Repeated "exited with code 1" after update

[OkWest6958]

Working as normal until some time in the last 24 hours. Jackettvpn docker-compose logs show the container starting correctly (web UI works fine). After around 10 seconds docker logs show the container exiting with code 1 and then restarting. This loops continuously

[Gitoffomalawn]

I'm experiencing this issue as well. You can prevent the constant reboot by adding -e RESTART_CONTAINER=no to the docker command. However, this only prevents the container from restarting. It will still fail setting up a network connection.

The reason for the constant restarts seems to be something that's preventing it from setting up a network connection. I keep seeing this error in the container logs just before the container restarts: [ERROR] Network is possibly down. I've tried pinging 8.8.8.8 from within the container, which works. Pinging google.com does not work.

I suspected it might be my .ovpn file, since that had a habit of breaking this container in the pas due to certificate rotation and server changes. I switched files, but the problem persists regardless of the new config file.

For now, I've found that the only way to run the current version of the container is by disabling the VPN functionality (-e VPN_ENABLED=no), which kinda defeats the purpose of it. I'll check back to see if it gets fixed, but if it doesn't I may have to ditch it for something else (sadly).

For future reference, It may be helpful to keep older versions of the container on Docker Hub as separate tags, in order to enable rollbacks for when issues like this arise.

EDIT: I tried to get it to work with the buster and dev tags. Both conclude that [ERROR] Network is possibly down. and exit the container. I don't understand why this is happening even with these 2 year old images. Nothing has changed in my setup other than the update to this container, and all my other containers work just fine.

[Gitoffomalawn]

Ok, some troubleshooting later, I seem to have found the problem: it was DNS (it's always DNS). I had this container with the -e NAME_SERVERS= option enabled, pointing to my internal DNS and my VPN DNS. Something in the new release (or maybe a new Docker version?) seems to have broken this option. Removing this option solves the problem for me. The container works normally now.

It's worth mentioning that I had the exact same problem with the arch-deluge container, which promped me to look at what they had in common. This eventually lead me down this path. Hope this helps you as well.