From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21188 invoked by alias); 14 May 2012 16:29:55 -0000 Received: (qmail 21164 invoked by uid 9447); 14 May 2012 16:29:52 -0000 Date: Mon, 14 May 2012 16:29:00 -0000 Message-ID: <20120514162952.21162.qmail@sourceware.org> From: agk@sourceware.org To: lvm-devel@redhat.com, lvm2-cvs@sourceware.org Subject: LVM2/doc example.conf.in Mailing-List: contact lvm2-cvs-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: lvm2-cvs-owner@sourceware.org X-SW-Source: 2012-05/txt/msg00017.txt.bz2 CVSROOT: /cvs/lvm2 Module name: LVM2 Changes by: agk@sourceware.org 2012-05-14 16:29:51 Modified files: doc : example.conf.in Log message: Refer to details of snapshot of raid problem. Patches: http://sourceware.org/cgi-bin/cvsweb.cgi/LVM2/doc/example.conf.in.diff?cvsroot=lvm2&r1=1.49&r2=1.50 --- LVM2/doc/example.conf.in 2012/05/10 00:18:49 1.49 +++ LVM2/doc/example.conf.in 2012/05/14 16:29:50 1.50 @@ -430,8 +430,12 @@ # "mirror" - The original RAID1 implementation provided by LVM2/DM. It is # characterized by a flexible log solution (core, disk, mirrored) # and by the necessity to block I/O while reconfiguring in the - # event of a failure. Snapshots of this type of RAID1 can be - # problematic. + # event of a failure. + # + # There is an inherent race in the dmeventd failure handling + # logic with snapshots of devices using this type of RAID1 that + # in the worst case could cause a deadlock. + # Ref: https://bugzilla.redhat.com/show_bug.cgi?id=817130#c10 # # "raid1" - This implementation leverages MD's RAID1 personality through # device-mapper. It is characterized by a lack of log options.