Robinson, Eric wrote on 2011-09-29: > We have a 3-node cluster running about 200 instances of MySQL. The way > we have our resource groups set up, the dependency stack looks like this: > > > Cluster_IP > Filesystem > MySQL_001 > > MySQL_002 > > MySQL_003 > > MySQL_004 > ... > MySQL_100 > > > It seems that MySQL_004 is dependent on MySQL_003, which is dependent on > MySQL_002, etc. When the server is rebooted, the MySQL services all get > stopped in reverse order, then started again in order. If we remove a > MySQL resource from the middle of the group, all of the resource afters > that one get stopped and restarted.
This is indeed the behavior described in the documentation for groups. > > How can we rewrite the config such that the dependencies look more like > this.. > > > Cluster_IP > Filesystem > MySQL_001, MySQL_002, MySQL_003... MySQL_1000 > > > This way, all the MySQL services are dependent on the filesystem and > cluster IP, bit they do not depend on each other. I believe you can't using groups. To achieve what you want I would link one by one every MySQL resource to the FS, then link the FS to the Cluster IP. This require (many) more configuration lines but with a little $EDITOR magic it's going to be easy. Something like that maybe... colocation MySQL_001-with-fs inf: fs MySQL_001 colocation MySQL_002-with-fs inf: fs MySQL_002 [...] colocation MySQL_XXX-with-fs inf: fs MySQL_XXX colocation fs-with-Cluster_IP inf: Cluster_IP fs order MySQL_001-after-fs inf: fs MySQL_001 order MySQL_002-after-fs inf: fs MySQL_002 [...] order MySQL_XXX-with-fs inf: fs MySQL_XXX order fs-after-Cluster_IP inf: Cluster_IP fs Hope I'm wrong though... ;) Greetings, Florian
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
