> > I prefer rolling upgrades; migrate service, stop cluster, update software > (reboot node if necessary), restart cluster. Repeat with next node until all > cluster nodes are updated. > > > Sounds complex and thus error prone to me. The problem seems to me > that e.g. in a release upgrade scenario the mysql binary must match its data > version for it to work correctly. > > The problem indeed arises when the upgraded service up-converts its data > so that a rollback to a previous version is not feasible. In that case, > taking a > snapshot before the upgrade is a good idea. > [>] Hi Lars,
my problem with the rolling upgrade is the drbd partition. If you migrate the service its data will move too. If you then restart the cluster and migrate back the data will not be in an upgraded state and thus not match the binary. Hence you can't be sure your service will start on the recently upgraded node with the old data version. So is rolling it really feasible? Stefan _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
