Zammad hangs once or twice a week

Technical assistance
1

Infos:

  • Used Zammad version: 6.4.1-1734476180.a187f3a5.bookworm
  • Used Zammad installation type: deb
  • Operating system: Debian 12 bookworm
  • Browser + version: any

Expected behavior:

  • Zammad runs

Actual behavior:

  • Zammad hangs and needs sudo systemctl restart zammad.service

Steps to reproduce the behavior:

  • Wait a week

How can I find out, what is happening?
Maybe elasticsearch 7.17.26 is eating my RAM?
Would upgrading to the latest versions help?

2

Update to the latest version is a good idea ever.

  • Do you use a monitoring system like CheckMK?
  • What says the Zammad Monitoring (Admin > Monitoring). Are there errors?
  • See the production.log and search for errors.
  • What your Hardware specs (CPU, RAM, HDD or SSD) and how many agents are online at the same time?
  • How many Sessions are online After some days (Admin > Sessions)? Maybe reduce it (exporation after 1 day or so)
  • Performance optimization implmemted or defaults?
1 Like
3

Thanks, I’ll check the logs. I didn’t know how huge they are :slight_smile:

I see ≈ 40 sessions, some of them pretty old and esp. sometimes multiple sessions of the same user. I reduced it now from four weeks to one week.

I can’t remember to apply any performance optimizations, so I guess, it’s all the defaults.

Will check the other things next week.

4

What you’re describing reeks of an unconfigured postgresql server.
The documentation will help you here:

2 Likes
5

Oh, max_connections was at only 100, indeed. Set to 2000 now. Thank you!

The other settings are for “bigger instances”, so I didn’t change those.

Let’s see, how behaviour will improve :slight_smile:

1 Like