Hello Sylvain, I will answer you inline, althought I must admit my responses might not solve your questions completely:
On Jan 17, 2008 12:26 PM, Sylvain MALAVIEILLE < [EMAIL PROTECTED]> wrote: > I'am testing heartbeat 2.1.2 (using crm) and I have some questions about > resources. > > My questions : > > 1) > I can see in traces (/var/log/ha-debug) lines like that : > > pengine[6084]: 2008/01/15_13:42:00 notice: StartRsc: anaisrh4-1 > Start resource_ip_alias_eth0 > others start resources ... > pengine[6084]: 2008/01/15_13:42:31 notice: StopRsc: anaisrh4-1 > Stop resource_ip_alias_eth0 > others stop resources ... > > My constatation is that start and stop orders are not respected in the > traces. > Is it normal ? > What does this lines mean ? > Is start / stop orders ar logged ? I cannot help you on this, since I don't know, and maybe I haven't quite correctly understood your question. Sorry. > > > 2) > "target_role" seams to be the future state to reach ; a dynamic variable. > In this case, why "target_role" appears into the CIB ? > Yes, "target_role" is the future state to reach. It is dynamic, but the CIB *IS* actually dynamic. In fact, node(s) score(s), pingd scores and such all go to the CIB, as far as I know. Thus, the CIB has a "fixed" part, which would be what is in the XML, and a dynamic part that is updated during runtime. > > 3) > Is there any way to link 2 resources in order to initiate a failover > only if the two resources have failed. > I think this might be accomplished having those resources on the same group. Probably you have to set a parameter or an attribute on the group meaning that upon failure of all resources, a failover is to occur, but not upon partial failure. Sorry I cannot give you more details on this, but I think this would be the approach. > 4) > On a resource failure on all nodes (the resource try to start on all the > cluster nodes before) is it possible to : > - stop native resources or a resources group > - start other resources or an other resource group > ? > The aim is to create fallback resources. > Again, I think I do not quite understand your question here :( Sorry. > > > Thanks, > > Sylvain > _______________________________________________ > Best Regards, Jonas Andradas. _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
