I have an HA setup with Pacemaker using heartbeat stack. I created a script that monitors opensips via the pid file. Eventhough the start event succeed and then opensips crash due to a module. The next monitor round will report an error, and the resource would be migrated to the next node.
I don't think the start stop daemon utility used to bring up opensips would succeed, and then opensips would crash starting the modules. On Mon, Jun 6, 2011 at 10:44 AM, Toyima Dias <[email protected]> wrote: > Ok, but it any case, the script should return 1 if opensips failed to > initialize; i'm thinking to build a HA infraestructure with opensips > (heartbeat, cluster glue...) and this opensips init script behavior > seems not to be appropriate; i will have to take a deeper look at > the opensips.init script > > 2011/6/6 Andrew Pogrebennyk <[email protected]> > >> On 06.06.2011 14:15, Toyima Dias wrote: >> >>> If opensips has not started (ps -ef | grep opensips doesn't show any >>> opensips pid), how is possible that /etc/init.d/opensips returns [OK] >>> when you execute /etc/init.d/opensips restart...is this beacuse of >>> mysql? is that what you mean...this script should return 1 (not >>> succesfull, different than 0) it no pid file has been created, am i >>> right? >>> >> >> Correct, but there can be a case when opensips starts but then some >> module's initialization function fails due to mysql connect error or >> incompatible table version and the opensips will exit. >> >> -- >> Sincerely, >> Andrew Pogrebennyk >> >> >> _______________________________________________ >> Users mailing list >> [email protected] >> http://lists.opensips.org/cgi-bin/mailman/listinfo/users >> > > > _______________________________________________ > Users mailing list > [email protected] > http://lists.opensips.org/cgi-bin/mailman/listinfo/users > > -- Robert
_______________________________________________ Users mailing list [email protected] http://lists.opensips.org/cgi-bin/mailman/listinfo/users
