From: Greg Banks Subject: Re: high latency NFS Date: Mon, 04 Aug 2008 19:25:54 +1000 Message-ID: <4896CB22.4000005@melbourne.sgi.com> References: <200807241311.31457.shuey@purdue.edu> <20080804003206.GB6119@disturbed> <20080804011158.GA8066@fieldses.org> <200808041118.19743.bs@q-leap.de> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Cc: "J. Bruce Fields" , Neil Brown , Michael Shuey , Shehjar Tikoo , linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org, rees@citi.umich.edu, aglo@citi.umich.edu To: Bernd Schubert Return-path: Received: from relay2.sgi.com ([192.48.171.30]:36754 "EHLO relay.sgi.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750998AbYHDJdI (ORCPT ); Mon, 4 Aug 2008 05:33:08 -0400 In-Reply-To: <200808041118.19743.bs-PKu+Ek1N2UGzQB+pC5nmwQ@public.gmane.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: Bernd Schubert wrote: > On Monday 04 August 2008 03:11:58 J. Bruce Fields wrote: > >> OK, so to summarize: when the rate of incoming rpc's is very high (and, >> I guess, when we're serving everything out of cache and don't have IO >> wait), all the nfsd threads will stay runable all the time. That keeps >> userspace processes from running (possibly for "minutes"). And that's a >> problem even on a server dedicated only to nfs, since it affects portmap >> and rpc.mountd. >> > > Even worse, it affects user space HA software such as heartbeat and everyone > with reasonable timeouts will see spurious 'failures'. > We're seeing that problem right now, even with the patch. -- Greg Banks, P.Engineer, SGI Australian Software Group. The cake is *not* a lie. I don't speak for SGI.