Gateway 504 Time-Out when importing big microsoft exchange mailbox via graph

Technical assistance
1
  • Used Zammad version: 6.5.0
  • Used Zammad installation type: docker-compose
  • Operating system: Ubuntu Server
  • Browser + version: MSEdge Version 138.0.3351.65

Expected behavior:

Successfully import a mailbox via the Microsoft Graph API.
After logging in with credentials, the setup window (to configure groups and the email address) should appear.

Actual behavior:

after logging in with credentials, it loads for a while and then returns a 504 Gateway Time-out.
After that the mailbox is imported as inactive, and if i click on edit (under the incoming section), it loads again until the same error appears.

Steps to reproduce the behavior:

import a fairly large mailbox (around 30GB) via graph api (microsoft exchange)
After authentication it loads for a while and then fails with a timeout error

Could we be hitting a Microsoft Graph API limit or something similar?

image
image663×457 14.8 KB

thanks

2

You’re probably hitting a 504 Gateway Timeout because the import of a very large mailbox (e.g. ~30 GB) takes longer than the default timeout limits of your web server or application server.

What I would do:
Increase Nginx timeouts
Edit your Nginx config (e.g. /etc/nginx/nginx.conf or you site-specific file) to include:

proxy_connect_timeout 240;
proxy_send_timeout    240;
proxy_read_timeout    240;
send_timeout  240;

Then restart Nginx.

3

Ferching mailboxes does not affect thr web part. It‘s handled by another process.

If I would have to guess, I‘d say the instance needs tuning. Ensure you have enough database connections available.

1 Like