Recovering From a Failed Server

Recovering From a Failed Server

When replacing a failed server in a high-availability configuration, you might not need to replace the ServeRAID adapter. However, if you replace your ServeRAID adapter, you must reconfigure the adapter after you have installed your new server.

-Important-
The following procedure requires specific configuration settings for the ServeRAID adapter.
If the server and adapter that are being replaced are functional, you can obtain these settings from the adapter.
However, if the adapter or the server is not functional, you will need a record of these settings, such as one that was created when the adapter was previously configured.
If you are replacing your ServeRAID adapter with your server, you must have correct configuration information to complete this procedure.

Use the following procedure to recover from a failed server:

Remove the failed server from your high-availability configuration.
Remove all hard disk drives from the disk drive array in the failed server.
As you remove your hard disk drives, be sure to note the bay in which each drive was installed.
If you are replacing your failed server with an identical server you can reinstall the drives in an identical configuration and get your server up and running quickly.
If the ServeRAID adapter is functional, remove it from the failed server.
As you remove the adapter from the failed server, be sure to:
- Note which SCSI cables are connected to the SCSI channel connectors on the adapter.
- Note which PCI slot has the adapter installed.
If you are replacing your failed server with an identical server, you can reinstall the ServeRAID adapter in an identical configuration and get your server up and running quickly.
Install the hard disk drives in the new server.
For information on how to install a hard disk drive, see the documentation that comes with your server.
If you are replacing the failed server with an identical server, install each hard disk drive in the same bay as the one it was removed from in the failed server.
Install the new ServeRAID adapter.
For instructions on how to install the ServeRAID adapter, see the IBM ServeRAID-3H and ServeRAID-3L Ultra2 SCSI Adapters Installation and User's Guide .
If you are installing a ServeRAID adapter that was previously installed in the failed server, install the adapter in the same PCI slot as it was installed in the failed server.
-Important- Do not reconnect the SCSI channel cables to the adapter at this time.
Configure the ServeRAID adapter in the new server.
Note: If you have installed the adapter in a server that is identical to the failed server and have installed it in the same PCI slot that it was installed in the failed system, you might not need to configure the adapter.
You will need the following information to configure your new ServeRAID adapter:
- SCSI Bus Initiator_IDs
- Adapter Host_ID
- Cluster Partner's Host_ID
If the ServeRAID adapter and server that you are replacing are functional, you can obtain this information by starting the server with the IBM ServeRAID Configuration Diskette and selecting the Display/Change Adapter Params item from the Advanced Functions menu.
If the ServeRAID adapter or server is not functional , you will need to refer to the record of the settings that you made when the adapter was previously configured.
If you do not have a record of the configuration information, the following hints might help you to assign the proper values.
- The SCSI Bus Initiator_IDs for a non-shared SCSI channel are set to 7 by default.
  For shared SCSI channels, the IDs will usually be 7 or 6 and must be different than the SCSI Bus Initiator_IDs for the corresponding SCSI channels of the cluster partner adapter.
  Therefore, if you can obtain the SCSI Bus Initiator_IDs for the corresponding cluster partner system, you can infer the correct SCSI Bus Initiator_IDs for this system.
  For example, if the cluster partner system's shared SCSI Bus Initiator_IDs were set to 7, then usually the adapter you are replacing would need to have its IDs set to 6.
  To obtain the SCSI Bus Initiator_IDs from the corresponding cluster partner adapter, start the cluster partner system with the ServeRAID Configuration Diskette and select the Display/Change Adapter Parameters option from the Advanced Functions menu.
- The proper settings for the Host_ID and Cluster Partner's Host_ID of the adapter being replaced can also be determined by reading the settings from the cluster partner system by using the Display/Change Adapter Params option.
  Ensure that the Host_ID for the new adapter is identical to the Cluster Partner's Host_ID value defined on the corresponding adapter in the cluster partner system, and that the Cluster Partner's Host_ID value for the new adapter is identical to the Host_ID value defined on the corresponding adapter in the cluster partner system.
Start the system from the IBM ServeRAID Configuration Diskette Version 3.00 or later.
Initialize the adapter configuration.
To initialize the adapter:
1. Select Advanced Functions from the main menu.
2. Select Init/View/Synchronize Config.
3. Select Initialize Config.
Ensure that the adapter is at the latest BIOS level.
The BIOS level of the adapter is displayed after system POST when the adapter BIOS loads.
The latest BIOS levels are available from the IBM website at
http://www.pc.ibm.com/support
Once you have connected to this URL, search for RAID BIOS.
Download and read the text file to determine the latest levels available.
If your adapter BIOS is downlevel, download and apply the BIOS update.
Update the configuration parameters.
To update the configuration parameters:
1. Start the system from the IBM ServeRAID Configuration Diskette Version 2.40 or later.
2. Select Advanced Functions from the main menu.
3. Select Display/Change Adapter Params.
4. Using the settings that were assigned to the ServeRAID adapter you are replacing,
  select and configure each of the following parameters:
  - SCSI Bus Initiator_IDs
  - Adapter Host_ID
  - Cluster Partner's Host_ID
5. Select Change RAID Parameters from the Advanced Functions menu and enable unattended mode.
Shut down the system and reconnect the SCSI channel cables to the adapter.
Be sure to connect the cables to the correct SCSI channels as noted in (above) step 3.
-Important- If the ServeRAID adapter being replaced is not the adapter that attaches to the server startup disk array or other non-shared disk arrays, you do not need to perform any of the following steps.
The system can now be restarted normally.
If the adapter that was replaced attaches to the operating system startup disk array for the system or if other non-shared disk arrays are attached to this adapter, start the system using the IBM ServeRAID Configuration Diskette Version 2.40 or later and then:
1. Select Advanced Functions from the main menu.
2. Select Merge Group Management.
3. Restore the adapter disk array configuration.
  - To restore non-shared disk array configurations:
    1) Select Merge Group Management from the Advanced Functions menu; then, press Enter.
    2) Select Merge/Unmerge Logical Drive and then press Enter.
    3) Select Merge Own Non-Shared Logical Drive.
    4) Type in the Group ID field the Merge Group ID for the array
    2xx
    where xx is the shared SCSI Bus Initiator_ID, and then press Enter.
    The Merge Group ID value is typically 206 or 207.
    A message appears at the bottom of the screen saying
    Merging own shared logical drive(s). Please wait...
    5) When the process completes, a message appears saying
    Merge/Unmerge operation completed successfully.
    Press any key to continue.
    6) Press Esc to return to the previous menu. Continue pressing Esc to return to the Main Menu. If the adapter you are replacing is the boot adapter, the system should now be able to startup the operating system properly.
  - To restore shared disk array configurations:
    Note: Usually all shared arrays will have failed over and will not need to be merged.
    1) Select Merge Group Management from the Advanced Functions menu; then, press Enter.
    A screen similar to the following appears.
    
    2) Select Merge/Unmerge Logical drive an then press Enter
    
    3) Select Merge Own Shared Logical Drive for each shared array
    (Merge Group IDs in the range from 1 to 8)
    that has not failed over to the cluster partner system
    (for example, RAID level 5 arrays in critical or degraded state)
    to restore the configuration of these shared arrays.
    4) Press Esc to return to the previous menu.
    Continue pressing Esc to return to the Main Menu.
    Repeat this process for each shared array
    (merge group IDs in the range from 1 to 8)
    that has not failed over to the cluster partner system
    (for example, RAID level 5 arrays in critical/degraded state)
    to restore the configuration of these shared arrays.
    -Important- The IBM ServeRAID Configuration Diskette must not be used to perform failover and failback to merge or unmerge drives belonging to the other server.
    Failover and failback to merge or unmerge drives belonging to the other server is normally handled by the operating system software and cluster support software.
Restart your server.
Once all array configurations have been restored, the server can be restarted normally.

Back to

Please see the LEGAL - Trademark notice.
Feel free - send a for any BUG on this page found - Thank you.