Domain no longer accessible on auto-deploy
I've got an express API server in dokploy, using a 'compose' service, and the docker-compose.yml has:
I have a domain configured which points to the service and port 3003.
When I make a new commit to the repo this auto-deploys, and then I can no longer connect to it via the domain.
HOWEVER, when I go back into domain, open it, save it (without any changes) and then redeploy, I can connect to the domain again.
Is there something I need to add to my docker-compose.yml that dokploy is doing when setting the domain?
1 Reply
For anyone who might stumble across this later, I fixed the issue by manually adding the appropriate labels in my docker-compose.yml:
labels:
- "traefik.enable=true"
- "traefik.docker.network=dokploy-network"
# HTTP → redirect to HTTPS
- "traefik.http.routers.fchq-api-web.entrypoints=web"
- "traefik.http.routers.fchq-api-web.rule=Host(
api.fchq.io
)"
- "traefik.http.routers.fchq-api-web.middlewares=redirect-to-https@file"
# HTTPS
- "traefik.http.routers.fchq-api-websecure.entrypoints=websecure"
- "traefik.http.routers.fchq-api-websecure.rule=Host(prodapi.fchq.io
)"
- "traefik.http.routers.fchq-api-websecure.tls.certresolver=letsencrypt"
# point Traefik at your internal port
- "traefik.http.services.fchq-api-web.loadbalancer.server.port=3003"
- "traefik.http.services.fchq-api-websecure.loadbalancer.server.port=3003"
Unfortunately this didn't turn out to be a long term solution, it's still bugging out. Is there something else I need to add to my config to ensure it can restart and auto-assign whatever black magic is happening in the background when I assign a domain?