Traefik error 504

I followed the Full docker example with Traefik 2 and I’m getting a 504 timeout.

The Traefik status looks good for HTTP Routers and Services, certs assigned, no errors in the log (except the 504). The Vikunja logs look good for api-1, db-1, and frontend-1 (although there are no further log entries after the initial up). In the Vikunja docker-compose.yml, I’m using endpoints “websecure” and certResolver “letsencrypt”.

Using Traefik 2.10.7, docker compose 2.23.3. Traefik docker-compose.yml:

image: traefik:latest
- “–log.level=WARN”
- “–accesslog=true”
- “–api.dashboard=true”
- “–api.insecure=true”
- “–ping=true”
- “–ping.entrypoint=ping”
- “–”
- “–entryPoints.web.address=:80”
- “–entryPoints.websecure.address=:443”
- “–providers.docker=true”
- “–providers.docker.endpoint=unix:///var/run/docker.sock”
- “–providers.docker.exposedByDefault=false”
- “–certificatesresolvers.letsencrypt.acme.tlschallenge=true”
- “
- “–”
- “–global.checkNewVersion=true”
- “–global.sendAnonymousUsage=false”
- /var/run/docker.sock:/var/run/docker.sock
- /data/traefik/ssl:/etc/traefik/acme
- ./.env
- “80:80”
- “443:443”
test: [“CMD”, “wget”, “http://localhost:8082/ping","--spider”]
interval: 10s
timeout: 5s
retries: 3
start_period: 5s
- “traefik.enable=true”
- “traefik.http.routers.dashboard.rule=Host(”
- “traefik.http.routers.dashboard.service=api@internal”
- “traefik.http.routers.dashboard.entrypoints=websecure”
- “”
- “traefik.http.routers.dashboard.tls=true”
- “traefik.http.routers.dashboard.tls.certresolver=letsencrypt”
- “”
- “traefik.http.routers.dashboard.middlewares=authtraefik”
- “traefik.http.middlewares.authtraefik.basicauth.users=${BASICAUTH_USER}:${BASICAUTH_PASSWORD}”
- “traefik.http.routers.http-catchall.rule=HostRegexp({host:.+})”
- “traefik.http.routers.http-catchall.entrypoints=web”
- “traefik.http.routers.http-catchall.middlewares=redirect-to-https”
- “traefik.http.middlewares.redirect-to-https.redirectscheme.scheme=https”
restart: unless-stopped

Just to add: going to the site gives a 504 for vikunja-frontend@docker and going to /api/v1/info gives a 504 for vikunja-api@docker.

Are you able to use other services with a similar traefik config successfully?

I have BookStack up and running with that config.

Are the api and frontend in the same docker network as traefik?

Yes, they’re both in the web and default networks, as specified in the yml example. I created the web network manually in docker (‘docker network create web’), otherwise I get an error that it’s not found.

It looks like your traefik is not in the web network though?

You’re right:

  • I see vikunja-frontend-1 and vikunja-api-1 in the web network.
  • And vikunja-frontend-1, vikunja-api-1, and vikunja-db-1 in the vikunja_default network.
  • Traefik and BookStack are sharing their own network…

Hmm… Do you know off-hand how to get them on the same network? If not, I’ll do my homework and share the results here.

In the docker compose example you linked:

      - web
      - default

Add the network key to the docker compose file and it should work.

The network details above are from using those docker compose networks values. And they’re getting used, but traefik isn’t including itself in those networks for some reason.

Well it doesn’t do that because you didn’t specify it. Add the networks section from Vikunja’s docker compose to the traefik configuration and it should work.

Sorry, I must not be explaining myself well. This:

      - web
      - default

was already included in the docker-compose.yml file I used. I copied the Full Docker example with Traefik 2 verbatim (except, of course, the updated domain, entrypoint, and tls cert resolver).

As a result, Vikunja is in the web and default networks. But Traefik is not in those networks.

But the traefik docker-compose.yml you shared earlier does not contain that section. It won’t auto-join docker networks only because you specified them in the traefik config.

Ah, gotcha, I see what you mean now. Yeah, makes sense, let me try that.