The service stops after a few minutes of normal execution

Question

After migrating the Confluce database from Oracle11g to Oracle18c, the service has stopped after a few minutes of normal execution.  The oracle driver used is version 21 (ojdbc8.jar)

Configuration has been adjusted to increase heap memory to 4 GB, threads to a maximum of 200 and hibernate connections to 60. All plugins were disabled, except UPM. These actions were unsuccessful.

Nothing related to the service crash appears in the log.

Luis Vargas · Answer

Thanks for your answer. I found that they were the confluence scheduled tasks, I disabled them all via the web and now I'm looking for which one causes the service to crash.

Nic Brough -Adaptavist- · Answer

If a Confluence service just stops dead after a while, with no warning, then one of three things have happened: It has crashed but cannot write a crash log because the disk is full. It has crashed, but at a different layer, and written a report to a different log. Start with a check of /logs/atlassian-confluence.log, then move on to /log/Catalina.out. If there's no crash report in either of those, then it's going to be something in the operating system logs, Java should have reported something to them. An admin or process management system has chosen to kill the process stone dead, with the hardest of stops - no option to write anything, the process has to stop because they said so. (Kill -9 to people who know a bit of *nix)