On Mon, Nov 10, 2008 at 11:16, Junko IKEDA <[EMAIL PROTECTED]> wrote: >> > The latest Pacemaker 1.0 can help our problem which I posted to the >> > following entry. >> > http://developerbugs.linux-foundation.org/show_bug.cgi?id=1990 >> > >> > A split brain under 4 nodes circumstances can be recovered successfully! >> > It seems that these patches have the effect for this behavior. >> > http://hg.clusterlabs.org/pacemaker/stable-1.0/rev/fd4066e9888d >> > http://hg.clusterlabs.org/pacemaker/stable-1.0/rev/7508f211e536 >> > >> > Could it be possible to reflect these patches to Heartbeat 2.1.4 tree? >> >> No. The 2.1.4 tree is closed. >> Do we really need to keep having this conversation? > > No, I may sound like I'm repeating myself, but our customer plans to take > Heartbeat 2.1.4 soon, > so I have to ask it again. > sorry for bother you. > Do you believe that the above two patches have efficacy as recovering split > brain of joining some additional nodes?
It would prevent some extra (unnecessary and probably plain wrong) fencing operations, but wont have any impact on the underlying membership layer. > If so, it might be a good promotional point to recommend Pacemaker to the > customer. Its not really about Pacemaker - more the underlying cluster stack. If the cluster tells Pacemaker that the membership hasn't reformed, then there's not a lot it can do about it (other than possibly shoot the node again). If anything, it might be a reason to give OpenAIS a go... _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
