Hi all,
just tried 2.6.12-rc2 and I still have the same errors from my SATA
disks as with 2.6.11. The setup is a bit complex. The relevant parts (I
think) are:
Adaptec AHA-2940UW SCSI-controller, attached are:
1 harddisk /dev/sda
1 DDS3 streamer /dev/st0
Promise SATA150 TX4 controller, attached are:
2 identical hardisks /dev/sdb and /dev/sdc
/dev/sda consists of the root partition, a swap partition and 4 other
partitions that are physical volumes for dm volume group /dev/vg1
/dev/sdb and /dev/sdc have two partitions each, the first of both make
a RAID-0 array /dev/md0 and the second of both make a md RAID-1 array
/dev/md1
/dev/md0 and /dev/md1 are the physical volumes for dm volume groups
/dev/vg2 and /dev/vg3 resp.
To trigger the failure:
- For all logical volumes in /dev/vg1, /dev/vg2 and /dev/vg3 a snapshot is
created.
- All snapshots are mounted read-only in a "snapshot hierarchy" under
/snap.
- A backup to tape is taken using something like:
find /snap -print | cpio -oaH crc -F /dev/st0
Backup must go to tape, no problem with /dev/null
- At this point, some additional i/o on the SATA disks cause the whole
box to hang. Mostly some errors are written to syslog, they are
always similar:
Apr 9 01:30:35 bear kernel: ata2: status=0x51 { DriveReady SeekComplete Error }Apr 9 01:30:35 bear kernel: ata2: called with no error (51)!
Apr 9 01:30:35 bear kernel: SCSI error : <2 0 0 0> return code = 0x8000002
Apr 9 01:30:35 bear kernel: sdc: Current: sense key: Medium Error
Apr 9 01:30:35 bear kernel: Additional sense: Unrecovered read error - auto reallocate failed
Apr 9 01:30:35 bear kernel: end_request: I/O error, dev sdc, sector 43100350
Apr 9 01:30:35 bear kernel: raid1: Disk failure on sdc2, disabling device.
Apr 9 01:30:35 bear kernel: ^IOperation continuing on 1 devices
The errors are always reported for /dev/sdc2, the second device of a
RAID-1 array. After reboot I am able to raidhotadd the failed partition
without problems.
The problem is 100% reproducible.
The hang is not a "hard hang": X keeps running, the watchdog doesn't hit
but no new processes can be started. Syslog entries stop after some time
(from a few seconds to several minutes).
The problem appeared somewhere between 2.6.10 and 2.6.11.
2.6.10: ok
2.6.10-ac8: ok
2.6.10-ac11: failed
2.6.11: failed
2.6.12-rc2: failed
I'd be glad if there would be a solution for this problem as it prevents
me from using any newer kernel.
-jo
--
-rw-r--r-- 1 jo users 63 2005-04-09 09:31 /home/jo/.signature
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
