AIX

Hi all,

I ended up replacing a failed disk drive on my power 7. I had a few issues with unmirroring as it had a stale logical volume. I ended up finally getting it to unmirror the rootvg. I had an empty space in slot 3 so I ended up putting the new disk in slot 3. It configured to hdisk0 with no problems but I decided I wanted to move the new disk to slot 1 where the old bad drive was. I did a rmdev -rdl hdisk0 and ran diags. I took the drive out of slot 3 with no problems. I ran command to insert drive into slot 1. Now the drive is stuck in "populated" state. I decided to run cfgmgr again and the drive will not configure itself. I thought rmdev -Rdl scrubbed out hdisk0 when it was in slot3. I look in the errpt log and see sas fabric device errors when I inserted the drive in slot 1. I was wondering if maybe the ODM is conflicted or if I have possible bad slot?

on a side note, if I shut the machine down and move the drive to a different slot(same sas controller) while its mirrored(rootvg), will this cause any issues?

Thanks

Is this a hot swap disk? It sounds quite like it, but I can't be sure. If so, then the best replacement process is to use the diag panels. Fire up diag as the super-user account, choose "Task Selection" then towards the bottom of the list will be "Hot Plug Task" or something like that. Select the option to "Replace" disk/slots appropriate and follow the on screen prompts. You will get an identifying light on the device that can be removed.

Are the slots of the same type for a starter. Your not going for a PCI versus a SCSI disk are you? I'm not sure if the physical connection will let you actually plug it in, of course.


Does slot 1 still have a definition showing?

Could this be a backplane issue rather than the disk? To that end, have you tried the failed disk in slot 3 to see if it revives?


Do you have a hardware support contract in place? If so, then this is their issue.



I hope that this helps.


Robin
Liverpool/Blackburn
UK

I have been using the diags method the entire time. When the failed disk was removed, it shows empty in devices. I have hardware support replacing the backplane. I will see if this fixes it.

Thanks