Logging for "Dead" Processes
Today we had an issue where a couple of our IRIS "processes" had a status of "Dead".
How do I configure IRIS so that, when that happens, IRIS will log an event to messages.log or alerts.log?
We are forwarding the contents of both of those logs to Splunk for analysis, and I'd like to be able to see those events in there as well.
Product version: IRIS 2020.1
$ZV: IRIS for UNIX (Red Hat Enterprise Linux for x86-64) 2020.1 (Build 217_1U) Tue May 26 2020 20:57:10 EDT
Is forwarding cconsole.log an option? There should be a 0-severity message there like:
Thanks for the reply, @Timothy Leavitt . I don't see that logfile on any of our systems. Is there anything special that needs to be done to enable that log?
Ah - this is actually messages.log on an IRIS instance. (It used to be called cconsole.log on older products.)
Do you see any messages like this in your messages.log?
Unfortunately I don't see that in messages.log. I've never really understood how (if it is even possible) to configure IRIS to lot more exceptions than it does by default, like failed jobs, etc.
As far as I recall this isn't logged at the point the status becomes "dead" only when cleaned up.
We once wrote a scheduled task that looks for and logs these processes using IsGhost() as they can sit there a while waiting to be cleaned up at times.
There is a mention in the IRIS documentation that says these are logged to the Event Log but I am not sure what happens if you aren't using Interoperability or Productions in IRIS, I suspect there is no logging then as it is part of Production Monitoring.
Matthew, in which log or utility are you seeing the "Dead" status? That can offer a pointer about how you could detect the dead jobs.
Do you know the cause of the dead jobs and what is the impact? The WRC can help sort that with you.
Under what circumstances will these two processes die? Can you tell me more about it?