Return-Path: Received: from mail-qk0-f176.google.com ([209.85.220.176]:33135 "EHLO mail-qk0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751686AbbKNA7E (ORCPT ); Fri, 13 Nov 2015 19:59:04 -0500 Received: by qkas77 with SMTP id s77so65390162qka.0 for ; Fri, 13 Nov 2015 16:59:03 -0800 (PST) Date: Fri, 13 Nov 2015 19:58:59 -0500 From: Jeff Layton To: Olga Kornievskaia Cc: Christoph Hellwig , linux-nfs Subject: Re: lock reclaim failed! Message-ID: <20151113195859.6bd50799@tlielax.poochiereds.net> In-Reply-To: References: <20151113092254.GA5047@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Sender: linux-nfs-owner@vger.kernel.org List-ID: On Fri, 13 Nov 2015 10:47:32 -0500 Olga Kornievskaia wrote: > On Fri, Nov 13, 2015 at 4:22 AM, Christoph Hellwig wrote: > > I see the following running xfstestst on latest Linus' tree against > > as sever from the same tree: > > > > generic/089 [ 2000.358405] NFS: nfs4_reclaim_open_state: Lock reclaim failed! > > [ 2001.769939] NFS: nfs4_reclaim_open_state: Lock reclaim failed! > > [ 2001.770732] NFS: nfs4_reclaim_open_state: Lock reclaim failed! > > [ 2122.608136] NFS: nfs4_reclaim_open_state: Lock reclaim failed! > > Going against the NetApp server, we have seen this error logged when > the server returned BAD_STATEID to the IO on the delegated stateid. Is > there such error on the network trace? Yes, a capture is probably the best way to tell what's really going on. I just tried generic/test089 a couple of times on the fedora kernel here (on both server and client) and didn't see that pop: http://koji.fedoraproject.org/koji/buildinfo?buildID=699182 What's the top commit of your kernel? I assume that the server didn't restart or anything, right? -- Jeff Layton