* [linux-lvm] Errors from a mirrored LV
@ 2010-09-11 5:02 Fredrik Tolf
2010-09-13 1:50 ` Stuart D Gathman
0 siblings, 1 reply; 2+ messages in thread
From: Fredrik Tolf @ 2010-09-11 5:02 UTC (permalink / raw)
To: linux-lvm
Hi list,
Having a disk that's going awry, I'm glad I've got the data in a
mirrored LV so that I can replace it.
However, now that I still haven't got a new disk to replace it with, it
seems that LVM is nevertheless returning I/O errors for reads from the
broken sectors, as evidenced by logs like these:
ata4.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x0
ata4.00: BMDMA2 stat 0xc0009
ata4.00: cmd 25/00:08:a7:1c:84/00:00:27:00:00/e0 tag 0 dma 4096 in
res 51/40:00:aa:1c:84/00:00:27:00:00/e0 Emask 0x9 (media error)
ata4.00: status: { DRDY ERR }
ata4.00: error: { UNC }
ata4.00: configured for UDMA/100
sd 3:0:0:0: [sdd] Result: hostbyte=DID_OK
driverbyte=DRIVER_SENSE,SUGGEST_OK
sd 3:0:0:0: [sdd] Sense Key : Medium Error [current] [descriptor]
Descriptor sense data with sense descriptors (in hex):
72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
27 84 1c aa
sd 3:0:0:0: [sdd] Add. Sense: Unrecovered read error - auto reallocate
failed
end_request: I/O error, dev sdd, sector 662969514
ata4: EH complete
I/O error in filesystem ("dm-3") meta-data dev dm-3 block 0x27841ae8
("xfs_trans_read_buf") error 5 buf count 4096
Note that last line of log, emitted by XFS.
Shouldn't LVM, when a read fails in a mirror, try to fetch that data
from the second disk instead?
--
Fredrik Tolf
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [linux-lvm] Errors from a mirrored LV
2010-09-11 5:02 [linux-lvm] Errors from a mirrored LV Fredrik Tolf
@ 2010-09-13 1:50 ` Stuart D Gathman
0 siblings, 0 replies; 2+ messages in thread
From: Stuart D Gathman @ 2010-09-13 1:50 UTC (permalink / raw)
To: linux-lvm
On 09/11/2010 01:02 AM, Fredrik Tolf wrote:
> Having a disk that's going awry, I'm glad I've got the data in a
> mirrored LV so that I can replace it.
>
> sd 3:0:0:0: [sdd] Add. Sense: Unrecovered read error - auto reallocate
> failed
> end_request: I/O error, dev sdd, sector 662969514
> ata4: EH complete
> I/O error in filesystem ("dm-3") meta-data dev dm-3 block 0x27841ae8
> ("xfs_trans_read_buf") error 5 buf count 4096
>
> Note that last line of log, emitted by XFS.
>
> Shouldn't LVM, when a read fails in a mirror, try to fetch that data
> from the second disk instead?
>
Correct me if I'm wrong, but LVM mirror support is still experimental
and doesn't do a lot of things. If you need reliable mirroring now, use
the md driver for raid1, and add the /dev/md* devices as physical volumes.
In the meantime, you should probably remove the failing PV.
AIX is still the best LVM mirroring I've seen. I know ZFS is great too,
but it is more tied to a specific filesystem.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2010-09-13 1:50 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-09-11 5:02 [linux-lvm] Errors from a mirrored LV Fredrik Tolf
2010-09-13 1:50 ` Stuart D Gathman
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).