ingard
Jan 25, 2016Nimbostratus
irule to log slow and/or tcp profile timed out requests to backend
Hi
I'm trying to identify which requests are causing our backends to stop responding/hang/fail. I was thinking some sort of timer and/or catch the tcp timeout rst maybe? I've been trying to figure out what event to attach this rule to and maybe it would be CLIENT_CLOSED but I'm struggling to see how to go about this. I don't want to log every request that ends with a client_closed obviously, but only the ones where there is no response. Any ideas out there?
kind regards ingard