504 timeout on mailing delivery configuration

Use this forum to report bugs and to check for bugfixes and new releases of OpenEMM

Moderator: moderator

perun
Posts: 9
Joined: Wed Dec 04, 2019 12:39 pm

504 timeout on mailing delivery configuration

Post by perun »

Hi,

We have problems with timeouts on the mailing delivery. When we finish our configuration and confirm, it starts loading and times out after 60s (504 Gateway Timeout errors on the response). We have tried to adjust timeouts on Tomcat server.xml but it doesn't seem to take any effect.

When we come back to updated mailing, the configuration seems to be set up correctly so the backend appears to be working correctly, just the frontend breaks the connection.

Is it a known issue? Is there any way to adjust the timeout so the frontend will wait a little bit longer?

Thanks,

Matt
maschoff
Site Admin
Posts: 2596
Joined: Thu Aug 03, 2006 10:20 am
Location: Munich, Germany
Contact:

Re: 504 timeout on mailing delivery configuration

Post by maschoff »

No, there is no known issue. Maybe the ressources of your server are not sufficient for your OpenEMM instance? What is the load of the server?
OpenEMM Maintainer
perun
Posts: 9
Joined: Wed Dec 04, 2019 12:39 pm

Re: 504 timeout on mailing delivery configuration

Post by perun »

Hi,

The load average is as follows: 0.00, 0.01, 0.00

I am basically testing it now where there's literally nothing going on. We have a separate server just for OpenEMM.

I can see within access logs that the first POST request to /mailingsend.do;jsessionid= comes back almost immediately. We are still waiting for the second /mailingsend.do;jsessionid= on the frontend though and timeout after 1min. The request comes back after around 2mins (from the beginning) with 200 response code. I am not aware of anything that has changed before it started happening.

Could it be affected by number of entries we have within db? It keeps growing quickly (our recipient list is around 1mln users and we send emails to them regularly).

Is there anything we could look into to perhaps investigate it further?

In the meantime, is there any way to increase the timeout to 3mins?

Any help would be appreciated.

Thanks,

Matt
Post Reply