[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1219428818.27921.43.camel@localhost.localdomain>
Date: Fri, 22 Aug 2008 19:13:38 +0100
From: Ian Campbell <ijc@...lion.org.uk>
To: Trond Myklebust <trond.myklebust@....uio.no>
Cc: Grant Coady <gcoady.lk@...il.com>, linux-kernel@...r.kernel.org,
neilb@...e.de, bfields@...ldses.org, linux-nfs@...r.kernel.org,
Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
Jesse Brandeburg <jesse.brandeburg@...el.com>,
Bruce Allan <bruce.w.allan@...el.com>,
PJ Waskiewicz <peter.p.waskiewicz.jr@...el.com>,
John Ronciak <john.ronciak@...el.com>,
e1000-devel@...ts.sourceforge.net
Subject: Re: NFS regression? Odd delays and lockups accessing an NFS
export.
On Fri, 2008-08-22 at 11:08 -0700, Trond Myklebust wrote:
> On Fri, 2008-08-22 at 11:23 +0100, Ian Campbell wrote:
> > On Mon, 2008-08-18 at 15:20 -0400, Trond Myklebust wrote:
> > > Please try to reproduce the hang, then do
> > >
> > > echo 0 >/proc/sys/sunrpc/rpc_debug
> > >
> > > and send the output from 'dmesg'...
> >
> > I've also been seeing some NFS related lockups, although I'm not sure if
> > they are the same as the one in this thread or not. Client is 2.6.26
> > (Debian's kernel) and Server is 2.6.25 (also a Debian kernel, but from
> > backports.org).
> >
> > rpc_debug on the server gives nothing, on the client gives:
> > [144741.637997] -pid- proc flgs status -client- -prog- --rqstp- -timeout -rpcwait -action- ---ops--
> > [144741.637997] 3439 0004 0080 -11 f3f48200 100003 f7770000 0 xprt_sending fa0ae88e fa0bddf4
> > [144741.637997] 3438 0001 00a0 0 f77f2a00 100003 f77700d0 15000 xprt_pending fa0ae88e fa0bddf4
>
> That's probably also a networking device driver issue candidate: your
> RPC task is queued up waiting to be sent.
>
> What networking card+device driver are you using here?
# ethtool -i eth0
driver: e1000
version: 7.3.20-k2-NAPI
firmware-version: N/A
bus-info: 0000:01:0a.0
Adding CC's to peeps listed in MAINTAINERS
I have to reboot the system now (or else there will be no TV to watch
during my tea ;-). It will probably repro again quite soon though.
> > There are no processes running with pid 3439 3438 (I don't think it's
> > that sort of pid though).
>
> The 'pid' is an internal RPC cookie that just serves to identify and
> track specific RPC requests.
Right, thanks.
Ian.
--
Ian Campbell
Colors may fade.
Download attachment "signature.asc" of type "application/pgp-signature" (198 bytes)
Powered by blists - more mailing lists