HP 4100 Enterprise Virtual Array Manual de usuario Pagina 1

HP EVA 4000/6000/8000 and EVA4100/6100/8100 User GuideAbstractThis document is intended for customers who operate and manage the EVA 4000/6000/8000 an

Risks...152OpenVMS configuratio

◦ HP Controller Power Supply Replacement Instructions◦ HP Disk Enclosure Power Supply/Blower Replacement Instructions◦ HP Fibre Channel Disk Drive Rep

6 Support and other resourcesContacting HPFor worldwide technical support information, see the HP support website:http://www.hp.com/supportBefore cont

Document conventions and symbolsTable 31 Document conventionsElementConventionCross-reference links and e-mail addressesBlue text: Table 31 (page 102)

Customer self repairHP customer self repair (CSR) programs allow you to repair your product. If a CSR part needsreplacing, HP ships the part directly

A Regulatory notices and specificationsThis appendix includes regulatory notices and product specifications for the HP Enterprise VirtualArray family.

interference will not occur in a particular installation. If this equipment does cause harmfulinterference to radio or television reception, which can

Certification and classification informationThis product contains a laser internal to the Optical Link Module (OLM) for connection to the Fibrecommuni

This symbol on the product or on its packaging indicates that this product must not be disposedof with your other household waste. Instead, it is your

luonnonvaroja ja varmistamaan, että laite kierrätetään tavalla, joka estää terveyshaitat ja suojeleeluontoa. Lisätietoja paikoista, joihin hävitettävä

τον άχρηστο εξοπλισμό σας για ανακύκλωση, επικοινωνήστε με το αρμόδιο τοπικό γραφείο, τηντοπική υπηρεσία διάθεσης οικιακών απορριμμάτων ή το κατάστημα

1 Enterprise Virtual Array startupThis chapter describes the procedures to install and configure the Enterprise Virtual Array. Whenthese procedures ar

Ten symbol na produkcie lub jego opakowaniu oznacza, że produktu nie wolno wyrzucaćdo zwykłych pojemników na śmieci. Obowiązkiem użytkownika jest prze

La recogida y el reciclado selectivos de los residuos de aparatos eléctricos en el momento de sueliminación contribuirá a conservar los recursos natur

Country-specific certificationsHP tests electronic products for compliance with country-specific regulatory requirements, as anindividual item or as p

Table 33 Environmental specifications (continued)10% to 90% non-condensingHumidity5% to 90% non-condensingShipping HumidityUp to 8,000 ft. (2,400 m)Al

Table 36 EVA4x00 power specifications — 208 Volts2C4D2C3D2C2D2C1DSpecification176713901013638Total System WattageTypical15585430030141729Total System

This data represents fully populated drive enclosures with 15K RPM disk drives. Other drive types may vary slightly. Forexample, if you are using 10K

1Typical is described as a system in normal steady state operation. (i.e., both PDUs operating normally, the arrayreading/writing to disk drives in a

B EMU-generated condition reportsThis section provides a description of the EMU generated condition reports that contain the followinginformation:• El

Table 42 Assigned element type codesElementCodeDisk Drives0.1.Power Supplies0.2.Blowers0.3.Temperature Sensors0.4.Audible Alarm10.6EMU0.7.Controller O

If the EMU cannot determine the drive link rate, the EMU activates the drive bypass function forone minute. During this time the EMU continually check

EVA6000/6100 storage system connectionsFigure 2 (page 12) shows a typical EVA6000/6100 SAN topology:• The HSV200-A and HSV200-B controllers connect vi

This error remains active until the problem is corrected.Complete the following procedure to correct this problem:1. Record all six characters of the

2. Ensure that there is AC power to the rack PDU, and from the PDU to the PDMs, and that thePDU and PDM circuit breakers are not reset.If there is no

Figure 41 Blower element numberingCAUTION: A single blower operating at high speed can provide sufficient air flow to cool anenclosure and the element

0.3.en.06 UNRECOVERABLE condition—No blowers installedNOTE: IMPORTANTWhen this condition exists there will be two error messages.The first message wil

3. Ensure that nothing is obstructing the air flow at either the front of the enclosure or the rear ofthe blower.4. Ensure that both blowers are opera

temperature thresholds). Under these conditions the EMU starts a timer that will automatically shutdown the enclosure in seven minutes unless you corr

4. If resetting the EMU did not correct the problem, replace the EMU.5. If unable to correct the problem, contact your HP authorized service represent

0.7.01.12 NONCRITICAL condition—EMU cannot read NVRAM dataThe EMU is unable to read data from the NVRAM. This condition report remains active until th

0.7.01.17 UNRECOVERABLE condition—Power shutdown failureThe power supply did not respond to a controller, EMU, or power supply shut down command.Shutt

Figure 43 Transceiver element numbering2. Transceiver 021. Transceiver 014. Transceiver 043. Transceiver 030.F.en.01 CRITICAL condition—Transceiver in

Figure 3 EVA4000/4100 configuration7 Fabric 21 Network interconnection8 Controller A2 Management server9 Controller B3 Non-host10 Cache mirror ports4

0.F.en.05 CRITICAL condition—Invalid fibre channel characterThis symptom can occur under the following conditions:• The incoming data stream is corrup

1.1.03.03 INFORMATION condition—Overrun recoveryThis condition report notes automatic recovery initiated by the occurrence of too many data overrunswi

1.2.en.04 CRITICAL condition—Low voltageThis condition report indicates that an element voltage has reached the low voltage CRITICALthreshold. This co

To correct this problem, record all six characters of the condition report, then contact yourHP-authorized service representative.I/O Module condition

3. Contact your HP-authorized service representative.8.7.en.12 NONCRITICAL condition—I/O Module NVRAM read failureThe system is unable to read data fr

C Controller fault managementThis appendix describes how the controller displays events and termination event information.Termination event informatio

Figure 46 Typical HP P6000 Command View Event displayDescriptionEIP TypeCACEvt NoSWCIDTimeDateThe Event display provides the following information:• D

Interpreting fault management informationEach version of HP P6000 Command View includes an ASCII text file that defines all the codesthat the authoriz

D Non-standard rack specificationsThe appendix provides information on the requirements when installing the EVA4x00/6x00/8x00in a non-standard rack. A

Determining the CG of a configuration may be necessary for safety considerations. CGconsiderations for CG calculations do not include cables, PDU’s an

iSCSI connection configurationsThe EVA4x00/6x00/8x00 support iSCSI attach configurations using the HP MPX100. Both fabricconnect and direct connect ar

NOTE: Further testing is required to update the information in Tables 45-47. Once testing iscomplete, these tables will be updated in a future release

Table 49 100V Wattage and BTU/HourEVA8x00EVA6x00EVA4x00EnclosuresBTU/hWattsVAAmpsBTU/hWattsVAAmpsBTU/hWattsVAAmpsEVA8x00 not supported118553474354535.

Table 51 UPS operating time limits (continued)Minutes of operationLoad (percent)With 2 ERMsWith 1 ERMWith standby batteryR5500462471006031980106611950

Table 53 EVA 6x00 UPS loading (continued)% of UPS capacityWattsEnclosuresR12000R5500R30007.921.235.395324.812.821.45771Table 54 EVA 4x00 UPS loading%

Shock and vibration specificationsTable 56 (page 144) lists the product operating shock and vibration specifications. This informationapplies to produ

E Single Path ImplementationThis appendix provides guidance for connecting servers with a single path host bus adapter (HBA)to the Enterprise Virtual

Installation requirements• The host must be placed in a zone with any EVA worldwide IDs (WWIDs) that access storagedevices presented by the hierarchic

switches and EVA controllers. Whereas the dual HBA server has multi-path software that managesthe two HBAs and their connections to the switch (with t

HP-UX configurationRequirements• Proper switch zoning must be used to ensure each single path HBA has an exclusive path toits LUNs.• Single path HBA s

Windows Server (32-bit) configurationRequirements• Switch zoning or controller level SSP must be used to ensure each single path HBA has anexclusive p

Procedures for getting startedResponsibilityStepCustomer1. Gather information and identify all related storagedocumentation.Customer2. Contact an auth

7 Controller A3 Host 28 Controller B4 Management serverWindows Server (64-bit) configurationRequirements• Switch zoning or controller level SSP must b

Figure 52 Windows Server (64-bit) configuration6 SAN switch 11 Network interconnection7 SAN switch 22 Management server8 Controller A3 Host 19 Control

Limitations• HP P6000 Continuous Access is not supported with single path configurations.• Single path HBA server is not part of a cluster.• Booting f

NOTE: For additional risks, see Table 60 (page 160).Figure 54 Tru64 UNIX configuration5 SAN switch 11 Network interconnection6 SAN switch 22 Host 17 C

Limitations• HP P6000 Continuous Access is not supported with single path configurations.Figure 55 OpenVMS configuration5 SAN switch 11 Network interc

Limitations• HP P6000 Continuous Access is not supported with single path configurations.• Single path HBA server is not part of a cluster.• Booting f

Limitations• HP P6000 Continuous Access is not supported with single path configurations.• Single path HBA server is not part of a cluster.• Booting f

Risks• Single path failure may result in loss of data accessibility and loss of host data that has notbeen written to storage.• Controller shutdown re

Risks• Single path failure may result in data loss or disk corruption.NOTE: For additional risks, see Table 63 (page 161).Limitations• HP P6000 Contin

Table 57 HP-UX failure scenarios (continued)Failure effectFault stimulusShort term: Data transfer stops. Possible I/O errors.Long term: Job hangs, can

The OCP on either controller can be used to input the WWN and password data. For moreinformation about the OCP, see “Operator control panel” (page 43)

OpenVMS and Tru64 UNIXTable 60 OpenVMS and Tru64 UNIX failure scenariosFailure effectFault stimulusAll I/O operations halted. Possible data loss from

Table 61 Linux failure scenarios (continued)Failure effectFault stimulusShort: I/O suspended, possible data loss.Long: I/O halts with I/O errors, data

Table 63 VMware failure scenarios (continued)Failure effectFault stimulusShort: I/O suspended, possible data loss.Long: I/O halts with I/O errors, dat

GlossaryThis glossary defines terms used in this guide or related to this product and is not acomprehensive glossary of computer terms.Aactive member

baud The maximum rate of signal state changes per second on a communication circuit. If each signalstate change corresponds to a code bit, then the ba

controller A hardware/firmware device that manages communications between host systems and otherdevices. Controllers typically differ by the type of i

disk replacementdelayThe time that elapses between a drive failure and when the controller starts searching for sparedisk space. Drive replacement sel

Enclosure ServicesProcessorSee ESP.Enterprise VirtualArrayThe Enterprise Virtual Array is a product that consists of one or more storage systems. Each

fiber optic cable A transmission medium designed to transmit digital signals in the form of pulses of light. Fiberoptic cable is noted for its propert

IDX A 2-digit decimal number portion of the HSV controller termination code display that defines oneof 32 locations in the Termination Code array that

4. Press or until the first character of the WWN is displayed. Press to accept this characterand select the next.5. Repeat Step 4 to enter the remaini

MmanagementagentThe HP P6000 Command View software that controls and monitors the Enterprise storage system.The software can exist on more than one ma

Pparam That portion of the HSV controller termination code display that defines:• The 2-character parameter identifier that is a decimal number in the

read aheadcachingA cache management method used to decrease the subsystem response time to a read requestby allowing the controller to satisfy the req

SSN Storage System Name. An HP P6000 Command View-assigned, unique 20-character name thatidentifies a specific storage system.storage carrier See carr

virtual disksnapshotSee snapshot.Vraid0 A virtualization technique that provides no data protection. Data host is broken down into chunksand distribut

IndexSymbols+12.5 VDC for the drives, 25+5.1 VDC, 25AAC input missing, 120AC powerdistributing, 52frequency, 25voltage, 25accessingmultipathing, 72Sec

Corrective Action Code see CACCorrective Action Codes see CACcountry-specific certifications, 112coupled crash control codes, 137creating virtual disk

Event Information Packets see EIPevent number, 135Ffabric setup, 87failure, 132FATA drives, using, 59fault managementdetails, 136display, 46displays,

NONCRITICAL conditions, 131lpfc driver, 84LTEA, 136LUN numbers, 15MManagement Server, 17Management Server, HP P6000 Command View, 11missingAC input, 1

Rracknon-standard specifications, 138physical layout, 19rack configurations, 52regulatory compliance noticescables, 105Class A, 104Class B, 104Europea

See the HP P6000 Command View Installation Guide for information on installing the software.Installing optional EVA software licensesIf you purchased

Veritas Volume Manager, 87version informationController, 47displaying, 47firmware, 47OCP, 47software, 47XCS, 47version information: firmware, 47vgcrea

2 Enterprise Virtual Array hardware componentsThe Enterprise Virtual Array includes the following hardware components:• Fibre Channel drive enclosure

© Copyright 2005, 2012 Hewlett-Packard Development Company, L.P.The information contained herein is subject to change without notice. The only warrant

loop switches. The EVA6100 includes two HSV200-B controllers with two Fibre Channel loopswitches.• EVA4000/4100 — available in configurations ranging

8. Power supply 27. Blower 210. Status indicators (EMU, enclosure power, enclosurefault)9. I/O module AI/O modulesTwo I/O modules provide the interfac

I/O module status indicatorsThere are three status indicators on the I/O module. See Figure 7 (page 21). The status indicatorstates for an operational

Fiber Optic Fibre Channel cablesThe Enterprise Virtual Array uses orange, 50-µm, multi-mode, fiber optic cables for connection tothe SAN. The fiber op

Up to 14 disk drives can be installed in a drive enclosure.Disk drive status indicatorsThree status indicators display the drive operational status. F

Table 6 Non-operational disk drive status indicationsDescriptionFaultOnlineActivityIndicates no connection or the controllers are offline.Recommended

The output of each power supply is 499 W, with a peak output of 681 W. A single power supplycan support an enclosure with a full complement of disks.T

The EMU for Fibre Channel-Arbitrated Loop (FC-AL) drive enclosures is fully compliant with SCSI-3Enclosure Services (SES), and mounts in the left rear

• Providing enclosure status data to the controllers.• Reporting the WWN and the logical address of all disk drives.NOTE: Although the EMU can determi

EMU indicator displaysThe EMU status indicators are located above the alphanumeric display. See Figure 13 (page 27).These indicators present the same

Contents1 Enterprise Virtual Array startup...11EVA8000/8100 storage system connections

Table 11 EMU display groupsDescriptionDisplay groupDisplayThe enclosure number is the default display and is a decimal numberin the range 00 through 1

Table 12 Audible alarm sound patterns (continued)Cycle 2Cycle 1Condition typeNONCRITICALINFORMATIONLegendAlarm OffAlarm OnControlling the audible alar

NOTE: Er is displayed in the alphanumeric display when an error condition is present.1. Press and hold the bottom push-button until the status indicat

A display of 01 through 14 indicates that the enclosure is connected to the enclosure address busand can exchange information with other enclosures on

Enclosure address bus connectionsConnecting the enclosures to the enclosure ID expansion cables establishes the enclosure addressbus. The enclosures a

Error condition categoriesEach error condition is assigned to a category based on its impact on disk enclosure operation.The following four error cate

The most severe error in the queue always has precedence, regardless of how long less severeerrors have been in the queue. This ensures that the most

Figure 16 Displaying error condition valuesPress and hold top push-button to view first error in queue.1Press and release top push-button.2Press and h

The reporting group numbers are displayed on the EMU alphanumeric display as a pair of two-digitdisplays. These two displays are identified as rH and

Figure 17 30-10022-01 loop switch status indicators1. Ethernet activity• Flashing—the Ethernet port is receiving data.• Flashing rapidly—the traffic l

Enclosure address bus connections...34Error Condition Reporting...

Figure 18 30-10010-02 loop switch status indicators2. Bezel snaps1. Handle4. Walk-up RS232 port3. Alignment tabs6. Port Bypassed indicator5. SFP statu

Table 15 30-10010-02 loop switch port status indicatorsDescriptionPort bypass indicator(Amber)SFP statusindicator (Green)Indicates that the port does

• Four 4 Gbps Fibre Channel-Switched fabric host ports (two host ports in HSV200-A orHSV200-B controller)• Four 2 Gbps Fibre Channel drive enclosure d

Figure 20 HSV200-A/B controller—rear view2. CAB (cabinet address bus)1. Dual controller interconnect4. Power ON3. Unit ID6. FC cache mirror ports5. FC

Figure 22 Controller OCP1. Status indicators (see Table 17 (page 44)) and UID button2. 40-character alphanumeric display3. Left, right, top, and botto

Table 18 Controller port status indicatorsDescriptionPortFibre Channel host ports• Green—Normal operation• Amber—No signal detected• Off—No SFP1detect

The menu tree is organized into the following major menus:• System Info—displays information and configuration settings.• Fault Management—displays fa

Displaying system informationNOTE: The purpose of this information is to assist the HP-authorized service representative whenservicing your system.The

Shutting the controller downUse the following procedure to access the Shutdown System display and execute a shutdownprocedure.CAUTION: If you decide N

6. Press the arrow keys to navigate to the open field and type DELETE and then press ENTER.The system uninitializes.NOTE: If you do not enter the word

Windows 2003 MSCS cluster installation...60Connecting to C-series switches.

Power suppliesTwo power supplies provide the necessary operating voltages to all controller enclosure components.If one power supply fails, the remain

Cache batteryBatteries provide backup power to maintain the contents of the controller cache when AC poweris lost and the storage system has not been

The data connections are the interfaces to the disk drive enclosures or loop switches (dependingon your configuration), the other controller, and the

Figure 26 60-Hz and 50-Hz wall receptaclesNEMA L6-30R receptacle, 3-wire, 30-A, 60-HzIEC 309 receptacle, 3-wire, 30-A, 50-Hz• The standard power confi

PDUsEach Enterprise Virtual Array rack has either a 50- or 60-Hz, dual PDU mounted at the bottom rearof the rack. The 228481-002/228481-003 PDU placem

Each PDM has eight AC receptacles and one thermal circuit breaker. The PDMs distribute the ACpower from the PDUs to the enclosures. Two power sources

Figure 29 Rack AC power distribution2. PDM 21. PDM 14. PDU 13. PDM 36. PDM 55. PDM 48. PDU 27. PDM 6Rack System/E power distribution componentsAC powe

Moving the rack requires a clear, uncarpeted pathway that is at least 80 cm (31.5 in) wide forthe 60.3 cm (23.7 in) wide, 42U rack. A vertical clearan

Figure 31 Raising a leveler foot2. Leveler foot1. Hex nut3. Carefully move the rack to the installation area and position it to provide the necessary

3 Enterprise Virtual Array operationThis chapter presents the tasks that you might need to perform during normal operation of thestorage system.Best p

Setting preferred paths...83Oracle Solaris...

Enabling Boot from SAN for Windows direct connectTo ensure that Boot from SAN is successful for Windows hosts that are directly connected to anarray,

HP Care Pack Service or HP contractual support agreement. HP Insight Remote Support supplementsyour monitoring, 24x7 to ensure maximum system availabi

Failback preference setting for HSV controllersTable 25 (page 62) describes the failback preference behavior for the controllers.Table 25 Failback pre

Table 25 Failback preference behavior (continued)BehaviorPoint in timeSettingIf cache data for a LUN exists on a particularcontroller, the unit will b

1If preference has been configured to ensure a more balanced controller configuration, the Path A/B – Failover/Failback setting isrequired to maintain

4. Under System Shutdown click Power Down. If you want to delay the initiation of the shutdown,enter the number of minutes in the Shutdown delay field

NOTE: For more information on using SSSU, see the HP Storage System Scripting Utility reference.See “Related information” (page 101).1. Double-click o

Example 1 Saving configuration data using SSSU on a Windows HostTo save the storage system configuration:1. Double-click on the SSSU desktop icon to r

• Set the add disk option to manual. See “Changing the Device Addition Policy” (page 69) formore information.• When adding multiple disk drives, add a

Adding a disk driveThis section describes the procedure for adding a Fibre Channel disk drive.Removing the drive blank1. Grasp the drive blank by the

Certification and classification information...106Canadien notice (avis Canadie

Figure 35 Installing the disk driveChecking status indicatorsCheck the following to verify that the disk drive is operating normally:NOTE: It may take

Figure 36 Disk drive status indicators2. Online1. Activity3. FaultAdding the disk to a disk groupAfter replacing the disk, use HP P6000 Command View t

4 Configuring application serversOverviewThis chapter provides general connectivity information for all the supported operating systems.Where applicab

Testing connections to the EVAAfter installing the FCAs, you can create and test connections between the host server and theEVA. For all operating sys

1. From HP P6000 Command View, create a virtual disk on the EVA4x00/6x00/8x00.2. Specify values for the following parameters:• Virtual disk name• Vrai

State========================================================================================ba 3 0/6 lba

http://www.hp.com/support/downloadsIn the Search products box, enter MPIO, and then click AIX MPIO PCMA for HP Arrays. Select IBMAIX, and then select

LinuxDriver failover modeIf you use the INSTALL command without command options, the driver’s failover mode dependson whether a QLogic driver is alrea

# modprobe qla2400To reboot the server, enter the reboot command.CAUTION: If the boot device is attached to the SAN, you must reboot the host.7. To ve

Compiling the driver for multiple kernelsIf your system has multiple kernels installed on it, you can compile the driver for all the installedkernels

0.3.en.06 UNRECOVERABLE condition—No blowers installed ...123Temperature conditions...

This line identifies the location of the binary RPM.4. Copy the binary RPM to the production servers and install it using the following command:# rpm

Table 29 Comparing console LUN to OS unit IDSystem DisplayID type$1$GGA100:Console LUN ID set to 100$1$DGA50:OS unit ID set to 50Adding OpenVMS hostsT

NOTE: The EVA4x00/6x00/8x00 console LUN can be seen without any virtual disks presented.The LUN appears as $1$GGAx (where x represents the console LUN

2. Enter the following command to mount the disk:MOUNT/SYSTEM name-of-virtual-disk volume-labelNOTE: The /SYSTEM switch is used for a single stand-alo

Update instructions depend on the version of your OS:• For Solaris 9, install the latest Oracle StorEdge SAN software with associated patches. Tolocat

3. If using a single FCA and no multipathing, edit the following parameter to reduce the risk ofdata loss in case of a controller reboot:nodev-tmo=120

1. Ensure that you have the latest supported version of the qla2300 driver.You must sign up for an HP Passport to enable access. For more information

6. If the qla2300 driver is version 4.15 or later, verify that the following or a similar entry ispresent in the /kernel/drv/sd.conf file:name="s

1. Go to http://support.veritas.com.2. Enter Storage Foundation for UNIX/Linux in the Product Lookup box.3. Enter EVA in the Enter keywords or phrase

Example 4 Setting the iopolicy# vxdmpadm getattr arrayname EVA8100 iopolicyENCLR_NAME DEFAULT CURRENT======================================

C Controller fault management...135Using HP P6000 Command View ...

50001fe1002709e9,5• Emulex (lpfc)/QLogic (qla2300) drivers:You can retrieve the WWPN by checking the assignment in the driver configuration file(the e

Example 5 Format command# formatSearching for disks...donec2t50001FE1002709F8d1: configured with capacity of 1008.00MBc2t50001FE1002709F8d2: configure

7. For each new device, use the disk command to select another disk, and then repeat Step 1through Step 6.8. Repeat this labeling procedure for each n

Configuring an ESX serverThis section provides information about configuring the ESX server.Loading the FCA NVRAMThe FCA stores configuration informat

ESX 4.x commands• The # esxcli nmp device setpolicy --devicenaa.6001438002a56f220001100000710000 --psp VMW_PSP_MRU command setsdevice naa.6001438002a5

Verifying virtual disks from the hostTo verify that the host can access the virtual disks, enter the more /proc/scsi/scsi command.The output lists all

5 Customer replaceable unitsThis chapter describes customer replaceable units. Information about initial enclosure installation,ESD protection, and co

level. The replacement component revision level must be the same as, or greater than, the numberon the element being replaced. The higher the revision

Table 30 Hardware component CSR support (continued)CSRSpare part number (nonRoHS/RoHS)Description✓366023-001/366023-002Disk drive – 300 GB 10K✓518736-

Table 30 Hardware component CSR support (continued)CSRSpare part number (nonRoHS/RoHS)Description✓390853-001, 70-41140-S1/411632-005, 70-41140-S3Front