HP StorageWorks 4400 Enterprise Virtual Array user guide Part number: 5697-7710 Fourth edition: December 2008
Legal and notice information © Copyright 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein.
Contents About this guide ................................................................................. 11 Intended audience .................................................................................................................... Related documentation .............................................................................................................. Document conventions and symbols .............................................................................................
Operating tips and information .................................................................................................. 37 Reserving adequate free space ............................................................................................ 37 Using FATA disk drives ........................................................................................................ 37 Failback preference setting for HSV controllers ..........................................................................
HBA configuration .............................................................................................................. Risks ................................................................................................................................. Limitations ......................................................................................................................... OpenVMS configuration ...........................................................................................
Laser device ...................................................................................................................... 95 Laser safety warnings ................................................................................................... 95 Compliance with CDRH regulations ................................................................................ 95 Certification and classification information .............................................................................
Figures 1 Storage system hardware components (back view) ...................................................... 15 2 Disk drive bay numbering ........................................................................................ 17 3 Disk enclosure (front view without bezel ears) ............................................................. 17 4 Disk enclosure (rear view) ........................................................................................ 17 5 I/O module detail .......................
33 NetWare configuration ............................................................................................ 67 34 Linux (32-bit) configuration ....................................................................................... 69 35 Linux (Itanium) configuration ..................................................................................... 70 36 IBM AIX Configuration .............................................................................................
Tables 1 Document conventions ............................................................................................. 12 2 Port status LEDs ....................................................................................................... 18 3 I/O module status LEDs ........................................................................................... 18 4 Disk status indicator LED descriptions .........................................................................
About this guide This user guide provides the following information: • Description of the HP StorageWorks 4400 Enterprise Virtual Array and its components • Operating your EVA4400 storage system Intended audience This book is intended for use by Enterprise Virtual Array customers involved in the installation, operation, and management of EVA4400 storage systems. Related documentation You can find related documents from the Manuals page of the HP Business Support Center website: http://www.hp.
Document conventions and symbols Table 1 Document conventions Convention Element Blue text: Table 1 Cross-reference links and e-mail addresses Blue, underlined text: http://www.hp.
HP technical support For worldwide technical support information, see the HP support website: http://www.hp.com/support Before contacting HP, collect the following information: • • • • • • Product model names and numbers Technical support registration number (if applicable) Product serial numbers Error messages Operating system type and revision level Detailed questions Customer self repair HP customer self repair (CSR) programs allow you to repair your StorageWorks product.
HP websites For additional information, see the following HP websites: • • • • • http://www.hp.com http://www.hp.com/go/storage http://www.hp.com/service_locator http://www.hp.com/support/manuals http://h20293.www2.hp.com Documentation feedback HP welcomes your feedback. To make comments and suggestions about product documentation, please send a message to storagedocsFeedback@hp.com. All submissions become the property of HP.
1 EVA4400 hardware The EVA4400 contains the following hardware components: • EVA controller enclosure — Contains power supplies, cache batteries, fans, and HSV controllers • Fibre Channel disk enclosure — Contains disk drives, power supplies, fans, midplane, and I/O modules. • Fibre Channel Arbitrated Loop cables — Provides connectivity to the EVA controller enclosure and the Fibre Channel disk enclosures • Rack—Several free standing racks are available.
M6412 disk enclosures The M6412 disk enclosure contains the disk drives used for data storage; a storage system contains multiple disk enclosures. The major components of the enclosure are: • • • • • • 12-bay enclosure Dual-loop, Fibre Channel drive enclosure I/O modules Copper Fibre Channel cables Fibre Channel disk drives and drive blanks Power supplies Fan modules NOTE: An EVA4400 requires a minimum of one disk shelf with eight disk drives.
1 2 3 4 5 6 7 8 9 10 11 12 15815 Figure 2 Disk drive bay numbering 1. Rack-mounting thumbscrew 4. Enclosure status LEDs 2. UID push button 5. Drive LEDs 3. Disk drive release Figure 3 Disk enclosure (front view without bezel ears) 1. Power supply 1 5. Fan 1 status LED 9. Enclosure status LEDs 2. Power supply 1 status LED 6. I/O module A 10. Fan 2 3. Fan 1 7. I/O module B 11. Power push button 4. Enclosure product number and serial number 8. Rear UID push button 12.
1. Double 7–segment display: enclosure ID 4. Manufacturing diagnostic port 2. 4 Gb I/O ports 5. I/O module status LEDs 3. Port 1 (P1), Port 2 (P2) status LEDs Figure 5 I/O module detail Each I/O module has two ports that can transmit and receive data for bidirectional operation. Activating a port requires connecting a Fibre Channel cable to the port. The port function depends upon the loop. I/O module status indicators There are five status indicators on the I/O module. See Figure 5.
Status LED Description • Fault indicator: • Flashing amber — Warning condition (not visible when solid amber showing). • Solid amber — Replace FRU. • Amber off — Normal operation. Fiber optic Fibre Channel cables The Enterprise Virtual Array uses orange, 50-µm, multi-mode, fiber optic cables for connection to the SAN or the host, where there is a direct connection to the host. The fiber optic cable assembly consists of two 2-m fiber optic strands and small form-factor connectors on each end. See Figure 6.
8164A-ST Figure 7 Copper Fibre Channel cable Fibre Channel disk drives The Fibre Channel disk drives are hot-pluggable and include the following features: • Dual-ported 4-Gbps Fibre Channel controller interface that allows up to 96 disk drives to be supported per array controller enclosure • Compact, direct-connect design for maximum storage density and increased reliability and signal integrity • Both online high-performance disk drives and FATA disk drives supported in a variety of capacities and spindl
Disk drive blank To maintain the proper enclosure air flow, a disk drive or a disk drive blank must be installed in each drive bay. The disk drive blank maintains proper airflow within the disk enclosure. Controller enclosures The EVA4400 contains either the HSV300 or HSV300–S controller enclosure. Two interconnected controllers ensure that the failure of a controller component does not disable the system.
1. Battery 1 status LEDs 4. Battery 2 status LEDs 2. Fan 1 status LEDs 5. Enclosure status LEDs 3. Fan 2 status LEDs 6. Front UID push button Figure 9 Controller enclosure (front bezel) 1. Rack-mounting thumbscrew 8. Fan 1 normal operation LED 2. Enclosure product number and serial number 9. Fan 1 fault LED 3. World Wide Number (WWN) 10. Fan 2 4. Battery 1 11. Battery 2 5. Battery normal operation LED 12. Enclosure status LEDs 6. Battery fault LED 13. Front UID push button 7.
1. Power supply 1 9. Enclosure power push button 2. HSV300 controller 1 10. Power supply 2 3. Management module status LEDs 11. Host ports, FP1, FP2, connection to front end (host or SAN) 4. Ethernet port 12. DP1–A port, back-end connection to A loop 5. Management module 13. DP1–B port, back-end connection to B loop 6. HSV300 controller 2 14. Manufacturing diagnostic port 7. Rear UID push button 15. HSV300 controller status and fault LEDs 8.
Figure 12 shows the back view of the HSV300–S controller enclosure Management module The web-based operator control panel (WOCP) provides a direct interface to the management module within each controller. From the WOCP you can display storage system status and configuration information, shut down the storage system, and manage the password. For tasks to perform with the WOCP, see the WOCP online help.
HSV300 controller status LEDs Figure 13 shows the location of the controller status LEDs; Table 7 describes them. 3 2 1 4 5 6 gl0114 Figure 13 Controller status LEDs Table 7 Controller status LEDs Item LED Indication 1 Blue LED used to identify a certain controller within the enclosure. 2 Controller health OK. Solid green LED after boot. 3 Solid amber LED used to identify a controller failure either from a voltage issue or driven by firmware. 4 N/A Not used. 5 Cache status.
1. Power supply 4. Status indicator (solid green on — normal operation; solid amber — failure or no power) 2. AC input connector 5. Handle 3. Latch Figure 14 Power supply Fan module Fan modules provide the cooling necessary to maintain the proper operating temperature within the controller enclosure. If one fan fails, the remaining fan is capable of cooling the enclosure. 1. Green — Fan normal operation LED 2.
Battery module Battery modules provide power to the controllers in the enclosure. 1. Green — normal operation LED 2. Amber — fault LED Figure 16 Battery module pulled out Each battery module provides power to the controller directly across from it in the enclosure. Table 9 Battery status indicators Status indicator On left — green Fault indicator Description Solid green Normal operation. Blinking Maintenance in progress. Off Amber is on or blinking, or the enclosure is powered down.
to the loops in the main cabinet. Copper cables are used the controllers (mirror port) and the controllers and the drive enclosures or loops. Storage system racks All storage system components are mounted in a rack. Each configuration includes one controller enclosure holding both controllers (the controller pair), FC cables the controller and the disk enclosures. Each controller pair and all associated drive enclosures form a single storage system.
NEMA L6-30R receptacle, 3-wire, 30-A, 60-Hz CXO7549A IEC 309 receptacle, 3-wire, 30-A, 50-Hz CXO5409B Figure 17 60-Hz and 50-Hz wall receptacles • The standard power configuration for any Enterprise Virtual Array rack is the fully redundant configuration. Implementing this configuration requires: • Two separate circuit breaker-protected, 30-A site power sources with a compatible wall receptacle (see Figure 17). • One dual PDU assembly. Each PDU connects to a different wall receptacle.
voltages are then routed to PDMs, sometimes referred to as AC power strips, mounted on the two vertical rails in the rear of the rack. 1 2 3 5 4 5 0130a 1. PDU 1 4. AC receptacles 2. PDU 2 5. Mounting hardware 3. Circuit breakers Figure 18 Dual PDU assembly PDU 1 PDU 1 connects to AC power distribution source 1. A PDU 1 failure: • • • • Disables the power distribution circuit. Removes power from PDMs 1, 2, and 3. Disables PS 1 in the drive enclosures. Disables the Controller 1 power supply.
CAUTION: The AC power distribution within a rack ensures a balanced load to each PDU and reduces the possibility of an overload condition. Changing the cabling to or from a PDM could cause an overload condition. HP supports only the AC power distributions defined in this user guide. 1 2 3 0131a 1. Power receptacles 2. Thermal circuit breakers 3. AC power connector Figure 19 Rack PDM Rack AC power distribution The power distribution in an Enterprise Virtual Array rack is the same for all variants.
1. PDM 1 5. PDM 4 2. PDM 2 6. PDM 5 3. PDM 3 7. PDM 6 4. PDU 1 8. PDU 2 Figure 20 Rack AC power distribution Rack System/E power distribution components AC power is distributed to the Rack System/E rack through Power Distribution Units (PDU) mounted on the two vertical rails in the rear of the rack. Up to four PDUs can be mounted in the rack—two mounted on the right side of the cabinet and two mounted on the left side. Each of the PDU power cables has an AC power source specific connector.
Moving and stabilizing a rack WARNING! The physical size and weight of the rack requires a minimum of two people to move. If one person tries to move the rack, injury may occur. To ensure stability of the rack, always push on the lower half of the rack. Be especially careful when moving the rack over any bump (e.g., door sills, ramp edges, carpet edges, or elevator openings). When the rack is moved over a bump, there is a potential for it to tip over.
4 3 2 2 5 6 8 1 7 CXO8238A 1. Front door 5. Rear service area depth 300 mm 2. Rear door 6. Rack depth 1000 mm 3. Rack width 600 mm 7. Front service area depth 406 mm 4. Service area width 813 mm 8. Total rack depth 1706 mm Figure 21 Single rack configuration floor space requirements If the feet are not fully raised, complete the following procedure: 1. 34 Raise one foot by turning the leveler foot hex nut counterclockwise until the weight of the rack is fully on the caster (see Figure 22).
2. Repeat Step 1 for the other feet. 1 2 CXO7589A 1. Hex nut 2. Leveler foot Figure 22 Raising a leveler foot 3. Carefully move the rack to the installation area and position it to provide the necessary service areas (see Figure 21). To stabilize the rack when it is in the final installation location: 1. Use a wrench to lower the foot by turning the leveler foot hex nut clockwise until the caster does not touch the floor. Repeat for the other feet. 2.
EVA4400 hardware
2 EVA4400 operation Best practices For useful information on managing and configuring your storage system, see the HP StorageWorks Enterprise Virtual Array configuration best practices white paper available from http:// h18006.www1.hp.com/storage/arraywhitepapers.html. Operating tips and information Reserving adequate free space To ensure efficient storage system operation, reserve some unallocated capacity, or free space, in each disk group.
Setting Point in time Behavior On controller failback All LUNs remain on the surviving controller. There is no failback except if a host moves the LUN using SCSI commands. At initial presentation The units are brought online to Controller 1. On dual boot or controller resynch If cache data for a LUN exists on a particular controller, the unit will be brought online there. Otherwise, the units are brought online to Controller 1.
Setting Point in time Behavior On dual boot or controller resynch If cache data for a LUN exists on a particular controller, the unit will be brought online there. Otherwise, the units are brought online to Controller 2. On controller failover All LUNs are brought online to the surviving controller. On controller failback All LUNs remain on the surviving controller. After controller restoration, the units that are online to Controller 1 and set to Path B are brought online to Controller 2.
When creating a virtual disk, one controller is selected to manage the virtual disk. Only this managing controller can issue I/Os to a virtual disk in response to a host read or write request. If a read I/O request arrives on the non-managing controller, the read request must be transferred to the managing controller for servicing.
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. The controllers complete an orderly shutdown and then power off. The disk enclosures then power off. Wait for the shutdown to complete. 5. If your management server is an SMA and you are not using it to manage other storage arrays, shut down the SMA. From the SMA user interface, click Settings > Maintenance > Shutdown.
Connecting to the management module You can connect to the management module through a public or a private network. NOTE: The management module has an MDI-X port that supports straight-through or crossover Ethernet cables. Use a Cat 5e or greater cable. Connecting through a public network 1. 2. 3. Initialize the EVA4400 storage system, using HP SmartStart EVA Storage or HP Command View EVA.
Connecting through a private network Use this procedure when the management module will not be connected through a public network or the array has not yet been initialized 1. 2. 3. 4. 5. Press and hold the recessed Reset button for 4–5 seconds (2 management module). This sets an IP address of 192.168.0.1. The amber indicator on the top right corner of the Ethernet jack (1, management module) will flash momentarily when the reset operation is completed. Browse to https://192.168.0.
Figure 24 WOCP Configure controller host ports page • • • • In the Topology box, select Direct from the dropdown menu. Click Save Changes. Change the default operating mode for the other controller. Close the WOCP and remove the Ethernet cable from the server. Accessing the WOCP through HP Command View EVA 1. Select the EVA4400 storage system to access. 2. Select Hardware > Controller Enclosure > Management Module. 3. Click Launch WOCP.
NOTE: For more information on using the utility, see the HP StorageWorks Storage System Scripting Utility reference. See “Related documentation” on page 11. 1. Double-click the SSSU desktop icon to run the application. When prompted, enter Manager (management server name or IP address), User name, and Password. 2. Enter LS SYSTEM to display the EVA storage systems managed by the management server. 3. Enter SELECT SYSTEM system name, where system name is the name of the storage system.
Example 2. Restoring configuration data on a Windows host If it is necessary to restore the storage system configuration, it can be done using the following procedure. 1. 2. Double-click on the SSSU desktop icon to run the application. Enter FILE pathname\filename, where pathname identifies the location where the configuration files are be saved and filename is the name of the first configuration file: for example, FILE c:\EVAConfig\LargeEVA_Step1A.txt 3.
To minimize the risk of contamination or damage, do the following: • Dust covers—Remove and set aside the dust covers and dust caps when installing an I/O module, a transceiver or a cable. Install the dust covers when disconnecting a transceiver or cable. One of the many sources for cleaning equipment specifically designed for fiber optic connectors is: Alcoa Fujikura Ltd.
EVA4400 operation
3 Replacing array components Customer self repair (CSR) Table 12 and Table 13 identify hardware components that are customer replaceable. Using WEBES, ISEE or other diagnostic tools, a support specialist will work with you to diagnose and assess whether a replacement component is required to address a system problem. The specialist will also help you determine whether you can perform the replacement. Parts-only warranty service Your HP Limited Warranty may include a parts-only warranty service.
• When you have confirmed that a component replacement is required, you may want to clear the Real Time Monitoring view. This makes it easier to identify additional hardware problems that may occur while waiting for the replacement part. • Before installing the replacement part, check the Real Time Monitoring view for any new hardware problems. If additional hardware problems have occurred, contact HP support before replacing the component.
Description Spare part number (non RoHS/RoHS) Array battery 460581–001 ✓ Array power supply 435740–001 ✓ Array fan module 460583–001 ✓ Array management module 460584–001 ✓ Array LED membrane display 461489–001 • Array midplane 461490–001 • Array riser assembly 461491–001 • Array power UID 466264–001 • Array front bezel 460585–001 ✓ CSR status Table 13 M6412 disk enclosure replaceable parts Description Spare part number (non RoHS/RoHS) 4Gb FC disk shelf midplane 461492–001 •
Description Spare part number (non RoHS/RoHS) SPS-CABLE ASSY, 4Gb COPPER, FC, 0.41m 496917-001 CSR status ✓ For more information about CSR, contact your local service provider or see the CSR website: http://www.hp.com/go/selfrepair To determine the warranty service provided for this product, see the warranty information website: http://www.hp.com/go/storagewarranty To order a replacement part, contact an HP-authorized service provider or see the HP Parts Store online: http://www.hp.
• HP StorageWorks controller enclosure 4Gb array controller with embedded switch replacement instructions • HP StorageWorks controller enclosure battery replacement instructions • HP StorageWorks controller enclosure cache DIMM replacement instructions • HP StorageWorks controller enclosure fan module replacement instructions • HP StorageWorks controller enclosure LED display replacement instructions • HP StorageWorks controller enclosure management module replacement instructions • HP StorageWorks controll
Replacing array components
4 Single path implementation This chapter provides guidance for connecting servers with a single path host bus adapter (HBA) to the Enterprise Virtual Array (EVA) storage system with no multipath software installed. A single path HBA is defined as: • A single HBA port to a switch with no multipathing software installed • A single HBA port to a switch with multipathing software installed HBA LUNs are not shared by any other HBA in the server or in the SAN.
• On HP-UX, Solaris, Microsoft Windows Server 2003 (32-bit), Windows 2000, Novell NetWare, Linux and IBM AIX operating systems, the zones consist of the single path HBA systems and one HSV controller port. • On OpenVMS, the zones consist of the single HBA systems and two HSV controller ports. This results in a configuration where there are two paths per device, or multiple paths. NOTE: Windows 2000 Datacenter operating system is not currently supported.
controllers. See the HP StorageWorks SAN Design Reference Guide at the following HP website for additional information about zoning: http://h18006.www1.hp.com/products/storageworks/ san/documentation.html To connect a single path HBA server to a SAN switch: 1. Plug one end of the Fibre Channel cable into the HBA on the server. 2. Plug the other end of the cable into the switch.
1 4 3 2 5 7 6 8 10 9 26465a Figure 27 Single path HBA server with OpenVMS 1. Network interconnection 6. SAN switch 1 2. Single HBA server (Host 1) 7. SAN switch 2 3. Dual HBA server (Host 2) 8. Fabric zone 4. Management server 9. Controller A 5. Multiple single HBA paths 10. Controller B HP-UX configuration Requirements • Proper switch zoning must be used to ensure each single path HBA has an exclusive path to its LUNs.
• Host 2 is a multiple HBA host with multipathing software. See Figure 28. Risks • Disabled jobs hang and cannot umount disks. • Path or controller failure may results in loss of data accessibility and loss of host data that has not been written to storage. NOTE: For additional risks, see “HP-UX” on page 74. Limitations • HP Continuous Access EVA is not supported with single-path configurations. • Single path HBA server is not part of a cluster. • Booting from the SAN is not supported.
Windows Server 2003 (32-bit) and Windows 2000 configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • Single path HBA server can be in the same fabric as servers with multiple HBAs. • Single path HBA server cannot share LUNs with any other HBAs.
1 4 3 2 6 5 8 7 26464a Figure 29 Windows Server 2003 (32-bit) and Windows 2000 configuration 1. Network interconnection 5. SAN switch 1 2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8. Controller B Windows Server 2003 (64-bit) configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs.
NOTE: Single path HBA servers running the Windows Server 2003 (x64) operating system will support multiple single path HBAs in the same server. This is accomplished through a combination of switch zoning and controller level SSP. Any single path HBA server will support up to four single path HBAs. Risks • Single path failure will result in loss of connection with the storage system. • Single path failure may cause the server to reboot.
1 2 4 3 5 7 8 6 10 9 26466a Figure 30 Windows Server 2003 (64-bit) configuration 1. Network interconnection 6. SAN switch 1 2. Management server 7. Multiple single HBA paths 3. Host 1 8. SAN switch 2 4. Host 2 9. Controller A 5. Host 3 10. Controller B SUN Solaris configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • Single path HBA server can be in the same fabric as servers with multiple HBAs.
HBA configuration • Host 1 is a single path HBA host. • Host 2 is a multiple HBA host with multipathing software. See Figure 31. Risks • Single path failure may result in loss of data accessibility and loss of host data that has not been written to storage. • Controller shutdown results in loss of data accessibility and loss of host data that has not been written to storage. NOTE: For additional risks, see “Sun Solaris” on page 75.
2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8. Controller B OpenVMS configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • All nodes with direct connection to a disk must have the same access paths available to them. • Single path HBA server can be in the same fabric as servers with multiple HBAs.
1 4 3 2 6 5 8 7 26464a Figure 32 OpenVMS configuration 1. Network interconnection 5. SAN switch 1 2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8. Controller B NetWare configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • Single path HBA server cannot share LUNs with any other HBAs.
NOTE: For additional risks, see “NetWare” on page 76. Limitations • HP Continuous Access EVA is not supported with single-path configurations. • Single path HBA server is not part of a cluster. • Booting from the SAN is not supported on single path HBA servers. 1 4 3 2 6 5 8 7 26464a Figure 33 NetWare configuration 1. Network interconnection 5. SAN switch 1 2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8.
• Single path HBA server can be in the same fabric as servers with multiple HBAs. • In the use of snapshots and snapclones, the source virtual disk and all associated snapshots and snapclones must be presented to the single path hosts that are zoned with the same controller. In the case of snapclones, after the cloning process has completed and the clone becomes an ordinary virtual disk, you may present that virtual disk as you would any other ordinary virtual disk.
1 4 3 2 6 5 8 7 26464a Figure 34 Linux (32-bit) configuration 1. Network interconnection 5. SAN switch 1 2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8. Controller B Linux (Itanium) configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • All nodes with direct connection to a disk must have the same access paths available to them.
• Host 2 is a dual HBA host with multipathing software. See Figure 35. Risks • Single path failure may result in data loss or disk corruption. NOTE: For additional risks, see “Linux” on page 77. Limitations • HP Continuous Access EVA is not supported with single path configurations. • Single path HBA server is not part of a cluster. • Booting from the SAN is supported on single path HBA servers. 1 4 3 2 6 5 8 7 26464a Figure 35 Linux (Itanium) configuration 1. Network interconnection 5.
IBM AIX configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • Single path HBA server can be in the same fabric as servers with multiple HBAs. • Single path HBA server cannot share LUNs with any other HBAs. • In the use of snapshots and snapclones, the source virtual disk and all associated snapshots and snapclones must be presented to the single path hosts that are zoned with the same controller.
1 4 3 2 6 5 8 7 26464a Figure 36 IBM AIX Configuration 1. Network interconnection 5. SAN switch 1 2. Single HBA server (Host 1) 6. SAN switch 2 3. Dual HBA server (Host 2) 7. Controller A 4. Management server 8. Controller B VMware configuration Requirements • Switch zoning or controller level SSP must be used to ensure each single path HBA has an exclusive path to its LUNs. • All nodes with direct connection to a disk must have the same access paths available to them.
Risks • Single path failure may result in data loss or disk corruption. NOTE: For additional risks, see “VMware” on page 78. Limitations • HP Continuous Access EVA is not supported with single path configurations. • Single HBA path at the host server is not part of a cluster, unless in a VMware High Availability Cluster. • Booting from the SAN is supported on single path HBA servers. 1 4 3 2 6 5 8 7 26464a Figure 37 VMware configuration 1. Network interconnection 5. SAN switch 1 2.
Failure scenarios HP-UX Fault stimulus Failure effect Server failure (host power-cycled) Extremely critical event on UNIX. Can cause loss of system disk. Short term: Data transfer stops. Possible I/O errors. Switch failure (SAN switch disabled) Long term: Job hangs, cannot umount disk, fsck failed, disk corrupted, need mkfs disk. Short term: Data transfer stops. Possible I/O errors. Controller failure Long term: Job hangs, cannot umount disk, fsck failed, disk corrupted, need mkfs disk.
Sun Solaris Fault stimulus Failure effect Server failure (host power-cycled) Check disk when rebooting. Data loss, data that finished copying survived. Short term: Data transfer stops. Possible I/O errors. Switch failure (SAN switch disabled) Long term: Repeated error messages on console, no access to CDE. System reboot causes loss of data on disk. Must newfs disk. Short term: Data transfer stops. Possible I/O errors. Controller failure Long term: Repeated error messages on console, no access to CDE.
Fault stimulus Failure effect If the LUN is not shared, I/O is suspended or process is terminated across this HBA until path is restored. Server path failure If running OpenVMS 7.3-1 and the LUN is shared, another cluster node having direct access will take over serving the device, resulting in no loss of service. In either case, no data is lost or corrupted.
Linux Fault stimulus Failure effect Server failure (host power-cycled) OS reboots, automatically checks disks. HSV disks must be manually checked unless auto mounted by the system. Short: I/O suspended, possible data loss. Switch failure (SAN switch disabled) Long: I/O halts with I/O errors, data loss. HBA driver must be reloaded before failed drives can be recovered, fsck should be run on any failed drives before remounting. Short term: I/O suspended, possible data loss.
Fault stimulus Failure effect Short term: Data transfer stops. Possible I/O errors. Server path failure Long term: Repeated error messages in errpt output. System reboot causes loss of data on disk. Must crfs disk. Short term: Data transfer stops. Possible I/O errors. Storage path failure Long term: Repeated error messages in errpt output. System reboot causes loss of data on disk. Must crfs disk.
5 Error messages This list of error messages is in order by status code value, 0 to xxx. Table 14 Error Messages Status Code Value 0 Successful Status Meaning The SCMI command completed successfully. How to Correct No corrective action required. Delete the associated object and try the operation again. Several situations can cause this message: Presenting a LUN to a host: • Delete the current association or specify a different LUN number.
Status Code Value 7 Invalid Class 8 Invalid Function 9 Invalid Logical Disk Block State 10 Invalid Loop Configuration 11 Invalid parameter Meaning The supplied class code is of an unknown type. This can be caused by a user or program error. Report the error to product support. The function code specified with the class code is of an unknown type. Report the error to product support. The specified command supplied unrecognized values. This can indicate a user or program error.
Status Code Value 16 Invalid Target Id 17 Invalid Time 18 Media is Inaccessible 19 No Fibre Channel Port 20 No Image 21 No Permission 22 Storage system not initialized 23 Not a Loop Port 24 Not a Participating Controller Meaning How to Correct The supplied target identifier is invalid. This can indicate a user or program error. Report the error to product support. The time value specified is invalid. This can indicate a user or program error. Report the error to product support.
Status Code Value Meaning How to Correct Case 1: Either delete the associated object or resolve the in progress state. Case 2: . Report the error to product support. Case 3: Unpresent the LUNs before deleting this virtual disk. Case 4: Resolve the delay before performing the operation. Case 5: Delete any remaining virtual disks or wait for the used capacity to reach zero before the disk group can be deleted. If this is the last remaining disk group, uninitialize the storage cell to remove it.
Status Code Value Meaning How to Correct Several states can cause this message: Case 1: The operation cannot be performed because an association exists a related object, or the object is in a progress state. Derived unit create: Case 2: The supplied virtual disk handle is already an attribute of another derived unit. This may indicate a programming error Derived unit discard: Case 3: One or more LUNs are presented to EVA hosts that are based on this virtual disk.
Status Code Value Meaning How to Correct Case 15: GROUP discard: The Continuous Access group cannot be discarded as one or more virtual disk members exist. The operation cannot be performed because the object does not exist. This can indicate a user or program error. 26 Parameter Object Does Not Exist VOLUME set requested usage: The disk volume set requested usage cannot be performed because the disk group does not exist. This can indicate a user or program error. Report the error to product support.
Status Code Value Meaning How to Correct The operation could not be completed because the drive volume is in a missing state. Resolve the condition and retry the request. Report the error to product support. The supplied cursor or sequence number is invalid. This may indicate a user or program error. Report the error to product support. The specified target logical disk already has an existing data sharing relationship. This can indicate a user or program error. Report the error to product support.
Status Code Value Meaning How to Correct Security credentials needed. Please update your system's ID and password in the Storage System Access menu. The management software is unable to log in to the storage system. The storage system password has been configured. Use the management software to save the password specified so communication can proceed. The management software is unable to login to the device. The storage system password may have been re-configured or removed.
Status Code Value 59 Maximum Number of Objects Exceeded. 60 Max size exceeded Meaning Case 1: The maximum number of items allowed has been reached. Case 2: The maximum number of EVA hosts has been reached. Case 3: The maximum number of port WWNs has been reached. How to Correct Case 1: If this operation is still desired, delete one or more of the items and retry the operation. Case 2: If this operation is still desired, delete one or more of the EVA hosts and retry the operation.
Status Code Value 65 Bad image header 66 Bad image 67 The firmware image file is too large. Image too large Meaning How to Correct The firmware image file has a header checksum error. Retrieve a valid firmware image file and retry the request. The firmware image file has a checksum error. Retrieve a valid firmware image file and retry the request. Invalid status for logical disk. This error is no longer supported. Retrieve a valid firmware image file and retry the request.
Status Code Value Meaning How to Correct The controller is temporarily busy and it cannot process the request. Retry the request later. The controller is currently processing a firmware download. Retry the request once the firmware download process is complete. Retry the request once the firmware download process is complete. The disk volume specified is in a predictive failed state. Report the error to product support.
Status Code Value Meaning How to Correct Case 1: A virtual disk is in an inoperative state and the request cannot be processed. 86 Command View EVA has detected this array as inoperative. Contact HP Service for assistance. Case 2: The snapclone cannot be associated with a virtual disk that is in an inoperative state. 86 Command View EVA has detected this array as inoperative. Contact HP Service for assistance. Report the error to product support.
Status Code Value Meaning How to Correct 97 Storage system could not be locked. System busy. Try command again. Another process has already taken the SCMI lock on the storage system. Retry the request later. 98 While the request was being performed, an error occurred on the remote storage system. 'Resolve the condition and retry the request The request failed because the operation cannot be performed on a Continuous Access connection that is up. Resolve the condition and retry the request.
Error messages
A Regulatory notices This appendix includes regulatory notices for the HP StorageWorks Enterprise Virtual Array family. Regulatory notices Federal Communications Commission (FCC) notice Part 15 of the Federal Communications Commission (FCC) Rules and Regulations has established Radio Frequency (RF) emission limits to provide an interference-free radio frequency spectrum.
Class A equipment 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 instructions, may cause harmful interference to radio communications.
Laser device All Hewlett-Packard systems equipped with a laser device comply with safety standards, including International Electrotechnical Commission (IEC) 825. With specific regard to the laser, the equipment complies with laser product performance standards set by government agencies as a Class 1 laser product. The product does not emit hazardous light; the beam is totally enclosed during all modes of customer operation and maintenance.
Canadien notice (avis Canadien) Class A equipment This Class A digital apparatus meets all requirements of the Canadian Interference-Causing Equipment Regulations. Cet appareil numérique de la classe A respecte toutes les exigences du Règlement sur le matériel brouilleur du Canada. Class B equipment This Class B digital apparatus meets all requirements of the Canadian Interference-Causing Equipment Regulations.
This symbol on the product or on its packaging indicates that this product must not be disposed of with your other household waste. Instead, it is your responsibility to dispose of your waste equipment by handing it over to a designated collection point for recycling of waste electrical and electronic equipment.
Finnish notice Laitteiden hävittäminen kotitalouksissa Euroopan unionin alueella Jos tuotteessa tai sen pakkauksessa on tämä merkki, tuotetta ei saa hävittää kotitalousjätteiden mukana. Tällöin hävitettävä laite on toimitettava sähkölaitteiden ja elektronisten laitteiden kierrätyspisteeseen. Hävitettävien laitteiden erillinen käsittely ja kierrätys auttavat säästämään luonnonvaroja ja varmistamaan, että laite kierrätetään tavalla, joka estää terveyshaitat ja suojelee luontoa.
Greek notice Hungarian notice Készülékek magánháztartásban történ selejtezése az Európai Unió területén A készüléken, illetve a készülék csomagolásán látható azonos szimbólum annak jelzésére szolgál, hogy a készülék a selejtezés során az egyéb háztartási hulladéktól eltér módon kezelend . A vásárló a hulladékká vált készüléket köteles a kijelölt gy jt helyre szállítani az elektromos és elektronikai készülékek újrahasznosítása céljából.
ka š s iek rtas tiks otrreiz ji p rstr d tas t d veid , lai pasarg tu vidi un cilv ku vesel bu. Lai uzzin tu, kur nolietot s iek rtas var izmest otrreiz jai p rstr dei, j v ršas savas dz ves vietas pašvald b , sadz ves atkritumu sav kšanas dienest vai veikal , kur izstr d jums tika nopirkts. Lithuanian notice Vartotoj iš priva i nam ki rangos atliek šalinimas Europos S jungoje Šis simbolis ant gaminio arba jo pakuot s rodo, kad šio gaminio šalinti kartu su kitomis nam kio atliekomis negalima.
ur ené na recykláciu vyradených elektrických a elektronických zariadení. Separovaný zber a recyklácia vyradených zariadení prispieva k ochrane prírodných zdrojov a zabezpe uje, že recyklácia sa vykonáva spôsobom chrániacim udské zdravie a životné prostredie. Informácie o zberných miestach na recykláciu vyradených zariadení vám poskytne miestne zastupite stvo, spolo nos zabezpe ujúca odvoz domového odpadu alebo obchod, v ktorom ste si výrobok zakúpili.
Normaler Betrieb (normal operation) Nach ISO 7779:1999 (Typprüfung) Japanese notice Harmonics conformance (Japan) Taiwanese notice Japanese power cord notice Country-specific certifications HP tests electronic products for compliance with country-specific regulatory requirements, as an individual item or as part of an assembly. The product label (see Figure 38) specifies the regulations with which the product complies.
Figure 38 Typical enclosure certification label NOTE: The certification symbols on the label depend upon the certification level. For example, the FCC Class A certification symbol is not the same as the FCC Class B certification symbol.
Regulatory notices
B Non-standard rack specifications The appendix provides information on the requirements when installing the EVA4400 in a non-standard rack. All the requirements must be met to ensure proper operation of the storage system. Internal component envelope EVA component mounting brackets require space to be mounted behind the vertical mounting rails. Room for the mounting of the brackets includes the width of the mounting rails and needed room for any mounting hardware, such as screws, clip nuts, etc.
EVA cabinet measures and tolerances EVA component rack mount brackets are designed to fit cabinets with mounting rails set at depths from 27.5 inches to 29.6 inches, inside rails to inside rails. Weights, dimensions and component CG measurements Cabinet CG dimensions are reported as measured from the inside bottom of the cabinet (Z), the leading edge of the vertical mounting rails (Y), and the centerline of the cabinet mounting space (X).
Configuration Standards EVA configurations are designed considering cable length, configuration CG, serviceability and accessibility, and to allow for easy expansion of the system. If at all possible, it is best to configure non HP cabinets in a like manner. UPS Selection This section provides information that can be used when selecting a UPS for use with the EVA. The four HP UPS products listed in Table 15 are available for use with the EVA and are included in this comparison.
Minutes of operation Load (percent) With standby battery With 1 ERM With 2 ERMs R5500 100 7 24 46 80 9 31 60 50 19 61 106 20 59 169 303 R12000 100 5 11 18 80 7 15 24 50 14 28 41 20 43 69 101 Shock and vibration specifications Table 17 lists the product operating shock and vibration specifications. This information applies to products weighing 45 Kg (100 lbs) or less.
Glossary This glossary defines terms used in this guide or related to this product and is not a comprehensive glossary of computer terms. µm A symbol for micrometer; one millionth of a meter. For example, 50 µm is equivalent to 0.000050 m. 3U A unit of measurement representing three “U” spaces. “U” spacing is used to designate panel or enclosure heights. Three “U” spaces is equivalent to 5.25 inches (133 mm). See also rack-mounting unit.
asynchronous Events scheduled as the result of a signal requesting the event or that which is without any specified time relation. backplane An electronic printed circuit board that distributes data, control, power, and other signals among components within an enclosure. bad block A data block that contains a physical defect. bad block replacement A replacement routine that substitutes defect-free disk blocks for those found to have defects.
communication logical unit number (LUN) See console LUN. console LUN A SCSI-3 virtual object that makes a controller pair accessible by the host before any virtual disks are created. Also called a communication LUN. console LUN ID The ID that can be assigned when a host operating system requires a unique ID. The console LUN ID is assigned by the user, usually when the storage system is initialized. See also console LUN.
disk drive enclosure A unit that holds storage system devices such as disk drives, power supplies, fans, I/O modules, and transceivers. disk failure protection A method by which a controller pair reserves drive capacity to take over the functionality of a failed or failing physical disk. For each disk group, the controllers reserve space in the physical disk pool equivalent to the selected number of physical disk drives. disk group A physical disk drive set or pool in which a virtual disk is created.
EMI Electromagnetic Interference. The impairment of a signal by an electromagnetic disturbance. enclosure A unit used to hold various storage system devices such as disk drives, controllers, power supplies, I/O modules, or fans. Enterprise Virtual Array The Enterprise Virtual Array is a product that consists of one or more storage systems. Each storage system consists of a pair of HSV controllers and the disk drives they manage.
FCA Fibre Channel Adapter. An adapter used to connect the host server to the fabric. Also called a Host Bus Adapter (HBA) or a Fibre Channel Host Bus Adapter (FC HBA). See also FC HBA. FCC Federal Communications Commission. The federal agency responsible for establishing standards and approving electronic devices within the United States. FCP Fibre Channel Protocol. The mapping of SCSI-3 operations to Fibre Channel. fiber The optical media used to implement Fibre Channel.
Gb Gigabit. A measurement of the rate at which the transfer of bits of data occurs. Sometimes referred to as Gbps. Nominally, a Gb is a transfer rate of 1,000,000,000 (109) bits per second. For Fibre Channel transceivers or FC loops the Gb transfer rates are: • 1 Gb is a transmission rate of 1,062,500,000 bits per second. • 2 Gb is a transmission rate of 2,125,000,000 bits per second. GB Gigabyte. A unit of measurement defining either: • A data transfer rate.
one or more of the virtual disks that are created and presented by the controller pair. host-side ports See host ports. hot-pluggable A method of element replacement whereby the complete system remains operational during element removal or insertion. Replacement does not interrupt data transfers to other elements. hub A communications infrastructure device to which nodes on a multi-point bus or loop are physically connected. It is used to improve the manageability of physical cables.
License Key A WWN-encoded sequence that is obtained from the license key fulfillment website. light emitting diode See LED. link A connection ports on Fibre Channel devices. The link is a full duplex connection to a fabric or a simplex connection loop devices. logon Also called login, it is a procedure whereby a user or network connection is identified as being an authorized network user or participant. loop pair A Fibre Channel attachment a controller and physical disk drives.
mirrored caching A process in which half of each controller’s write cache mirrors the companion controller’s write cache. The total memory available for cached write data is reduced by half, but the level of protection is greater. mirroring The act of creating an exact copy or image of data. MTBF Mean Time Failures. The average time from start of use to first failure in a large population of identical systems, components, or devices.
PDU Power Distribution Unit. The rack device that distributes conditioned AC or DC power within a rack. petabyte A unit of storage capacity that is the equivalent of 250, 1,125,899,906,842,624 bytes or 1,024 terabytes. physical disk A disk drive mounted in a drive enclosure that communicates with a controller pair through the device-side Fibre Channel loops. A physical disk is hardware with embedded software, as opposed to a virtual disk, which is constructed by the controllers.
than from the disk drives. Reading data from cache memory is faster than reading data from a disk. The read cache is specified as either On or Off for each virtual disk. The default state is on. read ahead caching A cache management method used to decrease the subsystem response time to a read request by allowing the controller to satisfy the request from the cache memory rather than from the disk drives. reconstruction The process of regenerating the contents of a failed member data.
Snapclone A virtual disk that can be manipulated while the data is being copied. Only an Active member of a virtual disk family can be snapcloned. The Snapclone, like a snapshot, reflects the contents of the source virtual disk at a particular point in time. Unlike the snapshot, the Snapclone is an actual clone of the source virtual disk and immediately becomes an independent Active member of its own virtual disk family.
UID Unit identification. uninitialized system A state in which the storage system is not ready for use. See also initialization. unwritten cached data Also called unflushed data. See also dirty data. UPS Uninterruptible Power Supply. A battery-operated power supply guaranteed to provide power to an electrical device in the event of an unexpected interruption to the primary power supply.
the physical disks. If one of the data chunks becomes corrupted, the data can be reconstructed from the parity chunk and the remaining data chunks. WOCP Web-based Operator Control Panel. The virtual operator control panel implemented as a web-based component. World Wide Name See WWN. write back caching A controller process that notifies the host that the write operation is complete when the data is written to the cache. This occurs before transferring the data to the disk.
Glossary
Index customer self repair, 13 parts list, 50 A AC power distributing, 28 D B bad image header, 88 bad image segment, 88 bad image size, 88 bays locating, 16 numbering, 16 bidirectional operation, 18 C cables FCC compliance statement, 94 cabling controller, 27 Cache batteries failed or missing, 86 CDRH, compliance regulations, 95 Center for Devices and Radiological Health See CDRH certification product labels, 102 Class A equipment, Canadian compliance statement, 96 Class B equipment, Canadian complianc
Fibre Channel Disk Enclosure defined, 15 maximum size exceeded, 87 media inaccessible, 81 H N hardware components, 15 harmonics conformance Japan, 102 help obtaining, 13 high availability HSV Controllers, 21 HP technical support, 13 HSV Controllers defined, 15 no image, 81 no FC port, 81 no logical disk for Vdisk, 85 no more events, 85 no permission, 81 non-standard rack, specifications, 105 not a loop port, 81 not participating controller, 81 I I/O modules bidirectional, 18 image already loaded, 88 i
regulatory compliance notices cables, 94 Class A, 94 Class B, 94 European Union, 96 laser devices, 95 modifications, 94 regulatory notices, 93 websites customer self repair, 13 HP , 14 HP Subscriber's Choice for Business, 13 WEEE recycling notices, 96 S security credentials invalid, 86 Security credentials needed, 86 slots See enclosures, bays status, disk drives, 20 storage connection down, 86 storage not initialized, 81 storage system racks, defined;, 28 Subscriber's Choice, HP, 13 symbols in text, 12 s
