My events history contains something like 100,000 new events DAILY looking like "Maintenance window starting ... for ..." and "Maintenance window stopping ... for ...". About 2/3 of my alerting rules that have a window appear in the events history database many times per second, sometimes for hours at a stretch. I have to run a .SQL script every day to delete the events to keep the database from getting too big and slowing the whole system down intolerably.
I have a nearly identically configured Zenoss system (with much less RAM) running the same version that is monitoring the same set of devices with identically-configured device classes and event mappings. That system is not doing this at all. One significant difference between the systems is that the zenactions daemon in the misbehaving system is not configured with a valid SMTP config, so it is never successful at sending alert emails.
Both systems are running on CentOS Linux 5.2 and Zenoss 2.4.5 (sorry to impose such an old setup on this group). Any ideas on what's causing the events? I am just about to the point where I am going to delete and recreate the alerting rules and their windows because I can't think of anything else to try. That will take a couple hours and I am trying to avoid that.
Matt Woodling
