From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Anderson Subject: Re: restart multipathd automatically Date: Tue, 29 Jan 2008 11:39:06 -0800 Message-ID: <20080129193906.GB27870@linux.vnet.ibm.com> References: <200801211454.m0LEs2p15801@tug.org> <20080129100445.GD15191@percy.comedia.it> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20080129100445.GD15191@percy.comedia.it> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: dm-devel@redhat.com List-Id: dm-devel.ids Luca Berra wrote: > On Mon, Jan 21, 2008 at 03:54:02PM +0100, Stephen Mike wrote: >> Hi, >> >> To restart multipathd automatically after it dies, is it ok to put, in >> /etc/inittab, a line like this? >> # run multipathd at level 4 (restart) >> mphd:4:respawn:/etc/rc.d/init.d/multipathd start >> > no, it is not. To add a little (hopefully)... If multipathd is dieing it is better to determine why it is dieing. Running multipathd in foreground mode with verbosity may provide info. "/sbin/multipathd -d -v4" If the issue is multipathd causing a segfault you would need to build a non-strip version and then use ulimit -c unlimited to allow creation of a core file. You could then use gdb post getting a segfault to run "bt full" on the core file to hopefully determine the area of the problem. -andmike -- Michael Anderson andmike@linux.vnet.ibm.com