Dell Unity: Reimaging a Storage Processor

The process to reimage a Dell Unity Storage Processor (SP) can be started using the Command-Line Interface (CLI) or Unisphere (UI).

This task can be performed after:

• Validating that a software issue exists with the Storage Processor (SP), such as software corruption.
• Validating that no issues with the Storage Processor (SP) are in place, such as internal component faults.

Unisphere (UI)

Step 1: Transition the SP into Service Mode.

• From the UI, go to SYSTEM > Service > Service Tasks.

• On the Service Tasks page, scroll to the SP that is intended to be reimaged and select the option Enter Service Mode > Execute.

• Enter the credentials of the Service user and select OK.

• The Storage Processor selected reboots to Service Mode.

Step 2: Begin the reimage task.

• After the reboot of the SP into Service Mode, navigate back to SYSTEM > Service > Service Tasks.
• On the Service Tasks page, scroll to the SP that is intended to be reimaged and select the option Reimage > Execute.

• Enter the credentials of the Service user and select OK.

• The Storage Processor selected reboots to Service Mode.
• The reimage process for the selected SP begins.

COMMAND LINE INTERFACE (CLI)

Step 1: Transition the SP into Service Mode.

• Open an SSH session to the array and log In with the Service user credentials.
• Confirm that you are on the correct SP where the reimage process is intended to take place.
  • Run ssh peer if you must move to the other SP.
• Set the flag to boot into Service Mode. This option manually forces the SP to boot into Service Mode the next time it is rebooted.
  • svc_rescue_state -s
• Confirm that the SP has the boot flag set correctly. This option details why the current SP has booted into Service Mode.
  • svc_rescue_state -l
• Reboot the SP.
  • svc_shutdown -r

Step 2: Begin the reimage task.

• After the reboot of the SP into Service Mode, open or restart the SSH session, using the Service user credentials, and verify that the SP shows the following banner:
  • *** WARNING *** This Storage Processor is in Service Mode  *** WARNING ***
• Alternatively, you can run svc_diag to verify the status of the Storage Processor.

• Start the reimage processor by running svc_reimage.
  • Awareness: This task does not complete the reimage process. An SP reboot is required.

• Confirm that all flags are cleared
  • svc_rescue_state -l
• Clear and reset all the boot counters. This option resets all the boot control counters, clears all BMC faults, and instructs the SP to attempt to boot into Normal Mode on the next reboot. Note that errors or faults may still cause the system to boot back into Service Mode. In most instances, you must first address the failure condition which put the SP into service mode before proceeding with clearing the counters.
  • svc_rescue_state -c
• Reboot the Storage Processor
  • svc_shutdown -r
• Wait for the reimage task to complete.
  • This should take around 20-30 minutes.

• Run svc_diag to verify the status of the Storage Processor and confirm it is back to Normal Mode. 

• The reimage task requires the SP being in Service Mode. When an SP is in Service Mode, it stops servicing I/O to the hosts.
• In physical deployments, and if the peer SP is healthy, all NAS servers on the SP failover to the peer SP.
  • By default, when the reimage process completes, the NAS servers failback to the original SP. 
  • If the Failback Policy is disabled, the NAS servers do not automatically failback to the original SP and they remain on the peer SP. A manual failback of the NAS servers is required.
  • Performance can degrade when all NAS servers reside on a single SP.
• An example of the CLI method is attached under the Related tab.
  • The file name is: Example of a Unity SP Reimage using CLI.docx