On Wed, Nov 23, 2011 at 8:31 AM, Andreas Kurz <[email protected]> wrote: > On 11/22/2011 01:27 AM, [email protected] wrote: >> Dear sirs: >> >> This is Yang. I set up 2 database server using heartbeat >> (heartbeat-2.1.3-3.el5.centos.rpm). They are running for over 40 days
/me falls over > > Please consider an update to Corosync & Pacemaker ... > >> very well. But suddently master Datebase server roll over to slave >> Database server. When I check the master server, everything is good. >> and I copy the ha-log as follows: >> ++++++++++++++++++++++++++++++++ >> heartbeat[16316]: 2011/11/21_10:46:40 info: killing >> /usr/local/bin/check_service process group 18673 with signal 15 >> heartbeat[16316]: 2011/11/21_10:46:40 info: killing >> /usr/lib/heartbeat/mgmtd -v process group 16383 with signal 15 >> mgmtd[16383]: 2011/11/21_10:46:40 info: mgmtd is shutting down >> heartbeat[16316]: 2011/11/21_10:46:40 info: killing >> /usr/lib/heartbeat/crmd process group 16382 with signal 15 >> crmd[16382]: 2011/11/21_10:46:40 info: crm_shutdown: Requesting shutdown >> crmd[16382]: 2011/11/21_10:46:40 info: do_state_transition: State >> transition S_IDLE -> S_POLICY_ENGINE [ input=I_SHUTDOWN cause=C_SHUTDOWN >> origin=crm_shutdown ] > > This looks like someone simply shutdown Heartbeat ... if this is the > reason you should find clear indication in logs. Yep, signal 15 is SIGTERM which means a normal shutdown. _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
