From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757598Ab0DRUNw (ORCPT ); Sun, 18 Apr 2010 16:13:52 -0400 Received: from mail-out2.uio.no ([129.240.10.58]:43651 "EHLO mail-out2.uio.no" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752529Ab0DRUNu (ORCPT ); Sun, 18 Apr 2010 16:13:50 -0400 Subject: Re: 2.6.34rc4 NFS writeback regression (bisected): client often fails to delete things it just created From: Trond Myklebust To: Nix Cc: linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org In-Reply-To: <87mxx0a5pc.fsf@spindle.srvr.nix> References: <87tyr9dfvv.fsf@spindle.srvr.nix> <1271618484.8049.1.camel@localhost.localdomain> <87vdboa7e0.fsf@spindle.srvr.nix> <1271620750.8049.3.camel@localhost.localdomain> <87mxx0a5pc.fsf@spindle.srvr.nix> Content-Type: text/plain; charset="UTF-8" Date: Sun, 18 Apr 2010 16:13:44 -0400 Message-ID: <1271621624.8049.6.camel@localhost.localdomain> Mime-Version: 1.0 X-Mailer: Evolution 2.28.3 (2.28.3-1.fc12) Content-Transfer-Encoding: 7bit X-UiO-Ratelimit-Test: rcpts/h 3 msgs/h 1 sum rcpts/h 3 sum msgs/h 1 total rcpts 66 max rcpts/h 12 ratelimit 0 X-UiO-Spam-info: not spam, SpamAssassin (score=-5.0, required=5.0, autolearn=disabled, UIO_MAIL_IS_INTERNAL=-5, uiobl=NO, uiouri=NO) X-UiO-Scanned: A5E6533DA78B086518B3EE1AF4A8A14710F18EB8 X-UiO-SPAM-Test: remote_host: 68.40.206.115 spam_score: -49 maxlevel 80 minaction 2 bait 0 mail/h: 1 total 28 max/h 4 blacklist 0 greylist 0 ratelimit 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sun, 2010-04-18 at 21:03 +0100, Nix wrote: > On 18 Apr 2010, Trond Myklebust said: > > > On Sun, 2010-04-18 at 20:27 +0100, Nix wrote: > >> On 18 Apr 2010, Trond Myklebust verbalised: > >> tip-of-tree includes that commit, and it's still happening for me there. > >> (Just verified again.) > > > > Does a 'sync' clear out the sillyrenamed files? > > Yes. OK. Next question is how these files are being written. If you 'strace' the process, do you see ordinary write() calls being used, or is the file perhaps being written via mmap()?