We ran into an issue where a scheduled batch job ran out of band about 45 minutes earlier than it should which caused issues.
I looked at the logs but I wasn’t able to find anything - is there a special place where I could determine why something would run out of band? I saw that there was a server re-election around the same time, but I don’t know why that would have triggered this behavior - I am at a loss for troubleshooting as the journald logs for nomad only showed so much with the default logging.
My questions:
- Is there another place to find why this job was scheduled?
- Is it recommended to set a specific log level in order to see why this occurred?