From: Andrew Morton <akpm@linux-foundation.org>
To: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: nfs@lists.sourceforge.net, linux-pm@lists.linux-foundation.org,
linux-kernel@vger.kernel.org, a.p.zijlstra@chello.nl
Subject: Re: A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?)
Date: Fri, 28 Sep 2007 13:10:12 -0700 [thread overview]
Message-ID: <20070928131012.4a03c53e.akpm@linux-foundation.org> (raw)
In-Reply-To: <1191009148.6702.46.camel@heimdal.trondhjem.org>
On Fri, 28 Sep 2007 15:52:28 -0400
Trond Myklebust <trond.myklebust@fys.uio.no> wrote:
> On Fri, 2007-09-28 at 12:26 -0700, Andrew Morton wrote:
> > On Fri, 28 Sep 2007 15:16:11 -0400 Trond Myklebust <trond.myklebust@fys.uio.no> wrote:
> > > Looking back, they were getting caught up in
> > > balance_dirty_pages_ratelimited() and friends. See the attached
> > > example...
> >
> > that one is nfs-on-loopback, which is a special case, isn't it?
>
> I'm not sure that the hang that is illustrated here is so special. It is
> an example of a bog-standard ext3 write, that ends up calling the NFS
> client, which is hanging. The fact that it happens to be hanging on the
> nfsd process is more or less irrelevant here: the same thing could
> happen to any other process in the case where we have an NFS server that
> is down.
hm, so ext3 got stuck in nfs via __alloc_pages direct reclaim?
We should be able to fix that by marking the backing device as
write-congested. That'll have small race windows, but it should be a 99.9%
fix?
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
NFS maillist - NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Morton <akpm@linux-foundation.org>
To: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: chakriin5@gmail.com, linux-pm@lists.linux-foundation.org,
linux-kernel@vger.kernel.org, nfs@lists.sourceforge.net,
a.p.zijlstra@chello.nl
Subject: Re: A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?)
Date: Fri, 28 Sep 2007 13:10:12 -0700 [thread overview]
Message-ID: <20070928131012.4a03c53e.akpm@linux-foundation.org> (raw)
In-Reply-To: <1191009148.6702.46.camel@heimdal.trondhjem.org>
On Fri, 28 Sep 2007 15:52:28 -0400
Trond Myklebust <trond.myklebust@fys.uio.no> wrote:
> On Fri, 2007-09-28 at 12:26 -0700, Andrew Morton wrote:
> > On Fri, 28 Sep 2007 15:16:11 -0400 Trond Myklebust <trond.myklebust@fys.uio.no> wrote:
> > > Looking back, they were getting caught up in
> > > balance_dirty_pages_ratelimited() and friends. See the attached
> > > example...
> >
> > that one is nfs-on-loopback, which is a special case, isn't it?
>
> I'm not sure that the hang that is illustrated here is so special. It is
> an example of a bog-standard ext3 write, that ends up calling the NFS
> client, which is hanging. The fact that it happens to be hanging on the
> nfsd process is more or less irrelevant here: the same thing could
> happen to any other process in the case where we have an NFS server that
> is down.
hm, so ext3 got stuck in nfs via __alloc_pages direct reclaim?
We should be able to fix that by marking the backing device as
write-congested. That'll have small race windows, but it should be a 99.9%
fix?
next prev parent reply other threads:[~2007-09-28 20:11 UTC|newest]
Thread overview: 106+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-28 6:32 A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?) Chakri n
2007-09-28 6:50 ` Andrew Morton
2007-09-28 6:59 ` Peter Zijlstra
2007-09-28 6:59 ` Peter Zijlstra
2007-09-28 8:27 ` Chakri n
2007-09-28 8:27 ` Chakri n
2007-09-28 8:27 ` Chakri n
2007-09-28 8:40 ` Peter Zijlstra
2007-09-28 8:40 ` Peter Zijlstra
2007-09-28 9:01 ` Chakri n
2007-09-28 9:01 ` Chakri n
2007-09-28 9:12 ` Peter Zijlstra
2007-09-28 9:20 ` Chakri n
2007-09-28 9:20 ` Chakri n
2007-09-28 9:20 ` Chakri n
2007-09-28 9:23 ` Peter Zijlstra
2007-09-28 9:23 ` Peter Zijlstra
2007-09-28 9:23 ` Peter Zijlstra
2007-09-28 10:36 ` Chakri n
2007-09-28 10:36 ` Chakri n
2007-09-28 10:36 ` Chakri n
2007-09-28 9:12 ` Peter Zijlstra
2007-09-28 9:01 ` Chakri n
2007-09-28 13:28 ` Jonathan Corbet
2007-09-28 13:28 ` Jonathan Corbet
2007-09-28 13:35 ` Peter Zijlstra
2007-09-28 13:35 ` Peter Zijlstra
2007-09-28 16:45 ` Alan Stern
2007-09-28 16:45 ` [linux-pm] " Alan Stern
2007-09-28 16:45 ` Alan Stern
2007-09-29 1:27 ` Daniel Phillips
2007-09-29 1:27 ` Daniel Phillips
2007-09-29 1:27 ` Daniel Phillips
2007-09-28 18:04 ` Andrew Morton
2007-09-28 18:04 ` Andrew Morton
2007-09-28 18:04 ` Andrew Morton
2007-09-28 13:28 ` Jonathan Corbet
2007-09-28 17:00 ` Trond Myklebust
2007-09-28 18:49 ` Andrew Morton
2007-09-28 18:49 ` Andrew Morton
2007-09-28 18:49 ` Andrew Morton
2007-09-28 18:48 ` Peter Zijlstra
2007-09-28 18:48 ` Peter Zijlstra
2007-09-28 19:16 ` Andrew Morton
2007-09-28 19:16 ` Andrew Morton
2007-09-28 19:16 ` Andrew Morton
2007-10-02 13:36 ` Peter Zijlstra
2007-10-02 15:42 ` Randy Dunlap
2007-10-03 9:28 ` [PATCH] lockstat: documentation Peter Zijlstra
2007-10-03 9:35 ` Ingo Molnar
2007-09-28 19:16 ` A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?) Trond Myklebust
2007-09-28 19:26 ` Andrew Morton
2007-09-28 19:26 ` Andrew Morton
2007-09-28 19:26 ` Andrew Morton
2007-09-28 19:52 ` Trond Myklebust
2007-09-28 19:52 ` Trond Myklebust
2007-09-28 19:52 ` Trond Myklebust
2007-09-28 20:10 ` Andrew Morton
2007-09-28 20:10 ` Andrew Morton [this message]
2007-09-28 20:10 ` Andrew Morton
2007-09-28 20:32 ` Trond Myklebust
2007-09-28 20:32 ` Trond Myklebust
2007-09-28 20:32 ` Trond Myklebust
2007-09-28 20:43 ` Andrew Morton
2007-09-28 20:43 ` Andrew Morton
2007-09-28 21:36 ` Chakri n
2007-09-28 23:33 ` Chakri n
2007-09-28 23:33 ` Chakri n
2007-09-28 23:33 ` Chakri n
2007-09-28 21:36 ` Chakri n
2007-09-28 20:24 ` Daniel Phillips
2007-09-28 20:24 ` Daniel Phillips
2007-09-29 1:51 ` KDB? Daniel Phillips
2007-09-29 1:51 ` KDB? Daniel Phillips
2007-09-29 1:51 ` KDB? Daniel Phillips
2007-09-28 19:16 ` A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?) Trond Myklebust
2007-09-28 17:00 ` Trond Myklebust
2007-09-29 0:46 ` Daniel Phillips
2007-09-29 0:46 ` Daniel Phillips
2007-09-29 0:46 ` Daniel Phillips
2007-09-28 6:50 ` Andrew Morton
2007-09-29 11:04 ` Fengguang Wu
2007-09-29 11:04 ` Fengguang Wu
2007-09-29 11:04 ` Fengguang Wu
2007-09-29 11:48 ` Peter Zijlstra
2007-09-29 12:28 ` Fengguang Wu
2007-09-29 12:28 ` Fengguang Wu
2007-09-29 12:28 ` Fengguang Wu
2007-09-29 14:43 ` Peter Zijlstra
2007-09-29 14:43 ` Peter Zijlstra
2007-09-29 11:48 ` Peter Zijlstra
2007-10-01 15:57 ` Chuck Ebbert
2007-10-01 15:57 ` Chuck Ebbert
2007-10-02 2:00 ` [PATCH] writeback: avoid possible balance_dirty_pages() lockup on a light-load bdi Fengguang Wu
2007-10-02 2:00 ` Fengguang Wu
2007-10-02 2:14 ` Andrew Morton
2007-10-02 2:14 ` Andrew Morton
2007-10-02 12:13 ` Fengguang Wu
2007-10-02 12:13 ` Fengguang Wu
2007-10-02 12:13 ` Fengguang Wu
2007-10-02 13:27 ` Fengguang Wu
2007-10-02 13:27 ` Fengguang Wu
2007-10-02 13:27 ` Fengguang Wu
2007-10-02 18:35 ` Chuck Ebbert
2007-10-02 18:35 ` Chuck Ebbert
2007-10-02 2:00 ` Fengguang Wu
2007-10-03 12:46 ` richard kennedy
2007-10-04 1:50 ` Fengguang Wu
2007-10-04 1:50 ` Fengguang Wu
2007-10-04 1:50 ` Fengguang Wu
2007-10-03 12:46 ` richard kennedy
-- strict thread matches above, loose matches on Subject: below --
2007-09-28 6:32 A unresponsive file system can hang all I/O in the system on linux-2.6.23-rc6 (dirty_thresh problem?) Chakri n
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=20070928131012.4a03c53e.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@lists.linux-foundation.org \
--cc=nfs@lists.sourceforge.net \
--cc=trond.myklebust@fys.uio.no \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.