-
Hi, I'm trying to deploy multiple istances of NextCloud AiO and as the GitHub post says "The normal way is creating multiple VMs, installing AIO in reverse proxy mode in each of them and having one reverse proxy in front of them that points to each VM (you also need to use a different TALK_PORT for each of them). So, i'm using VMware to deploy multiple VM of NextCloud AiO, nc01.domain.com, nc02.domain.com, ecc...in reverse proxy mode and another VM for NGINX Proxy Manager. Everything works except Talk app, it keeps restarting the call when you are doing it, and in the settings it says there is a problem with STUN/TURN serverI have two theories:- in NGINX Proxy Manager, I used Streams on 1002 port (I change the default TALK_PORT to this one), so i'm forwarding requests that arrives on the proxy manager, that is hosting my domain nc01.domain.com, to the ip address of my nc01 istance, TCP and UDP...and maybe for some reason isn't working (?)- I'm not understanding what is the meaning of "multiple VM with one reverse proxy in front of them"...maybe I have to install NGINX Proxy Manager on the same machine that is hosting NextCloud? But if I do this, what is the purpose of the reverse proxy? Having it or not having it is the same thing, it is all in the same vm and also I have to reserve a public IP address for each nextcloud aio istance... Here is my NGINX Proxy Manager docker-compose.yml:
Here is my docker-run command:
Can you help me understand this? Thank you guys <3 |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 13 replies
-
Hi, in general things look correct but the Talk port is not allowed to be proxied. It must be directly forwarded to each VM that is why you need to choose a different talk port for each VM and open that in your router udp and tcp. You can check this also in the talk admin settings for each instance. Edit: apparently a port over 1024 was required. |
Beta Was this translation helpful? Give feedback.
Hi, in general things look correct but the Talk port is not allowed to be proxied. It must be directly forwarded to each VM that is why you need to choose a different talk port for each VM and open that in your router udp and tcp. You can check this also in the talk admin settings for each instance.
Edit: apparently a port over 1024 was required.