Markus Schulte ([EMAIL PROTECTED]) wrote on 14 July 1999 19:50:
...
>Jul 14 19:17:42 raid kernel: sdd1's event counter: 00000003
>Jul 14 19:17:42 raid kernel: sde1's event counter: 00000004
>Jul 14 19:17:42 raid kernel: sdc1's event counter: 00000005
>Jul 14 19:17:42 raid kernel: sdb1's event counter: 00000005
>Jul 14 19:17:42 raid kernel: md: superblock update time inconsistency --
...
>Jul 14 19:17:42 raid kernel: md: kicking non-fresh sdd1 from array!
...
>Jul 14 19:17:42 raid kernel: md0: kicking faulty sde1!
...
>Jul 14 19:17:42 raid kernel: raid5: not enough operational devices for
>md0 (2/4 failed)
You have two drives inconsistent... raid5 gives protection against only
one failure at a time.
This may happen because of scsi cable/controller problems. If you have
two or more disks on the same controller you lose. This happened to
me. IF your data are still good you can do
mkraid --really-force /dev/md0
PROVIDED your /etc/raidtab matches what's on the disks, including scsi
ordering, etc. This command will not erase your data.
If the problem is with the disks, you obviously have backups, no? :-)
