From mboxrd@z Thu Jan 1 00:00:00 1970 From: Brian Candler Subject: Fix for freeze on SATA hotplug Date: Wed, 19 Sep 2012 08:01:57 +0000 Message-ID: <20120919080157.GA34192@nsrc.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from b-pb-sasl-quonix.pobox.com ([208.72.237.35]:53117 "EHLO smtp.pobox.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751728Ab2ISIB7 (ORCPT ); Wed, 19 Sep 2012 04:01:59 -0400 Content-Disposition: inline Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: James.Bottomley@HansenPartnership.com Cc: linux-scsi@vger.kernel.org I had an issue where a failing drive which generated SATA hotplug events caused the whole I/O subsystem to deadlock. Jack Wang on the gluster-users mailing list identified the problem as a regression, and gave me a patch which has successfully fixed it: http://gluster.org/pipermail/gluster-users/2012-September/011355.html However, I see this patch is not in mainline kernel yet. Has any progress been made towards this, e.g. including it in subsystem maintainer's tree? Thanks, Brian. P.S. Original problem description at http://gluster.org/pipermail/gluster-users/2012-September/011354.html The issue is now also recorded at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1049013 since it exists in Ubuntu kernel 3.2.0-30