tricky raid-1 recovery problem
Jack Howarth
howarth at bromo.msbb.uc.edu
Mon Aug 16 20:57:13 UTC 2004
In case it helps here is some specific information from
mdadm regarding the problem /dev/md4 partition I am attempting
to remount from a spare that was partially synced (the master
raid drive has bad blocks). The spare (which was /dev/sdb6 but
now is /dev/sda6 showed no errors itself in e2fsck).
With the master sata drive (which was sda) disconnected all
the other md partitions mount except the one which had the bad
block (/dev/md4 which was composed of /dev/sda6 and /dev/sdb6).
For one of the md's which mount fine I now see...
/sbin/mdadm --examine /dev/sda3
/dev/sda3:
Magic : a92b4efc
Version : 00.90.00
UUID : fa51c3e7:6a8b8980:187ab924:9a8bdecc
Creation Time : Tue Jun 1 07:11:19 2004
Raid Level : raid1
Device Size : 3068160 (2.93 GiB 3.14 GB)
Raid Devices : 2
Total Devices : 1
Preferred Minor : 2
Update Time : Mon Aug 16 16:55:37 2004
State : clean, no-errors
Active Devices : 1
Working Devices : 1
Failed Devices : 0
Spare Devices : 0
Checksum : 432c4f3f - correct
Events : 0.418524
Number Major Minor RaidDevice State
this 1 8 3 1 active sync /dev/sda3
0 0 0 0 0 removed
1 1 8 3 1 active sync /dev/sda3
However for the problem partition I get...
/sbin/mdadm --examine /dev/sda6
/dev/sda6:
Magic : a92b4efc
Version : 00.90.00
UUID : 781bf9dd:6aebc140:89932fde:b710e16a
Creation Time : Tue Jun 1 07:10:33 2004
Raid Level : raid1
Device Size : 81280768 (77.52 GiB 83.23 GB)
Raid Devices : 2
Total Devices : 2
Preferred Minor : 4
Update Time : Mon Aug 16 13:53:59 2004
State : clean, no-errors
Active Devices : 1
Working Devices : 2
Failed Devices : 1
Spare Devices : 1
Checksum : 5396c022 - correct
Events : 0.198153
Number Major Minor RaidDevice State
this 2 8 22 2 spare /dev/sdb6
0 0 8 6 0 active sync /dev/sda6
1 1 0 0 1 faulty removed
2 2 8 22 2 spare /dev/sdb6
I need somehow to force /dev/md4 to be updated to the superblock
information for the current /dev/sda6 (which is the drive that was
/dev/sdb before I detached /dev/sda...which had bad blocks).
Also I guess I need to get rid of the line showing an active sync
on /dev/sdb6. Any ideas how to get out of this muddle short of
deleting the /dev/md4 device and creating it from /dev/sda6 anew?
Jack
More information about the fedora-list
mailing list