Hello fellow selfhosters! I changed my server OS from fedora server to alma linux as I wanted a more stable base with less frequent updates for my docker containers, so I backed everything up and installed almalinux. It all went smoothly, and now all the containers seem to work as before except for nginx proxy manager: on fedora, I set all the proxy hosts like this (using jellyfin for instance): jellyfin.mydomain.duckdns.org > http://myserverhostname:8096 now, for some strange reasons, it does not work anymore, and I get a bad gateway error. I tried with and without SSL (that I had from the previous installation, do I need to regenerate them? but without it should work anyway…), pointing to localhost:8096, to myserverhostname:8096, to jellyfin:8096, but nothing. The only way I could make it work was with the resolved IP address of my server:

# host myserverhostname
myserverhostname.station has address 192.168.1.13

pointing nginx to 192.168.1.13:8096 correctly shows the website. why does this happens? am I doing something wrong?

EDIT: the only one that’s working with localhost is nginx itself (http://localhost:81)

======

SOLUTION: instead of myserverhostname, in nginx proxy host configuration (in the gui) I had to set myserverhostname.station to get it working. I don’t know why, in the previous installation only the hostname was enough…

  • tubbadu@lemmy.kde.socialOP
    link
    fedilink
    English
    arrow-up
    1
    ·
    11 months ago

    on the server host myserverhostname correctly resolves, but if I enter the container (docker exec -it nginx-app-1 bash) it does not work anymore:

    [root@docker-11e3869f946f:/app]# host tserver
    Host tserver not found: 3(NXDOMAIN)
    

    (I had to install dnsutils before)

    it seems a nginx issue then

    • Oisteink@feddit.nl
      link
      fedilink
      English
      arrow-up
      1
      ·
      11 months ago

      Could also be docker network-config. Docker should by default use the hosts resolver config if there’s nothing in /etc/resolve.conf

      You can also supply dns server on the docker command or in your compose file if you’re using compose.

      As a last resort you can enter server and ip i the container’s /ets/host file if the ip is static. But that’s gone once you rebuild the image.

      Or maybe there’s env on the container you use for dns

      • tubbadu@lemmy.kde.socialOP
        link
        fedilink
        English
        arrow-up
        1
        ·
        11 months ago

        I found a solution: use myserverhostname.station instead of just the hostname. I really have no idea why, on the previous installation it worked well with just the hostname… ahh, whatever.

        thank you very much for the help!