Replies: 1 comment 4 replies
|
Hi @Stephan-P, The One thing I noticed: in your description you mention both For further troubleshooting, you could try logging into the AliasVault container itself and running a curl against localhost:80 and localhost:443. This should confirm whether nginx is actually responding from inside the container. If the curl works internally, but 192.168.2.211:1080 / :1443 still gives connection refused, then the issue is likely with Docker/OMV port forwarding rather than AliasVault itself. |
4 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
My NAS is an up-to-date openmediavault 8 system that runs a fair number of docker container apps behind NPMplus (Nginx Proxy Manager Plus).
After failing to get AliasVault (AV) up and running based upon the documented guidelines, I discovered the docker-compose.all-in-one.nas.yml with modified ports upon which my compose is now based.
The Compose log suggests that all should be well, be it that there's no mention of ports 1080 and 1443 as specified in the compose file:
Attempts to open the admin web page via http://192.168.2.211:1080/admin or https://192.168.2.211:1443/admin fail: ERR_CONNECTION_REFUSED
In NPMplus, I've created a Proxy Host for https://aliasvault.mydomain.tld with TLS certificate, pointing towards 192.168.2.11, but whichever port I point it to, it always ends up with a 502 Bad Gateway, which I reccon is consistent with the failures mentioned earlier.
As far as I can tell, there are no docker ports conflicts. The container would not have been allowed to start, if there were.

Any ideas as to why AliasVault is not responding to http://192.168.2.211:1080/admin as it is supposed to?
All reactions