Something I find suspicious, on Aug 13 11:31:12, this happened:
kjournald starting. Commit interval 15 seconds
EXT3-fs (sdb1): using internal journal
EXT3-fs (sdb1): mounted filesystem with ordered data mode
kjournald starting. Commit interval 15 seconds
EXT3-fs (sda1): using internal journal
EXT3-fs (sda1): mounted filesystem with ordered data mode
I really hope this did not succeed! sda & sdb are supposed to be partitionless!
In the recent upgrade from Fedora 9 to Opensuse 11.4, sda & sdb pairwise
swapped device names with sdc & sdd. So, I can well imagine trying to mount
these as usual, before noticing the new device names.
Since md superblock is at the end of the device, it could be there is old
leftover partition table at the beginning of the device (ext3/4 does not use
first KB of the device). However most likely you wouldn't be able to mount a
filesystem from such partition because superblock would not be at the place of
https://bugzilla.novell.com/show_bug.cgi?id=714201
https://bugzilla.novell.com/show_bug.cgi?id=714201#c6
Jan Kara changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |NEEDINFO
InfoProvider| |andreas_nordal_4@hotmail.co
| |m
--- Comment #6 from Jan Kara 2011-09-03 00:23:26 UTC ---
(In reply to comment #5)
the partition where filesystem expects it. So I'd rather think that sda1/sdb1
were your backup disks at that time. Names like sda, sdb are no really stable
(depend on the order of discovery in kernel). That's why it's better to use
labels or /dev/disk/by-id/ if that's not what you are already doing.
Now to the rest of the log. I can see there:
Aug 24 18:49:26 nerdvar kernel: [942179.968166] EXT4-fs (md0): mounted
filesystem with ordered data mode. Opts: (null)
Aug 24 18:52:52 nerdvar kernel: [942386.451762] EXT4-fs (md0): mounted
filesystem with ordered data mode. Opts: (null)
..
Aug 24 20:15:09 nerdvar kernel: [947323.146397] EXT4-fs error (device md0):
ext4_lookup:1044: inode #105136936: comm find: deleted inode referenced:
105137415
and then the log continues with 26 errors of this type for that directory. Then
you unmounted the filesystem (Is this step 4? What was happening with the
filesystem during these 4 hours?):
Aug 24 22:52:39 nerdvar kernel: [956772.689363] EXT4-fs error (device md0):
ext4
_put_super:730: Couldn't clean up the journal
And tried to mount it again which failed:
Aug 24 22:58:07 nerdvar kernel: [957100.881615] EXT4-fs (md0):
ext4_check_descriptors: Block bitmap for group 1024 not in group (block 0)!
And then you apparently run fsck for a minute (since the fs is no longer marked
as with errors and descriptors are fixed) and tried mounting the filesystem
again (step 7.?):
Aug 24 22:59:09 nerdvar kernel: [957163.467233] EXT4-fs (md0): warning:
mounting unchecked fs, running e2fsck is recommended
Aug 24 22:59:09 nerdvar kernel: [957163.480937] EXT4-fs (md0): mounted
filesystem with ordered data mode. Opts: (null)
And then things get interesting again:
Aug 24 23:00:13 nerdvar kernel: [957227.297242] ata1.00: exception Emask 0x0
SAct 0x0 SErr 0x0 action 0x6
Aug 24 23:00:13 nerdvar kernel: [957227.297248] ata1.00: BMDMA stat 0x5
Aug 24 23:00:13 nerdvar kernel: [957227.297253] ata1.00: failed command: READ
DMA EXT
Aug 24 23:00:13 nerdvar kernel: [957227.297262] ata1.00: cmd
25/00:00:d0:13:14/00:04:01:00:00/e0 tag 0 dma 524288 in
Aug 24 23:00:13 nerdvar kernel: [957227.297263] res
51/84:5f:71:11:14/84:02:01:00:00/e0 Emask 0x10 (ATA bus error)
Aug 24 23:00:13 nerdvar kernel: [957227.297267] ata1.00: status: { DRDY ERR }
Aug 24 23:00:13 nerdvar kernel: [957227.297270] ata1.00: error: { ICRC ABRT }
Aug 24 23:00:13 nerdvar kernel: [957227.297281] ata1: soft resetting link
Aug 24 23:00:13 nerdvar kernel: [957227.469035] ata1.00: configured for
UDMA/133
Aug 24 23:00:13 nerdvar kernel: [957227.469055] ata1: EH complete
(and there are lot more messages like this for both drives). This error message
means we were not able to read some block. Given that it happens for both
drives, I don't think it's actually a drive failure. The most likely problem
could be in cabling or motherboard having some issues... Can you attach here
/var/log/boot.msg so that I can see some details about your HW?
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.