jslowik wrote: > Hey everyone, > > I have installed zenoss, and got some basic monitoring going (a few machines, > a few processes on each, etc). But now I'm starting to look to set up > specific alerts and I'm having trouble figuring out how to do so. Basically > the alert I'd like to set up is to alert if ANY process on a particular > machine goes above 50% cpu for X amount of minutes. I have looked at > thresholds for TOTAL cpu usage...but I don't know how I would go about > getting the CPU usage for a process without defining each and every process > individually. > > Any ideas or pointers to help me set this up? Thanks in advance.
Due to the dynamic nature of processes, this isn't an easy task. Why are you wanting to alert on individual processes? The process is eating a percentage of the total CPU, so why not just alert on total CPU use? Another approach would be to watch the CPU use of your system, find a normal range and then set a CPU threshold/alert accordingly. You could also setup an event command to SSH to the box, grab the output of top and parse the top x offenders. You can then build some logic in your script to restart the service(s) or email you a list of top CPU hogs. Otherwise, just log in and start troubleshooting from there. -------------------- m2f -------------------- Read this topic online here: http://community.zenoss.com/forums/viewtopic.php?p=24347#24347 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
