lvdisplay showing ???? under PV1 field

[Scott Guy_2]

Recently, I had a disk (two disks in a RAID 1 format in an FC60) fail. (Yep, believe it or not, both disks failed due to an overheat situation). Now, in one logical volume, I have 5 extents that were on that disk that show as ???? under the PV1 field. I've tried using the lvreduce with the -k option, but it's just not working. I've resigned myself to the fact that data is gone (and we're getting some residual I/O errors), but I want to clean this up. Any ideas? I've looked through other threads and seen similar issues, but just haven't seen the same scenario.

[Tim Nelson]

This is the best place to start.

http://docs.hp.com/en/5991-1236/When_Good_Disks_Go_Bad_WP.pdf

[Scott Guy_2]

Will try some of those steps tonight when I have sys downtime. I'm afraid I'm too far gone in that the failed disks have been replaced but still having problems. I think I missed something when I replaced the disks.

[Henry Fauni]

If both disks failed on that LUN, it's possible the LUN is gone. Check status of FC60 LUN in question by:

amdsp -l

Check complete status of FC60 first before restoring LVM config by:

amdsp -a

To get of your FC60 you can use:

amdsp -i

To recreate the LUN, you have to use "amcfg".