Skip to main content

Recovering a failed server after deploying a server pattern

If a server fails after you deploy a server pattern, you can recover the server by unassigning the profile from the failed server and then reassigning that profile to a standby server.

Procedure

Complete the following steps to recover the failed server that uses Lenovo XClarity Administrator managed authentication.

  1. Identify the failed server.
  2. Unassign the server profile from the failed server (see Deactivating a server profile).
    Attention
    The failed server must be powered off to deactivate the assigned virtual addresses before you re-assign the profile. When you unassign the server profile, select Power off server on the Unassign Server Profile dialog to power off the failed server (see Powering on and off a server).
  3. Assign the server profile to a standby server (see Activating a server profile).
  4. Activate the profile by either powering on the standby server if it is currently powered off or by restarting the standby server if it is currently powered on (see Powering on and off a server).
  5. Migrate the VLAN settings on the attached switches to the standby server.
  6. Ensure that the failed server is powered off.
  7. Replace or repair the failed server. If you repair the server, perform the following steps to ensure that the newly repaired server is reset to default settings:
    1. Reset the BMC to factory defaults by using the management web interface for the server. For information about resetting the BMC, see Recovering ThinkSystem, Converged, NeXtScale, or System x M5 or M6 server management after a management server failure by resetting the management controller.
    2. Clear the Unified Extensible Firmware Interface (UEFI) information, including any I/O adapter virtual addresses by using the UEFI menus. For information, see the UEFI documentation.