linux-ide.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Allan Wind <allan_wind@lifeintegrity.com>
To: IDE/ATA development list <linux-ide@vger.kernel.org>,
	linux-scsi@vger.kernel.org
Subject: Re: MD/RAID time out writing superblock
Date: Thu, 17 Sep 2009 12:36:47 -0400	[thread overview]
Message-ID: <20090917163647.GA6663@lifeintegrity.com> (raw)
In-Reply-To: <4AB2596D.10809@kernel.org>

On 2009-09-18T00:44:45, Tejun Heo wrote:
> Hello,
> 
> Chris Webb wrote:
> > It's quite hard for us to do this with these machines as we have
> > them managed by a third party in a datacentre to which we don't have
> > physical access.  However, I could very easily get an extra 'test'
> > machine built in there, generate a work load that consistently
> > reproduces the problems on the six drives, and then retry with an
> > array build from 5, 4, 3 and 2 drives successively, taking out the
> > unused drives from chassis, to see if reducing the load on the power
> > supply with a smaller array helps.
> 
> Yeap, that also should shed some light on it.

I have a SuperMicro X8DT3-F motherboard with 2 (2 TB) WDC drives 
of the 8 bays available in the machine.  They are on a different 
controller LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS
which was flashed into "Integrated Target Mode" to get it running 
under Linux.

Disabling smartmontools seems to have helped in terms of failure 
frequency.  It is almost always the 2nd drive that is kicked out 
of the mirror although the last time it was the primary after 
disabling smart.  hddtemp was never running on this host.

[2256003.055451] end_request: I/O error, dev sdb, sector 3907028974
[2256003.055674] md: super_written gets error=-5, uptodate=0
[2256003.055677] raid1: Disk failure on sdb2, disabling device.
[2256003.055678] raid1: Operation continuing on 1 devices.
[2256003.437315] RAID1 conf printout:
[2256003.437318]  --- wd:1 rd:2
[2256003.437321]  disk 0, wo:0, o:1, dev:sda2
[2256003.437323]  disk 1, wo:1, o:0, dev:sdb2
[2256003.440542] RAID1 conf printout:
[2256003.440545]  --- wd:1 rd:2
[2256003.440548]  disk 0, wo:0, o:1, dev:sda2

[3880879.007618] end_request: I/O error, dev sda, sector 3907028974
[3880879.007839] md: super_written gets error=-5, uptodate=0
[3880879.007842] raid1: Disk failure on sda2, disabling device.
[3880879.007843] raid1: Operation continuing on 1 devices.
[3880879.028518] RAID1 conf printout:
[3880879.028521]  --- wd:1 rd:2
[3880879.028524]  disk 0, wo:1, o:0, dev:sda2
[3880879.028527]  disk 1, wo:0, o:1, dev:sdb2
[3880879.031607] RAID1 conf printout:
[3880879.031610]  --- wd:1 rd:2
[3880879.031613]  disk 1, wo:0, o:1, dev:sdb2

There is barely any load on this box.  Disabling NCQ did not help 
for me. 


/Allan
-- 
Allan Wind
Life Integrity, LLC
<http://lifeintegrity.com>


  reply	other threads:[~2009-09-17 16:36 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <004e01ca25e4$c11a54e0$434efea0$@ca>
     [not found] ` <9cfb6af689a7010df166fdebb1ef516b.squirrel@neil.brown.name>
     [not found]   ` <4A948A82.4080901@redhat.com>
     [not found]     ` <b585ed9f13649050bbc984869d081315.squirrel@neil.brown.name>
     [not found]       ` <4A94905F.7050705@redhat.com>
     [not found]         ` <005101ca25f4$09006830$1b013890$@ca>
     [not found]           ` <4A94A0E6.4020401@redhat.com>
     [not found]             ` <005401ca25ff$9ac91cc0$d05b5640$@ca>
     [not found]               ` <4A950FA6.4020408@redhat.com>
     [not found]                 ` <92cb16daad8278b0aa98125b9e1d057a@localhost>
     [not found]                   ` <4A95573A.6090404@redhat.com>
2009-08-26 18:12                     ` MD/RAID: what's wrong with sector 1953519935? Andrei Tanas
2009-08-27  0:07                       ` Mark Lord
2009-08-27  1:37                         ` Andrei Tanas
2009-08-27  2:33                       ` Robert Hancock
     [not found]                       ` <d086b110526f8bac2f562850dfc70b03@localhost>
2009-08-27 21:57                         ` MD/RAID time out writing superblock Ric Wheeler
2009-08-31  8:10                           ` Tejun Heo
2009-08-31 12:04                             ` Ric Wheeler
2009-08-31 12:20                               ` Tejun Heo
2009-09-07 11:44                                 ` Chris Webb
2009-09-07 11:59                                   ` Chris Webb
2009-09-09 12:02                                     ` Chris Webb
2009-09-14  7:41                                       ` Tejun Heo
2009-09-14  7:44                                         ` Tejun Heo
2009-09-14 12:48                                           ` Mark Lord
2009-09-14 13:05                                             ` Tejun Heo
2009-09-14 14:25                                               ` Mark Lord
2009-09-16 23:19                                                 ` Chris Webb
2009-09-17 13:29                                                   ` Mark Lord
2009-09-17 13:32                                                     ` Mark Lord
2009-09-17 13:37                                                     ` Chris Webb
2009-09-17 15:35                                                     ` Tejun Heo
2009-09-17 16:16                                                       ` Mark Lord
2009-09-17 16:17                                                         ` Mark Lord
2009-09-18 17:05                                                           ` Chris Webb
2009-09-21 10:26                                                             ` Chris Webb
2009-09-21 19:47                                                               ` Mark Lord
2009-09-22  6:16                                                               ` Robert Hancock
2009-09-20 18:36                                                         ` Robert Hancock
2009-09-14 13:11                                           ` Henrique de Moraes Holschuh
2009-09-14 13:24                                             ` Tejun Heo
2009-09-14 14:02                                               ` Henrique de Moraes Holschuh
2009-09-14 14:34                                                 ` Tejun Heo
2009-09-14 13:14                                         ` Gabor Gombas
2009-09-07 16:55                                   ` Allan Wind
2009-09-07 23:26                                     ` Thomas Fjellstrom
2009-09-14  7:46                                       ` Tejun Heo
2009-09-14 21:13                                         ` Thomas Fjellstrom
2009-09-14 22:23                                           ` Tejun Heo
2009-09-16 22:28                                 ` Chris Webb
2009-09-16 23:47                                   ` Tejun Heo
2009-09-17  0:34                                     ` Neil Brown
2009-09-17 12:00                                       ` Chris Webb
2009-09-17 11:57                                     ` Chris Webb
2009-09-17 15:44                                       ` Tejun Heo
2009-09-17 16:36                                         ` Allan Wind [this message]
2009-09-18  0:16                                           ` Tejun Heo
2009-09-18  2:47                                             ` Allan Wind
2009-09-18 17:07                                         ` Chris Webb
2009-09-20 18:46                                         ` Robert Hancock
2009-09-21  0:02                                           ` Kyle Moffett
2009-09-17 13:35                                     ` Mark Lord
2009-09-17 15:47                                       ` Tejun Heo
2009-08-31 12:21                             ` Mark Lord
2009-08-31 23:45                               ` Mark Lord
2009-09-01 13:07                                 ` Andrei Tanas
2009-09-01 13:15                                   ` Mark Lord
2009-09-01 13:30                                     ` Tejun Heo
2009-09-01 13:47                                       ` Ric Wheeler
2009-09-01 14:18                                         ` Andrei Tanas
2009-09-02 21:58                                   ` Allan Wind
2009-09-04 19:39                                     ` Andrei Tanas

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=20090917163647.GA6663@lifeintegrity.com \
    --to=allan_wind@lifeintegrity.com \
    --cc=linux-ide@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    /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).