Explanation
The solid state drive has detected faults that indicate that the drive is likely to fail soon. The drive should be replaced. The SAN Volume controller (SVC) error log will identify a managed disk ID for the solid state drive that caused the error.
Action
Perform the maintenance procedure for the 1215 error to identify the managed disk ID for the solid state drive that caused the error.
If the managed disk has gone offline since this error has occurred, the managed disk has failed and you must follow the solid state drive replacement procedure in MAP 'Replacing an offline SSD'.
If the managed disk is still online, perform the following procedure to replace the solid state drive without any data loss:
- Submit the command 'svctask rmmdisk –force (mdisk name/id)', where (mdisk name/id) is the name or ID of the managed disk identified in the error log. This command migrates all of the data from the failing managed disk into the free extents in the rest of the managed disk group. If the command fails with a message that indicates that there are not enough free extents, create more free extents in the managed disk group and resubmit the command. If you cannot create a sufficient number of free extents so that the command completes without error, you must use MAP 'Replacing a failed SSD that is a member of a managed disk group' to replace the drive. You can use any of the following three options to increase the number of free extents: The first option is to remove some of the VDisk copies that exist in this managed disk group. The second option is to migrate some of the VDisk copies into other managed disk groups. The third option is to temporarily add more managed disks to the managed disk group.
- Wait until the status of the managed disk that must be replaced is ‘Unmanaged’.
- Submit the command 'svcinfo lsmdisk (mdisk id)', where (mdisk id) is the name or the ID of the managed disk that is identified in the error log. Record the ‘controller_name’, ‘node_name’ and ‘location’ properties of the MDisk.
- Submit the command 'svcinfo lsnodevpd (node name/id)', where (node name/id) is the node name displayed by the lsmdisk command in step 3. Record the front_panel_id property of this node.
- Perform the solid state drive remove/replace instructions in the Hardware Maintenance guide to replace the solid state drive. To Identify the correct SSD to replace, use the following information: The 'front_panel_id' is on a label on the front of the node, and 'location' identifies the specific drive bay of the node. The drive bays are labeled in red numerals to the right of the drive slot.
- Submit the command ‘svctask detectmdisk’ to discover the new solid state drive. Verify that a new managed disk is discovered and that it has the correct slot number.
- Add the new managed disk into the managed disk group.
- Reverse any procedures that might have been performed in step 1 to create free extents in the managed disk group.
- This step is optional. Run the ‘balance.pl’ script that is available from the alphaworks package called SVCTools to redistribute the VDisk extents equally across all of the managed disks in the managed disk group.
Possible Cause-FRUs or other: