered failed Mismatched The two controllers installed have different Do not use controllers from different vendors CNTLR personality profiles Unique controller in the same disk array enclosure Remove the personalities are created for different vendors incompatible controller and replace it with a These personalities ar
194. ontroller X SCSI Connector 2 3 Controller Y SCSI Connector 1 cs 11 CODICE 0000000000 fou NC _ 00000000000000000 4 Controller Y SCSI Connector 2 5 P1 P3 AC Power Connectors Product Description Product Features 17 uondrnbsoq Product Description Product Features Figure 4 Disk Array Controls and Indicators Description 1 Power Module Status Light 5 Control Panel Display 2 Disk Module Status Light 6 Control Panel Status Light 3 Fan Module Status Light 7 Controller Module Status Light 4 Power Standby Switch 18 Product Description Product Features Figure 5 Disk Module Slots F1 F3 Fan modules P1 P3 Power modules A1 A6 and B1 B6 Disk modules X and Controller module s 19 uondrbsoq Description Product Description Connecting Power Cords Connecting Power Cords You should connect power cords to all ac power connectors see Figure 3 Disk array availability is enhanced if each power cord is connected to a separate ac branch circuit WARNING To avoid a fire hazard the ac branch service must be properly current protected by either a fuse or a circuit breaker Use only UL CSA approved power cord SVT type rated for suitabl
195. oo Press or then press ENTER to select values for a task enable disable and confirm cancel Returning to the Main Menu To return to the main menu press Cancel 70 Operating the Control Panel Using the Control Panel Buttons Using the Control Panel Buttons Viewing Array Settings 1 2 3 4 5 Select the View Settings from the main menu Use or to view the current settings Select the SCSI values menu to view settings for Unit Attn Term Power Parity WDTR SDTR DRR and VEB Select the Disk State menu to view the state of disks Al to and B1 to Select the L Drv State menu to view the capacity of logical drives in the array Creating a Logical Drive 1 Select the Logical Drv Changes from the main menu Select Create Logical Drv Select a drive number 0 to 7 Select a capacity size 100 Megabytes to Maximum Select Confirm Renumbering a Logical Drive NOTE Before renumbering a logical drive you should know the impact this will have on the operating system After you renumber a logical drive make sure you perform the necessary steps to configure the renumbered logical drive into the operating system 1 2 3 4 5 Select the Logical Drv Changes from the main menu Select Renumber L Drv Select From drive number 0 to 7 Select To drive number 0 to 7 Select Confirm 71 pued 10002 Control Panel Operating the Control Panel Using the
196. oouso qno Troubleshooting Troubleshooting Event Code Descriptions activity Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 208 0xd0 Internal SCSI Y Single Controller A disk drive responded Bus State Occurrence Disk Drive incorrectly The result was an Event Internal SCSI internal SCSI bus reset and a Bus hot plug recovery sequence to regenerate the bus state The control panel display may indicate verifying drives as a result of the internal bus reset Umso Y SM eme ced ECC Error 6 months if Controller power on This is expected in During no operator any of the following Initialization situations The batteries have been fully discharged The batteries were disconnected from the memory New SIMM memory was added other situations this event represents a true memory error 178 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication The controller experienced an 210 0xd2 Correctable Y More than 1 ii unrecoverable error during ECC Error 6 months if Controller power on This is expected in During no operator any of the following Initialization activity situations The batteries have been fully discharged The batteries were disconnected from the memory New SIMM
197. or Call the support line and remove the disk Unknown unexpected state module Disk States This type of disk module is unsupported and CAUTION Replace the disk module with one Unsupported has not been tested that is supported Inclusion of this disk module may cause unpredictable results or loss of data Disk States More than one log disk is installed This log None Information only Unused Log Disk disk will not be used Error Message Troubleshooting Control Panel Error Messages Table 12 Control Panel Error Messages Explanation Action Abterm Abnormal process termination has occurred Record the error code and call the support line Balancing The array is redistributing data among the disk None Information only modules for better performance Battery Charging The battery in a controller module is not fully If the array stays in this mode for more than charged but has enough power to allow normal one hour check the battery connection If the operation Also the battery may not be battery is properly connected replace the connected properly battery or replace the controller module Caution Always perform a successful Shutdown prior to replacing the batteries Battery The controller module batteries are discharged Wait 15 minutes for the battery to recharge Discharged or low Each set of batteries provides one week Controller batteries should be replaced every of back
198. or redundant solutions some power is likely being delivered to array enclosures from the remaining PDUs All power to the fan assembly has failed Also it is likely that two PDUs have failed and are not providing power to the arrays 254 Appendix D Back to Back Racking Problem Resolution Problem Resolution No LEDs lit power to the fan assembly has failed This is an indication the two of the PDUs and possibly the third are not supplying power to the arrays thus indicating that none of the array enclosures are operating This could be caused by a general power failure at the site tripped circuit breakers for the circuits delivering power to the rack or unplugged power cables If these are not the causes problem diagnosis of the rack is as follows 1 Are the PDU circuit breakers tripped Check the PDU circuit breakers The PDUs are located at the back of the rack The circuit breakers are black buttons located at the bottom of each PDU When circuit breaker is tripped the black buttons will be popped out Locate and remove the cause of any overload Reset the circuit breaker by pushing in on the black buttons 2 Are both plugs of the exhaust fan assembly plugged in Although unlikely it is possible that the exhaust fan assembly is not plugged in If necessary plug the exhaust fan assembly power cords into their receptacles CAUTION This condition is critical as the fans are not cooling the disk storage uni
199. ower Modules Troubleshooting Procedure If the disk array contains three power modules and the status lights on all three power modules are amber follow this procedure 1 Cycle the disk array power 2 After power on the power module status lights will be amber If all fan module status lights do not turn green before the array turns off replace the fan modules If at least one fan status light turns green after power on either a power module has failed or an over current condition exists Follow the steps under Power Module General Troubleshooting Procedure in this chapter Disk Module If a disk module failure occurs check the following 1 Asolid amber status light at any time other than during a self test indicates a disk module fault A flashing or pulsing green status light indicates read write activity 2 Replace the disk module Host SCSI Bus Check for SCSI bus problems as follows 1 Check the SCSI cable connections and also look for bent pins 2 Check the SCSI terminator connections A single ended terminator on a differential bus will cause the bus to hang 3 Check SCSI IDs for conflicting ID numbers 4 Check to see if the Host Bus Adapter includes on board termination and termination power or if these must be supplied externally 120 Troubleshooting Power On Sequence Test Power On Sequence Test When the disk array is powered on it goes through a power on sequence test during which time various
200. owing terms used in this column NA This event code is not applicable because it is never logged Ignore This event code is not relevant to predictive diagnostics Single Occurrence A single occurrence of this error is sufficient to warrant maintenance See accompanying errors This refers to other log entries which occur in conjunction with the event This indicates that a given event should always be accompanied by other events which lend it the appropriate significance Host reported error Indicates that there should be additional information in host logs describing the occurrence of an event if it is significant Ignore unless Indicates that this event normally accompanies other events or operator activity If the event appears in the log without accompanying events or activity then a single occurrence is sufficient to warrant maintenance Note that exceptions may exist based on drive type Recurs After Recovery This indicates that the error is expected under certain circumstances and a recovery procedure should be invoked The recovery procedure should be described in the description of the event If recovery is unsuccessful then maintenance is warranted 135 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Table 14 Controller Log Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 0 0x00 No
201. p traffic to the array Remove the failed controller After Initialization display panel shows Ready CAUTION It is very important that the array be in a Ready state The array should not present any warning other than a single controller warning statements at this point 59 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 6 Shutdown the array using the front panel KEY CANCEL CANCEL MENU ENTER ENTER DISPLAY Ready Language View Settings Shutdown Confirm Shutdown Complete After shutdown has completed push the on off button to the off position completing the shutdown This will turn off power to all modules 7 Disconnect all disk drive modules from the back plane You do not need to remove them completely Just pull them out so that the cam handle is fully extended 8 Pullthe existing good controller and reset disconnect and then re connect both batteries at the same time This will mean that the only copy of valid maps reside on the disk drive modules 9 Install both controllers Put the controller with the desired firmware into controller slot X 10 Push the on off button to the on position and allow the array to complete its initialization process After the initialization has completed the front panel should say Not Enough Disks Check the system state using the key strokes below The System state should be Not Eno
202. ported by the disk array include RAID 0 1 and RAID 5 RAID 0 1 is a disk array operating mode that provides high performance but is somewhat inefficient in its use of disk space RAID 0 1 implements Data Redundancy by keeping a separate copy of all data This mirroring technique consumes half of the disk capacity for Data Redundancy but provides maximum performance for servicing disk writes Distributing data across all disks called Block Mirroring provides a performance boost A disk array operating mode that is efficient in its use of disk space but suffers a performance penalty when performing write I Os As the disk array begins to fill up less frequently updated data is moved to RAID 5 space A diagnostic test that checks the integrity of the disk media by reading data from random locations on the disk A Read Verify test is non destructive that is it will not alter or destroy any of the data on the disk The disk undergoing the test remains in use by the disk array while the test is in progress A Rebuild is the process of recovering data that was on a failed disk The disk array reconstructs the data that was on the failed disk using redundant data from the remaining disks Until a Rebuild is complete the disk array is operating in a non redundant mode and is vulnerable to a second disk failure See also Auto Rebuild Rebuild Priority sets the priority of a Rebuild equal to high or lower than low host I Os The Rebu
203. ppendix D Back to Back Racking Product Description Figure 49 Back to Back Rack Configurations 2 0 meter rack no side access 2 0 meter rack with side access Sup exX 237 Racking Appendix D Back to Back Racking Product Description Other Documentation This document constitutes an overview of the back to back racking installation Detailed installation instructions accompany the other kits Other documentation includes E7687 90001 7687 A HP 200 240V Exhaust Fan Assembly Installation Guide A4915 96000 HP SureStore E Disk Array 12H Power Upgrade Kit Quick Installation Guide E7670 90001 Power Distribution Unit Kit Installation Instructions C2785 90002 Computer Support Rail Kit Installation Instructions Features The exhaust fan module contains three cooling fans The module operates in a redundant power configuration that is the module draws power from two of the rack s PDUs one fan draws from one PDU the other two fans draw power from another PDU If power from one of the PDUS fails one or two fans drawing power from the other PDU will continue to provide cooling for the rack If one of the fans fails physically the remaining fans in the fan module will continue to cool the rack However after a fan failure it is recommended that the failed unit be replaced within 48 hours so that disk drive reliability is not
204. re without the need for any action on your part This maintains performance because while the disk array is operating in non redundant or critical mode performance decreases Active Hot Spare provides an additional performance benefit because it is used as RAID 0 1 space until a disk fails Amount of cache DRAM Increasing the amount of cache DRAM can alleviate bottlenecks that may form in the controller data path This increases the rate at which the disk array can service host I Os Proper disk module installation Installing the disk modules in the correct slots in the enclosure distributes the disks across the internal SCSI busses and balances the I O load properly Disk modules should be installed from left to right top to bottom Keep the disk capacity small Having twelve disks with smaller capacities will improve the performance as compared to having twelve disks with larger capacities 105 sjdo uo Concepts Concepts and Management Optimizing Performance Controller Data Path The data path through the controller is designed to transfer data between the disks and the host as efficiently as possible However under heavy I O loads bottlenecks may form in the data path These bottlenecks can frequently be reduced by adding more hardware resources such as cache RAM or disks Figure 26 Data Path READ WRITE DECODER CACHE
205. reStore E Disk Array 12H the controller firmware of the disk array must be at revision HP54 or higher The use of 36 Gigabyte disk drives on controller firmware revision previous to HP54 is not supported If a 36 Gigabyte disk drive is manually included under older controller firmware it will be treated as if its physical capacity is about 24 Gigabytes AutoRAID Capacity Planning Tool For planning the maximum usable capacity that can be achieved refer to the AutoRAID Capacity Planning Tool on the World Wide Web as follows On the World Wide Web http www hp com go autoraid tool http www hp com essd modell2H autoraid tool html http www hp com essd capacity html Inside the HP firewall HP qualified personnel only 102 http essd boi hp com products DiskArrays autoraid capacityplanningtool index htm http essd boi hp com products DiskArrays autoraid capacityplanningtool capacity html Concepts and Management Managing the Disk Array Capacity Capacity Management Strategies The way you decide to configure your disk array capacity will be determined by your system needs and your disk array management strategy Are you willing to sacrifice maximum protection against disk failure for increased capacity Is performance an issue Table 3 describes the steps you can take to achieve your primary goal in managing the disk array The trade offs associated with each strategy are also described Table 3 Disk Arra
206. ree to fifty seconds and then go off Reset Cntrl resets controller x or controller y and runs internal diagnostics on the controller 76 Operating the Control Panel Using Control Panel Menus Format Array CAUTION Once Logical Drives LUNs have been created on your disk array you will not be permitted to Format the disk array again You must delete all of the existing Logical Drives before a disk array can be formatted Deleting Logical Drives LUNs and formatting the disk array destroys all data on the disk array Before formatting a disk array make backup copies of the entire disk array or you will lose all data Format Array clears all disk sets in the array data is lost all disk sets are lost and all NVRAM information is reset You must delete all existing Logical Drives before formatting the disk array Parity NOTE This command requires a SCSI Reset or power cycle to take effect Parity switches SCSI Parity on default or off for the primary array controller When parity is on it enables error detection in the primary array controller SDTR NOTE This command requires a SCSI Reset or power cycle to take effect Synchronous Data Transfer Request SDTR enables default or disables SCSI SDTR Synchronous Data Transfer Request SDTR controls the data rate on the SCSI bus used by the host and the array SDTR also determines the negotiation protocol of the host If SDTR is enabled the array will initi
207. res accompanyin recovery indicate that at least Suspected g errors one failure or hot plug During occurred in addition to the loss Recovery of NVRAM that the recovery This implies that data integrity likely to have been compromised in ways that are detectable outside of the occurrence of this log or sense data report It is likely that the controller does not have sufficient information to determine the extent of the damage 94 0x5e Volume Set Y Ignore None The volume set referenced in a Missing operator error recover command is not During present in the subsystem If a Recovery NULL volume set was referenced then this error indicates that no members of any volume set were present This error can also result from incompatible stamp versions 95 0x5f RAM Y Single Controller An attempt was made to Configuration Occurrence SIMM recover an image after RAM Mismatch operator error loss but the image would not During fit in the non failing memory available on the controller Either the SIMM configuration has been modified or one or more SIMMs have failed Recovery may be possible using another controller as the primary controller Recovery 155 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 96 0x60 Read Recovered With RAID 5 Retries Ignore See acco
208. roller If you did not select the primary controller use the TESTDISK and TESTLEVEL commands as detailed above to change to the primary controller To Download 15 16 17 18 19 20 21 Type DOWNLOAD You ll be presented with two options 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 Select option 1 Enter the firmware filename that is on the tape to download Type 2 to redisplay the firmware files At Do you want to do the update enter DO NOT INTERRUPT THIS PROCESS OR THE DEVICE COULD BE RENDERED INOPERATIVE You will receive several messages letting you know what is happening After a few minutes you will receive a message Firmware Downloaded Successfully Repeat this process for all other arrays on the system When finished downloading firmware to all arrays at ARDIAG gt type EXIT ALL At ISL type RESET 55 J9 0 ju0 Controller HP SureStore E Disk Array 12H Controller Firmware Download Procedure Download D Downloading Disk Array Firmware from a File on a V Class system using dfdutil Utility Assumptions Use this procedure when the desired version of controller firmware is available as a DFDUTIL LIF file for the V Class system CAUTION Controller firmware HP4x and any later versions have a dif
209. rray enclosures This solution is known as back to back racking because both front and back areas of the rack are used and the disk arrays are oriented back to back in the rack Back to Back installation of disk arrays requires several modifications be made to the rack prior to installing the additional arrays These modifications include Installation of power upgrade kit three 19 PDU assemblies replacing the existing 56 PDUs Installing a high volume exhaust fan assembly and Removing the rear rack door and hardware Field upgrade kits are available for the above modifications and need to be available prior to beginning the installation see Table 19 Upgrade Kits and Components Prior to modifying the rack for back to back installation you need to determine if the maximum number of disk arrays can be installed Six disk arrays completely fill a 2 0 meter rack and four disk arrays fill a 1 6 meter rack known as high density configuration Depending on the service access method the maximum number of arrays that can be installed varies If the rack is located to allow a side panel to be removed for service access for example if the rack is located at the end of a row or can be rolled out then the rack can be filled completely high density If the rack is situated such that side access is not possible then one position one disk enclosure space must remain open to allow for service access see Figure 49 NOTE The 1 6 meter ra
210. rray that IS NOT used as a boot device and the Version of Firmware on the Replacement Array Controller DOES NOT Match Version of Firmware on the Installed Array Controllers CAUTION Controller firmware HP4x and any later versions have a different data map format in NVRAM memory Once you have HP4x or later you can not downgrade to firmware version prior to HP4x without first completely backing up all data on the array including any meta data e g LUN and LVM configuration After you have downgraded the firmware and formatted the array all meta data and real data will need to be restored from backup NOTE This is an offline procedure All I O to the array will need to be suspended for duration of this procedure Assumptions Use this procedure if an existing array controller has failed if the disk array a non boot device and the replacement array controller DOES NOT have the same version of firmware as the installed array controllers 1 Ifreplacement controller is new install the controller batteries supplied 2 Ifreplacement controller is NOT new reset disconnect and then re connect both batteries on the replacement controller at the same time to erase NVRAM 3 Verify which controller has failed by examining the logs using the 1ogprint command This is to make sure you do not remove the functioning good controller 4 Quiet quiesce or stop all I Os deactivate volume set unmount file systems sto
211. s Language View Settings System State Not Enough Disks Primary Cntrl X 61 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 14 15 16 17 18 19 20 21 62 Verify that both controllers have the desired firmware using the front panel The following table shows the proper front panel commands to view the firmware revision installed KEY DISPLAY CANCEL CANCEL Not Enough Disks MENU Language View Settings ENTER System State Not Enough Disks Primary Cntrl X SCSI ID x _ SCSIIDy Firmware HP Push on off button to off position and wait for the power to turn off Remove both controllers from slot X and Y and reset disconnect and then re connect both batteries on each controller at the same time Re install both controllers It doesn t matter which controller goes into which slot at this time Plug in all drive modules Push on off button to the on position and wait for the initialization to complete The disk array should display the Ready state Check the system state There should be no other warnings KEY DISPLAY CANCEL CANCEL Ready MENU Language View Settings ENTER System State Ready Primary Cntrl X If the array does not show up using ARMServer arraydsp i you may have to re connect to see the array using the ARMServer arraydsp R com
212. s from RAID 0 1 so moving data that may only be read to RAID 5 has little impact on performance Figure 19 Moderate Data Storage BD on RAID 0 1 RD 00 0 1 RAID LRAIDOIT p RAID 0 1 lt READ RAIDS RAD S 5J 5 DISK 1 DISK 2 DISK 3 DISK 4 86 Concepts and Management Disk Array Features Stage 3 High Data Storage As the disk array nears its maximum storage capacity most data is stored in RAID 5 The array always reserves a certain amount of capacity for RAID 0 1 to maintain an adequate level of performance The disk array continues to try to keep the most recently updated data in RAID 0 1 but if the array is near maximum capacity this becomes more difficult As long as the disk array can continue to service most writes from RAID 0 1 performance is maintained But if the number of writes exceeds the current available RAID 0 1 space the disk array must begin servicing writes from RAID 5 and performance begins to suffer When data residing in RAID 5 must be updated it is usually necessary to move the data to RAID 0 1 before performing the write Figure 20 High Data Storage wew RAID 0 1 REID OI O14 om RAIDS RAIDS RAIDS DISK 1 DISK 2 DISK 3 DISK 4 87 sjdo uo Concepts Concepts and Management Disk Array Features Logical Drives LUNs The entire capacity of the array is divided into entities called logical drives or LUNs The host operating system treats each
213. s installed from the power upgrade kit are not wired through the rack power switch The rack power switch at the top of the rack is disconnected and no longer provides power switching To disconnect power to the rack you must unplug each of the three PDUs 252 Indicator LEDs Appendix D Back to Back Racking Troubleshooting The exhaust fan assembly has four indicator LEDs located at the top front of the rack The LEDs show the condition of the exhaust fan assembly and power to the fan assembly Figure 54 Rack Indicator LEDs LED 1 LED 3 d HEWLETT Q ouo Q LED 2 LED 4 LED 1 Green PDU 1 and PDU 1 fans normal LED 2 Amber PDU 1 fan failure LED 3 Green PDU 2 and PDU 2 fans normal LED 4 Amber PDU 2 fan failure 253 Sur e Racking Appendix D Back to Back Racking Troubleshooting Indicator LED States This is the rack s normal operating state Power is being delivered from both Both green PDUS and to all the fans Fans should be running normally LEDs lit One or two of the three fans are not operating probably due to a physical failure LED lit Two Two or three fans have failed This condition is unlikely to occur amber LEDs lit One bank of LEDs not lit When one bank of LEDs is not lit no green no amber power from one of the PDUS has failed One or two of the cooling fans are not spinning This is a likely indication that the PDU is not supplying power to the rack F
214. s not the same as actually freeing space within the disk array The file system data still occupies the same amount of space on the disk array so deleting files from the file system will not release capacity back to the disk array for use as RAID 0 1 space The key point is this Once data has been written to the disk array you cannot reclaim the space consumed by that data simply by deleting files from the operating system To truly erase the data and reclaim the capacity for use as RAID 0 1 space you must delete the logical drive on which the data resides Of course this will erase all the data on that particular logical drive so you must make sure you backup any critical data on a logical drive before you delete it 111 s3do uo Chapter 5 Troubleshooting This chapter describes how to troubleshoot the disk array if a failure is indicated by module status lights or control panel messages Module Slot Numbers Each module has a slot number which can be displayed in control panel messages see Figure 5 The index number is embossed near each module slot The indexes for the disk modules have two parts column letter A or B and row numbers 1 through 6 Module Status Lights Each module has a status light located on its front panel see Figure 4 Table 5 through Table 8 list the status light indications and actions for each of the modules NOTE The control panel status light changes from amber testing to green ready durin
215. s the operation of its partner Regardless of how the failure is detected it is ensured that operation is automatically handled by the second array controller if the first array controller ever fails Following are the firmware behavior scenarios if a controller fails If a controller fails in a dual controller configuration the redundant controller may be offline performing necessary initialization routines that check for data integrity The following scenarios depict the offline activities that will occur 1 Hot plug event of either controller This will cause the redundant controller to invoke a full initialization test cycle This test cycle is required in order to insure data integrity of the NVRAM contents as well as back end disk drives that were affected by removing a SCSI device the controller from the bus e The controller that is hot plugged will also issue a SCSI reset on the host side This will invoke host initialization routines as well e Time offline is approximately 2 minutes 2 Afailure in the secondary controller other than a hot plug event The primary controller will not need to issue an initialization test cycle e There should be NO offline time 3 A failure if the primary controller other than a hot plug event e This will cause the redundant controller to invoke a full initialization test cycle This test cycle is required in order to insure data integrity of the NVRAM contents and to set up the contr
216. s unterminated for High Availability A4915A 200 240 VAC International PDU Non US installations PDUs for Opt ABB 10 IEC 320 receptacles line cord is unterminated High Availability A4915A 200 240 VAC UPS PDU US or Non US installations Opt 024 10 IEC 320 receptacles 4 5M line cord is UPS PDUs for High availability 2 Upgrade the rack s cooling capacity Back to Back racking requires that the cooling capacity of the rack be upgraded The cooling upgrade requires installation of a new Exhaust Fan Assembly product E7687A This assembly contains three fans an LED fault indication circuit and a 1 EIA unit air scoop This is required to promote maximum airflow through the center of the rack Also to promote additional air flow the rear cabinet door will be removed and set aside To maintain cooling efficiency filler panels should be installed in any unfilled spaces in the front and back of the rack see Table 19 239 Sur ey Racking Appendix D Back to Back Racking Installation 3 Add rail kits The upgrade requires enough rail kits part C2788A to accommodate the number of disk storage enclosures to be racked A 2 0 meter rack requires three rail kits and a 1 6 meter rack requires two rail kits If the rack is completely filled with front racked disk storage enclosures then no additional rail kits are required 4 Add power cables and SCSI cables The upgrade requires ordering enough power and SCSI cables to serve all in
217. sed to store data recovery map information If the data maps in NVRAM memory are lost you may be able to recover them with the arrayrecover utility The arrayrecover utility and the data recovery map settings and parameters are described in more detail in the System Administrator s Guide which is shipped with each disk array The data resiliency feature for HP AutoRAID provides a means of recovering data after loss of NVRAM when the disk array is not shut down properly Such a loss can only occur due to dead battery battery mishandling a single failure in single controller configuration or a multiple failure in dual controller configuration The data maps in NVRAM memory are reconstructed using the latest copy of mapping information stored on the disk drives This reconstruction includes a combination of the maps which were written during Shutdown and the data recovery maps which are written at regular intervals Part of the recovery operation includes performing a parity scan on the contents of the entire disk array to validate the accuracy of the maps and to correct any parity inconsistencies This process can take up to several hours depending on the amount of data on the disk array Controller firmware places a limit on the amount of time either data or map information can reside in battery backed RAM before the process of writing that information to disk is begun The firmware can optionally eliminate this limited exposure which it automaticall
218. ssembly e The spring fits between the front of the enclosure and the power button The clock hand on the power button should point upward Removal and Replacement AC Power Receptacle HP qualified personnel only AC Power Receptacle HP qualified personnel only CAUTION The following procedures require that you perform a shutdown turn off the disk array and make it inaccessible Check with the system administrator before proceeding To replace the ac power receptacle Figure 34 Item 1 1 Remove appropriate power module and its power cord 2 Remove the two T 15 screws 2 from the ac power receptacle Remove the ac power receptacle Be careful not to drop the screws into the chassis since they are difficult to locate if they fall below the chassis tray 3 Replace in reverse order ensuring that the metal spacer is inserted between the ac power receptacle and the back of the chassis Also make sure that the larger end of the ac power receptacle is installed toward the outside of the chassis and that the ground pin on the ac power receptacle is facing upward 201 IEAo ur qI Removal 202 Removal and Replacement AC Power Receptacle HP qualified personnel only Figure 34 Replacing the AC Power Receptacle 1 AC Power Receptacle 2 Screw with Captive Lock Washer Removal Replacement Module Cam Lever HP qualified personnel only Module Cam Lever HP qualified personnel only To replace t
219. ssigned to a Logical Drive is available to the operating system The data contained in each Logical Drive is spread across all disks in the array so there is no direct correlation between a Logical Drive and any single physical disk in the array A unique number 0 7 assigned to each Logical Drive on the disk array This number is used by your operating system to identify each array Logical Drive The unique ID of a node in Fibre Channel loop topology sometimes referred to as a Loop ID Also a status type in the FC Status Menu showing the Loop Address of the FC SCSI MUX Logical Unit Number For consistency with UNIX terminology LUN is used interchangeably with Logical Drive See Logical Drive 265 JUSSO Glossary Glossary Maximum Environmental Limit The maximum limit of temperature humidity vibration shock and Migration Mirrored Disks Multiplexer MUX Offline Online Previously Used Disk Primary Array Controller 266 altitude which should not be exceeded during operation of an the product See Recommended Operating Range Migration is the movement of data between one RAID mode and another RAID mode A Mirrored Disk is an entire duplication of data Traditional mirrored disks require disks of identical capacity The disk array uses a special type of mirroring called RAID 0 1 mode or Block Mirroring A device that allows two or more signals to be transmitted simultaneously on a single
220. st fan assembly a b 4 Unplug the exhaust fan assembly power cords from the two PDUs Wait for the fans to stop spinning Remove the two mounting screws at the top rear of the rack Lift the back of the exhaust fan assembly slightly and then pull the assembly back a few inches Lift the exhaust fan assembly off the rack NOTE It is not necessary to remove the 1U air scoop when replacing the exhaust fan assembly 2 Install the new exhaust fan assembly a b 4 Working from the rear of the rack raise the exhaust fan assembly above the rack Tilt the front edge of the exhaust fan assembly the side that is farthest away from you into the top of the frame opening Align the exhaust fan assembly s edges with the rack frame Slide the exhaust fan assembly forward engaging the tabs of the exhaust fan assembly into the slots in the front of the rack Route the two power cords inside the rack frame When you have done this correctly the assembly s power cords should be oriented around any installed array enclosures and with the assembly s power cord plugs near the PDUs Lower the back end of the exhaust fan assembly into place Replace the two mounting screws in the exhaust fan assembly 257 Racking Appendix D Back to Back Racking Specifications Specifications Electrical e Input Voltage 200 230 240 VAC 16 A max per PDU e Input Frequencies 50 Hz 60 Hz singl
221. stalled disk enclosures Standard HP power cables 8120 6514 will meet all power cable requirements Standard HP SCSI cables will meet all SCSI cable requirements 5 Back to Back Array Density Depending on whether service accessibility is available through the side of the rack or not determines the number of disk arrays that can be installed Side access requires that the facility housing the racks allow HP service personnel sufficient room at either side of the rack This space should be approximately 2 feet This clearance will allow HP service personal to remove the side panel of the rack and gain access to the components inside If side access is not possible then the capacity of back to back racking is reduced Refer to Figure 49 240 Appendix D Back to Back Racking Installation Brief Summary of Installation Steps The sequence below describes the general approach for installing the field upgrade for back to back racking These steps may vary slightly depending on the particular installation The installation process is described in more detail in the following Installation Procedure section Bor Qvi Confirm power requirements Take disk arrays offline and power down rack Disconnect all power to the rack Remove the rack top sides and back door and door hinges and latch catch the door will not be re installed and can be set aside Remove existing 56 PDUs If necessary remove array enclosures array enclosures
222. t has been type tested and found to comply with the limits for a Class A digital device pursuant to part 15 of the FCC Rules See instructions if interference to radio reception is suspected For Europe this equipment is designed to meet EC EN55022 CISPR 22 Level A e Magnetic interference Magnetic Operating lt 5 Gauss peak to peak on all surfaces of product Magnetic Nonoperating lt 2 mGauss at 7 ft 2 m Acoustic Emissions 6 6 bels Environmental Requirements Temperature Ranges e Operating 5 to 40 C 41 to 104 F maximum rate of change 20 C hour 36 F hour Recommended Operating Range 20 to 25 C 68 to 78 F e Nonoperating Disk Module 40 to 65 C 40 to 149 F Humidity Ranges Maximum Wetbulb Temperature 28 C 82 e Operating Disk Module 10 to 8096 RH noncondensing e Nonoperating Disk Module 10 to 90 RH noncondensing 213 soodg Specs Appendix A Product Specifications Environmental Requirements Altitude Ranges e Operating 0 to 3 048 m 0 to 10 000 ft e Nonoperating 0 to 4 572 m 0 to 15 000 ft Vibration e Operating Random Vibration 0 21 Grms 5 to 500 Hz e Nonoperating Random Vibration 2 09 Grms 5 to 500 Hz e Nonoperating Vibration Sine Sweep 0 5 G 0 to peak 5 to 500 Hz Shock e Operating Shock 6 35 mm 0 25 in tilt drop independent of weight Line Input Requirements e Nominal Voltages 100 127 Vac 200 240 Vac e Maximum Input Power
223. t may be busy performing data retries 149 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Number DEC hex Event Logged Predictive Maintenance Implication Suspected Components Description 71 0x47 Internal SCSI N NA None A command failed due to a Hot Plug hotplug event or possibly due to an error that caused error recovery 72 0x48 Host Write Y Ignore see None An earlier host write was Collision With accompanying flushed from cache but was Failed Cache errors not written successfully to the Post disk This caused cache to mark the write cache entry as RETRY which means it is stuck in cache A subsequent host write involved the same address as the write stuck in cache and the controller failed the new write 73 0x49 Disk Drive Y Single Disk Drive The disk drive returned a Defect List Occurrence status indicating a disk drive Error defect list error The controller s retries are exhausted or the I O is not retryable A low level format of the disk drive may correct this problem 74 0x4a Disk Drive Y Single Disk Drive The disk drive returned a Format Error Occurrence status indicating a format error problem The controller did not attempt to send a Reassign Blocks command This may be cause for immediate failure A low level format of the disk drive may correct this problem This event may occur following the download of new firmware t
224. tallation Procedure Table 21 Disk Array Component Positioning EIA Space Component Component Rail Clip Nut Aray Enclosure Flange Installed in Installed ear Standard Clip Nut Front of Rack Back of Rack Hole Position ear Standard Hole Positions 2 meter Rack 1 13 Array 1 Array 4 3 7 33 14 PDU n a n a 15 27 Array 2 Array 5 45 49 75 28 40 Array 3 Array 6 84 88 114 41 Filler Air Scoop n a n a 1 6 Meter Rack 1 13 Array 1 Array 3 3 7 33 14 26 Array 2 Array 4 42 46 72 27 29 Filler s Filler s n a n a 30 Filler PDU n a n a 31 Filler Filler n a n a 32 Filler Air Scoop n a n a 1 BIA spaces and holes are numbered from the bottom of the rack up 2 Arrays are numbered from the bottom up beginning on the front of the rack 1 through 3 or 1 through 2 then from the back of the rack bottom up 4 through 6 or 3 through 4 244 Appendix D Back to Back Racking Installation Procedure Figure 50 Rack EIA Array and Component Positioning AIR 5 00 d 61 HP SureStore E Disk Array A3700A Location 6 HP SureStore E Disk Array A3700A Location 5 61 E7676A 19 PDU 2 HP SureStore E Disk Array A3700A Location 4 2 Meter Rack AIR SCOOP FILLER PANEL E7676A 19 PDU 2 FILLER PANEL FILLER PANEL FILLER PANEL dd 61
225. the ATE ius oris RR E RR E CR HERR ERIT ERE RR OR EEE 74 Shutting Down the Array spi e EE EU re Dr e Re er 74 Setting the Language i ge Tee dee EUH e iege pede 74 USING CONTROL PANEL 05 75 PCT BUG RCo EET 76 Cntr Ch nges iioi eee Pa edi e dete eerte ge pe 76 Disk Changes 5 cene ete rem edere Qa wa unu td 79 Logical DIV Changes i aset ere ede Rer e ede HRS e eee edges 61 Nanni mM 82 View Settings dt asse deae edd 82 CHAPTER 4 CONCEPTS AND MANAGEMENII I 83 DISK ARRAYS IN BRIEF eite 83 DISK ARRAY FEATURES eto iiir endete mr reet ix ee rect leone lee eden 84 Dyn micDat Migr tion x ene ee ep ee reve dpi ted i 64 Logical Drives di eei er RR RE E 88 AULO INCLU vases 89 Active Hot SPA re ade i n ota OQ eO RD UR D 89 90 UIT Em 91 Mme ERU 92 DISK ARBAY tiene pBOU I DICE DIU Fr OH XR PRIN TR e RP E Eia 93 Disk Array Controller Module lan asna anu aa yma netten tenet en nene enne EAEE a 93 Disk
226. the controller that was noted as primary in one of the previous steps Install the third controller into slot X This may mean that you have to move the existing controller into slot Y Push the on off button to the on position and allow the array to complete its initialization process After the initialization has completed the front panel should say Not Enough Disks Check the system state using the key strokes below The System state should be Not Enough Disks with a Firmware Needed warning KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed 47 J9 0 3u0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure 10 Verify if the primary controller currently installed has the desired firmware using the front panel display The following table shows the proper front panel commands to view which controller is the primary array controller and what version of firmware is on the primary array controller KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed Primary Cntrl X SCSIID x __ SCSIIDy Firmware HP 11 Copy the firmware from the primary controller to the secondary controller using the front panel KEY CANCEL CANC
227. total number of disk installed in the array plus any missing drives ever exceeds 16 this condition is logged The disk array always generates a data unavailable warning before this condition is reached Once the 16 disk limit is reached the controller will not recognize any additional drives installed in the array Because the controller cannot add the new disks to the drive list the status light on the newly installed disk will remain amber To correct his problem and maintain data integrity find and reinstall the missing disks If the data on the disk array is not critical the disk array can be reformatted with the new disks 160 Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 143 0x8f Duplicate Disk Y Single Controller As part of its stamp each Drive Identifier Occurrence Erroneous drive indicates its logical Operator position in the disk set to Activity which it is attached This event indicates that the controller has detected two disk drives with identical stamps indicating that they occupy the same logical position of the same disk set The most likely source of this error is the erroneous assignment of duplicate disk set ID s by the controller This would generally originate with incorrect controller serial number assignment and or loss of NVRAM This error would also occur if a
228. troller 131 Second installable SIMM on X controller 132 First installable SIMM on Y controller 133 Second installable SIMM on Y controller 134 Other non reporting controller 135 Battery 136 Fan 1 137 Fan2 138 Fan 3 139 Power Supply 1 140 Power Supply 2 141 Controller X 142 Controller Y 143 Power Supply 3 192 Internal SCSI Bus 0 193 Internal SCSI Bus 1 194 Internal SCSI Bus 2 195 Internal SCSI Bus 3 The reporting controller is the controller identified in the Component Specific ID field of the Controller Error Event specific data 133 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Code Descriptions Table 14 lists all event codes that can appear in the HP XLR1200 1255 Advanced Disk Array controller log event pages The table also includes events that are not logged but can still generate SCSI sense data Appropriate corrective action is included for all events that require it Many events are informational and require no action on the part of the user The Suspected Component column lists the hardware components potentially implicated by the event code In general the most likely failing component is indicated in the FRU field of the event descriptor The components are listed in decreasing order of likelihood Term Definitions The following terms used in the table may require some clarification 134 Disk Drive ID Number this is the uniqu
229. troller is set as the primary array controller If there is a controller fault it does not indicate the faulty controller 123 Sunoouso qno Troubleshooting Troubleshooting Control Panel Error Messages Error Message Table 10 Control Panel Error Messages System States Explanation Action Warning System States An Active Hot Spare request cannot be Either add more data capacity disks or disable ActivSpare honored the array no longer has the physical Active Hot Spare Warning capacity necessary to enable an Active Hot Spare System States The battery in a controller module is not fully If the array stays in this mode for more than Battery Charging charged but has enough power to allow normal one hour check the battery connection If the operation Also the battery may not be battery is properly connected replace the connected properly battery or replace the controller module Caution Always perform a successful Shutdown prior to replacing the batteries System States The capacity threshold alarm has occurred the Add more capacity disk modules and reset the Capacity threshold set for capacity consumption has capacity to a higher threshold or disable the Warning been exceeded capacity warning System States The X or Y controller has failed Replace the X or Y controller Cntrl Failed System States Dual controllers are not able to communicate Replace faulty contr
230. ts in the rack and disk drive reliability may be compromised When the power fail condition is corrected both green LEDs on the rack front should light indicating that at least two of the PDUs are functioning 255 Racking Appendix D Back to Back Racking Problem Resolution One bank of LEDs not lit When one bank of LEDs is not lit no green and no amber power to one of the PDUs has failed One or two of the cooling fans are not spinning The remaining two PDUS are likely to be providing power to the arrays This could be caused by tripped circuit breakers for the circuits delivering power to the rack or unplugged power cables If these are not the causes problem diagnosis of the rack is as follows 1 Isa PDU circuit breaker tripped Check the PDU circuit breakers The PDUs are located at the back of the rack The circuit breakers are black buttons located at the bottom of each PDU When the circuit breakers are tripped the black buttons will be popped out Locate and remove the cause of any overload Reset the circuit breakers by pushing in on the black buttons 2 Are both plugs of the exhaust fan assembly plugged in Although unlikely it is possible that one of the power cords for the exhaust fan assembly is not plugged in If necessary plug the exhaust fan assembly power cord into its receptacle Plug each of the fan assembly power cords into a different PDU When the power fail condition is corrected t
231. ugh Disks with a Firmware Needed warning KEY CANCEL CANCEL MENU ENTER 60 DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed HP SureStore E Disk Array 12H Controller Firmware Download Procedure 11 Verify if the primary controller currently installed has the desired firmware using the front panel display The following table shows the proper front panel commands to view which controller is the primary array controller and what version of firmware is on the primary array controller KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disks Language View Settings System State Not Enough Disks Warning Firmware Needed Primary Cntrl X SCSIID x _ SCSI ID y _ Firmware HP 12 Copy the firmware from the primary controller to the secondary controller using the front panel KEY CANCEL CANCEL MENU ENTER ENTER ENTER DISPLAY Not Enough Disks Language Cntrl Changes SCSI ID Copy Firmware Confirm Copying Firmware then Firmware Loading then array will initialize then Not Enough Disks 13 The front panel display should say Not Enough Disks Check the System State on the front panel display It should say Not Enough Disks with no other warnings KEY CANCEL CANCEL MENU ENTER DISPLAY Not Enough Disk
232. ules 7200 rpm Option 312 Twelve 36 4 Gigabyte Disk Modules 7200 rpm Option 164 Four 9 1 Gigabyte Disk Modules 10 000 rpm Option 165 Five 9 1 Gigabyte Disk Modules 10 000 rpm Option 168 Eight 9 1 Gigabyte Disk Modules 10 000 rpm Product Description Options Option 172 Twelve 9 1 Gigabyte Disk Modules 10 000 rpm Option 184 Four 18 2 Gigabyte Disk Modules 10 000 rpm Option 185 Five 18 2 Gigabyte Disk Modules 10 000 rpm Option 188 Eight 18 2 Gigabyte Disk Modules 10 000 rpm Option 192 Twelve 18 2 Gigabyte Disk Modules 10 000 rpm Option 314 Four 36 4 Gigabyte Disk Modules 10 000 rpm Option 315 Five 36 4 Gigabyte Disk Modules 10 000 rpm Option 318 Eight 36 4 Gigabyte Disk Modules 10 000 rpm Option 322 Twelve 36 4 Gigabyte Disk Modules 10 000 rpm Option 200 One 96 Megabyte Controller Module Option 203 Two 96 Megabyte Controller Modules Option 801 0 9 m SCSI Cable high density 68 pin to 68 pin Option 802 2 5 m SCSI Cable high density 68 pin to 68 pin Option 803 5 0 m SCSI Cable high density 68 pin to 68 pin Option 804 10 0 m SCSI Cable high density 68 pin to 68 pin Option 806 1 0 m SCSI Cable high density LP 68 pin to 68 pin Option 807 2 5 m SCSI Cable high density LP 68 pin to 68 pin Option 808 5 0 m SCSI Cable high density LP 68 pin to 68 pin Option 809 10 0 m SCSI Cable high density LP 68 pin to 68 pin Option 840 2 0 m SCSI Cable high density V Cable to 68 pin Option 841 2
233. up time The controller can operate on three years or sooner if the display module only one of the two batteries installed within indicates a constant Battery Discharged each controller the second battery is for message If the message persists after replacing redundancy the batteries call the support line Cntrl Failed The primary array controller failed Replace the primary array controller Cntrl x y Failed The controller module in slot X or Y failed Replace controller module X or Y If the array has a single controller install a replacement controller and power on the array before removing the failed controller Disk Not The SCSI command to Include Disk failed The disk module may be malfunctioning or Included unsupported View settings menu of front panel If the state shows a disk module failure replace the disk module Otherwise call the support line Disk Xn Failed The disk module in slot Xn failed where Replace the disk module X A B n 1 6 Display Failed control panel display was temporarily Call the support line Replace the display unable to fully display a character module 127 Sunoouso qno Troubleshooting Troubleshooting Control Panel Error Messages Error Message Fan Failed Explanation A fan in slot F1 F2 or F3 failed Action Check the fan status lights to determine which fan module failed then replace the fan module Caution Do not remove a f
234. ure Up to four disk array enclosures can be installed into a 1 6 meter cabinet and up to six disk array enclosures can be installed into a 2 0 meter cabinet The disk array is also available in a deskside cabinet configuration The base models available are A3700A HP SureStore E Disk Array 12H Field Integrated Rackmount Configuration A3700AZ HP SureStore E Disk Array 12H Factory Installed Rackmount Configuration A3700AD HP SureStore E Disk Array 12H Deskside Cabinet Configuration uondrpsoq Description Product Description Options Options Options include the following items 10 Option 002 Third Power Module Option 104 Four 4 3 Gigabyte Disk Modules 7200 rpm Option 105 Five 4 3 Gigabyte Disk Modules 7200 rpm Option 108 Eight 4 3 Gigabyte Disk Modules 7200 rpm Option 112 Twelve 4 3 Gigabyte Disk Modules 7200 rpm Option 124 Four 9 1 Gigabyte Disk Modules 7200 rpm Option 125 Five 9 1 Gigabyte Disk Modules 7200 rpm Option 128 Eight 9 1 Gigabyte Disk Modules 7200 rpm Option 132 Twelve 9 1 Gigabyte Disk Modules 7200 rpm Option 144 Four 18 2 Gigabyte Disk Modules 7200 rpm Option 145 Five 18 2 Gigabyte Disk Modules 7200 rpm Option 148 Eight 18 2 Gigabyte Disk Modules 7200 rpm Option 152 Twelve 18 2 Gigabyte Disk Modules 7200 rpm Option 304 Four 36 4 Gigabyte Disk Modules 7200 rpm Option 305 Five 36 4 Gigabyte Disk Modules 7200 rpm Option 308 Eight 36 4 Gigabyte Disk Mod
235. user made an image copy of a drive The solution to this difficult problem involves two steps 1 Try to determine which of the disk drives share the conflicting stamp This can be done by removing disk drives until the problem disappears Concentrate on any previously used or unincluded disks first 2 When you have isolated the two conflicting disks determine which one contains valid data 161 Sunoouso qno Troubleshooting Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Number Logged Maintenance Components DEC hex Implication 144 0x90 Disk Drive Y Ignore None Installed Operator Activity 145 0x91 Disk Drive Y Ignore None Added To Disk Operator Set activity 162 Description The controller has detected the installation of a new disk drive This event occurs regardless of the disposition of the drive toward the disk set The disk drive ID number assigned to the drive installed appears in the component identifier field of this log entry A disk drive ID number consisting of the controller serial number followed by a 0 indicates that the disk drive ID number has not yet been assigned The controller has included a new disk drive in the disk set This event occurs only when a disk drive that was not previously a member of the disk set becomes a member It does NOT occur when a down failed or missing disk is returned to read
236. using the following command dd ifz the DFDUTIL VHPxx LIF of dev rmt tapedrive bs 2k 3 Run dfdutil and download HPxx firmware 57 J9 0 ju0 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Replace A Replacing an Array Controller Version of Firmware on the Replacement Array Controller Matches Version of Firmware on the Installed Array Controllers Assumptions Use this procedure if an existing array controller has failed if disk array is either a boot or non boot device and the replacement array controller has the same version of firmware as the installed array controllers 1 Ifreplacement controller is new install the controller batteries supplied 2 Ifreplacement controller is NOT new reset disconnect and then re connect both batteries on the replacement controller at the same time to erase NVRAM 3 Verify which controller has failed by examining the logs using the Logprint command This is to make sure you do not remove the functioning good controller 4 Remove failed controller After Initialization display panel shows Ready 5 Install the replacement controller in the disk array After initialization display panel shows Ready 6 disk array will display the Ready state when both versions of firmware match 58 HP SureStore E Disk Array 12H Controller Firmware Download Procedure Replace B Replacing an Array Controller In an A
237. ver utility A front panel request to begin a rebuild occurred The controller has posted the NVRAM to the disks and the array is ready to be turned off The array is malfunctioning Call the support line The array is malfunctioning Call the support line The array is malfunctioning Call the support line None Information only None Information only Shutting Down The controller is preparing to enter the Shutdown state None Information only SIMM 1 2 x y Failed SIMM 1 or 2 on controller module X or Y failed Replace the controller SIMM Count Error The controller modules each have a different number of SIMMs installed You must have the same amount of memory on both controller modules Add or remove SIMMs from the NEW controller module so that each controller has the same amount of memory DO NOT remove any SIMMs from the original single controller module you must always have the same amount of memory or more on each controller module as you had on the original single controller module SIMM Error Single Controller The primary controller does not have enough SIMMs to upload the present disk volume set Resiliency parameters are not set to single controller mode but the array has a single controller installed Add more SIMMs to the controller module s Each controller must have the same amount of memory Either put the d
238. want to save or move the files to another logical drive Delete L Drv Deletes a logical drive SCSI LUN number 0 to 7 When you delete a logical drive its capacity is added to the unallocated capacity of the array CAUTION Before renumbering a logical drive you should know the impact this will have on the operating system After you renumber a logical drive make sure you perform the necessary steps to configure the renumbered logical drive into the operating system Renumber L Drv Renumbers any logical drive SCSI LUN number 0 to 7 to any logical drive SCSI LUN number 0 to 7 One reason for renumbering a logical drive is if you want to swap logical drive 0 with another logical drive you want to serve as the system boot drive 81 poueg 10002 Control Panel Operating the Control Panel Using Control Panel Menus Shutdown NOTE This command requires a SCSI Reset or power cycle for the controller to become ready Shutdown is a single command and has no menu items Shutdown makes the array unavailable to the host When a Shutdown is initiated the array completes any I Os in progress posts array information stored in controller NVRAM to disks in the array then goes offline A Shutdown should be performed prior to the following procedures e Powering off the disk array for more than a week e Replacing the batteries in a controller module e Moving all of the disk modules to a new array and retaining all data
239. with correctable or uncorrectable memory errors or with parity errors If no other condition is present this event is reported to indicate the malfunction 174 Event Number DEC hex 195 0xc3 Event Predictive Logged Maintenance Implication Event Name Incompatible Y Address Table In RAM Image Ignore Operator error Suspected Components None Troubleshooting Event Code Descriptions Description The controller has detected that the amount of NVRAM available for the address table has been reduced The disk array uses all available NVRAM to create the largest possible address table which increases the total amount of usable capacity the controller can support Once the size of the address table has been established it cannot be reduced This event is typically caused by removing a SIMM from the controller Reinstall the SIMM to correct the problem 196 0xc4 Non Volatile Y SIMM Contents Lost Ignore unless no operator activity SIMM Controller This error occurs when NVRAM has failed or has been removed at a time when its contents were critical to disk array operation not shutdown The NVRAM on the affected controller is invalidated due to missing contents in the additional NVRAM 197 0xc5 Battery Failed Y Discharge Test Single Occurrence Batteries A battery has failed the long discharge test This test occurs following power on and every 30 days therea
240. wn the disk array turn off the disk array disconnect the ac power cords and SCSI interface cables Remove all modules fans disks power supplies and controllers from all bays Remove the two rear cover captive screws as shown in Figure 33 Item 6 Tilt back the top of the rear cover 5 and remove the rear cover completely from the disk array 1 Remove the four standoffs 3 and their lock washers Note carefully the position of the standoffs for later reassembly Remove the two T 15 M4 screws 4 that secure the display module to the backplane assembly Remove the display module from the front of the disk array Remove the eleven M4 nuts 8 from the backplane assembly Note that the three nuts across the top of the backplane assembly also have flat washers 7 beneath the nuts Do NOT remove any Pozidriv or Phillips screws from the backplane Remove the backplane assembly 2 Reassemble in reverse order Align the push rod with the switch on the backplane assembly and check for proper operation of the power button before attaching all hardware Removal and Replacement Backplane Assembly HP qualified personnel only SCSI Bus Reset Switches HP qualified personnel only CAUTION After replacing the backplane assembly make sure that the SCSI bus reset switches are set as shown in Figure 33 Altering the default factory settings on the SCSI bus reset switches may cause operating system conflicts for the host computer s The
241. x is the default when only one array controller is installed 22 Product Description Controller Module Suppressing Single Controller Warning Messages When a single controller disk array is powered on for the first time the control panel may display a System Warning error message which occurs because only one controller is installed By default the disk array assumes that there should be two controllers installed and therefore it will generate single controller error messages until you suppress them If a System Warning error message is displayed on the front panel you must first verify that it is caused by the detection of a single controller Use the control panel keys to verify the source of the message by using the following key sequence KEY DISPLAY CANCEL CANCEL System Warning MENU Language View Settings ENTER System State System Warning Warning Single Controller To avoid future single controller warning messages you can suppress them using the arraymgr ARM command line utility The arraymgr utility is described in further detail in the System Administrator s Guide which is shipped with each disk array 23 uondrnbsoq Description Product Description Controller Module 24 Figure 7 Installing a Controller Module Product Description SCSI Cabling Requirements SCSI Cabling Requirements The following are the SCSI cabling requirem
242. y Management Strategy optimize You should ut be aware that Capacity e Make sure all capacity is allocated to e Disabling Active Hot logical drives Do not leave any capacity Spare to increase the unallocated available capacity has the disadvantage of reducing protection from multiple disk failures e Avoid using Active Hot Spare thus freeing up more capacity for logical drives Disk Array Leave some capacity unallocated for use Je Leaving capacity Performance as RAID 0 1 space This maintains unallocated and enabling performance as your disk array begins to Active Hot Spare both reach its maximum storage capacity reduce the amount of Keep Active Hot Spare Enabled to ci operating system maintain performance following a disk failure Because performance suffers when the disk array is operating in critical or non redundant mode rebuilding quickly will maintain disk array performance The array also uses the Active Hot Spare as RAID 0 1 space until needed Data Redundancy Leave Active Hot Spare and Auto e Enabling Active Hot Rebuild enabled This offers the best Spare decreases the protection from a disk failure amount of capacity available to the host 103 sjdo uo Concepts Concepts and Management Upgrading the Capacity of a Fully Loaded Disk Array Upgrading the Capacity of a Fully Loaded Disk Array It is easy to increase the capacity of your disk array even if it is fu
243. y does during single controller operation If the data maps in NVRAM memory are lost all data except that retained only in NVRAM can be recovered using the Recover command via the front panel or the arrayrecover utility In single controller operation the write cache is disabled resulting in no data exposure to NVRAM loss Disabling of the write cache is accompanied by a very serious degradation in performance under some workloads This feature is supported on controller firmware versions later than HP40 94 Concepts Management Disk Array Hardware Disk Cache The array controller implements caching during both reads and writes Separate read and write caches are maintained in DRAM The amount of cache DRAM and NVRAM can be expanded to improve performance Controller Batteries CAUTION Controller batteries are a vital and integral part of maintaining the memory maps during a power loss or any time your disk array has not successfully completed a Shutdown before power off If power is lost and you are not using an Uninterruptable Power Supply UPS the batteries in the controller store the current memory maps of your data in NVRAM Each controller module contains dual backup batteries which provide power to the controller NVRAM when ac power to the disk array enclosure is off Many factors affect battery life including not only time of battery use but also time of battery storage Controller batteries should be replac
244. y status The disk drive ID number assigned to the drive added appears in the component identifier field of this log entry Troubleshooting Event Code Descriptions Event Event Name Event Predictive Suspected Description Number Logged Maintenance Components DEC hex Implication 146 0x92 Disk Drive Y Ignore unless Disk Drive The controller has detected Removed no operator Internal SCSI that a disk drive has been activity Bus removed This event occurs Controller for disk set members only The disk drive ID number assigned to the drive removed appears in the component identifier field of this log entry A disk drive ID number consisting of the controller serial number followed by a 0 indicates that the disk drive ID number has not yet been assigned 147 0x93 Disk Drive Y Ignore None The controller has deleted a Deleted From operator disk set member This occurs Disk Set activity or only when a drive that was associated previously a member becomes errors a non member The drive may or may not be present in the disk array at the time The disk drive ID number assigned to the drive deleted appears in the component identifier field of this log entry 148 0x94 Logical Unit Y Ienore None The controller has Created operator or successfully created a new host activity LUN The component identifier field of this log entry contains the number of the new LUN first 8 bytes and its capacity last 8 bytes 16
245. ze the array performance Power Off The primary array controller attempted to switch Either the secondary array controller or the Failure off the power to the power module but it was primary array controller failed and did not unsuccessful allow the power off to occur or the power module itself may be malfunctioning Power Off The disk array will continue to function if one If you remove a fan module whether it is Pending fan module fails however when a fan module is working or has failed from the disk array enclosure it must be replaced in ten minutes or less or a power off will occur Power 1 Failed The power module in slot P1 failed Replace power module Power 2 Failed The power module in slot P2 failed Replace power module P2 Power 3 Failed 130 The power module in slot P3 failed Replace power module P3 Error Message Explanation Troubleshooting Control Panel Error Messages Action Ready No errors or warnings to report the array is None Information only ready to transfer data Rebuild The command to Rebuild did not start The array is malfunctioning Not Started Call the support line Rebuild Not Stopped Recover Failed Recover Warning Rebuild Requested Shutdown Complete The SCSI command to Rebuild did not stop Data was not fully recovered during the recover utility Problem occurred while using the reco

HP SureStore E Disk Array 12H User`s and Service Manual

Contents

Download Pdf Manuals

Related Search

Related Contents