Desktop client sometimes fails to connect due to timeout

Hi community,

Nextcloud 13.0.1

In the past month, my users have had trouble connecting with the Windows desktop client regularly.
They get the message “No connection to server at [URL]. Time out”

The issue seems to be appearing sometimes for several hours, pretty much randomly to a dozen of users (out of ~60).
Precisions :
- When it occurs, they can still access the web client on chrome normally.

  • Server load is below 50%
  • Php.ini was configured to have a timeout of 300
  • Bandwidth usage is fine
  • User does not take any specific action for the error to come up, if might work for 2 hours, and suddenly, no be able to connect anymore.
  • Users access nextcloud through internal network

I tried the following :

  • reset dns cache
  • restart vm
  • restart apache
  • disconnect from desktop client, and reconnect
  • restart pc running the desktop client
  • reallocate priority of network cards
  • reinstall nextcloud

Logging:

  • The desktop client shows no log of the attempts.
  • The admin panel has no relevant errors to the situation (only random encryption errors)

Frankly at this point, I don’t even know where to look from extra information on debuging this. I peered through the apache, php, proxy logs and didn’t find anything relevant.

I am pretty confused by the lack of trace of the error in the logs of the desktop client.

Thanks for the help !

possibly you have a database issue.
check your nextcloud.log if there appears something like
.... 2006 MySQL server has gone away

I checked both in the admin interface and the nextcloud.log in data folder, and didn’t find a similar line. Maybe it is linked to this issue ? Timeout while trying to connect to Nextcloud