From: Andre Noll <maan@systemlinux.org>
To: Steve Ungerer <steve@ungerer.us>
Cc: linux-raid@vger.kernel.org
Subject: Re: Read errors on raid5 device; array is still clean
Date: Thu, 14 Jan 2010 20:59:27 +0100 [thread overview]
Message-ID: <20100114195927.GN7517@skl-net.de> (raw)
In-Reply-To: <4B4E261E.7070508@ungerer.us>
[-- Attachment #1: Type: text/plain, Size: 5208 bytes --]
On 14:59, Steve Ungerer wrote:
> Jan 13 10:50:28 RAID kernel: [3126305.778753] ata3.00: cmd 60/00:30:3f:39:4d/01:00:64:00:00/40 tag 6 ncq 131072 in
> Jan 13 10:50:28 RAID kernel: [3126305.778754] res 41/40:34:3f:39:4d/40:00:64:00:00/40 Emask 0x9 (media error)
> Jan 13 10:50:28 RAID kernel: [3126305.778799] ata3.00: status: { DRDY ERR }
> Jan 13 10:50:28 RAID kernel: [3126305.778812] ata3.00: error: { UNC }
> Jan 13 10:50:28 RAID kernel: [3126305.778828] ata3: hard resetting link
> Jan 13 10:50:29 RAID kernel: [3126306.680039] ata3: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> Jan 13 10:50:29 RAID kernel: [3126306.720221] ata3.00: configured for UDMA/133
> Jan 13 10:50:29 RAID kernel: [3126306.720269] sd 2:0:0:0: [sdc] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
> <snip>
> Jan 13 10:50:29 RAID kernel: [3126306.720534] end_request: I/O error, dev sdc, sector 1682783039
> Jan 13 10:50:29 RAID kernel: [3126306.720573] sd 2:0:0:0: [sdc] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
> Jan 13 10:50:29 RAID kernel: [3126306.720576] sd 2:0:0:0: [sdc] Sense Key : Medium Error [current] [descriptor]
> Jan 13 10:50:29 RAID kernel: [3126306.720578] Descriptor sense data with sense descriptors (in hex):
> Jan 13 10:50:29 RAID kernel: [3126306.720580] 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
> Jan 13 10:50:29 RAID kernel: [3126306.720586] 64 4d 39 3f
> Jan 13 10:50:29 RAID kernel: [3126306.720588] sd 2:0:0:0: [sdc] Add. Sense: Unrecovered read error - auto reallocate failed
> Jan 13 10:50:29 RAID kernel: [3126306.720591] end_request: I/O error, dev sdc, sector 1682782783
> Jan 13 10:50:29 RAID kernel: [3126306.720631] sd 2:0:0:0: [sdc] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE,SUGGEST_OK
> Jan 13 10:50:29 RAID kernel: [3126306.720633] sd 2:0:0:0: [sdc] Sense Key : Medium Error [current] [descriptor]
> Jan 13 10:50:29 RAID kernel: [3126306.720636] Descriptor sense data with sense descriptors (in hex):
> Jan 13 10:50:29 RAID kernel: [3126306.720637] 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
> Jan 13 10:50:29 RAID kernel: [3126306.720643] 64 4d 39 3f
> Jan 13 10:50:29 RAID kernel: [3126306.720646] sd 2:0:0:0: [sdc] Add. Sense: Unrecovered read error - auto reallocate failed
> Jan 13 10:50:29 RAID kernel: [3126306.720648] end_request: I/O error, dev sdc, sector 1682782527
> Jan 13 10:50:29 RAID kernel: [3126306.720683] ata3: EH complete
> Jan 13 10:50:29 RAID kernel: [3126306.720720] sd 2:0:0:0: [sdc] 1953525168 512-byte hardware sectors: (1.00 TB/931 GiB)
> Jan 13 10:50:29 RAID kernel: [3126306.720734] sd 2:0:0:0: [sdc] Write Protect is off
> Jan 13 10:50:29 RAID kernel: [3126306.720736] sd 2:0:0:0: [sdc] Mode Sense: 00 3a 00 00
> Jan 13 10:50:29 RAID kernel: [3126306.720755] sd 2:0:0:0: [sdc] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
> Jan 13 10:50:29 RAID kernel: [3126306.733783] __ratelimit: 182 callbacks suppressed
> Jan 13 10:50:29 RAID kernel: [3126306.733786] raid5:md0: read error corrected (8 sectors at 1682781184 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733790] raid5:md0: read error corrected (8 sectors at 1682781192 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733793] raid5:md0: read error corrected (8 sectors at 1682781200 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733795] raid5:md0: read error corrected (8 sectors at 1682781208 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733798] raid5:md0: read error corrected (8 sectors at 1682781216 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733800] raid5:md0: read error corrected (8 sectors at 1682781224 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733802] raid5:md0: read error corrected (8 sectors at 1682781232 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733809] raid5:md0: read error corrected (8 sectors at 1682781240 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733811] raid5:md0: read error corrected (8 sectors at 1682781248 on sdc1)
> Jan 13 10:50:29 RAID kernel: [3126306.733814] raid5:md0: read error corrected (8 sectors at 1682781256 on sdc1)
> </snip>
>
> My first question: what exactly is going on here? /dev/sdc reports an
> unrecovered read error, md tries to reset the link, reattempts the
> read which still fails, recovers parity from the other drives in the
> array?
Yes (but it's the (S)ATA layer that resets the link).
> Does anything happen to these bad sectors on sdc?
md computes the data the read should have returned by reading all
other component devices of the array. Then it writes that data back
to the bad sector on sdc1 in the hope the drive will reassign the
bad sector. The "read error corrected" message indicates that this
write succeeded.
> A check of the md array still shows it as clean with no drives failing.
This is how it is supposed to be :)
> Is there the possibility I'm replacing a perfectly good drive and
> these errors are due to some software problem?
Unlikely, since you mentioned the smart log also contains error
messages.
Andre
--
The only person who always got his work done by Friday was Robinson Crusoe
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]
prev parent reply other threads:[~2010-01-14 19:59 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-13 19:59 Read errors on raid5 device; array is still clean Steve Ungerer
2010-01-14 19:59 ` Andre Noll [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20100114195927.GN7517@skl-net.de \
--to=maan@systemlinux.org \
--cc=linux-raid@vger.kernel.org \
--cc=steve@ungerer.us \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).