So every now and then (once in a month or two) one of our processes, running a critical piece of code that we don't want to touch, is being killed by the Out Of Memory killer. It says so in
We are going to address the cause separately. I want to monitor and address the effects right now. I want us (the sysops) to be sent a notification whenever this happens, so that we can wait for an hour or so for the traffic on the cluster to die down a bit and restart the task that was killed.
How do I arrange for such a notification to happen?
I think we run Red Hat Linux on our boxes if it matters.