Apologies for the delay On Tue, Sep 22, 2009 at 3:13 AM, <renayama19661...@ybb.ne.jp> wrote: > Hi Andrew, > >> Ok, but I'm trying to understand why. > > Thank you. > >> What function are these daemons performing? Why do they need to keep >> running even when the node is in standby? > > For example, respawn can detect the abnormality of the node before becoming > online by detecting > trouble in standby.
Even with respawn, the node still has to become a full cluster member and complete resource probing before the respawned processes can have any effect. > In a sense it has the same effect to start pingd in respawn. But we don't really encourage that anymore either. Certainly no-one else running openais based clusters has seemed overly concerned about being forced to run it as a resource. You just have to write your rules so that the cluster produces the "correct" behavior when no value exists for your attribute(s). Or are you looking for the ability to do more than simply prevent resources from running? > In respawn which users made in imitation of pingd, what a function to detect > in this standby state is > included in will be expected. > > Best Regards, > Hideo Yamauchi. > _______________________________________________ Pacemaker mailing list Pacemaker@oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker
