From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Anderson Subject: Re: [PATCH 1/9] blk: Do not abort requests if queue is stopped Date: Wed, 5 May 2010 11:18:29 -0700 Message-ID: <20100505181829.GA4053@linux.vnet.ibm.com> References: <1272944228-30511-1-git-send-email-andmike@linux.vnet.ibm.com> <1272944228-30511-2-git-send-email-andmike@linux.vnet.ibm.com> <20100504104537.GN27497@kernel.dk> <20100505045231.GA25395@linux.vnet.ibm.com> <4BE1B55B.9090007@cs.wisc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <4BE1B55B.9090007@cs.wisc.edu> Sender: linux-scsi-owner@vger.kernel.org To: Mike Christie Cc: Jens Axboe , linux-scsi@vger.kernel.org, James Bottomley , dm-devel@redhat.com List-Id: dm-devel.ids Mike Christie wrote: > All iscsi drivers should have blocking and timer reset now, so if a > transport problem casues a IO error to be sent to dm, then when > blk_abort_request is called that should prevent the scsi layer from > sending the whole host into recovery. > > > I do not remember the problem with the lack of eh blocking though. > Did we need that too? My comment was not on need I was just comparing the transports. -andmike -- Michael Anderson andmike@linux.vnet.ibm.com