| [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] |
On Fri, 2003-08-01 at 03:54, Stanley Wang wrote: > From: Greg Freemyer <freemyer-ml@NorcrossGroup.com> > > To: opengfs-devel@lists.sourceforge.net <opengfs-devel@lists.sourceforge.net> > > Subject: Re: [ogfs-dev]Recovery Race conditions > > Date: 01 Aug 2003 01:35:42 -0400 > > > > Updated list after sig. > > > > Greg > > -- > > Greg Freemyer > > > > ====== > > > > > > Recovery is comprised of: > > > > Cluster Lock Recovery > > Journal Replay > > Granting of queued lock requests to waiting nodes. > > Abort and retry any ongoing mounts (if required) > > > > Any other major steps above? > Stonith? It's cluster manager's duty to fence the died node. We may not > care about it when using OpenDLM plus HA Heartbeat. Very good and important point. <I Added STONITH to my list> The current lock manager invokes STOMITH which supports multiple nodes. How do other dlm's do fencing? I have not used STONITH from heartbeat, but I'm almost positive heartbeat only invokes it for 2-node HA clusters, not for 3+ node non-HA clusters. I think stonith has a command-line invocation, so ogfs could invoke it directly if required, but it seems like fencing is something that needs to be handled prior to ogfs even knowing that a problem exists. Greg -- Greg Freemyer ------------------------------------------------------- This SF.Net email sponsored by: Free pre-built ASP.NET sites including Data Reports, E-commerce, Portals, and Forums are available now. Download today and enter to win an XBOX or Visual Studio .NET. http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01 _______________________________________________ Opengfs-devel mailing list Opengfs-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opengfs-devel