Return-Path: Received: from fieldses.org ([173.255.197.46]:59901 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750944AbbCITp7 (ORCPT ); Mon, 9 Mar 2015 15:45:59 -0400 Date: Mon, 9 Mar 2015 15:45:59 -0400 From: "J. Bruce Fields" To: Christoph Hellwig Cc: Dave Chinner , Eric Sandeen , linux-nfs@vger.kernel.org, xfs@oss.sgi.com Subject: Re: panic on 4.20 server exporting xfs filesystem Message-ID: <20150309194559.GC2627@fieldses.org> References: <20150304222709.GI1627@fieldses.org> <20150304224557.GY4251@dastard> <54F78BE5.1020608@sandeen.net> <20150304225623.GZ4251@dastard> <20150305040849.GJ1627@fieldses.org> <20150305131731.GA16235@lst.de> <20150305150138.GA15674@fieldses.org> <20150305170217.GC15674@fieldses.org> <20150305204749.GA17934@fieldses.org> <20150308153056.GA24037@lst.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20150308153056.GA24037@lst.de> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Sun, Mar 08, 2015 at 04:30:56PM +0100, Christoph Hellwig wrote: > On Thu, Mar 05, 2015 at 03:47:49PM -0500, J. Bruce Fields wrote: > > nfsd: client 192.168.122.32 failed to respond to layout recall. Fencing.. > > nfsd: fence failed for client 192.168.122.32: -2! > > nfsd: client 192.168.122.32 failed to respond to layout recall. Fencing.. > > nfsd: fence failed for client 192.168.122.32: -2! > > There is no userspace elper to do the fencing, so unfortunately this > is expecvted. > > > receive_cb_reply: Got unrecognized reply: calldir 0x1 xpt_bc_xprt ffff88005639a000 xid c21abd62 > > Now this looks like some issue with the low-level callback path. I've never > seen tis before, but from looking at receive_cb_reply this happens if > xprt_lookup_rqst can't find a rpc_rqst structured for the xid. Looks like > we might be corrupting the request list / xid allocation somewhere? > > I can prepare a patch for you to aid with xid tracing if you want. I'll take a look when I get back. But before that I'd like to understand why the layout seems to be left here blocking writes forever, instead of getting cleaned up after a lease period with no layout return. --b.