I've clearly misconfigured something...

[BurdAlex]

As the title says, I've clearly misconfigured something. I keep getting this set of lines right before docker closes it.

Edit: I forgot to say I'm using the Docker Compose setup. Was probably obvious, but just in case.

TransmissionVPN  | sd_bus_open_system: No such file or directory
TransmissionVPN  | 2026-09-25 12:15:35 WARNING: Failed running command (--up/--down): external program exited with error status: 1
TransmissionVPN  | 2026-09-25 12:15:35 Exiting due to fatal error

This is my setup:

version: 3.7.1
services:
    transmission-openvpn:
        container_name: TransmissionVPN
        restart: on-failure:2
        cap_add:
            - NET_ADMIN
        volumes:
            - ./protonvpn/:/etc/openvpn/custom/
            - ./:/config # where transmission-home is stored
            - /home/someone/Downloads/:/data # where transmission will store the data
        environment:
            - OPENVPN_PROVIDER=custom
            - OPENVPN_CONFIG=[REDACTED]
            - OPENVPN_USERNAME=[REDACTED]
            - OPENVPN_PASSWORD=[REDACTED]
            - LOCAL_NETWORK=192.168.0.0/24
        logging:
            driver: json-file
            options:
                max-size: 10m
        ports:
            - 9091:9091
        image: haugene/transmission-openvpn

Any and all help is appreciated. Also, I apologize if this has come up before. I couldn't find it.

[jmrplens]

It is not your compose file, it is two lines in the ProtonVPN .ovpn. ProtonVPN's configs come with

up /etc/openvpn/update-resolv-conf
down /etc/openvpn/update-resolv-conf

for desktop Linux, and those do not work in this container. In the image, resolvconf is a link to resolvectl, the client for systemd-resolved, which talks to it over the system D-Bus, and the container runs neither. When OpenVPN runs that script as its up step, it hands the DNS server Proton pushes to resolvconf, which fails with sd_bus_open_system: No such file or directory, and OpenVPN treats a failing up script as fatal and exits.

Delete those two lines (or put a # in front of them) in the .ovpn inside ./protonvpn and start the container again. You do not lose DNS by doing that: the container sets it itself, since its tunnelUp.sh, which OpenVPN runs through --route-up, writes the DNS servers the VPN pushes into /etc/resolv.conf.

One thing so it does not look like the fix failed: on the current image (OpenVPN 2.7) you will still see one sd_bus_open_system: No such file or directory line after the change. It comes from OpenVPN's own dns-updown hook, which runs when there is no up script, and it is harmless: the log says dns up command exited with status 0 and the tunnel comes up. If it bothers you, add dns-updown disable to the .ovpn.

I checked it on haugene/transmission-openvpn:latest (5.5.2): /usr/sbin/resolvconf points to ../bin/resolvectl, and running the image's /etc/openvpn/update-resolv-conf with a pushed DNS option and a tun interface present gives exactly your sd_bus_open_system: No such file or directory and exits with 1.

The same error with ProtonVPN is in #3016, where removing the two lines was confirmed to fix it. There is also an open PR that would strip them automatically, #3007; until it is merged and released, removing them by hand is the fix. The other way out is the built-in provider, OPENVPN_PROVIDER=PROTONVPN with one of its configs, whose files already have those lines commented out, at the cost of picking from its per-country configs instead of a specific P2P node.

[BurdAlex]

Thank you! That was EXACTLY the issue. Thank you!