Hi Guys,
we have SCOM 2012 SP1 in production.
I'm receiving complaints from the customer that SCOM has failed to alert on an event. Example: though the agent service is running and the threshold is breached, logical disk monitor, service monitor (many others) wont generate alerts.
When i check the health explorer, the status of monitor will be "critical" but the time stamp will be very old, say months. i then i manually flush the agent cache and restart the agent service. it works then on.
we are monitoring nearly 600 agents. one or the other agent will be having this issue. it's getting hard for me to identify which agent is having this cache problem. i want to find out such agents, so that i can take some proactive action on it.
i faced this problem in SCOM 2007, 2012 & now in SP1.
has anyone came across this type of problem? Is there any solutoin to this issue? what's the best practice to overcome this issue?
Thanks in advance,
Venu
Best regards, Venu.