8100 Updating Product Software Guide (XCS 6.250) (5697-1716, February 2012)

NOTE: This procedure may cause data to become temporarily unavailable to the host. Normally

this will not happen, but if a hardware failure occurred while performing this procedure, the host

may lose access to data until the procedure is complete. HP recommends that you schedule a

maintenance window to perform these steps.

1. Disconnect the enclosure address bus cable from each EMU.

2. Unseat/remove all EMUs in the array, and then restart the top controller.

3. When the top controller has restarted and is back online, restart the bottom controller.

4. When both controllers are back online, insert the EMUs one at a time, in any order, as follows:

a. Insert the EMU, and wait for the LED to display 00. When it does, connect the enclosure

address bus cable and make sure the correct enclosure address replaces 00 in the display.

HP strongly recommends that you wait one minute between reseating each EMU.

b. Repeat this process for all EMUs.

5. When all the EMUs are reinstalled and display their correct enclosure IDs, restart the top

controller.

6. When the top controller is back online, restart the bottom controller.

For additional information, see the user guide for your array as recommended in “Reference

documentation” (page 7).

Multiple resynchronizations

During the upgrade process, both controllers perform a resynchronization simultaneously. This

invokes a fast restart of the controllers, bypassing execution of lengthy hardware diagnostics. The

fast restart allows the controllers to complete the resynchronization sequence before the host timeout

period elapses and the host becomes aware that the storage system is unavailable.

If an event occurs that causes a second resynchronization while the upgrade resynchronization is

still in progress, the controller will terminate with a code of 0b 08 01 00. The following events

cause the controller to resysnchronize automatically:

• Initialization of the storage system

• Setting debug flags through the command line interface or OCP

• Disk group hardware problem on a Vraid1 virtual disk, or attempted reconnection of a disk

when there is a disk group hardware problem

• Bad Block Replacement (BBR) performed on a disk

Every effort should be made to ensure that these events do not occur during the upgrade process.

If controller termination occurs due to multiple resynchronizations, contact your HP-authorized

service representative for assistance.

Offline storage system

The controller software upgrade should take approximately 10 minutes to complete. If the storage

system does not come back online within 10 minutes, check the following:

• Determine if the problem is with HP P6000 Command View by answering the questions at

the start of this chapter.

• Check any Insight Remote Support messages.

• Check the storage system event logs:

1. Run HP P6000 Command View in your browser.

2. Select the storage system whose logs you want to view.

3. Click View events to view the event logs that can be displayed.

60 Troubleshooting