[Spacewalk-list] Flood of messages: Possible Notification Meltdown!!!

Van Deman, Quint CTR US USJFCOM J7 quint.vandeman at att.jfcom.mil
Thu Jan 29 15:07:42 UTC 2009


Hello all--

Since enabling some of the monitoring features in Spacewalk v0.4, I've
been getting a flood of the following messages:

Subject: Possible Notification Meltdown!!!

ALERTS queue size is 1088 at Thu Jan 29 09:45:02 2009

<< Sent by /usr/bin/monitor-queue line 45 on host SpaceWalk.mydomain.com
(sat bf8270bc82e8, "RHN Monitoring Satellite").
   1 of these messages were filtered since Thu Jan 29 14:30:02 2009 GMT
>>

In reviewing the message contents I was able to eliminate several of
them by:
touch /var/log/notificiation/notifier.log
touch /var/log/notificiation/notif-escalator.log
touch /var/log/notificiation/generate_config.log
touch /var/log/notificiation/notif-launcher.log
chown nocpulse:nocpulse /var/log/notification/*.log

However, I can't eliminate those that contain the message: XXXX has
stale heartbeat file -- respawning, where XXXX might be any one of the
monitoring/notification components.

The box has properly sync'ed NTP time & a proper timezone setting, can
anyone shed any light on this issue?

The installation was a fresh install of 0.4...not an upgrade from 0.3 if
that is relevant.

Thanks,

Quint




More information about the Spacewalk-list mailing list