On Fri, Jun 15, 2007 at 12:02:05AM +0200, Yan Fitterer wrote: > Personally, my preferred method is to make all resources unmanaged > (is_managed = false), then stop hb on all nodes, install upgrade, restart > heartbeat, leave to settle, then put resources back to managed state. > > I think it's possible to do a rolling upgrade, one node at a time, but I > prefer to down the cluster, so that the status section gets cleared. > > Yan
Yeah, I have a couple of tools: "update_managed.pl" which allows me to mark nodes unmanaged, and "scorched_earth.pl" which nukes all heartbeat's processes on all nodes before it has a chance to figure out what's going on, then wipes out all on-disk config and state, builds and installs the latest config from svn and brings everything back up unmanged. It fits nicely with my general policy of "everything can be rebuilt from version controlled config files" (except user data obviously, but that's pretty handily segregated on external drive units and the install scripts are capable of enumerating /sys/, finding which drive units are connected to which machines and rebuilding fstabs and mount points for them) Bron. _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
