I have just wondered about this as well. Not having confirmed this myself, I suspect that only the "real" master actually writes perfdata.
A possible approach might be to make the perfdata directory centrally available, e.g. via NFS. I also assume that this approach will only work in an active/passive scenario, where the secondary master will only become active when the first one fails. I got no idea for a "real" active-active scenario though. Regards, Markus Sent from my iPhone > On 30 Nov 2015, at 23:55, Madhusmita Gochhait <mgochh...@juniper.net> wrote: > > Hi , > > Currently I have a setup of a single instance of icinga2 and icingaweb2 > running on a single node . It has the perfdata enabled with pnp4nagios for UI > graphs . > > I want to setup HA for icinga2 and icingaweb2 with Active/Active cluster . > But not sure , if the perfdata will be synced between the nodes , so that > each UI displays same graphs . > Is there any way to achieve this . > > > Thanks, > mgochhait > > _______________________________________________ > > icinga-users mailing list > > icinga-users@lists.icinga.org > > https://lists.icinga.org/mailman/listinfo/icinga-users >
_______________________________________________ icinga-users mailing list icinga-users@lists.icinga.org https://lists.icinga.org/mailman/listinfo/icinga-users
