On (latest and olders) deb package running zammad behind nginx-reverse proxy-server i have an issue if i restart/backup the zammad-server: in the browser UI i have an error-message that does not go away and prevents zammad to recover when the backend is back online. as i backup at night, my computer is standby then, but on getting back to work, zammad does not recover and needs klicks to work again properly
, which is thrown by nginx, as the backend is really down for some minutes. but is working again, when i’m back at work.
how do i prevent zammad getting stuck only solvable by klicking manually? i want zammad to recover correctly and remove all warnings and get to work, if the backend is there again, so i don’t need to click around every morning…
not possible, as the nginx-reverse-proxy is standalone on another vm and servers lots of other backends. in my opinion, those errors should recover when the backend is back and remove those overlays…
This issue occurs if your Zammad WebApp tries to retreive a lot of ticket updates in a moment the instance is not responding. As the webserver is still available, it will throw error 504 (because it can’t reach the backend).
There are two problems with this:
The Zamamd WebApp is unable to interpret above HTML text and thus throws the modal you see.
If you have a lot of requests, you can actually see how the modals stack over each other. The application doesn’t notice there’s already another modal shown.
You can solve this issue via two ways:
reload the webapp (easy right :D)
click dozens times into the background (the background will get lighter with each click)
This is hacky, but currently I couldn’t find another way.
A solution would be to ensure your webserver responses in JSON for this particular error.
