IBM 7133 Fault Check Method
1. Check ERRPT error messages.
An error message is generated when read/write operations to the hard disk fail. Hot spare disks or disks in the array that have been rejected due to failure typically generate only one error message, while failed disks in the array or disks with intermittent errors during read/write operations will generate consecutive error messages.
2. diag -> Task -> ssa service aids -> Link Verification: Check the LINK status. If it displays 'failed', there is a connection issue or fault with the disk. (If it displays 'GOOD', the connection is good, but this does not guarantee the hard disk itself is functioning correctly.)
3. smitty ssaraid->List All Defined SSA RAID Arrays
List all RAID arrays. If all display 'GOOD', then all disks within the RAID are functioning properly. If any RAID displays 'degraded', a disk within that RAID has failed. Combine this information with ERRPT error messages to identify the specific faulty disk.
smitty ssaraid->List/Identify SSA Physical Disks
-> List Rejected Array Disks: Lists disks that have been REJECTED. These are faulty disks removed from the RAID and require replacement.
Since a failed hot spare disk typically generates only one ERRPT error message and does not affect the RAID status, it can easily be overlooked if the logs are old or have been cleared by the customer. Therefore, we must determine the ownership status of all disks to confirm if there are any hidden faulty disks.
Method: 1) Run lsdev -Cc pdisk to list all pdisk devices.
2) Run smitty ssaraid -> ... -> List Disks in an SSA RAID Array to list all disks within the RAID.
3) Run smitty ssaraid -> ... -> List Hot Spares to list all hot spare disks.
4) Run smitty ssaraid -> ... -> List Rejected Array Disks to list all rejected disks.
Disks that have no assigned ownership but physically exist are considered hidden faulty disks.
4. If the RAID status displays as 'GOOD' but ERRPT shows consecutive error messages for a disk within that RAID, the disk is still usable but unsafe. It is recommended to replace it after communicating with the customer.