#New Sidecar doesn't start

1 messages · Page 1 of 1 (latest)

prisma narwhal
#

Hello,
i have created to docker container, with similar compose files. The first one works fine, but the second has the attached logs.

I've followed the tutorial from https://af3556.github.io/Compute/Vaultwarden-❤️-Tailscale

Docker-compose:
services:
ts-opencloud:
image: tailscale/tailscale:latest
container_name: ts-opencloud
environment:
- TS_HOSTNAME=opencloud
- TS_AUTHKEY=tskey-client-...
- TS_STATE_DIR=/var/lib/tailscale
volumes:
- /media/piDocker/opencloud/tailscale/state:/var/lib/tailscale
- /media/piDocker/opencloud/tailscale/config:/config
devices:
- /dev/net/tun:/dev/net/tun
cap_add:
- net_admin
- sys_module
restart: unless-stopped

I'm working on a Pi5 and mounted the /media/piDocker
The compose files are handled over portainer.
When i try the recommended docker exec -it ts-opencloud tailscale --socket="/tmp/tailscaled.sock" i get following message:
Error response from daemon: Container 0c968e9371984a7028c7b1052c210e10bf0906d2defaafafd5afd3d6e324d311 is restarting, wait until the container is running

Does anyone have a clue, how i can fix the problem?

Odds & Sods

The following is one of the simpler1 and more-secure ways of setting up a self-hosted Vaultwarden server, using Tailscale for private (and public if desired) connectivity, all in Docker.

onyx locust
#

You have two different docker compose files with tailscale with different hostnames and Auth keys but one works and the other doesn't?

prisma narwhal
#

I have multiple compose files with tailscale and a defined TS_HOSTNAME and different Auth Keys. But this mentioned case is the only one that doesn't work.

onyx locust
#

How odd, I wonder if the permissions in the state directory are wrong/confused? Might be worth deleting the state directory, removing it from the console, and getting a new Auth key to start over?

prisma narwhal
#

What exactly do you mean by removing from the console?
This is the log from the system.
The permissions from the other docker folder are identical.

I don't know where this error message came from:
{"logtail":{"client_time":"2026-03-17T22:35:22.026929016Z","proc_id":1186706129,"proc_seq":34},"v":1,"text":"dnsfallback: SetCachePath error reading "/var/lib/tailscale/derpmap.cached.json": open /var/lib/tailscale/derpmap.cached.json: no such file or directory\n"}

onyx locust
#

If you log in to tailscale.com, you can delete devices from your tailnet so you can start over

prisma narwhal
#

There appears no device. Because there is no opencloud device created

prisma narwhal
#

And I've created several keys, with all rights and none worked

prisma narwhal
#

The problem was in front of the pc🫣

onyx locust
#

Sorry I missed this, glad you figured it out though. Docker isn't the easiest to get to grips with