Hi all,
there was some talk recently about certain raid devices not being assembled
properly. I also have such situation, so this might shed some light into
this.
What i have is a raid0 array made of 14 mirrors. All were configured at the
same time in exactly the same way, same tools etc. That's why this is
interesting.
Partitions are type 83 and there is a script that takes care of assembling
devices in order on boot. The script uses mdadm --assemble /dev/mdX --uuid
<uuid>.
But after every boot, /dev/md9 comes up as:
md9 : active raid1 sdi1[0]
35842944 blocks [2/1] [U_]
The missing component is sdw1.
So here's what mdadm 1.2 says about the disks in question:
# mdadm -Q /dev/md9
/dev/md9: 34.18GiB raid1 2 devices, 0 spares. Use mdadm --detail for more
detail.
/dev/md9: device 4 in 14 device active raid0 md19. Use mdadm --examine for
more detail.
# mdadm -D /dev/md9
/dev/md9:
Version : 00.90.00
Creation Time : Mon Apr 28 16:46:07 2003
Raid Level : raid1
Array Size : 35842944 (34.18 GiB 36.70 GB)
Device Size : 35842944 (34.18 GiB 36.70 GB)
Raid Devices : 2
Total Devices : 0
Preferred Minor : 9
Persistence : Superblock is persistent
Update Time : Sat Aug 21 02:06:04 2004
State : dirty, no-errors
Active Devices : 1
Working Devices : 1
Failed Devices : -1
Spare Devices : 0
Number Major Minor RaidDevice State
0 8 129 0 active sync /dev/sdi1
1 0 0 1 faulty removed
UUID : 5f68a534:97eb094d:bcc5de8d:225eb1f7
Events : 0.214
# mdadm -Q /dev/sdi1
/dev/sdi1: is not an md array
/dev/sdi1: device 0 in 2 device active raid1 md9. Use mdadm --examine for
more detail.
# mdadm -Q /dev/sdw1
/dev/sdw1: is not an md array
/dev/sdw1: device 1 in 2 device mismatch raid1 md9. Use mdadm --examine for
more detail.
# mdadm -E /dev/sdi1
/dev/sdi1:
Magic : a92b4efc
Version : 00.90.00
UUID : 5f68a534:97eb094d:bcc5de8d:225eb1f7
Creation Time : Mon Apr 28 16:46:07 2003
Raid Level : raid1
Device Size : 35842944 (34.18 GiB 36.70 GB)
Raid Devices : 2
Total Devices : 0
Preferred Minor : 9
Update Time : Sat Aug 21 02:06:04 2004
State : dirty, no-errors
Active Devices : 1
Working Devices : 1
Failed Devices : -1
Spare Devices : 0
Checksum : 19b4cf3 - correct
Events : 0.214
Number Major Minor RaidDevice State
this 0 8 129 0 active sync /dev/sdi1
0 0 8 129 0 active sync /dev/sdi1
1 1 0 0 1 faulty removed
# mdadm -E /dev/sdw1
/dev/sdw1:
Magic : a92b4efc
Version : 00.90.00
UUID : 5f68a534:97eb094d:bcc5de8d:225eb1f7
Creation Time : Mon Apr 28 16:46:07 2003
Raid Level : raid1
Device Size : 35842944 (34.18 GiB 36.70 GB)
Raid Devices : 2
Total Devices : 1
Preferred Minor : 9
Update Time : Tue Apr 20 20:28:11 2004
State : dirty, no-errors
Active Devices : 2
Working Devices : 2
Failed Devices : -1
Spare Devices : 0
Checksum : fa277e - correct
Events : 0.213
Number Major Minor RaidDevice State
this 1 65 97 1 active sync /dev/sdw1
0 0 8 129 0 active sync /dev/sdi1
1 1 65 97 1 active sync /dev/sdw1
There are obvious some differences ... beggining with mdadm -Q for sdw1
saying it is a 'mismatch raid1' and continuing with the contents of the
superblocks on the two disks.
I'm interested in theoires how this might have happened and in suggestions
on how to fix it. Is mdadm -U the right thing here?
--
Jure PeÄar
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
