topic Re: Insight Manager disabled servers becoming re-enabled in Server Management - Systems Insight Manager

Insight Manager disabled servers becoming re-enabled

Peter G Brown — Mon, 10 Jan 2011 19:14:31 GMT

We've got a few servers (clusters) that are becoming excessively noisy and I've gone into SIM and disabled them. However, something is re-enabling them.

What could be doing that if I've disabled them indefinitely from the Insight console?

Also, is there any way I can turn off the Disk, CPU and System notifications for clusters, but still leave on the predictive hardware failure notifications from the physical nodes?

Thanks in advance!!!!

Re: Insight Manager disabled servers becoming re-enabled

Ralph Frampton — Mon, 10 Jan 2011 19:20:22 GMT

Have you set Disk and/or CPU thresholds? If so then go & set them back to below 5%, that disables traps related to threshold triggering.

Re: Insight Manager disabled servers becoming re-enabled

Peter G Brown — Mon, 10 Jan 2011 19:24:33 GMT

Yes, we disabled the Disk and CPU thresholds on each physical node.

We still will get a notification from a physical node ("Monitor Status Change - Node") from the physical node. Then 5 minutes later (default settings) we get a "System is reachable" email from the cluster itself.

Re: Insight Manager disabled servers becoming re-enabled

Ralph Frampton — Mon, 10 Jan 2011 19:50:07 GMT

Have you looked at the Cluster Admin directly on the server to see what is going on with the nodes/resources that might be triggering these events/traps?

Your last reply sounds more like something is changing state (node/resource fail-over?).

You could configure the event handler(s) to not send unreachable/reachable messages relating to that but then if/when you DO have a problem then you'd never receive an email about it (you should still be able to see it in HPSIM though).

Re: Insight Manager disabled servers becoming re-enabled

Ralph Frampton — Mon, 10 Jan 2011 20:00:15 GMT

I just saw something in another thread, do you have daily Automatic Discovery running? Do the traps/emails correspond to the time this runs?

We don't use it in our site (everything is added manually to HPSIM) so maybe that's a clue to your error source or something that you can adjust if you are so willing (ie. stop completely, exclude your cluster ip ranges etc.).