My system is a RAID6 with 6 1T SATA (barracuda) drives and has been running without a hitch for over a year now and is up to date to the latest 7.5.1 but last monday we had some major power fails in the area. I was not there when it happened so I am not sure if there were only one or multiple power interruptions but it lasted for several hours into the night.
Anyway, the next morning I found the server not responding and on the console I found it failed with a
Kernel panic - not syncing: Attempted to kill init!
The other error messages on the console are:
EXT3-fs error (device dm-0): ext_check_descriptors: Block bitmap for group 3840 is not in group (block 267435456)
EXT3-fs: group descriptors corrupted!
Above that there were
mdadm: /dev/md2 has been started with 4 drives (out of 6)
and above that a list of 5 drives 0, 1, 3, 4, 5 (/dev/sdc2 missing)
2 logical volume(s) in volume group "main" now active
So it looks like the LVM is up and running but the problem is in the EXT3 fs.
I booted with sme rescue and tried to follow
http://wiki.contribs.org/Recovering_SME_Server_with_lvm_drives. I got it to rebuild the array (without /dev/sdc2) but after rebooting it still fails with the same or very similar error.
There were other suggestions in the Recovering_SME_Server_with_lvm_drives and other forum message but a lot of those commands are not available on the sme rescue disk.
I really don't know anymore how I can recover from this situation bar reformatting but I really don't what to do that at this stage.
I was about to burn a Knoppix CD but have no idea how to use this to fix this EXT3-fs corruption problem.
What can I do?
Marco
1 Replies
988 Views