Operating system: CentOS Linux release 7.8.2003 (Core)
Browser + version: (FF, Chrome, not browser issue)
Expected behavior:
We expect Zammad to run for weeks on end, without having error in /api/v1/monitoring/health_check
(returning health: false)
Actual behavior:
Every second or third day we get health: false response from health_check script
with error :
scheduler may not run (last execution of OnlineNotification.cleanup about 5 hours over) - please contact your system administrator
Steps to reproduce the behavior:
We dont know, we just use the app day or two and it hangs.
We installed zammad-3.2.0-1575624531.d79c85cf.centos7.x86_64 on 12th of December 2019
and all was running fine until we installed update 3.4.0-1594825410.4cacfa4b.centos7.x86_64 on 17th of July 2020. Then the problem started. attachment plugin is lost with every update and we need to manually update it. But the worst is, that it stops pulling emails and needs to be restarted quite often.
Is the only log file production.log? because I cannot seem to find much there…
This is not a direct issue of Zammad, but you updating elasticsearch.
If that happens after Zammads update, you’ll have to manually install the plugin again as it becomes not supported after an upgrade. Usually our postinstall scripts to handle that for you.
Sounds like performance issues.
How is the load on your system? E.g. does “scheduler.rb” stick with 100% permanently?
Also please provide more information on the host the instance runs on.
Yes, production log usually should hold all information related.
You may check the scheduler* logs in the same directory as well.