From: Charles-Edouard Ruault Subject: Re: Re: Linux 2.4.25, nfs client hangs when talking to a MacOS nfs server. Date: Wed, 17 Mar 2004 20:28:52 +0100 Sender: nfs-admin@lists.sourceforge.net Message-ID: <4058A6F4.6070408@idtect.com> References: <117F2C39-7800-11D8-93C6-000A95CFFC9C@idtect.com> <1079542707.3047.12.camel@lade.trondhjem.org> <4825B7F5-783A-11D8-97E9-000A95CFFC9C@idtect.com> <1079546462.3769.249.camel@d158140025182.Cadence.COM> <1079547364.3047.27.camel@lade.trondhjem.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Cc: csc@cadence.com, nfs@lists.sourceforge.net Return-path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list2.sourceforge.net with esmtp (Exim 4.30) id 1B3gj2-0004de-1Z for nfs@lists.sourceforge.net; Wed, 17 Mar 2004 11:29:12 -0800 Received: from charlus.dyndns.org ([81.57.109.127]) by sc8-sf-mx1.sourceforge.net with esmtp (TLSv1:AES256-SHA:256) (Exim 4.30) id 1B3gj0-0002fq-LV for nfs@lists.sourceforge.net; Wed, 17 Mar 2004 11:29:10 -0800 To: Trond Myklebust In-Reply-To: <1079547364.3047.27.camel@lade.trondhjem.org> Errors-To: nfs-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: Discussion of NFS under Linux development, interoperability, and testing. List-Post: List-Help: List-Subscribe: , List-Archive: Trond Myklebust wrote: >P=E5 on , 17/03/2004 klokka 13:01, skreiv Chris Croswhite: > =20 > >>I have seen similar issues, but with slowlaris NFS server to linux.=20 >>Never was able to resolve the issue and so moved to linux NFS server. =20 >> >>I am willing to spend sometime on this issue if Trond can give me some >>assistance? I can detail the issue if you have you are available. >> =20 >> > >It's very simple: if you using UDP as the transport mechanism, then >packets can (and *will*) be lost. > =20 > yep, that's why i tried switching to TCP ... but with no better result. >This is particularly true if you are working across a net which mixes >network speeds, since the switches have to queue data when going from >the fast to the slower net: once this queue has built up to the point >where the switch runs out of memory, it will start dropping incoming >packets. > >IOW: this is not an NFS client bug, it is a network design bug. > > =20 > We'll see if this disappears after we've finished migrating the network=20 to 1Gbps >In these environments you *must* use TCP, since that has congestion >control capabilities baked into the protocol... > >For more detailed info on how to deal with this sort of problem, I >suggest the NFS FAQ and HOWTO: see http://nfs.sourceforge.net/ > > =20 > I'll definitely have a look and try again with TCP just in case i can=20 capture a different error log. Thanks again for the info. >Cheers, > Trond > =20 > ------------------------------------------------------- This SF.Net email is sponsored by: IBM Linux Tutorials Free Linux tutorial presented by Daniel Robbins, President and CEO of GenToo technologies. Learn everything from fundamentals to system administration.http://ads.osdn.com/?ad_id=1470&alloc_id=3638&op=click _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs