Return-Path: Received: from mx1.redhat.com ([209.132.183.28]:33292 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751410Ab1GHGF3 (ORCPT ); Fri, 8 Jul 2011 02:05:29 -0400 Content-Type: text/plain; charset=us-ascii Message-ID: <19990.40467.487035.962976@regina.usersys.redhat.com> Date: Fri, 8 Jul 2011 16:05:07 +1000 From: Max Matveev To: Trond Myklebust Cc: Chuck Lever , linux-nfs@vger.kernel.org Subject: Re: NFS/TCP timeout sequence In-Reply-To: <1310048213.3863.37.camel@lade.trondhjem.org> References: <19989.27202.793003.725608@regina.usersys.redhat.com> <1310046439.3863.30.camel@lade.trondhjem.org> <5F749FAD-94B0-4D9D-84F6-F7D9662A1CF6@oracle.com> <1310048213.3863.37.camel@lade.trondhjem.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Thu, 07 Jul 2011 10:16:53 -0400, Trond Myklebust wrote: > Anyway, why shouldn't we back off if the server is failing to respond? Wasn't it the no-backoff/drop-connection approach what Mike Eisler was advocating back in '06 during Connectathon? http://www.connectathon.org/talks06/eisler.pdf I think this was the trigger to go from exponential backoff to linear. max