From time to time in production environment (Windows server) some Java batches will exit with code 143. Before it happened yesterday, last time was 3 months ago. So it's pretty random, as most of the batches are daily processes.
Java batch is embedded in a .bat , in which a couple of variables are defined before a "call java ...." command. Batch is launched by Autosys scheduler, using a generic production account for batches.
From what I've read, JVM gets some kind of termination signal from outside, and exits abruptly. People who have solved the issue have found the root cause randomly : antivirus, user getting disconnected, external program killing the process...
Since I can't reproduce the issue, I'd like to find a way to track the root cause of the termination signal : is there a way to do this ? catching the signal from within my batch and log some useful info ? or maybe catch the exit code returned by the "call java" and do something useful to find the culprit ?
My primary suspect is the Autosys agent, but the Autosys team in my company won't investigate that issue unless I give them at least some kind of proof...
Thanks in advance for your help