Hi Guys,
I've created a number of Monitor Scripts each with a 5 minute collection interval that essentially run a "ps | grep" within the target (Linux) OS to return a value of 0 or 1 depending on if a particular process is running.
This works ok 99.9% of the time, and I can observe availability metrics for this object, fail the process and see the valid alert etc.
However, I am seeing a small number of Alerts being generated every few days of type Object is not available that is then cancelled usually within <1 minute, and I don't see any other symptoms (VM and OS appear idle, EPOPS adapter is still showing 100% Resource Availability) etc.
Any ideas?
Thanks,
Dan