From: Grant Coady Subject: Re: NFS regression? Odd delays and lockups accessing an NFS export. Date: Thu, 21 Aug 2008 09:17:07 +1000 Message-ID: References: <1219087258.7192.19.camel@localhost> Reply-To: Grant Coady Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Grant Coady , linux-kernel@vger.kernel.org, neilb@suse.de, bfields@fieldses.org, linux-nfs@vger.kernel.org To: Trond Myklebust Return-path: Received: from relay03.mail-hub.dodo.com.au ([123.2.6.238]:45347 "EHLO relay03.mail-hub.dodo.com.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751491AbYHTXRj (ORCPT ); Wed, 20 Aug 2008 19:17:39 -0400 In-Reply-To: <1219087258.7192.19.camel@localhost> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Mon, 18 Aug 2008 15:20:58 -0400, Trond Myklebust wrote: >On Mon, 2008-08-18 at 12:02 +1000, Grant Coady wrote: >> Hi there, >> >> I've been using NFS here for years, lately there's something odd going on >> since about a month or so. Previously reported last month: >> http://www.gossamer-threads.com/lists/linux/kernel/951419?page=last >> >> Now with 2.6.27-rc3 on one of the client boxes I get a complete stall >> at odd times when accessing the server's exported directory, cannot >> see a pattern to it. Eventually recovers after a Ctrl-C. Nothing in >> the server or client log files. Not easy to reproduce either. >> >> The server runs 2.6.26.2 at the moment. >> >> Server config, etc: http://bugsplatter.id.au/kernel/boxen/deltree/ >> Client config, etc: http://bugsplatter.id.au/kernel/boxen/pooh/ > >Please try to reproduce the hang, then do > > echo 0 >/proc/sys/sunrpc/rpc_debug > >and send the output from 'dmesg'... It's not NFS, I just had WinXP box stall completely (no KB NumLock response) on a PuTTY (ssh) terminal to a linux box running 2.6.26.3 for about five or ten seconds, when it came back to operation the PuTYY toolbar icon was flashing. Dunno which terminal as I had four open at the time, but they all to a server box running 2.6.26.3 with multiple Intel pro/100 (e100) NICs. I think this may be to do with the e100 network driver instead, as a few weeks ago I started compiling it in rather than as a module like I've been doing for years (in response to the firmware (microcode) being dropped on the floor last month issue). This box is the server (deltree) that all linux boxes mount a /home/common NFS export from. All machines have the Intel pro/100 NIC, going to recompile the lot back to modules, see if that fixes the problem. Grant.