From: Hannes Reinecke <hare@suse.de>
To: James Bottomley <James.Bottomley@SteelEye.com>
Cc: Laurent Riffard <laurent.riffard@free.fr>,
Andrew Morton <akpm@linux-foundation.org>,
linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org,
linux-scsi@vger.kernel.org
Subject: Re: 2.6.24-rc3-mm1: I/O error, system hangs
Date: Mon, 26 Nov 2007 08:54:26 +0100 [thread overview]
Message-ID: <20071126075426.GC11261@ochil.suse.de> (raw)
In-Reply-To: <1195926253.3195.16.camel@localhost.localdomain>
On Sat, Nov 24, 2007 at 07:44:13PM +0200, James Bottomley wrote:
> Probing intermittent failures in Domain Validation, even with the fixes
> applied leads me to the conclusion that there are further problems with
> this commit:
>
> commit fc5eb4facedbd6d7117905e775cee1975f894e79
> Author: Hannes Reinecke <hare@suse.de>
> Date: Tue Nov 6 09:23:40 2007 +0100
>
> [SCSI] Do not requeue requests if REQ_FAILFAST is set
>
> The essence of the problems is that you're causing REQ_FAILFAST to
> terminate commands with error on requeuing conditions, some of which are
> relatively common on most SCSI devices. While this may be the correct
> behaviour for multi-path, it's certainly wrong for the previously
> understood meaning of REQ_FAILFAST, which was don't retry on error,
> which is why domain validation and other applications use it to control
> error handling, but don't expect to get failures for a simple requeue
> are now spitting errors.
>
> I honestly can't see that, even for the multi-path case, returning an
> error when we're over queue depth is the correct thing to do (it may not
> matter to something like a symmetrix, but an array that has a non-zero
> cost associated with a path change, like a CPQ HSV or the AVT
> controllers, will show fairly large slow downs if you do this). Even if
> this is the desired behaviour (and I think that's a policy issue),
> DID_NO_CONNECT is almost certainly the wrong error to be sending back.
>
> This patch fixes up domain validation to work again correctly, however,
> I really think it's just a bandaid. Do you want to rethink the above
> commit?
>
Given the amounted error, yes, I'll have to.
But we still face the initial problem that requeued requests will be
stuck in the queue forever (ie until the timeout catches it), causing
failover to be painfully slow.
Anyway, I'll think it over.
Cheers,
Hannes
--
Dr. Hannes Reinecke zSeries & Storage
hare@suse.de +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N�rnberg
GF: Markus Rex, HRB 16746 (AG N�rnberg)
next prev parent reply other threads:[~2007-11-26 7:54 UTC|newest]
Thread overview: 128+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-11-21 4:45 2.6.24-rc3-mm1 Andrew Morton
2007-11-21 5:51 ` 2.6.24-rc3-mm1 Dave Young
2007-11-21 6:00 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-21 6:03 ` 2.6.24-rc3-mm1 Dave Young
2007-11-21 6:15 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-21 6:22 ` 2.6.24-rc3-mm1 Dave Young
2007-11-21 18:35 ` 2.6.24-rc3-mm1 Kirill A. Shutemov
2007-11-21 22:25 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-26 18:48 ` 2.6.24-rc3-mm1 Rik van Riel
2007-11-26 19:33 ` 2.6.24-rc3-mm1 Jiri Slaby
2007-11-21 5:56 ` 2.6.24-rc3-mm1 - Build Failure on S390x Kamalesh Babulal
2007-11-21 6:04 ` Andrew Morton
2007-11-21 5:58 ` 2.6.24-rc3-mm1 KAMEZAWA Hiroyuki
2007-11-21 6:08 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-21 12:49 ` 2.6.24-rc3-mm1 Rene Herman
2007-11-21 6:11 ` 2.6.24-rc3-mm1 - Kernel Panic on IO-APIC Kamalesh Babulal
2007-11-21 6:18 ` Andrew Morton
2007-11-21 9:22 ` Kamalesh Babulal
2007-11-21 9:29 ` Andrew Morton
2007-11-21 9:43 ` Kamalesh Babulal
2007-11-21 19:33 ` Torsten Kaiser
2007-11-22 10:04 ` Kirill A. Shutemov
2007-11-21 19:22 ` Len Brown
2007-11-21 19:48 ` Torsten Kaiser
2007-11-24 0:49 ` Alexey Dobriyan
2007-11-26 19:39 ` Rik van Riel
2007-11-26 20:33 ` Andrew Morton
2007-11-26 20:45 ` Ingo Molnar
2007-11-26 22:08 ` Jiri Slaby
2007-11-26 22:17 ` Andrew Morton
2007-11-26 22:22 ` Jiri Slaby
2007-11-26 23:14 ` Jiri Slaby
2007-11-26 23:28 ` Andrew Morton
2007-11-27 17:50 ` Rik van Riel
2007-11-26 20:54 ` Rik van Riel
2007-11-26 20:56 ` Christoph Lameter
2007-11-21 8:06 ` 2.6.24-rc3-mm1- powerpc link failure Kamalesh Babulal
2007-11-21 22:52 ` Stephen Rothwell
2007-11-21 8:24 ` 2.6.24-rc3-mm1 make headers_check fails Kamalesh Babulal
2007-11-21 0:32 ` Andrew Morton
2007-11-21 8:41 ` Kamalesh Babulal
2007-11-21 8:44 ` Avi Kivity
2007-11-21 8:52 ` Robert P. J. Day
2007-11-21 9:04 ` Andrew Morton
2007-11-21 9:06 ` Robert P. J. Day
2007-11-21 9:58 ` Sam Ravnborg
2007-11-21 10:00 ` Avi Kivity
2007-11-21 10:17 ` Avi Kivity
2007-11-21 10:31 ` Robert P. J. Day
2007-11-28 5:02 ` Andrew Morton
2007-12-02 8:56 ` Avi Kivity
2007-11-24 14:34 ` Adrian Bunk
2007-11-21 8:42 ` 2.6.24-rc3-mm1 (sync is slow ?) KAMEZAWA Hiroyuki
2007-11-21 8:49 ` Andrew Morton
2007-11-22 3:06 ` KAMEZAWA Hiroyuki
2007-11-24 12:04 ` kosaki
2007-11-24 18:04 ` Gabriel C
2007-11-26 7:06 ` KAMEZAWA Hiroyuki
2007-11-21 8:49 ` KAMEZAWA Hiroyuki
2007-11-21 18:23 ` 2.6.24-rc3-mm1: usb mouse doesn't work Kirill A. Shutemov
2007-11-21 22:22 ` Andrew Morton
2007-11-22 10:17 ` Kirill A. Shutemov
2007-11-22 17:07 ` [linux-usb-devel] " Alan Stern
2007-11-22 17:41 ` Marin Mitov
2007-11-23 2:51 ` Alan Stern
2007-11-23 5:19 ` Kirill A. Shutemov
2007-11-23 16:21 ` Alan Stern
2007-12-31 21:06 ` Alan Stern
2007-11-21 21:45 ` 2.6.24-rc3-mm1: I/O error, system hangs Laurent Riffard
2007-11-21 22:41 ` Andrew Morton
2007-11-23 7:29 ` Laurent Riffard
2007-11-23 7:51 ` Hannes Reinecke
2007-11-23 11:38 ` Hannes Reinecke
2007-11-23 17:52 ` Laurent Riffard
2007-11-24 6:42 ` James Bottomley
2007-11-24 12:57 ` Laurent Riffard
2007-11-24 13:26 ` James Bottomley
2007-11-24 17:54 ` Gabriel C
2007-11-24 18:04 ` James Bottomley
2007-11-24 18:08 ` Gabriel C
2007-11-24 18:28 ` Gabriel C
2007-11-24 22:59 ` Laurent Riffard
2007-11-25 7:37 ` James Bottomley
2007-11-25 20:39 ` Laurent Riffard
2007-11-28 21:38 ` Laurent Riffard
2007-11-24 17:44 ` James Bottomley
2007-11-26 7:54 ` Hannes Reinecke [this message]
2007-11-22 10:22 ` 2.6.24-rc3-mm1 Kirill A. Shutemov
2007-11-23 0:18 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-23 0:48 ` 2.6.24-rc3-mm1 Thomas Gleixner
2007-11-23 6:05 ` 2.6.24-rc3-mm1 Kirill A. Shutemov
2007-11-23 8:59 ` 2.6.24-rc3-mm1 Andreas Herrmann
2007-11-23 1:39 ` 2.6.24-rc3-mm1 Gabriel C
2007-11-23 4:12 ` 2.6.24-rc3-mm1 Andrew Morton
2007-11-23 5:55 ` 2.6.24-rc3-mm1 Gabriel C
2007-11-27 6:15 ` 2.6.24-rc3-mm1 Andrew Morton
2007-12-11 16:33 ` 2.6.24-rc3-mm1 James Bottomley
2007-12-12 10:08 ` 2.6.24-rc3-mm1 Boaz Harrosh
2007-12-12 11:03 ` [PATCH] REQ-flags to/from BIO-flags bugfix Boaz Harrosh
2007-12-12 15:18 ` Matthew Wilcox
2007-12-12 15:54 ` Matthew Wilcox
2007-12-13 5:36 ` David Chinner
2007-12-12 16:06 ` Boaz Harrosh
2007-12-12 16:33 ` Matthew Wilcox
2007-12-12 11:36 ` 2.6.24-rc3-mm1 Jens Axboe
2007-12-14 9:00 ` 2.6.24-rc3-mm1 Hannes Reinecke
2007-12-14 14:26 ` 2.6.24-rc3-mm1 James Bottomley
2008-01-07 14:05 ` Multipath failover handling (Was: Re: 2.6.24-rc3-mm1) Hannes Reinecke
2008-01-07 17:57 ` James Bottomley
2008-01-07 18:24 ` Mike Christie
2007-11-26 19:13 ` 2.6.24-rc3-mm1 Randy Dunlap
2007-11-26 19:34 ` 2.6.24-rc3-mm1 Christoph Lameter
2007-11-26 20:40 ` 2.6.24-rc3-mm1 Randy Dunlap
2007-11-26 20:56 ` 2.6.24-rc3-mm1 Christoph Lameter
2007-11-26 20:47 ` [PATCH -mm] x86 allnoconfig memory model Randy Dunlap
2007-11-26 21:00 ` Christoph Lameter
2007-11-26 21:17 ` Randy Dunlap
2007-11-26 21:20 ` Andrew Morton
2007-11-26 21:52 ` Christoph Lameter
2007-11-26 21:57 ` Andrew Morton
2007-11-26 23:19 ` Christoph Lameter
2007-11-27 7:16 ` 2.6.24-rc3-mm1 - brick my Dell Latitude D820 Valdis.Kletnieks
2007-11-27 7:27 ` Andrew Morton
2007-11-27 7:54 ` Valdis.Kletnieks
2007-11-27 8:17 ` Andrew Morton
2007-11-27 10:25 ` Ingo Molnar
2007-11-27 8:25 ` Dave Young
2007-11-27 8:46 ` Valdis.Kletnieks
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=20071126075426.GC11261@ochil.suse.de \
--to=hare@suse.de \
--cc=James.Bottomley@SteelEye.com \
--cc=akpm@linux-foundation.org \
--cc=laurent.riffard@free.fr \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@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