Return-Path: Received: from bombadil.infradead.org ([198.137.202.9]:51874 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932903AbbD2POR (ORCPT ); Wed, 29 Apr 2015 11:14:17 -0400 Date: Wed, 29 Apr 2015 08:14:04 -0700 From: Christoph Hellwig To: Chuck Lever Cc: Christoph Hellwig , Trond Myklebust , Linux NFS Mailing List Subject: Re: [PATCH, RFC] backchannel overflows Message-ID: <20150429151404.GA12936@infradead.org> References: <20150428202157.GA23972@infradead.org> <1C0C92C2-FBCF-49D8-BB31-3C23A520B075@oracle.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1C0C92C2-FBCF-49D8-BB31-3C23A520B075@oracle.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, Apr 29, 2015 at 10:55:10AM -0400, Chuck Lever wrote: > > On Apr 28, 2015, at 4:21 PM, Christoph Hellwig wrote: > > > Currently the client will just crap out if a CB_NULL comes in at the > > same time as a slot controlled CB_COMPOUND that includes a CB_SEQUENCE. > > Under what circumstances does the server send a CB_NULL while a CB_COMPOUND > is in flight? When a client is under heavy loads from fsx or aio-stress, and we lose the connection (nfsd4_conn_lost is called) while doing heavy recalls. xfstests against a server offering pnfs layouts for which the client can't reach the storage devices is an easy reproducer.