NC-AIO + Tailscale + NGNIX = Talk_issues

Hello, everyone!

Few years ago I’ve discovered Nextcloud for myself. I love it so much, and using it later I’ve found AIO bundle. I started to use it with home HW and docker-compose stuf. And in some day it just got broken auto updating at night, and I decided to give a shot to install with NGNIX Proxy Manager, as far as now I need some other instances too on the same host.

So, now I have set it up in the following way - I have domain (and using subdomain for my NC), my own Tailnet with Headscale, VPS with static IP and NPM proxying with LE certs to my NC instance connected with Tailscale and dynamic IP from provider. And after some time I have found how to set up all the stuff to make it look like it works) Almost.

The main issue for now, is I’m experiencing problems with Talk. In it’s logs I’ve found that somehow it is trying to connect my NPM instance with primary network interface, not tailscale0. But my NPM is only allowed to proxy from my Tailscale addresses. Then I decided to recreate nextcloud-aio docker bridge network in order to attach it to tailscale0 interface and let NPM do it’s business, but somehow it doesn’t works.

So, now I’m trying to figure out how to make my Talk run all the signaling, STUN, TURN and all the other stuff it basically needs to let me have a calls inside my Tailnet. I hope some of you have a good idea. I can share with you all the logs and additional details by your request, and will appreciate any help. Thank you for your attention for my message. Cheers!