0

Last night my server crashed, after reboot I have found it has a disk failure

The configuration is 2 x 500GB sata on Centos 6

  • sda3/sdb3 is software RAID1
  • sda1/sdb1 = ext3 for /
  • sda2/sdb2 = swap
  • sda3/sdb3 = LVM
  • LVM = /var, /usr, /home, /opt, /backup

disk sdb has failed, it is currently running only on sda

Now for the magic...

after rebooting, the entire filesystem has gone back a year and a half?

today is 14/07/15, the last file updated is 15/02/14

So my question is, how can a LVM on top of RAID1 go back in time???

no backups, well at least not in date anymore :( they were all on the LVM /backup partition which has also gone back in time

Many thanks

Stuart

  • 8
    A wild guess is that the RAID sda/sdb3 had failed a long time ago, running only on sdb. Now that sdb has failed, perhaps it came back online with only sda? – Halfgaar Jul 14 '15 at 8:21
  • +1 for that guess. Same thing happened to a friend's server some time ago... – etagenklo Jul 14 '15 at 8:58
  • That's not a wild guess, that's the correct answer most likely. Time to test if at least your backup was working (or to ship the disks to a recovery firm). – Sven Jul 14 '15 at 9:04
  • Halfgaar, can you add that as answer so I can mark as correct, thanks – Stuart Brand Jul 14 '15 at 9:14
1

As requested, now as an answer:

A wild guess is that the RAID sda/sdb3 had failed a long time ago, running only on sdb. Now that sdb has failed, perhaps it came back online with only sda?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.