Maintenance and Service Guide HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Product Version: 8.8-1 Third Edition (March 2005) Part Number: EK–G80MS–SA. C01 This guide provides step-by-step hardware and firmware installation instructions for HP StorageWorks HSG60 and HSG80 array controllers. It also serves as a reference for the operation, troubleshooting, and future upgrades of these array controllers.
© Copyright 2000–2005 Hewlett-Packard Development Company, L.P. Hewlett-Packard Company makes no warranty of any kind with regard to this material, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. Hewlett-Packard shall not be liable for errors contained herein or for incidental or consequential damages in connection with the furnishing, performance, or use of this material.
contents Contents About this Guide. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Intended audience. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents Model 2100 and 2200 enclosure EMU switches and LEDs . . . . . . . . . . . . . . . . . . . . . . 41 Model 2100 and 2200 enclosure ECB LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Model 2100 and 2200 enclosure fan LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 2 Common Replacement Information and Procedures . . . . . . . . . . . . . . . . . . . . . . . . .45 Required tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents Replacing a failed JBOD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Replacing a storage unit with partitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Replacing a disk drive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Removing and replacing all disk drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents Replacing a PVA module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Replacing an I/O module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Handling a failed DIMM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 4 Upgrading the Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents Upgrading firmware on a device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 Upgrading to a dual-redundant controller configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Upgrading cache memory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 A Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 8 Using thin needle nose pliers to disconnect a fiber optic cable . . . . . . . . . . . . . . . . . . . . 59 Controller Reset button and Port #5 button on the OCP . . . . . . . . . . . . . . . . . . . . . . . . . 65 Reset button and first three LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 Cache module locations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Removing or installing a DIMM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Install Options screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents 94 Example of the screen displayed after issuing the SHOW RAIDSETS SPECIAL_FUNCTION_ONE command (top controller) . . . . . . . . . . . . . . . . . . . . . . 214 95 Example of the screen displayed after issuing the SHOW RAIDSETS SPECIAL_FUNCTION_ONE command (bottom controller) . . . . . . . . . . . . . . . . . . . . 215 96 Example of a RAIDset needing repair on the top controller (following rolling upgrade v87P-6 to V88-1P) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents Tables 1 Related Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2 Document conventions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 3 HSG80 subsystem components and part numbers for BA370 enclosures. . . . . . . . . . . . 26 4 HSG60 and HSG80 subsystem components and part numbers for Model 2100 and 2200 enclosures. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents 12 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide
about this guide About this Guide This maintenance and service guide provides information to help you: ■ Identify, remove, and replace StorageWorks BA370, M2100, and M2200 About this HP Guide enclosure components. ■ Upgrade HP StorageWorks HSG60 or HSG80 array controller firmware.
About this Guide Overview This section covers the following topics: ■ Intended audience ■ Prerequisites ■ Related documentation Intended audience This book is intended for use by users who are experienced with the following: ■ HSG60 and HSG80 array controllers ■ HP StorageWorks Array Controller Software (ACS), V8.
About this Guide ■ Determine the model and types of components installed in your enclosure. The procedures in this document are specific to HSG60 and HSG80 array controllers in BA370, Model 2100, and Model 2200 enclosures. Component part numbers are detailed in the “General Description” section, which starts on page 23. Related documentation Other documentation relative to HSG60 and HSG80 hardware, software, and firmware is listed in Table 1.
About this Guide Table 1: Related Documentation (continued) Item Document Name Document Part Number 12. HP StorageWorks HSG80 Enterprise/Modular Storage RAID Array Fibre Channel Solution Software Version 8.8 for IBM AIX Release Notes AA-RV1JA-TE 13. HP StorageWorks HSG80 Enterprise/Modular Storage RAID Array Fibre Channel Solution Software Version 8.8 for Linux X86 and Alpha Release Notes AA-RV1KA-TE 14. HP StorageWorks HSG80 ACS Solution Software Version 8.
About this Guide Table 1: Related Documentation (continued) Item Document Name Document Part Number 26. Compaq StorageWorks 64-Bit PCI-to-Fibre Channel Host Bus Adapter User Guide AA-RKPDB-TE 27. Digital StorageWorks UltraSCSI RAID Enclosure (DS-BA370-Series) User’s Guide EK-BA370-UG. B01 28. HP StorageWorks HSG80 ACS Solution Software Version 8.8 for Windows Installation and Configuration Guide AA- RV1XA -TE 29. HP StorageWorks HSG80 ACS Solution Software Version 8.
About this Guide Conventions Conventions consist of the following: ■ Document conventions ■ Text symbols ■ Equipment symbols Document conventions This document follows the conventions in Table 2.
About this Guide Tip: Text in a tip provides additional help to readers by providing nonessential or optional techniques, procedures, or shortcuts. Note: Text set off in this manner presents commentary, sidelights, or interesting points of information. Equipment symbols The following equipment symbols may be found on hardware for which this guide pertains.
About this Guide Power supplies or systems marked with these symbols indicate the presence of multiple sources of power. WARNING: To reduce the risk of personal injury from electrical shock, remove all power cords to completely disconnect power from the power supplies and systems. Any product or assembly marked with these symbols indicates that the component exceeds the recommended weight for one individual to handle safely.
About this Guide Rack stability Rack stability protects personnel and equipment. WARNING: To reduce the risk of personal injury or damage to the equipment, be sure that: ■ The leveling jacks are extended to the floor. ■ The full weight of the rack rests on the leveling jacks. ■ In single rack installations, the stabilizing feet are attached to the rack. ■ In multiple rack installations, the racks are coupled. ■ Only one rack component is extended at any time.
About this Guide Getting help If you still have a question after reading this guide, contact an HP authorized service provider or access our website: http://www.hp.com. HP technical support Telephone numbers for worldwide technical support are listed on the following HP website: http://www.hp.com/support/. From this website, select the country of origin. Note: For continuous quality improvement, calls may be recorded or monitored.
General Description 1 This chapter illustrates and describes HSG60 and HSG80 subsystems and their components: ■ Subsystem components—exploded views, page 24 ■ Connectors, switches, and LEDs, page 35 Information regarding these components apply to the following enclosures: ■ HP StorageWorks BA370 enclosure ■ HP StorageWorks Model 2100 enclosure ■ HP StorageWorks Model 2200 enclosure Refer to the Fibre Channel switch and host bus adapter (HBA) documentation that came with the switch kit for specif
General Description Subsystem components—exploded views This section illustrates the various subsystem enclosures and components, and it describes specific attributes of each component. Spare part numbers are included, as a convenience, under two part numbering systems (the 6–3 system and the 2–5–2 system). HSG80 subsystem in a BA370 enclosure Figure 1 on page 25 and Table 3 on page 26 identify an HSG80 subsystem in a BA370 enclosure.
General Description 1 18 2 17 15 16 3 14 4 13 12 9 11 5 9 10 6 9 9 8 7 CXO7180A 1 2 3 4 5 6 7 8 9 BA370 rack-mountable enclosure Cooling fan Power cable kit I/O module Fibre channel hub, 7-port Fibre channel hub, 12-port Fibre channel hum, 8 port Fibre channel switch, 16-port Gigabit interface converter (GBIC) q w e r t y u i Cache module HSG8 array controller Power verification and addressing module Environmental monitoring unit (EMU) AC input box 180-watt power supply Disk drive Power
General Description Table 3: HSG80 subsystem components and part numbers for BA370 enclosures Item 26 Description 6–3 Part Number 2–5–2 Part Number 1 BA370 rack-mountable enclosure 401914–001 DS–BA370–MA 2 Cooling fan, blue Cooling fan, gray 400293–001 402602–001 FC–BA35X–MK FC–BA35X–ML 3 Power cable kit, white 401916–001 17–03718–10 4 Input/output (I/O) module, blue I/O module, gray 400294–001 401911–001 FC–BA35X–MN 70–32856–S2 5 Fibre Channel hub, 7-port 234454–001 FE–09061–01 6
General Description Table 3: HSG80 subsystem components and part numbers for BA370 enclosures (continued) Item y Description Disk drives 6–3 Part Number various1 2–5–2 Part Number various (4 GB to 18 GB drives) u Power cable, gray Power cable, black 401915–001 401916–001 17–03718–09 17–03718–10 i External cache battery (ECB), dual (shown) ECB, single (not shown) 400291–001 FC–HS35X–BD 400292–001 FC–HS35X–BC 1. Contact HP support for disk drive part numbers.
General Description HSG60 and HSG80 subsystems in Model 2100 and 2200 enclosures Figure 2, Figure 3 (on page 29) and Table 4 (on page 30) identify HSG60 and HSG80 subsystems in Model 2100 and 2200 enclosures.
General Description 1 2 10 3 4 2 3 5 14 3 9 14 11 6 7 8 7 8 6 12 14 14 13 CXO7176A 1 2 3 4 5 6 7 Model 2100 and 2200 rack-mountable enclosure ECB Blank bezel Fan EMU 180-watt power supply Array controller 8 9 Cache module I/O module q w e r External cache battery Blank bezel Fibre channel hub, 7-port Fibre channel hub, 12-port Fibre channel switch, 16-port Figure 3: HSG80 subsystem in a Model 2200 enclosure HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software
General Description Table 4: HSG60 and HSG80 subsystem components and part numbers for Model 2100 and 2200 enclosures Item 30 Description 6–3 Part Number 2–5–2 Part Number 1 Model 2100 and 2200 rack-mountable enclosure 126314–001 70–33725–S1 2 ECB without a battery pack ECB battery pack 126312–001 147514–001 70–33547–S1 12–44670–S4 3 Blank bezel 173406–001 74–60460–01 4 Fan 126310–001 70–33538–S1 5 EMU 126315–001 70–40081–S1 6 180-watt power supply 400288–001 FC–BA35X–HH 7 HSG
General Description BA370 enclosure EMU Figure 4 and Table 5 identify the Environmental Monitoring Unit (EMU) and EMU-EMU communication cable for BA370 enclosures.
General Description HSG60 and HSG80 array controllers Figure 5 and Table 6 (on page 33) identify an array controller, the location of the program card or PCMCIA (Personal Computer Memory Card International Association) card, and how to connect a fiber optic cable and maintenance port cable. This setup applies to BA370, Model 2100, and Model 2200 enclosures.
General Description Table 6: HSG60 and HSG80 array controller assembly Item Description 1 Gigabit link module (GLM) 2 Program card: - ACS V8.8-1F - ACS V8.8-1G - ACS V8.8-1L (HSG60 only) - ACS V8.8-1P - ACS V8.
General Description Cache module Figure 6 and Table 7 identify the cache module and the location of the dual inline memory module (DIMM). The bezel of the upgraded cache module is labeled HSx80 CACHE. Caution: An older version of the HP StorageWorks cache module (part number 70-33256-01) cannot be used in Model 2100 and 2200 enclosures. These enclosures require the updated version (part number 70-33256-11, spare part number 70-33256-S1) to function properly.
General Description Connectors, switches, and LEDs This section provides connector, switch, and LED information for the following components: ■ HSG60 and HSG80 array controller front panel, OCP, and GLM ■ LEDs for a 180–watt power supply ■ BA370 enclosure PVA module connector and switches ■ BA370 enclosure EMU connectors, switches, and LEDs ■ Model 2100 and 2200 enclosure EMU switches and LEDs ■ Model 2100 and 2200 enclosure ECB LEDs ■ Model 2100 and 2200 enclosure fan LEDs Note: For detailed
General Description HSG60 and HSG80 array controller front panel, OCP, and GLM This section describes the array controller front panel, operator control panel (OCP) switches and LEDs, and gigabit link module (GLM) components. Front panel Figure 7 identifies various physical parts of the array controller.
General Description GLM connectors and components Figure 9 identifies various GLM connectors and components on the array controller.
General Description LEDs for a 180–watt power supply Figure 10 identifies power supply LEDs for BA370, Model 2100, and Model 2200 enclosures. Note: Refer to the specific enclosure user guide for detailed information.
General Description BA370 enclosure PVA module connector and switches Figure 11 identifies the power verification and addressing (PVA) module connector and switches for BA370 enclosures. Note: Refer to the BA370 enclosure user guide for detailed information.
General Description BA370 enclosure EMU connectors, switches, and LEDs Figure 12 identifies EMU connectors, switches, and LEDs for BA370 enclosures. Note: Refer to the BA370 enclosure user guide for detailed information.
General Description Model 2100 and 2200 enclosure EMU switches and LEDs Figure 13 identifies EMU switches and LEDs for Model 2100 and 2200 enclosures. Note: Refer to the Model 2100 and 2200 enclosure user guide for detailed information.
General Description Model 2100 and 2200 enclosure ECB LEDs Figure 14 identifies ECB LEDs for Model 2100 and 2200 enclosures. Note: Refer to the Model 2100 and 2200 enclosure user guide for detailed information.
General Description Model 2100 and 2200 enclosure fan LEDs Figure 15 identifies fan LEDs for Model 2100 and 2200 enclosures. Note: Refer to the Model 2100 and 2200 enclosure user guide for detailed information.
General Description 44 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide
Common Replacement Information and Procedures 2 This chapter contains the following replacement information and procedures: ■ Required tools, page 46 ■ Field replacement strategy, page 47 ■ Precautions, page 48 ■ Establishing a local connection to the controller, page 50 ■ Fiber optic cable cleaning instructions, page 54 ■ Cleaning a GLM, page 55 ■ Replacing a GLM, page 57 ■ Replacing a fiber optic cable, switch, or hub, page 67 ■ Replacing DIMMs, page 70 ■ Handling storagesets, mirrors
Common Replacement Information and Procedures Required tools The following tools may be needed to service the array controller, cache module, ECB, PVA module, GLM, and I/O module: 46 ■ Flathead screwdriver for loosening and tightening the I/O module retaining screws ■ Small Phillips screwdriver for loosening and tightening the GLM access door screws ■ Pair of thin needle-nose pliers to remove fiber optic cables from the array controller ■ Antistatic wrist strap ■ Antistatic mat on which to place
Common Replacement Information and Procedures Field replacement strategy HP-authorized service providers troubleshoot HSG60 and HSG80 subsystem problems to the field replaceable unit (FRU) and replace the defective unit. Replacement procedures for subsystem FRUs addressed in this guide include HSG60 and HSG80 array controllers, cache modules, DIMMs, ECBs, and GLMs. One way to replace these components on running subsystems is by using the Field Replacement Utility (FRUTIL).
Common Replacement Information and Procedures Precautions To prevent accidental damage to subsystem components, always follow the precautions in this section when carrying out the procedures in this guide. Electrostatic discharge precautions Static electricity collects on all non-conducting material, such as paper, cloth, and plastic. An electrostatic discharge (ESD) can easily damage the array controller or other subsystem components even though the discharge might not be seen or felt.
Common Replacement Information and Procedures Very-high-density cable interface cable precautions All cables that connect to the array controller, cache module, and ECB use very-high-density cable interface (VHDCI) connectors. VHDCI connectors have extraordinarily small mating surfaces that can be damaged by dust and cable movement. Always take the following VHDCI precautions when servicing any subsystem component: ■ Clean the mating surfaces with only a blast of compressed air or freon.
Common Replacement Information and Procedures Establishing a local connection to the controller Communication with an array controller is established locally or remotely. Use a local connection to configure the array controller for the first time. Use a remote connection to the host system for all subsequent configuration tasks. Refer to the controller installation and configuration guide that shipped with the platform kit for details.
Common Replacement Information and Procedures Tip: Following is an example of how the baud rate is set while replacing a controller: Example: A controller pair is operating with ACS V8.8-x with the top and bottom at 4800 baud. The bottom controller is to be replaced. As such, you would complete the following steps: 1. Issue the SET FAILOVER command on the top controller. 2. Use FRUTIL to replace the bottom controller with a factory fresh controller and a V8.8-x program card. 3.
Common Replacement Information and Procedures 1 2 1 2 3 4 5 6 Maintenance port cable for a PC connection Maintenance port Adapters (not shown), 9-pin D-sub to 25-pin D-sub: 1 2 ■ ■ ■ Male to female (null modem) Male to male (null modem) Male to male (modem) CXO7181A Figure 16: PC and terminal to maintenance port connection Caution: Connecting a maintenance port cable between the array controller and the PC (or terminal) can cause radio and television interference and cause the subsystem to op
Common Replacement Information and Procedures 4. Press Enter or Return. The CLI prompt appears, indicating that a local connection was established with the array controller. Note: The default data transfer rate of a new HSG80 array controller using ACS V8.6 and earlier is 9600 baud. (The default data transfer rate for controllers using ACS V8.7 may be set at 19200.) The maximum transfer rate is 19200. If the current configuration uses 19200, use step 5 to establish this rate.
Common Replacement Information and Procedures Fiber optic cable cleaning instructions To ensure optimum performance of the cable, clean the fiber optic cables while replacing the array controller. Use the polyester cloth from the cleaning kit that came with the cable. Figure 17 shows the ferrule on a fiber optic cable. 1 Ferrule 1 CXO6503B Figure 17: Ferrule on a fiber optic cable Caution: HP recommends cleaning the fiber optic cable only when replacing the array controller.
Common Replacement Information and Procedures Cleaning a GLM If you are replacing a GLM, clean the GLM receptacles to ensure optimum performance. Use the following steps to clean a GLM: 1. Take the polyester swab from the cleaning kit that came with the fiber optic cable. Caution: Do not exert excessive force when the swab reaches a stopping point, and do not repeat this procedure frequently.
Common Replacement Information and Procedures 3. Carefully remove dust from the cavity by rotating the swab tip back and forth one or two times. 4. Repeat step 1 (on page 55) through step 3 for the receiving side of the optical GLM cavity.
Common Replacement Information and Procedures Replacing a GLM This section shows you how to replace a GLM in an HSG60 or HSG80 array controller. Figure 19 shows the location and orientation of the GLMs. 1 3 2 4 5 6 7 CXO6245C 1 2 3 4 Access door Port 1 GLM Release lever Locking tab 5 6 7 Guide holes GLM connector Port 2 GLM Figure 19: Location of GLMs inside an array controller (HSG80 shown) Caution: ESD can easily damage the array controller or the GLM.
Common Replacement Information and Procedures Replacing a GLM in single-controller configurations Use the steps in “Removing a GLM” and “Installing a GLM” on page 60 to replace a GLM in a single-controller configuration. Caution: Static electricity can easily damage a controller or GLM. Wear a snug-fitting, grounded electrostatic discharge (ESD) wrist strap. Removing a GLM Use the following steps to remove a GLM in single-controller configurations: 1. Check the controller to make sure it is operating.
Common Replacement Information and Procedures 1 2 1 2 3 4 5 6 CXO6991A 1 Reset button 2 First three port LEDs Figure 20: Controller Reset button and first three port LEDs on the OCP 6. Disconnect all host bus cables from the controller. Caution: For fiber optic cables without extender clips, thin needle nose pliers must be used to remove the cable from the controller without potentially damaging the cable (see Figure 21).
Common Replacement Information and Procedures 7. If connected to an operational controller, disconnect the PC or terminal from the controller maintenance port. 8. Disengage both retaining levers and remove the controller containing the failed GLM from the enclosure, and then place the controller on an antistatic bag or a grounded antistatic mat. 9. Remove the screw that secures the access door on top of the controller. 10. Remove the access door and set the door aside. 11.
Common Replacement Information and Procedures Note: After fully seated, the controller starts automatically—the Reset LED turns on. A controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. Note: If the controller did not restart, use the following steps: a. Press and hold the controller Reset button. b. Reseat the controller program card. c. Release the Reset button. 6.
Common Replacement Information and Procedures Removing a GLM Use the following steps to remove a GLM in dual-redundant configurations: 1. Connect a PC or terminal to the maintenance port of the operational controller. Note: The controller connected to the PC or terminal becomes “this controller,” and the controller being removed becomes the “other controller.” 2.
Common Replacement Information and Procedures 8. Remove the “other controller” by using the following substeps: a. Disconnect all host bus cables or terminators from the controller. Caution: For fiber optic cables without extender clips, thin needle nose pliers must be used to remove the cable from the controller without potentially damaging the cable (see Figure 21 on page 59). b. Disengage both retaining levers and remove the controller containing the failed GLM from the enclosure. c.
Common Replacement Information and Procedures 4. Install the access door on top of the controller and secure the door with the screw. 5. Connect a PC or terminal to the maintenance port of the operational controller. Note: The controller connected to the PC or terminal becomes “this controller,” and the controller being installed becomes the “other controller.” 6.
Common Replacement Information and Procedures 12. Insert the controller and engage the retaining levers. Note: In step a, ensure the program card is not installed in the replacement controller. a. Insert the controller (without the program card installed) into the appropriate bay, and engage the retaining levers. b. Press and hold the controller Reset button (see Figure 22), insert the program card, and continue holding the Reset button. c.
Common Replacement Information and Procedures 13. Press Enter or Return within 3 minutes of completing substep 12e on page 65 to exit FRUTIL, and then wait 1 minute to allow the controller to restart. Note: If Enter or Return is not pressed within 3 minutes in step 13 above, the operational controller issues an automated command to cancel the installation of the replaced controller. This cancellation causes all the port LEDs on the controller to go off.
Common Replacement Information and Procedures Replacing a fiber optic cable, switch, or hub This section details how to replace a fiber optic cable, switch, or hub. Removing a fiber optic cable, switch, or hub Use the following steps to remove a fiber optic cable connected to either side of your switch or hub, or to remove the switch or hub: 1. Shut down the host system (refer to applicable host documentation for more details). 2.
Common Replacement Information and Procedures Tip: If you are replacing several cables, a switch, or a hub, label each cable to facilitate installation of the replacement item. Otherwise, subsystem errors can develop from improper connections. 3. If replacing a fiber optic cable without extender clips, disconnect the failed cable at each end using thin needle-nose pliers (see inset on Figure 24). If replacing a switch or hub, disconnect all cables connected to the switch or hub.
Common Replacement Information and Procedures 3. Restart each array controller by pressing the Reset button. The array controller automatically restart and the subsystem is now ready for operation. Tip: If the array controller did not restart, use step a through step c. a. Press and hold the Reset button. b. Reseat the array controller program card. c. Release the Reset button. 4. Restart the host system using host documentation.
Common Replacement Information and Procedures Replacing DIMMs This section shows you how to replace DIMMs in a cache module. DIMM locations are shown in Figure 25 and supported configurations in Table 8.
Common Replacement Information and Procedures Caution: ESD can easily damage a cache module or a DIMM. Wear a snug-fitting, grounded ESD wrist strap while replacing DIMMs. Note: If a DIMM fails, note which DIMM needs replacement based on the diagram displayed on the console. ACS displays the following CLI messages if a DIMM failure is detected: 1.%CER--DebugTop> --18-MAY-2005 10:50:47-- Cache module DIMM 2 failed 2.
Common Replacement Information and Procedures Figure 26 illustrates how to remove and install DIMMs. 1 2 1 2 3 DIMM DIMM slot DIMM retaining clip 3 5025 Figure 26: DIMM components Replacing an HSG60 and HSG80 DIMM in single-controller configurations Use the steps in “Removing a GLM” and “Installing a GLM” to replace DIMMS in a cache module. Caution: Static electricity can easily damage a cache module or a DIMM. Wear a snug-fitting, grounded electrostatic discharge (ESD) wrist strap.
Common Replacement Information and Procedures ■ If the controller is operational, connect a PC or terminal to the controller maintenance port. ■ If the controller is not operational, Go to step 6 on page 74. 2. From the host console, stop all host activity to the controller, and then dismount the logical units in the subsystem. 3. If using a Microsoft® Windows 2000® or Windows NT® platform, shut down the server. 4. Run the Fault Management Utility (FMU) to obtain the last failure codes, if desired. 5.
Common Replacement Information and Procedures Caution: You must disable the ECB before disconnecting the ECB Y-cable from the cache module. Failure to disable the ECB can damage the cache module. 6. Disable the ECB by pressing in the battery disable switch while removing the ECB Y-cable from the cache module. 7. Release the battery disable switch. 8. Disengage both retaining levers, remove the cache module, and then place the cache module on an antistatic bag or a grounded antistatic mat.
Common Replacement Information and Procedures 4 1 2 3 CXO6577B 1 2 DIMM DIMM slot 3 4 DIMM retaining clips Rail guide Figure 28: Removing or installing a DIMM 11. Gently remove the DIMM from the DIMM slot, and then place it in an antistatic bag or on a grounded antistatic mat. 12. Repeat step 10 and step 11 for each DIMM being replaced.
Common Replacement Information and Procedures Installing DIMMs Use the following steps to install DIMMs in a cache module in single-controller configurations: Note: Before installing DIMMs, note the following: ■ New cache modules have no DIMMs installed; the DIMMs are packaged separately. Unpack the DIMMs and install them into the cache module as required. ■ Both cache modules must contain the same memory configuration for the ACS version.
Common Replacement Information and Procedures Caution: You must disable the ECB before disconnecting the ECB Y-cable from the cache module. Failure to disable the ECB can damage the cache module. 8. If not already done, disable the ECB by pressing the battery disable switch for approximately 5 seconds. 9. Connect the ECB Y-cable to the cache module. 10. If not already connected, connect a PC or terminal to the controller maintenance port. 11. Restart the controller by pressing the Reset button.
Common Replacement Information and Procedures Replacing an HSG60 and HSG80 DIMM in dual-redundant controller configurations Use the steps in “Removing a GLM” and “Installing a GLM” to replace DIMMS in a cache module: Note: If you are upgrading cache memory, be sure to install a valid memory configuration (see Figure 25 on page 70). Failure to do so results in an inoperable controller and increases subsystem down-time. Both cache modules must contain the same memory configuration for the ACS version.
Common Replacement Information and Procedures 7. Enter Y(es) to confirm the intent to remove the “other controller” cache module. 8. Wait for FRUTIL to quiesce the device ports—indicated by an “All device ports quiesced” message. Caution: Failure to allow the ports to quiesce can result in data loss. Quiescing can take several minutes. The ECB must be disabled before disconnecting the ECB cable from the cache module. Failure to disable the ECB might result in cache module damage.
Common Replacement Information and Procedures 13. Press down on the DIMM retaining clips at both ends of the DIMM being removed (see Figure 28 on page 75). Note: To facilitate pressing down on the DIMM retaining clips, use the eraser end of a pencil or a small screwdriver. 14. Gently remove the DIMM from the DIMM slot, and then place it in an antistatic bag or on a grounded antistatic mat. 15. Repeat step 13 and step 14 for each DIMM being replaced.
Common Replacement Information and Procedures The controller connected to the PC or terminal becomes “this controller”; the controller for the cache module being installed becomes the “other controller.” 7. Start FRUTIL with the following command: RUN FRUTIL 8. Enter N(o) to the question about replacing the cache battery. 9. Enter 2 for the install a controller or cache module option. 10. Enter 3 for the install the “other controller” cache module option. 11.
Common Replacement Information and Procedures Note: If the “other controller” did not restart: a. Press and hold the “other controller” Reset button. b. Reseat the “other controller” program card. c. Release the Reset button. Tip: To verify that the “other controller” restarts, connect a PC or terminal to the maintenance port, and then verify that it passed the cache diagnostic test. Caution: In step 14, entering the appropriate SET command is critical.
Common Replacement Information and Procedures Handling storagesets, mirrorsets, stripesets, RAIDsets and JBODs The following subsections provide information on handling complex storageset, mirrorset, stripeset, RAIDset and JBOD issues. Replacing failed storageset members If a disk drive fails in a RAIDset or mirrorset, array controllers automatically place the disk drive into the failedset.
Common Replacement Information and Procedures Removing a failed RAIDset or mirrorset member Use the following guidelines when replacing disk drives: ■ Never remove more than one disk drive at a time. If a disk drive is replaced, array controllers use data from the other disk drives in the array to reconstruct data on the replacement disk drive. If more than one disk drive is removed, a complete data set is not available to reconstruct data on the replacement disk drives, and permanent data loss may occur.
Common Replacement Information and Procedures Installing a new RAIDset or mirrorset member To install a new RAIDset or mirrorset member, insert a new disk drive that satisfies the replacement policy of the reduced storageset into the PTL location of the failed disk drive. Note: HSG60 and HSG80 array controllers automatically initialize a new disk drive and place this disk drive into the spareset if the AUTOSPARE switch is enabled. (By default, array controllers are set to NOAUTOSPARE.
Common Replacement Information and Procedures 1. Verify which RAIDset is to be moved (see Figure 29).
Common Replacement Information and Procedures 3. Add the RAIDset on the “other controller” (see Figure 31). zBot> add disk disk20400 2 4 0 3 Members of the original RAIDset zBot> add disk disk50300 5 3 0 zBot> add disk disk60200 6 2 0 New disk that is not a member zBot> add disk disk10300 1 3 0 zBot> init disk10300 Figure 31: Adding the RAIDset to the “other controller” and adding a new disk 4. Add the new disk, initialize it (see Figure 31), and then issue the SHOW command (see Figure 32).
Common Replacement Information and Procedures %EVL--zTop> --28-JAN-2004 16:35:39-- Instance Code: 02675201 < The device specified in the Device Locator field has been removed from the RAIDset associated with the logical unit. The removed device is now in the Failedset. The RAIDset is now in Reduced state. > Template: 81.(51) Occurred on 28-JAN-2004 at 16:35:39 Power On Time: 0. Years, 340. Days, 21. Hours, 31. Minutes, 40.
Common Replacement Information and Procedures 6. Verify that the RAIDset is moved and configured correctly, and that the one disk is moved to the failedset.
Common Replacement Information and Procedures zBot> show d5 LUN Uses Used by ---------------------------------------------------------------------------- D5 R5 LUN ID: 6000-1FE1-0007-A220-0009-1040-2700-00A9 IDENTIFIER = 5 Switches: RUN NOWRITE_PROTECT READAHEAD_CACHE READ_CACHE WRITEBACK_CACHE MAX_READ_CACHED_TRANSFER_SIZE = 32 MAX_WRITE_CACHED_TRANSFER_SIZE = 32 Access: ALL State: ONLINE to this controller Not reserved NOPREFERRED_PATH Size: 106643109 blocks Geometry (C/H/S): ( 20993 / 20 /
Common Replacement Information and Procedures 7. Delete the failedset, and then place the disk of choice into the RAIDset so that the unit can be Normalized: zBot> delete FAILDSET disk10300 Moving storagesets Move storagesets that are in Normal mode only. Never move storagesets that are reconstructing (content regeneration after a failed disk to a spare) or were reduced (missing member from a mirrorset or RAIDset).
Common Replacement Information and Procedures 7. Add a new disk to the array controller configuration with one of the following commands: ADD DISKS DISKXXXX P T L or RUN CONFIG Note: If you submit the RUN CONFIG CLI command when a bad disk drive exists, ACS skips the bad disk drive while running the RUN CONFIG command and reports the following error: DEVICE AT P1:T4:L0 failed initialization, Skipping Device 8. Add a unit with the following command: ADD UNITS DXXX DISKXXXX 9.
Common Replacement Information and Procedures the replacement (backup and restore) of the data contained in all partitions on that device. This is particularly true of a partitioned JBOD and stripeset. Use great care if you are planning the use of partitions for anything other than temporary data storage.
Common Replacement Information and Procedures Replacing a disk drive Disk drive hot swapping is supported under the circumstances described below. Abide by the following rules related to disk drive removal and disk drive insertion into HSG80 subsystems.
Common Replacement Information and Procedures ■ If at any time, disk drive errors occur and you must contact an HP service representative, submit the SHOW DEVICE_ERRORS FMU command prior to contacting an HP service representative. Submitting the SHOW DEVICE_ERRORS FMU command generates a report that assists the HP service representative in resolving device problems.
Common Replacement Information and Procedures If a disk drive is replaced after the array controller is off, install the replacement disk drive before restoring power to the array controller. After the power is restored, use the DELETE container-name CLI command to remove the disk from the configuration. Then, use the ADD DISKS CLI command to add the new disk drive.
Common Replacement Information and Procedures swap process. This involves quiescing the device bus for the device to be removed and replaced. This activity provides a momentary stall on that bus, while work continues on the adjacent bus. Using the following steps to complete a warm swap: 1. Press the appropriate port button on the array controller front panel until the I/O quiesces on the bus. 2. Remove the disk drive. 3. Repeat the above steps to replace a disk drive.
Common Replacement Information and Procedures CLI Commands to HSG80 or Actions Wait for Port to quiesce. Comments Go to the side of cab from where you remove the disk drive and wait approximately 10-15 seconds. The port is quiesced, when all the disk drives on that port show a flashing amber LED.
Common Replacement Information and Procedures Disk drive auto-read-reallocate bit activation Select disk drives use an auto-read-reallocate (ARRE) function that allows drives to resolve recoverable errors. All disk drives with a model number beginning with B (for example, B00721937) implement ARRE functionality.
Common Replacement Information and Procedures 100 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide
Replacing BA370, M2100 & M2200 Enclosure Elements 3 This chapter describes the following topics: ■ Shutting down the subsystem, page 102 ■ Restarting the subsystem, page 105 ■ Replacing array controllers and cache modules, page 107 ■ Replacing an ECB, page 149 ■ Replacing a PVA module, page 158 ■ Replacing an I/O module, page 161 ■ Handling a failed DIMM, page 165 Refer to your enclosure documentation for information about replacing power supplies, power cables, AC input boxes, cooling fans,
Replacing BA370, M2100 & M2200 Enclosure Elements Shutting down the subsystem Use the following steps to shut down a subsystem: 1. Connect a PC or terminal to the maintenance port of one of the subsystem array controllers. 2. From a host console, stop all host activity to the array controllers and dismount the logical units in the subsystem. 3. If you are using a Windows NT or Windows 2000 platform, shut down the server. 4.
Replacing BA370, M2100 & M2200 Enclosure Elements 6. Remove power to the subsystem. Caution: After shutting down an array controller in a BA370 enclosure for longer than one day, perform the steps in the next section, “Disabling the external cache batteries (BA370 enclosures only)” to prevent these write-back cache batteries from discharging.
Replacing BA370, M2100 & M2200 Enclosure Elements 1 3 4 5 2 5023 1 2 3 ECB 1 ECB 2 Power connector 4 5 Status LED Battery disable switch (SHUT OFF) Figure 39: ECB battery disable switch location (dual ECB example) 2. Repeat step 1 on page 103 for all ECBs in the subsystem. Note: The batteries are no longer protecting the cache module memory. Enabling the external cache batteries (BA370 enclosures only) To return the subsystem to normal operation, power on the subsystem.
Replacing BA370, M2100 & M2200 Enclosure Elements Restarting the subsystem If you need to restart a subsystem, refer to your enclosure documentation for specific procedures for restarting the subsystem. Note: Powering on the subsystem automatically starts the array controller and turns on the ECBs. An array controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. If the array controller does not restart, use the following steps: 1.
Replacing BA370, M2100 & M2200 Enclosure Elements 5022 Figure 40: Using thin needle-nose pliers to disconnect a fiber optic cable Note: For single-controller configurations, the only supported array controller and cache module bays are controller A and Cache A. For array controllers in a M2200 enclosure, do not use controller B and Cache B bays instead of controller A and Cache A bays.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing array controllers and cache modules The following subsections provide information on the following: ■ Special considerations and prerequisites, page 107 ■ Replacing array controllers in a single-controller configurations, page 112 ■ Replacing array controllers in dual-redundant configurations, page 119 ■ Replacing cache modules in single-controller configurations, page 132 Special considerations and prerequisites The following special cons
Replacing BA370, M2100 & M2200 Enclosure Elements ■ Cache Module Memory Configuration—The replacement cache module normally uses DIMMs from the cache module being replaced. When replacing DIMMs, you must install DIMMs in the replacement cache module in same position as they were in the cache module that is being replaced. ■ Facility Baud Rate—Record the baud rate of the operational controller.
Replacing BA370, M2100 & M2200 Enclosure Elements Figure 41 provides a graphical illustration of the controller replacement process for DRM configurations. Caution: In DRM configurations, use a factory-provided controller (configuration-free) or one that is reset with the Port #5 button reset action to avoid mirrored cache corruption on the replacement controller or recursive bugchecks (OCP code 25) on both initiator controllers.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing a controller with an unknown status Perform the following procedures to initialize your controller configuration if you are unsure1 of your replacement controller configuration: 1. Determine whether you have an offline or non-DRM-configured subsystem, and then complete one of the following options: ■ Complete the following substeps if you have an offline or non-DRM-configured subsystem: a.
Replacing BA370, M2100 & M2200 Enclosure Elements d. Initialize and install the new controller by completing the instructions in the “Installing array controllers in dual-redundant controller configurations” section (which starts on page 126). e. Reconnect the host port Fibre cables to all of the DRM controllers (initiator and target). f. Remount any units that were dismounted. g. Start all jobs that were halted in substep a on page 110.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing a controller with a known status Perform the following procedures if the configuration of the replacement controller is known: ■ Determine whether the controller to be replaced is a target controller, and then complete one of the following options: — If the controller to be replaced is not a target controller, proceed to the “Replacing array controllers in dual-redundant configurations” which starts on page 119.
Replacing BA370, M2100 & M2200 Enclosure Elements 1 2 3 4 5 CXO7072A 1 2 3 EMU HSG80 array controller in controller A bay Fiber optic cable with extender clip 4 5 Cache module in cache A bay PVA module Figure 42: Single-controller configuration in a BA370 enclosure 1 2 3 CXO7178A 1 2 Controller A Cache A 3 Fiber optic cable with extender clip Figure 43: Single-controller configuration in a Model 2200 enclosure HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Ma
Replacing BA370, M2100 & M2200 Enclosure Elements Removing controllers in single-controller configurations Perform the following steps to remove a controller in a single-controller configuration: 1. Connect a PC or terminal to the controller maintenance port (see Figure 44) if the controller is operational. If the controller is not operational, proceed to step 7 on page 116. CXO8276A Figure 44: Connecting PC and terminal cables to the controller maintenance port (dual-redundant configuration shown) 2.
Replacing BA370, M2100 & M2200 Enclosure Elements Enter the following CLI command from the controller prompt to save the current device configuration for “this controller” if the disk was initialized with the SAVE_CONFIGURATION switch: ■ CONFIGURATION SAVE 6. Shut down “this controller” with the following command: SHUTDOWN THIS_CONTROLLER Note: After the controller shuts down, the Reset button and the first three LEDs turn on (see Figure 45).
Replacing BA370, M2100 & M2200 Enclosure Elements 7. Remove the program card ESD cover (see Figure 46) and the program card, and save them in an antistatic bag or on a grounded antistatic mat for the replacement controller. 1 2 3 5024 1 2 Program card slot Eject button Program card 3 ESD cover Figure 46: Removing the controller ESD cover 8. Disconnect all host bus cables or terminators from the controller.
Replacing BA370, M2100 & M2200 Enclosure Elements 10. Disengage both retaining levers (see Figure 47), remove the controller, and then place the controller in an antistatic bag or on a grounded antistatic mat.
Replacing BA370, M2100 & M2200 Enclosure Elements 4. Without releasing the Reset button, press and hold the Port #5 button on the controller. 5. Release the Reset button. The controller resets its internal configuration cache and halts with an OCP code 33. LEDs 6, 5, 2, and 1, and the Reset button LED light up. The controller also resets its baud rate to 9600. 6. Release the Port #5 button. 7. Press and release the Reset button. 8. The controller restarts. This process may take up to three minutes.
Replacing BA370, M2100 & M2200 Enclosure Elements ■ Automatically restore the configuration by entering the following command below if the current device configuration is saved on a disk drive: CONFIGURATION RESTORE 13. Determine whether configuration restore process completed successfully, and then perform one of the following options: ■ Perform the following substeps if the configuration restore process did not complete successfully: a.
Replacing BA370, M2100 & M2200 Enclosure Elements Figure 48 and Figure 49 provide an illustration of the BA370 and Model 2200 and Model 2200 enclosures in a dual-redundant configurations.
Replacing BA370, M2100 & M2200 Enclosure Elements To differentiate on which component activity is being completed, symbols are displayed in the margin of text for your convenience. Table 9 describes those symbols. Table 9: Replacement Procedure Symbols Symbol Description Indicates that the procedural step must be completed for the operational controller, which is not being replaced.
Replacing BA370, M2100 & M2200 Enclosure Elements f. Restart the controller by entering the following command: RESTART THIS_CONTROLLER Note: Restarting the controller ensures that the unit persistent reservation flags transfer to the operational controller. g. Disconnect the PC or terminal from the maintenance port. 2. For the operational controller ( ), complete the following substeps: a.
Replacing BA370, M2100 & M2200 Enclosure Elements e. Exit VTDPY by pressing Ctrl+Y. f. Disable Failover mode, and take the controllers out of the dual-redundant configuration by entering the following command: SET NOFAILOVER 3. For the controller to be replaced ( ), complete the following substeps: a. Ensure that the Reset button (see Figure 45 on page 115) is a constant green. b. Remove the program card ESD cover (see Figure 46 on page 116). c.
Replacing BA370, M2100 & M2200 Enclosure Elements FRUTIL Main Menu 1. Replace or remove a controller or cache module 2. Install a controller or cache module 3. Replace an I/O module 4. Exit Enter choice: 1,2,3,4 -> Figure 51: FRUTIL Main Menu for a controller in M2100 and M2200 enclosures c. Enter option 1 to remove the controller. The Replace or remove Options screen is displayed (see Figure 52). Replace or remove Options: 1. Other controller and cache module 2. Other controller module 3.
Replacing BA370, M2100 & M2200 Enclosure Elements [ [ [ [ [ Slot Designations M2100/M2200 (back view) Port1 ] [ Port3 ] [ Port5 ] Port2 ] [ Port4 ] [ Port6 ] --------Controller A -------- ] --------Controller B -------- ] Cache Module A ] [ Cache Module B ] Figure 54: Slot Designations screen for controllers in M2100 and M2200 enclosures e. Ensure that the program card was removed from the controller to be replaced. Do not proceed until it is removed. f.
Replacing BA370, M2100 & M2200 Enclosure Elements 6. For the operational controller, complete the following substeps: a. Observe that after the “other controller” is removed that FRUTIL restarts all device ports and asks if a replacement controller is available. b. Enter N(o) after FRUTIL asks you if a replacement controller is available, and then disconnect the PC or terminal from the controller maintenance port. FRUTIL exits.
Replacing BA370, M2100 & M2200 Enclosure Elements c. Start FRUTIL by entering the following command: RUN FRUTIL d. Enter N(o) to the question about replacing the cache battery. The FRUTIL Main Menu is displayed. e. Enter option 2 to install a controller or cache module. If both the controller and cache are missing, the Install Options screen is displayed (see Figure 55). Install Options: 1. Other controller and cache module 2. Other controller module 3. Other cache module 4.
Replacing BA370, M2100 & M2200 Enclosure Elements Note: If FRUTIL times out before a replacement controller is installed, restart FRUTIL. \ H 2. For the replacement controller, complete the following substeps: H a. Remove the program card in the replacement controller if it is installed. H b. Insert the replacement controller (without the program card installed) into the appropriate bay, and engage the controller retaining levers. H c.
Replacing BA370, M2100 & M2200 Enclosure Elements 3. For the operational controller, complete the following substeps: a. Press Enter or Return within 3 minutes of completing substep 2f above to exit FRUTIL, and then wait 1 minute to allow the replacement controller to restart. Note: If Enter or Return is not pressed within 3 minutes in substep 3a above, the operational controller issues an automated command to cancel the installation of the replacement controller.
Replacing BA370, M2100 & M2200 Enclosure Elements H 4. For the replacement controller, complete the following substeps: H a. Connect a PC or terminal to the maintenance port of the replacement controller. H b. Perform the following: H i. H ii. Press Enter or Return several times to get a prompt. H iii. Proceed to the next substep. H Set the terminal baud rate to 9600. c.
Replacing BA370, M2100 & M2200 Enclosure Elements 5. For the operational controller, complete the following substeps: a. Connect a PC or terminal to the maintenance port of the operational controller. Caution: In substep 5b below, entering the appropriate SET command is critical. Enabling an incorrect Failover mode can cause loss of data and incur system downtime. b. Restore the Failover mode and re-establish the dual-redundant controller configuration with one of the following commands.
Replacing BA370, M2100 & M2200 Enclosure Elements 6. Disconnect the PC or terminal from the controller maintenance port of the operational controller. H 7. Connect the host bus cables to the replacement controller. H 8. Install the program card ESD cover on the controller being replaced. Replacing cache modules Replacing a cache module involves transferring the cache memory DIMMs from the failed cache module to the replacement cache module.
Replacing BA370, M2100 & M2200 Enclosure Elements Figure 56 and Figure 57 provide general information about the cache module. 5 1 4 2 3 5028 1 2 3 Cache memory power LED External cache battery (ECB) cable Retaining lever 4 5 Dual inline memory module (DIMM) Backplane connector Figure 56: Cache module Note: In Figure 57, the environmental monitoring unit (EMU) and power verification assembly (PVA) modules are present in only HP StorageWorks BA370 enclosures.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing cache modules in single-controller configurations Complete the steps in the following subsections to replace HSG60 and HSG80 cache modules in single-controller configurations. Caution: ESD can easily damage the cache module or a DIMM. Wear a snug-fitting, grounded ESD wrist strap when completing controller cache module removal and installation procedures.
Replacing BA370, M2100 & M2200 Enclosure Elements 7. For the cache module to be replaced, complete one of the following two options: ■ For cache modules in an M2100 or M2200 enclosure, disengage both cache module retaining levers, and remove the cache module. ■ For cache modules in a BA370 enclosure, complete the following substeps: a. Disengage both cache module retaining levers, and partially remove the cache module—about halfway. Caution: You must disable the ECB before disconnecting the ECB Y-cable.
Replacing BA370, M2100 & M2200 Enclosure Elements 8. Note the location for each DIMM in the cache module that is being replaced, and install the DIMMs in the corresponding locations in the replacement cache module. Use the following substeps to remove the DIMMs: Note: To facilitate pressing down on the DIMM retaining clips, consider using the eraser end of a pencil or a small screwdriver. a. Press down on the DIMM retaining clips (see Figure 58) at both ends of the DIMM being removed.
Replacing BA370, M2100 & M2200 Enclosure Elements Installing cache modules in single-controller configurations Perform the following steps to install HSG60 and HSG80 cache modules in single-controller configurations. Note: The replacement cache modules must contain the same memory configuration as that of the cache module which was removed. 1. Insert each DIMM (from the cache module that was removed) straight into the appropriate slot of the replacement cache module (see Figure 58 on page 136).
Replacing BA370, M2100 & M2200 Enclosure Elements 5. Restart the controller by pressing the Reset button. Note: A controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. Note: If the controller did not restart, use the following steps: a. Press and hold the Reset button. b. Reseat the controller program card. c. Release the Reset button. 6.
Replacing BA370, M2100 & M2200 Enclosure Elements Removing cache modules in dual-redundant controller configurations Perform the following steps to remove HSG60 and HSG80 cache modules in dual-redundant controller configurations: 1. For the controller associated with the cache module to be replaced, complete the following substeps: a. Connect a PC or terminal to the maintenance port. b. Run VTDPY by entering the following command: RUN VTDPY c. Record which units have I/O activity. d.
Replacing BA370, M2100 & M2200 Enclosure Elements b. Display and note the configuration information (for example, the Failover mode, cache status, serial numbers, SCSI mode, and so forth) for the operational controller by entering the following command: SHOW THIS_CONTROLLER c. Disable failover and take the controllers out of the dual-redundant configuration by entering the following command: SET NOFAILOVER d.
Replacing BA370, M2100 & M2200 Enclosure Elements Note: A countdown timer allows you a total of 2 minutes to remove the cache module. After 2 minutes, “this controller” exits FRUTIL, restarts the "other controller,” and resumes operations. If this happens, perform the following steps: a. Connect the PC or terminal to the "other controller." b. Shut down the "other controller" by issuing the command: SHUTDOWN THIS_CONTROLLER c. Disconnect the PC or terminal, and reconnect it to the operational controller.
Replacing BA370, M2100 & M2200 Enclosure Elements 4. For the operational controller, observe that FRUTIL restarts I/O ports, disables writeback caching and mirrored writes, indicates that the mirrored cache is disabled, and terminates. 5. Enter N(o) to the question for a replacement cache module. FRUTIL exits. 6. For the operational controller, place the cache module on an antistatic bag or a grounded antistatic mat. 7.
Replacing BA370, M2100 & M2200 Enclosure Elements Note: If you are installing a cache module after resolving a failed cache issue (see Table 14 through Table 16 of the “Controller and Cache Replacement Troubleshooting” appendix that starts on page 253), the operational controller and the replacement controller references that appear in this section may have been reversed. The operational controller is the one attached to the terminal or PC and the one from which FRUTIL is executed. H 1.
Replacing BA370, M2100 & M2200 Enclosure Elements Install Options: 1. Other controller and cache module 2. Other controller module 3. Other cache module 4. Exit NOTE: OPTION 1 DISABLED (So: Do cache, then controller.) Enter choice: 1,2,3,4 -> Figure 59: Install Options screen e. Enter 3 to install the other cache module. The Slot Designations screen is displayed. f. Enter Y(es) to confirm the intent to install the slot x cache module.
Replacing BA370, M2100 & M2200 Enclosure Elements Caution: If your cache module is in a BA370 enclosure, you must disable the ECB before proceeding to the next step. Failure to disable the ECB can result in cache module damage. To disable the ECB, complete the following steps: a. Disable the ECB by pressing the battery disable switch on the battery module. b. Connect the ECB Y-cable to the cache module. H b.
Replacing BA370, M2100 & M2200 Enclosure Elements H ■ If the controller does not restart, complete the following substeps: H a. Press and hold the controller Reset button on the controller. H b. Reseat the program card in the controller. H c. Release the controller Reset button. 7. For the operational controller, complete the following substeps: a. Observe that FRUTIL restarts the device ports and then exits. b.
Replacing BA370, M2100 & M2200 Enclosure Elements H c. Clear invalid or failed cache conditions, if any exist. See Table 14 through Table 16 in the “Controller and Cache Replacement Troubleshooting” appendix that starts on page 253 for more information on clearing invalid or failed cache conditions. Caution: Failure to clear an invalid cache message (especially a Cache is FAILED message) before setting the Failover mode can result in a recursive bugcheck error message that renders the controller unusable.
Replacing BA370, M2100 & M2200 Enclosure Elements c. Verify the failover configuration by entering the following commands: SHOW THIS_CONTROLLER SHOW OTHER_CONTROLLER 148 d. Verify that the controller cache and, if enabled, mirrored cache are good. If either the cache or mirrored cache is not good, see Table 14 through Table 16 in the “Controller and Cache Replacement Troubleshooting” appendix that starts on page 253 for more information on clearing invalid or failed cache conditions. e.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing an ECB The ECB can be replaced with subsystem power on or off. Note: HP recommends that you replace the ECB once every 3 ½ years. A dual ECB (for and HP StorageWorks RA8000) is shown in Figure 60 and contains two batteries. A single ECB contains only one battery.
Replacing BA370, M2100 & M2200 Enclosure Elements 1 2 3 4 1 3 2 B1 supports Cache B B2 supports Cache B A2 supports Cache A A1 supports Cache A 4 CXO6864A Figure 61: ECB and cache module locations in Model 2100 and 2200 enclosures (front view) 1 1 2 3 1 2 3 4 Controller A Controller B Cache A Cache B CXO5608B Figure 62: ECB and cache module locations in Model 2100 and 2200 enclosures (rear view) CXO6922B Figure 63: Removing an ECB that supports cache module B in Model 2100 and 2200 enclosures
Replacing BA370, M2100 & M2200 Enclosure Elements WARNING: The ECB is a sealed, rechargeable, lead acid battery that must be recycled or disposed of properly according to local regulations or policies after replacement. Do not incinerate the battery.
Replacing BA370, M2100 & M2200 Enclosure Elements 4. Enter Y(es) to confirm the intent to replace the “this controller” ECB with power on. Caution: Ensure that at least one ECB is connected to the ECB Y-cable at all times during this procedure. Otherwise, cache memory data is not protected and is subject to loss. The ECB Y-cable has a 12-volt and a 5-volt pin. Improper handling or misalignment while connecting or disconnecting can cause these pins to contact ground, resulting in cache module damage.
Replacing BA370, M2100 & M2200 Enclosure Elements 1 2 3 1 2 3 ECB charged LED ECB charging LED ECB fault LED 5019 Figure 64: ECB status LEDs c. Remove the old ECB and install the blank bezel in this vacant bay. d. Proceed to step 7 on page 154. 6. Replace the ECB: Note: Do not wait for the battery status light on the replacement ECB to turn solid green. ■ For a single ECB: a. Remove the old ECB and insert the replacement ECB into the same location. b.
Replacing BA370, M2100 & M2200 Enclosure Elements 7. Press Enter or Return. The ECB expiration date and deep discharge history are updated. FRUTIL exits. 8. If you are replacing the ECB for an array controller in a M2200 enclosure, proceed to step 10. 9. For a dual-redundant controller configuration with a dual ECB installed: a. If the “other controller” cache module is also to be connected to the replacement ECB, connect the PC or terminal to the “other controller” maintenance port.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing an ECB with the BA370 subsystem powered off Use the following steps to replace the ECB with the BA370 subsystem powered off: Note: You can replace the ECB storage building block (SBB) while the storage system is shut down by taking the following actions. First, issue the SET NOFAILOVER CLI command from the operational controller. This takes the units and moves them from the controller that is to be replaced to the operational controller.
Replacing BA370, M2100 & M2200 Enclosure Elements 4. Remove power to the subsystem by turning off both AC input boxes for all BA370 enclosures, and then disable the ECBs. Caution: The ECB cable has a 12-volt and a 5-volt pin. Improper handling or misalignment while connecting or disconnecting it can cause these pins to contact ground, resulting in cache module damage. 5. Replace the ECB: ■ For a single ECB: a. Remove the old ECB and insert the replacement ECB into the same location. b.
Replacing BA370, M2100 & M2200 Enclosure Elements 10. Press Enter or Return. The ECB expiration date and deep discharge history are updated. FRUTIL exits. 11. For dual-redundant controller configurations, complete the following substeps to replace the ECB for both cache modules: a. If the “other controller” cache module is also to be connected to the replacement ECB, connect the PC or terminal to the “other controller” maintenance port. The connected controller now becomes “this controller.” b.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing a PVA module Use the following steps to replace a PVA module in a master enclosure or expansion enclosure. The master enclosure contains array controllers and cache modules; expansion enclosures do not. Note: Array controllers can support up to three BA370 enclosures: a master enclosure and two expansion enclosures. A PVA can be replaced in either a single-controller or a dual-redundant controller configuration using this procedure. 1.
Replacing BA370, M2100 & M2200 Enclosure Elements 7. Enter Y(es) to confirm the intent to install the PVA module. 8. Set the replacement PVA SCSI ID address: ■ Set to 0 for a master enclosure. ■ Set to 2 for the first expansion enclosure. ■ Set to 3 for the second expansion enclosure. 9. Press Enter or Return. Caution: Wait for FRUTIL to quiesce the device ports—indicated by an All device ports quiesced message. Failure to allow the ports to quiesce can result in data loss.
Replacing BA370, M2100 & M2200 Enclosure Elements b. Insert the replacement PVA and engage both retaining levers. 11. Press Enter or Return to resume device port activity, and then restart the “other controller.” Note: A controller restart is indicated by the temporary cycling of the port LEDs and a flashing Reset button. 12. If the “other controller” did not restart, press the Reset button on the “other controller.” 13.
Replacing BA370, M2100 & M2200 Enclosure Elements Replacing an I/O module Figure 66 shows a rear view of the BA370 enclosure and the relative location of the six I/O modules (also referred to as ports). Figure 67 shows the six I/O modules, the location of the connectors and securing screws, and a bay-to-bus numbering correlation. Note: An I/O module can be replaced in either a single-controller or a dual-redundant controller configuration using this procedure.
Replacing BA370, M2100 & M2200 Enclosure Elements Note: The array controller can function with one failed I/O module. Use the following steps to replace an I/O module: 1. Connect a PC or terminal to the maintenance port of an operational controller. 2. In a dual-redundant controller configuration, disable Failover and take the controllers out of dual-redundant configuration with one of the following commands: SET NOFAILOVER or SET NOMULTIBUS_FAILOVER 3.
Replacing BA370, M2100 & M2200 Enclosure Elements 5. Enter option 4 to replace an I/O module. The example in Figure 68 shows the generated I/O module status display and indicates cabinet 0, port 5 as missing or bad.
Replacing BA370, M2100 & M2200 Enclosure Elements Note: If the “other controller” did not restart, press the Reset button. 9. Enable Failover and re-establish the dual-redundant configuration with one of the following commands: SET FAILOVER COPY=THIS_CONTROLLER or SET MULTIBUS_FAILOVER COPY=THIS_CONTROLLER This command copies the subsystem configuration from “this controller” to the “other controller.” 10.
Replacing BA370, M2100 & M2200 Enclosure Elements Handling a failed DIMM If a DIMM fails, ACS displays the message shown in Figure 69. If a DIMM failure occurs, note which DIMM needs replacement based on the display. 1.%CER--DebugTop> failed 2.
Replacing BA370, M2100 & M2200 Enclosure Elements 166 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide
Upgrading the Subsystem 4 This chapter provides the following instructions: ■ Upgrading from ACS V8.6-x and V8.7-x to V8.8, page 168 ■ After-upgrade maintenance checks, page 213 ■ Downgrading ACS V8.8-x to V8.6-1 or V8.
Upgrading the Subsystem Upgrading from ACS V8.6-x and V8.7-x to V8.8 This section covers the following topics: ■ Upgrading controller software, page 168 ■ Upgrading the software image, page 169 ■ Upgrading to ACS V8.8-xF, V8.8-xG, V8.8-xL, and V8.8-xS, page 171 ■ ■ Rolling upgrade procedure for dual-redundant configurations, page 171 ■ Shutdown upgrade procedure for dual-redundant and single-controller configurations, page 181 Upgrading to ACS V8.
Upgrading the Subsystem Upgrading the software image This section describes the supported upgrade methods and procedures required to successfully upgrade the array controller ACS firmware from V8.6-x or V8.7-x to V8.8-x. ACS V8.8-x supports both dual-redundant controller rolling upgrade and dual-redundant controller shutdown upgrade methods. The rolling upgrade method allows the host system to continue I/O activity, with minimal impact, while each controller is upgraded.
Upgrading the Subsystem Before upgrading, determine which version of ACS is currently in use (by issuing a SHOW THIS CONTROLLER and SHOW OTHER CONTROLLER command, as applicable, from a CLI prompt), and then determine which upgrade method to use: ■ Rolling Upgrade—This method allows I/O activity to continue during the upgrade by using the failover capabilities of the controllers.
Upgrading the Subsystem Caution: Windows users who are upgrading their firmware version should perform the following steps during the upgrade process: a. Shut down the host system. b. Perform a shutdown upgrade of the controllers. c. Restart the host system. Failure to perform the above actions prevents units from being presented with the original disk letter and can cause you to lose access to unit data. Upgrading to ACS V8.8-xF, V8.8-xG, V8.8-xL, and V8.
Upgrading the Subsystem Note: The following steps must be followed carefully for the upgrade to work properly. This procedure takes approximately 5 to 10 minutes depending upon the complexity of the configuration. If you are upgrading from V8.6-x, the units involved are briefly unavailable twice during the procedures for approximately 10 to 20 seconds in step 21 on page 178 and step 26 on page 179. Caution: Before upgrading, verify that the subsystem is in a stable state.
Upgrading the Subsystem Note: The VTDPY display refreshes every ten seconds. 4. Press Ctrl+C and then enter the following command to exit VTDPY: VTDPY> EXIT 5. Temporarily, move the PC or terminal to the maintenance port of controller B to repeat step 2 through step 4 on controller B. 6. Return PC or terminal to the maintenance port of controller A. 7. If the controller is running in SCSI-2 mode and the Command Console LUN (CCL) is enabled, you must disable the CCL prior to upgrading: a.
Upgrading the Subsystem 8. If upgrading from ACS V8.6-xS or 8.7-xS, delete all Snapshot units: a. Identify all Snapshot units by entering the following command: HSGA> SHOW UNITS FULL b. Record the configuration for each Snapshot unit for restoration in step 34 on page 181. c. Individually delete all Snapshot units by entering the following command: HSGA> DELETE snapshot-unit-name d.
Upgrading the Subsystem Note: The cache flush timer value is displayed in the caching parameters section. This parameter is modified during the procedure and must be restored later in this procedure. 11. For each unit, identify and record the unit writeback, read ahead, read, and preferred path characteristics by entering the following command: HSGA> SHOW UNITS FULL These characteristics are modified during the upgrade procedure and must be restored in step 32 on page 180.
Upgrading the Subsystem Figure 73 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=1 SECOND Figure 73: Screen display indicating cache status and whether data remains in cache Note: Repeat step 15 (on page 175) on both controllers (“this controller” and the “other controller”) until no unwritten data remains in either cache module memory.
Upgrading the Subsystem Note: After controller B shuts down, the Reset button and the first three LEDs turn on (see Figure 74). Proceed only after the Reset button stops flashing and remains on. 1 2 1 2 1 2 3 4 5 Reset button First three LEDs 6 5021 Figure 74: Reset button and first three LEDs 17.
Upgrading the Subsystem Figure 75 shows only a portion of the resulting display—an arrow points to the pertinent status indication. State: ONLINE to this controller Not reserved NOPREFERRED_PATH Figure 75: Screen display indicating unit online status 18. Run VTDPY to verify that all units are online and that expected I/O activity is resumed by entering the following command: HSGA> RUN VTDPY 19.
Upgrading the Subsystem e. After the card is fully inserted, release the button. Controller B restarts. Note: A controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. Disregard messages pertaining to misconfigured controllers or failover status. f. Install the program card ESD cover on controller B. 22. Verify that controller B completed initialization by completing the following substeps: a.
Upgrading the Subsystem b. Press and hold the controller Reset button while ejecting the old program card. c. After ejecting the program card, release the Reset button. d. Press and hold the Reset button while inserting the new program card. e. After the card is fully inserted, release the Reset button. Note: A controller restart can take as long as 60 seconds, indicated by the temporary cycling of port LEDs and a flashing Reset button.
Upgrading the Subsystem Note: If I/O activity is congested, additional time may be required for cache settings to completely restore. 34. Restore all Snapshot units removed in step 8 on page 174. 35. If the CCL was disabled in step 7 on page 173, enable it by entering the following command: HSGA> SET THIS_CONTROLLER COMMAND_CONSOLE_LUN 36. Verify that the CCL setting was restored by entering the following command: HSGA> SHOW THIS_CONTROLLER 37.
Upgrading the Subsystem Note: This procedure is written for dual-redundant controller configurations. If upgrading a single-controller configuration, disregard references to multiple controllers and controller B. 1. From a host console, stop all host activity to the controllers and dismount the logical units in the subsystem. 2. Connect a PC or terminal to the maintenance port of controller A. 3.
Upgrading the Subsystem 4. If upgrading from ACS V8.6-xS OR V8.7-xS, delete all Snapshot units by performing the following substeps: a. Identify all Snapshot units by entering the following command: HSGA> SHOW UNITS FULL b. Record the configuration for each Snapshot unit for restoration in step 13 on page 185. c. Individually delete all Snapshot units by entering the following command: HSGA> DELETE snapshot-unit-name d.
Upgrading the Subsystem Note: The cache flush timer value is displayed in the caching parameters section. This parameter is modified during the procedure and must be restored later in this procedure. 7. Set the CACHE_FLUSH_TIMER to 1 second by entering the following commands: HSGA> SET THIS_CONTROLLER CACHE_FLUSH_TIMER=1 HSGA> SET OTHER_CONTROLLER CACHE_FLUSH_TIMER=1 8.
Upgrading the Subsystem 10. Upgrade the software on both controllers by performing the following substeps: a. Remove the program card ESD cover from controller A. b. Press and hold the Reset button while ejecting the old program card. c. After the program card is ejected, release the Reset button. d. Repeat substep a through substep c for controller B.
Upgrading the Subsystem 16. Mount the logical units on the host. 17. Disconnect the PC or terminal from the maintenance port of controller A. 18. Refer to the “After-upgrade maintenance checks” section that starts on page 213 for additional after-upgrade steps if you are upgrading from ACS V8.7-2 or later. Upgrading to ACS V8.8-xP software ACS V8.8-xP implements the Data Replication Manager (DRM) feature, which can also be upgraded by using either a rolling or shutdown upgrade method.
Upgrading the Subsystem Caution: Before attempting the upgrade procedure, verify that the subsystem is in a stable state. This can be verified by examining the controller Reset buttons and the adjacent LEDs. The Reset button should be flashing, while all LEDs should be off (no light). Failure to perform this verification can result in data loss or leave the controller in an unusable state. The following procedure is categorized into four parts. To fully complete the ACS upgrade for V8.
Upgrading the Subsystem c. Press Ctrl+C and then enter the following command to exit VTDPY: VTDPY> EXIT d. Temporarily move the PC or terminal to the maintenance port of controller B. e. Run VTDPY on controller B at by entering the following command: HSGB> RUN VTDPY The controller’s percent idle is the first item displayed on the second line of the VTDPY display. f. If the controller is less than 50% idle, reduce the processing load. g.
Upgrading the Subsystem Figure 79 shows only a portion of the resulting display—arrows point to the pertinent status indication.
Upgrading the Subsystem 4. Identify and record the current cache flush timer value of the “this controller.” After issuing the following command, the subsequent screen displays the cache flush timer value: HSGA> SHOW THIS_CONTROLLER Figure 80 shows only a portion of the resulting display—an arrow points to the pertinent status indication.
Upgrading the Subsystem 9. Determine if all data is flushed from the cache module by entering the following command: HSGA> SHOW THIS_CONTROLLER Figure 81 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=1 SECOND Figure 81: Screen display indicating whether flushed data remains in cache 10.
Upgrading the Subsystem Note: After controller B shuts down, the Reset button and the first three LEDs turn on (see Figure 82). Proceed only after the Reset button stops flashing and remains on. 1 2 1 2 1 2 3 4 5 Reset button First three LEDs 6 5021 Figure 82: Reset button and first three LEDs 12.
Upgrading the Subsystem Note: In step 16 and step 17, controller B is updated first. After controller B restarts, it automatically shuts down controller A if you are upgrading from ACS V8.6-x to V8.8-x. If you are upgrading from ACS V8.7-x to V8.8-x, manually shut down controller A by entering the SHUTDOWN OTHER_CONTROLLER command. 16. Upgrade the software on controller B by performing the following substeps: Note: After step 16 is performed, the previous ACS version cannot be restored to this subsystem.
Upgrading the Subsystem c. Perform one of the following options: — If you are upgrading from ACS V8.6-x to V8.8-x, verify controller A is shut down and that the Reset button and the first three LEDs turn on (See Figure 82 on page 192). Proceed only after the Reset button stops flashing and remains on. — If you are upgrading from ACS V8.7-x to V8.8-x, shut down controller A with the following CLI command: HSGB> SHUTDOWN OTHER_CONTROLLER 18.
Upgrading the Subsystem Caution: Stop at this point when upgrading the initiator site controllers and proceed to upgrade the target site controllers. Failure to stop at this point, without fully upgrading the target site controllers, can cause the initiator and target site controllers to become dysfunctional and prevent the host from accessing data storage in these subsystems.
Upgrading the Subsystem Figure 84 shows only a portion of the resulting display—arrows point to the pertinent status indication.
Upgrading the Subsystem Figure 85 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=DEFAULT (10 seconds) Figure 85: Screen display indicating cache flush timer value 5. Identify and record the current cache flush timer value of the “other controller.
Upgrading the Subsystem 9. Determine if all data is flushed from the cache module by entering the following commands: HSGA> SHOW THIS_CONTROLLER HSGA> SHOW OTHER_CONTROLLER Figure 86 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=1 SECOND Figure 86: Screen display indicating whether flushed data remains in cache 10.
Upgrading the Subsystem Figure 87 shows only a portion of the resulting display—an arrow points to the pertinent status indication. State: ONLINE to this controller Not reserved Figure 87: Screen display indicating online status of the unit 14. Run VTDPY to verify that all units are online and that expected I/O activity is resumed by entering the following command: HSGA> RUN VTDPY 15.
Upgrading the Subsystem e. After the card is fully inserted, release the Reset button. Controller B restarts. Note: A controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. Disregard messages pertaining to misconfigured controllers or failover status. f. Install the program card ESD cover on controller B. Controller A shuts down. 18.
Upgrading the Subsystem 22. Upgrade the software on controller A by performing the following substeps: a. Remove the program card ESD cover from controller A. b. Press and hold the controller Reset button while ejecting the old program card. c. After ejecting the program card, release the Reset button. d. Press and hold the Reset button while inserting the new program card. e. After the card is fully inserted, release the button.
Upgrading the Subsystem 31. Disconnect the PC or terminal from the maintenance port of controller A. Upgrade on the target site controllers is complete. Part 4: Initiator site upgrade (final steps) Continue the interrupted upgrade of the initiator site controllers (see step 16 on page 193) as follows: 1.
Upgrading the Subsystem Shutdown upgrade procedures for DRM configurations Specific controllers are referred to as controller A or controller B during the procedure. For clarity, the CLI prompts illustrated in this procedure use HSGA> and HSGB> to indicate which controller (A or B) is used. Note: The following steps must be followed carefully for the upgrade to work properly. This procedure takes approximately 5 to 10 minutes. The units involved are unavailable during the upgrade procedure.
Upgrading the Subsystem 3. Determine if the controller is in SCSI-2 mode and the CCL is enabled by entering the following command: HSGA> SHOW THIS_CONTROLLER Figure 88 shows only a portion of the resulting display—arrows point to the pertinent status indication.
Upgrading the Subsystem d. Verify that all Snapshot units were deleted by entering the following commands, as required: HSGA> SHOW UNITS FULL Note: If any Snapshot unit exists, repeat step 5 on page 204. 6. Identify and record the current cache flush timer value of the “this controller.” The subsequent screen displays the cache flush timer value: HSGA> SHOW THIS_CONTROLLER Figure 89 shows only a portion of the resulting display—an arrow points to the pertinent status indication.
Upgrading the Subsystem Figure 90 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=1 SECOND Figure 90: Screen display indicating whether flushed data remains in cache Note: Repeat step 6 (on page 183) on both controllers (“this controller” and the “other controller”) until no unwritten data remains in either cache module memory. 10.
Upgrading the Subsystem e. Simultaneously press and hold the Reset button on both controllers, and insert a new program card into each controller. f. Simultaneously release the Reset buttons. Both controllers restart. Note: A controller restart can take as long as 60 seconds, indicated by the temporary cycling of the port LEDs and a flashing Reset button. Disregard messages pertaining to misconfigured controllers or failover status. g. Install a program card ESD cover on each controller.
Upgrading the Subsystem Figure 91 shows only a portion of the resulting display—arrows point to the pertinent status indication.
Upgrading the Subsystem Figure 92 shows only a portion of the resulting display—an arrow points to the pertinent status indication. Cache: 256 megabyte write cache, version 0022 Cache is GOOD No unflushed data in cache CACHE_FLUSH_TIMER=DEFAULT (10 seconds) Figure 92: Screen display indicating cache flush timer value 6. Identify and record the current cache flush timer value of the “other controller.
Upgrading the Subsystem 9. Repeat step 8 (on page 209) on both controllers (“this controller” and the “other controller”) until no unwritten data remains in either cache module memory. 10. Shut down both controllers by entering the following commands: HSGA> SHUTDOWN OTHER_CONTROLLER HSGA> SHUTDOWN THIS_CONTROLLER Note: After the controllers shut down, the Reset buttons and the first three LEDs turn on.
Upgrading the Subsystem 12. After the controllers restart, restore the CACHE_FLUSH_TIMER to the values recorded in step 5 and step 6, which starts on page 208, by entering the following commands: HSGB> SET THIS_CONTROLLER CACHE_FLUSH_TIMER=n HSGB> SET OTHER_CONTROLLER CACHE_FLUSH_TIMER=n 13. Verify that the CACHE_FLUSH_TIMER values have been updated: HSGA> SHOW THIS_CONTROLLER HSGA> SHOW OTHER_CONTROLLER 14. Restore all Snapshot units removed in step 4 on page 208. 15.
Upgrading the Subsystem 6. Mount the logical units on the host. 7. Disconnect the PC or terminal from the maintenance port of controller A. The upgrade of the initiator site controllers is complete. Refer to the “After-upgrade maintenance checks” section that starts on page 213 for additional after-upgrade steps if you are upgrading from ACS V8.7-2 or later.
Upgrading the Subsystem After-upgrade maintenance checks ACS V8.7-2 potentially allows for silent metadata inconsistencies to develop if a disk drive was spared into a RAIDset (either manually or automatically). With ACS V8.8-1, at the time the controller is first upgraded, the firmware provides metadata inconsistency corrections within the controller memory. However, corrected information is not written to the disk devices during the upgrade process.
Upgrading the Subsystem SHOW RAIDSETS SPECIAL_FUNCTION_ONE An example of the subsequent screen (for the top controller) is displayed in Figure 94. As you review Figure 94, note that units D11, D12, and D13 of RAIDset R1 are partitioned RAIDsets. Also, RAIDset R6 has no units above it.
Upgrading the Subsystem An example of the subsequent screen (for the bottom controller) is displayed in Figure 95. As you review Figure 95, note that units D11, D12, and D13 of RAIDset R1 are partitioned RAIDsets. Also, RAIDset R5 has no units configured above it.
Upgrading the Subsystem Handling RAIDset containers with at least one unit above the container (recommended offline method) Note: Although the above procedures are recommended for repairing RAIDsets, you can also use the procedures described in “Manually manipulating disk devices needing repair (alternative procedure).” If any RAIDset shows a status of Maintenance Recommended, and the RAIDset is currently associated with a unit (including a unit above a partition), the ACS V8.
Upgrading the Subsystem Note: If only one or two RAIDsets are impacted and only a single device within each RAIDset needs maintenance, you can use this alternative procedure. 1. Identify the RAIDset unit needing maintenance by issuing the following CLI command: HSG> SHOW RAIDSET SPECIAL_FUNCTION_ONE Figure 96 and Figure 97 (on page 218) show an example of an affected RAIDset.
Upgrading the Subsystem HSG_TOP> SHOW R1 Name Storageset Uses Used by R1 raidset DISK10200 D7 DISK20200 DISK30200 Switches: POLICY (for replacement) = BEST_PERFORMANCE RECONSTRUCT (priority) = NORMAL CHUNKSIZE = 256 blocks State: NORMAL DISK20200 (member 0) is NORMAL DISK10200 (member 1) is NORMAL DISK30200 (member 2) is NORMAL Size: 16751956 blocks Figure 97: Example showing the unit affected 2.
Upgrading the Subsystem HSG_TOP> REINITIALIZE R1 SPECIAL_FUNCTION_ONE=INFO Unit: D7, Type: Un-partitioned Raidset: R1 Device: DISK10200, Marked partitioned: No, Affected: No, mdata_version (vsi): 11 Structure version Device: DISK30200, Marked partitioned: No, Affected: No, mdata_version (vsi): 11 Structure version Device: DISK20200, Marked partitioned: Yes, Structure version Affected: Yes, mdata_version (vsi): 11 Figure 98: Example of a device showing inconsistent metadata attributes that differ from
Upgrading the Subsystem 4. Identify RAIDset replacement policy (see Figure 99).
Upgrading the Subsystem Figure 99 on page 220 shows an example of how to remove a device from a RAIDset membership. An event (Instance Code 02675201), reported on the CLI, indicates that the device specified in the Device Locator field has been removed from the RAIDset associated with the logical unit. Note: The drive displayed in Figure 99 on page 220, shows a drive with partition attributes different from the attributes at the RAIDset level. The drive metadata is inconsistent. 7.
Upgrading the Subsystem 9. Verify that the procedure was successful by ensuring that the affected RAIDset is GOOD and placed in the appropriate RAIDset (see Figure 100 and Figure 101).
Upgrading the Subsystem Handling RAIDset containers without a unit above the container For RAIDsets without a unit above the container, HP recommends that you take immediate steps to either initialize RAIDsets if data on the media is not needed, or invoke the following CLI command with the PARTITION or NOPARTITION qualifier, as appropriate: container-name SPECIAL_FUNCTION_ONE= After issuing the above command, allow the controller to update the media on the disk with corrected metadata elements from control
Upgrading the Subsystem Note: Although the above procedures are recommended for repairing RAIDsets, you can also use the procedures described in “Repairing disk devices (offline and alternative methods)” on page 215. Handling initiator site remote copy sets made of RAIDsets that need repair After you complete the repair steps outlined in the “Handling target site remote copy sets made of RAIDsets that need repair,” complete the following steps on the initiator site: 1.
Upgrading the Subsystem Handling concatenated storagesets with RAIDset containers needing repair Figure 102 on page 225 shows an example of the steps that must be taken in order to clean up a RAIDset member of a concatenated storageset. This example shows units D1, C1, R1, and R2 (the added member) with the affected RAIDset containers. During this check, verify that no reconstructs are in progress.
Upgrading the Subsystem Downgrading ACS V8.8-x to V8.6-1 or V8.7-1 This document describes the procedures necessary to downgrade the controller from the ACS V8.8-x to V8.6-1 or V8.7-1. Downgrading controllers to the previous version should only be completed by HP authorized service personnel. Failure to properly follow the downgrade procedures may result in the loss of data. Prior to downgrading, ensure that the following items have been completed: ■ Dismount all units from the host operating system.
Upgrading the Subsystem ■ The SCSI_FAIRNESS, SMART_ERROR_EJECT and DEFAULT_ACCESS_MASK are cleared in both the downgrade cases. ■ If you are downgrading ACS V8.8-x, and pre-existing patches from V8.6 or V8.7 remain in controller memory, the system downgrades your system to the highest patch level that was previously installed. Running the downgrade program Perform the following procedure to downgrade ACS V8.8-x to V8.7-1 or V8.6-1: 1.
Upgrading the Subsystem A - Downgrade ACS version 8.8 to ACS version 8.7-1. B - Downgrade ACS version 8.8 to ACS version 8.6-1. Which downgrade do you want to run (A or B)? Figure 104: Downgrade selection A confirmation screen is displayed. 4. Enter (Y)es to the question asking whether you want to complete the downgrade to continue with the downgrade. 5. Insert the PCMCIA card into the top controller. 6.
Upgrading the Subsystem The following host information will be lost during this downgrade: Name -------- Host Id ------------------- Adapter Id ------------------- Control ------- Port ---- Offset ------ RStyle ------ Alpha1 0123-4567-8901-1234 0123-4567-8901-1234 this 1 0 Std NtSrv05 9876-5432-0543-1234 9876-5432-0543-1234 this 1 0 Std You can exit if you need to reconfigure your systems or record this information.
Upgrading the Subsystem 9. Enter (Y)es to proceed with the downgrade. The system indicates that the downgrade is to begin (see Figure 108). This controller will now be downgraded to run ACS Version 8.6. You must also downgrade the other controller. When both controllers have been downgraded, install the ACS Version 8.6 program cards and restart both controllers at the same time. Continue (Y/N)? Downgrade procedure complete. Halting.
Upgrading the Subsystem Installing, deleting, and listing software patches by using CLCP Use CLCP to manage software patches. These small programming changes are placed into the controller non-volatile memory and become active upon restarting the controller. Consider the following points while installing or deleting patches: ■ Patches are associated with specific software versions. CLCP verifies the patch against the currently installed version.
Upgrading the Subsystem 5. Enter option 2. The Code Patch Main Menu is displayed (see Figure 110). You have selected the Code Patch local program. This program is used to manage software code patches. Select an option from the following list: Type ^Y or ^C (then RETURN) at any time to abort Code Patch. Code Patch Main Menu 0: Exit 1: Enter a Patch 2: Delete Patches 3: List Patches Enter option number (0..3) [0] ? Figure 110: Code Patch Main Menu 6. Enter option 1.
Upgrading the Subsystem Deleting a software patch All installed ACS V8.7 (or earlier) patches should be removed after an ACS V8.8-x installation since they are no longer applicable. Removing older patches frees up patch memory for future ACS patches. Use the following steps to delete a software patch: 1. From a host console, quiesce all port activity. 2. Connect a PC or terminal to the controller maintenance port. 3.
Upgrading the Subsystem 5. Enter option 2. The Delete Patches option screen is displayed (see Figure 114). This is the Delete Patches option. The program prompts you for the software version and patch number you wish to delete. If you select a patch for deletion that is required for another patch, all dependent patches are also selected for deletion. The program lists your deletion selections and asks if you wish to continue. Type ^Y or ^C (then RETURN) at any time to abort Code Patch.
Upgrading the Subsystem The Code Patch Main Menu is displayed (see Figure 116). Code Patch Main Menu 0: Exit 1: Enter a Patch 2: Delete Patches 3: List Patches Enter option number (0..3) [0] ? Figure 116: Code Patch Main Menu 9. Enter option 0. 10. Press the controller Reset button to restart the controller. 11. For dual-redundant controller configurations, repeat step 1 (on page 233) through step 10 for the second controller. Listing software patches Use the following steps to list software patches: 1.
Upgrading the Subsystem 3. Enter option 2. The Code Patch Main Menu is displayed (see Figure 118). You have selected the Code Patch local program. This program is used to manage software code patches. Select an option from the following list: Type ^Y or ^C (then RETURN) at any time to abort Code Patch. Code Patch Main Menu 0: Exit 1: Enter a Patch 2: Delete Patches 3: List Patches Enter option number (0..
Upgrading the Subsystem 4. Enter option 3. A screen showing the patches currently install is displayed (see Figure 119).
Upgrading the Subsystem 5. Enter option 0. 6. For dual-redundant controller configurations, repeat step 1 (on page 235) through step 5 for the second controller.
Upgrading the Subsystem Upgrading firmware on a device Use the format and device code load utility (HSUTIL) to upgrade a device with firmware located in contiguous blocks at specific logical block numbers (LBNs) on a source disk drive. The source disk drive must be configured as a unit on the same controller. Upgrading firmware on a disk is a two-step process (see Figure 120): 1. Copy the new firmware from the host to a disk drive configured as a unit in the subsystem. 2.
Upgrading the Subsystem ■ HSUTIL cannot install firmware on devices configured as single-disk drive units or as members of a storageset, spareset, or failedset. If installing firmware on a device previously configured as a single-disk drive, delete the unit number and storageset name associated with it. ■ During the installation, the source disk drive is not available for other subsystem operations.
Upgrading the Subsystem 4. Start HSUTIL with the following command: RUN HSUTIL The HSUTIL Main Menu is displayed (see Figure 121). HSUTIL Main Menu: 0. Exit 1. Disk Format 2. Disk Device Code Load 3. Tape Device Code Load 4. Disaster Tolerance Backend Controller Code Load Enter function number: (0:4) [0]? Figure 121: HSUTIL Main Menu 5. Enter option 2. 6. Choose the single-disk unit as the source disk for the download. 7. Enter the starting LBN of the firmware image—usually LBN 0. 8.
Upgrading the Subsystem Upgrading to a dual-redundant controller configuration Use the following steps to upgrade a single-controller configuration subsystem to a dual-redundant configuration subsystem. To replace failed components, see the “Replacing BA370, M2100 & M2200 Enclosure Elements” chapter that starts on page 101 for more information.
Upgrading the Subsystem 6. Enter Y(es). A countdown timer allows a total of 4 minutes to install the controller and cache module. After 4 minutes, “this controller” exits FRUTIL and resumes operations. If this happens, return to step 2 (on page 242) and proceed. Caution: ESD can easily damage a cache module or controller. Wear a snug-fitting, grounded ESD wrist strap during this procedure. 7. Insert the new ECB: ■ For BA370 enclosures: a. Insert the ECB into an empty bay or on top of the enclosure.
Upgrading the Subsystem 9. Make sure that the program card is in the new controller, and then insert the controller into its bay. Engage its retaining levers. After fully seated, the newly installed controller automatically restarts. Press Enter or Return to continue. If the “other controller” did not restart, follow these steps: a. Press and hold the “other controller” Reset button. b. Reseat the “other controller” program card. c. Release the Reset button. 10.
Upgrading the Subsystem Upgrading cache memory Use Figure 122 and Table 10 to configure the cache module. 1 3 2 4 1 2 3 4 Cache memory slot 1 Cache memory slot 2 Cache memory slot 3 Cache memory slot 4 CXO6576B Figure 122: Cache module DIMM locations Table 10: Cache module memory configurations Memory DIMMs Quantity Location 128 MB 32 MB 4 1 2 3 4 256 MB 128 MB 2 1 3 512 MB 128 MB 4 1 2 3 4 Note: The cache size requirement for running ACS V8.8P or V8.8S is 512 MB. For ACS V8.8F, V8.
Upgrading the Subsystem To upgrade cache module memory, shut down its controller. Use the following steps to upgrade or add DIMMs: Caution: ESD can easily damage a cache module or a DIMM. Wear a snug-fitting, grounded ESD wrist strap during this procedure. 1. From the host console, dismount the logical units in the subsystem. 2. If using a Windows NT platform, shut down the server. 3. If the controller is operating, connect a PC or terminal to the controller maintenance port.
Upgrading the Subsystem 5. For BA370 enclosures only: a. Disable the ECB by pressing the battery disable switch until the status light stops blinking—approximately 5 seconds. b. Disconnect the ECB cable from the cache module. 6. Disengage the two retaining levers on the cache module, remove the DIMM from the enclosure, and place it onto a grounded antistatic mat. 7. If adding DIMMs (see Figure 123): a. Press the DIMM retaining clips down for the vacant DIMM slot. b.
Upgrading the Subsystem 8. If replacing DIMMs (see Figure 123): a. Press down on the DIMM retaining clip at both ends of the DIMM being removed. b. Gently remove the DIMM from the DIMM slot. c. Insert the new DIMM straight into the slot, ensuring that the notches in the DIMM align with the tabs in the slot. d. Repeat step a through step c for each DIMM. 9. In a dual-redundant controller configuration, repeat step 5 (on page 247) through step 8, as appropriate, for the other cache module.
Specifications A This appendix covers the following topics: ■ Physical and electrical specifications for HSG60 and HSG80 array controllers and cache modules, page 250 ■ Environmental specifications, page 251 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide 249
Specifications Physical and electrical specifications for HSG60 and HSG80 array controllers and cache modules Table 11 lists the physical and electrical specifications for HSG60 and HSG80 array controllers and cache modules. Voltage measurements in Table 11 are nominal measurements without tolerances. Table 11: HSG60 and HSG80 array controllers and cache module specifications Current at +5 VDC Hardware Width Depth Power HSG60 array controller 305 mm (12 in) 241 mm (9.5 in) 26.0 W 5.
Specifications Environmental specifications HSG60 and HSG80 array controllers are intended for installation in a Class A environment. The operating environmental specifications are listed in Table 12 and the non-operating environmental specifications are listed in Table 13. These specifications are the same for all HP StorageWorks storage devices. Table 12: Operating environmental specifications Condition Ambient temperature Specification +10 °C to +35 °C (+50 °F to +95 °F) Derate for altitude 0.
Specifications 252 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide
Controller and Cache Replacement Troubleshooting B This section provides instructions on resolving common problems that may occur while replacing controllers or cache modules.
Controller and Cache Replacement Troubleshooting Resolving invalid cache error messages—serial number mismatch error messages An invalid cache error message may display after the SHOW THIS_CONTROLLER command is submitted. The controller reports that the cache and mirrored cache with cache are invalid and that the cache should be installed in a specific controller with the serial number ZGxxxxxxxx.
Controller and Cache Replacement Troubleshooting Resolving failed cache error messages Caution: Failure to clear an invalid cache message (especially a Cache is FAILED message) before setting the Failover mode can result in a recursive bugcheck error message that renders the controller unusable.
Controller and Cache Replacement Troubleshooting mirrored cache is in a failed condition. When a SHOW THIS_CONTROLLER command is submitted, the replacement controller indicates that its mirrored cache is in a Failed state: Cache: Cache is GOOD Mirrored Cache: Cache is FAILED Table 16 on page 259 provides instructions for resolving the above event. If the failure cannot be cleared using the procedures in Table 16 on page 259, a defective hardware component may exist.
Controller and Cache Replacement Troubleshooting Table 15: Resolving a failed cache error message (event 1) Details Message %CER-Atop> This controller (replacement) has an invalid cache module Resolution Caution: If the cache failed, disable mirrored cache to clear the failed condition. Since unmirroring a controller pair destroys all persistent reservation bits in the controllers, you must halt all host activity to this controller pair before proceeding.
Controller and Cache Replacement Troubleshooting Table 15: Resolving a failed cache error message (event 1) (continued) Details Resolution (Cont’d) Caution: In step 8, you must break up mirrored cache. Before completing this step, note that breaking up mirrored cache configurations loses persistent reservation conditions on the units in these controllers. Before proceeding with this procedure, inform customers and your manager that you are about to break the mirror cache configuration. 8.
Controller and Cache Replacement Troubleshooting Table 16: Resolving a failed cache error message (event 2) Details Message Resolution %CER-Atop> This controller (replacement) has an invalid cache module Caution: If the cache failed, you must disable mirrored cache in order to clear the failed condition. Because unmirroring a controller pair destroys all persistent reservation bits in the controllers, you must halt all host activity to this controller pair before proceeding.
Controller and Cache Replacement Troubleshooting Table 16: Resolving a failed cache error message (event 2) (continued) Details Resolution (continued) e. After restoring defaults, complete the instructions for installing a controller in the “Installing array controllers in dual-redundant controller configurations” section that starts on page 126. f. 260 Return controller pair to service.
Controller and Cache Replacement Troubleshooting Resolving unexpected bugcheck condition error messages If the replacement controller restarts at the same time you press Enter or Return to terminate FRUTIL on the operational controller, an unexpected bugcheck condition error message can occur. Note: This condition cannot be cleared with CLI commands.
Controller and Cache Replacement Troubleshooting Table 17: Resolving an unexpected bugcheck condition error message (continued) Resolution Resolution (Cont’d) 2. Record all controller configuration information for both controllers, if possible. Otherwise, record controller configuration information for the operational controller only, and proceed to the next step. 3. Shut down the operational controller by entering the following command: SHUTDOWN THIS_CONTROLLER 4.
glossary Glossary This glossary defines terms used in this guide or related to this product and is not a comprehensive glossary of computer terms. Glossary ACS Array Controller Software. The software component of the array controller storage systems. ACS executes on the array controller and processes input and output requests from the host, performing the device-level operations required to satisfy the requests.
Glossary byte A binary character string made up of 8 bits operated on as a unit. cache memory A portion of memory used to accelerate read and write operations. The objective of caching data in a system is to improve performance by placing the most frequently used data in the highest performance memory. cache module A fast storage buffer. channel An interface which allows high speed transfer of large amounts of data. Another term for a SCSI bus. See also SCSI.
Glossary container (1) Any entity that is capable of storing data, whether it is a physical device or a group of physical devices. (2) A virtual, internal controller structure representing either a single disk or a group of disk drives linked as a storageset. Stripesets and mirrorsets are examples of storageset containers that the array controller uses to create units. See also storage unit.
Glossary dual data link See DDL. dual-redundant configuration An array controller configuration consisting of two active controllers operating as a single controller. If one controller fails, the other controller assumes control of the failing controller devices. ECB External Cache Battery. The unit that supplies backup power to the cache module in the event the primary power source fails or is interrupted. EIA Electronic Industries Association.
Glossary fiber optic cable A transmission medium designed to transmit digital signals in the form of pulses of light. Fiber optic cable is noted for its properties of electrical isolation and resistance to electrostatic contamination. flush The act of writing dirty data from cache to a storage media. See also dirty data. FMU Fault Management Utility. A utility that is run to provide fault or error reporting information. FRU Field Replaceable Unit.
Glossary hot-pluggable A replacement method that allows normal I/O activity on a device bus to remain active during device removal and insertion. The device being removed or inserted is the only device that cannot perform operations during this process. See also pluggable. hot swap A device remove and replace procedure using hot-pluggable method. HP StorageWorks The HP brand name for a family of modular data storage products that allows customers to design and configure their own storage subsystems.
Glossary local connection A connection to the subsystem, by way of the array controller serial maintenance port, to a maintenance terminal or the host terminal. A local connection enables you to connect to one subsystem controller to perform maintenance tasks. See also maintenance terminal and local terminal. local terminal A terminal plugged into the EIA-423 maintenance port located on the front bezel of the array controller. See also maintenance terminal and local connection.
Glossary N-m Newton-meter. The metric equivalent to inch-pounds of torque. node (3) In Fibre Channel, a device that has at least one N_Port or NL_Port. nonredundant controller configuration (1) A single-controller configuration. (2) An array controller configuration that does not include a second controller. OCP Operator control panel. The control and indicator panel associated with an array controller.
Glossary pluggable A replacement method that allows the complete system to remain online during device removal or insertion. The system bus must be halted, or quiesced, for a brief period of time during the replacement procedure. See also hot-pluggable. port In general terms, the port is: ■ A logical channel in a communications system. ■ The hardware and software used to connect a host controller to a communications bus, such as a SCSI bus or serial bus.
Glossary RAID-0 A RAID storageset that stripes data across an array of disk drives. A single logical disk spans multiple physical disks, allowing parallel data processing for increased I/O performance. While the performance characteristics of RAID-0 is excellent, this RAID level is the only one that does not provide redundancy. RAID-0 storagesets are sometimes referred to as stripesets.
Glossary reduced A term that indicates that a mirrorset or RAIDset is missing one member because the member failed or was physically removed. redundancy The provision of multiple interchangeable components to perform a single function in order to cope with failures and errors. A RAIDset is considered to be redundant when user data is recorded directly to one member and all of the other members include associated parity information.
Glossary single ended I/O module A 16-bit I/O module. See also I/O module. single-ended SCSI bus An electrical connection where one wire carries the signal and another wire or shield is connected to electrical ground. Each signal logic level is determined by the voltage of a single wire in relation to ground. This is in contrast to a differential connection where the second wire carries an inverted signal.
Glossary target ID number The address a bus initiator uses to connect with a bus target. Each bus target is assigned a unique target address. “this controller” The array controller that is serving your current CLI session through a local or remote terminal. See also “other controller.” transfer data rate The speed at which data may be exchanged with the central processor, expressed in thousands of bytes per second. Ultra SCSI bus A Fast-20 SCSI bus. uninterruptible power supply See UPS.
Glossary write-through cache A cache management technique for retaining host write requests in read cache. When the host requests a write operation, the array controller writes data directly to the storage device. This technique allows the array controller to complete some read requests from the cache, greatly improving the response time to retrieve data. The operation is complete only after the data to be written is received by the target storage device.
index 180-watt power supply LEDs 38 A adding cache memory 245 DIMMs 245 array controller. See controller.
Index HSG80 subsystem illustrated 25 table 26 I/O module bay numbering (illustrated) 161 locations (illustrated) 161 part number 26 maintenance port cable part number 33 power cable (black) part number 27 power cable kit (white) part number 26 power supply LEDs illustrated 38 program (PCMCIA) card part numbers 33 PVA module connectors and switches (illustrated) 39 part number 26 PVA module location (illustrated) 159 shutting down the subsystem 102 single-battery ECB part number 27 single-controller configu
Index local connection 50 nonoperating environmental specifications (table) 251 OCP switches and LEDs 36 switches and LEDs (illustrated) 36 operating environmental specifications (table) 251 removing dual-redundant controller configuration 121 replacing single-controller configuration 112 Reset button and first three LEDs (illustrated) 67, 102, 177, 192 shutting down 102 specifications (table) 250 table of parts 33 controller and cache module installing single-controller configuration 107 removing single-c
Index E ECB BA370 enclosure battery disable switch location (illustrated) 104 dual configuration (illustrated) 149 replacing dual-redundant controller configuration 149 single-configuration controller 149 disabling, shutting down a BA370 subsystem 103 DS–BA356–MW enclosure replacing subsystem powered on dual-redundant controller configuration 151 enabling, shutting down a BA370 subsystem 104 location in a Model 2100 enclosure (illustrated) 150 location in a Model 2200 enclosure (illustrated) 150 Model 2100
Index F G fan Model 2100 enclosure LEDs illustrated 43 Model 2200 enclosure LEDs illustrated 43 fiber optic cable installing dual-redundant controller configuration 68 single-configuration controller 68 removing dual-redundant controller configuration 67 single-configuration controller 67 replacing dual-redundant controller configuration 67 single-configuration controller 67 using thin needle-nose pliers to disconnect (illustrated) 68 Fibre Channel 8-/16-port switch part number 30 8-/16-port switch, part
Index I I/O module BA370 enclosure bay numbering (illustrated) 161 locations (illustrated) 161 replacing dual-redundant controller configuration 161 single-controller configuration 161 illustrated 180-watt power supply LEDs 38 BA370 enclosure dual ECB configuration 149 dual-redundant controller configuration 120 ECB battery disable switch location 104 EMU and its communication cable 31 EMU connectors, switches, and LEDs 40 PVA module connectors and switches 39 PVA module location 159 single-controller conf
Index single-controller configuration 113 PC/terminal to maintenance port connection 52 removing an ECB that supports cache module B 150 installing cache module dual-redundant controller configuration 142 controller dual-redundant controller configuration 126 controller and cache module single-controller configuration 107 controller, cache module, and ECB 242 dual-redundant controller configuration cache module 142 controller 126 fiber optic cable 68 hub 68 switch 68 fiber optic cable 68 dual-redundant con
Index controller front panel connectors, switches, and LEDs illustrated 36 OCP—switches and LEDs illustrated 36 DIMM part numbers 34 DIMM components (illustrated) 72 ECB battery pack part number 30 LEDs (illustrated) 42, 153 locations (illustrated) 150 part number 30 ECB and cache module locations (illustrated) 150 EMU part number 30 switches and LEDs (illustrated) 41 fan LEDs (illustrated) 43 part number 30 fiber optic cable part number 33 GLM connectors and components illustrated 37 part number 33 HSG60
Index fiber optic cable part number 33 GLM connectors and components illustrated 37 part number 33 HSG80 array controller fiber optic cabling and program card (illustrated) 32 part number 30 table of parts 33 HSG80 subsystem (illustrated) 29 HSG80 subsystem (table) 30 I/O module part number 30 maintenance port cable part number 33 part number, rack-mountable 30 power supply LEDs illustrated 38 program (PCMCIA) card part numbers 33 removing an ECB that supports cache module B (illustrated) 150 removing an E
Index program (PCMCIA) cards 33 rack-mountable enclosure 30 Model 2200 enclosure 180-watt power supply 30 blank bezel 30 cache module 30 DIMM 34 ECB 30 ECB battery pack 30 EMU 30 fan 30 fiber optic cable 33 GLM 33 HSG80 array controller 30 I/O module 30 maintenance port cable 33 program (PCMCIA) cards 33 rack-mountable enclosure 30 rack-mountable enclosure BA370 26 Model 2100 30 Model 2200 30 partitions 92 patches deleting 233 installing 231 listing 235 listing, installing, deleting 231 PC/terminal to main
Index fiber optic cable single-controller configuration 67 GLM 58 hub dual-redundant controller configuration 67 single-controller configuration 67 single-controller configuration fiber optic cable 67 GLM 58 hub 67 switch 67 switch dual-redundant controller configuration 67 single-controller configuration 67 replacement strategy, field 47 replacing cache module dual-redundant controller configuration 138 single-controller configuration 119 controller single-controller configuration 112 controller and cache
Index S shutting down the subsystem 102 disabling the ECBs 103 enabling the ECBs 104 single-controller configuration BA370 enclosure (illustrated) 113 illustrated Model 2100 enclosure 113 Model 2200 enclosure 113 Model 2100 enclosure (illustrated) 113 Model 2200 enclosure (illustrated) 113 removing controller and cache module 107 GLM 58 replacing BA370 enclosure ECB with subsystem powered off 155 on 151 cache module 119 controller 112 controller and cache module 107 DIMMs 70 ECB 149 GLM 57 I/O module 161 P
Index nonoperating environmental specifications 251 operating environmental specifications 251 Related documentation 15 technical support, HP 22 terminal connection. See maintenance port.
Index 290 HP StorageWorks HSG60 and HSG80 Array Controller and Array Controller Software Maintenance and Service Guide