Fix a failing healthcheck blocking the deploy on Coolify

Error: Healthcheck failed / Container is unhealthy

Coolify's healthcheck never reports the new container healthy, so Traefik won't route to it and the deploy stalls or rolls back, surfacing a 404 / "No available server".

Coolify runs the healthcheck inside the container using curl or wget; the official docs state the container must have one of them installed. The check fails when (a) the base image ships neither curl nor wget (common on alpine/slim/distroless/scratch, and on multi-stage builds that install them only in the build stage), (b) the configured port or path doesn't match a route the app actually answers with a 2xx, or (c) the app binds to a different interface than the check targets. Until the check passes, Traefik won't send traffic and Coolify keeps the old container live or rolls back.

The fix

  1. Open Configuration > Healthchecks and confirm the Path and expected response code match a route your app actually answers with 2xx (e.g. path /, code 200). A wrong path returns a non-2xx status and the check fails.

  2. Make sure the container's runtime stage has curl or wget; the Coolify docs require one of them for healthchecks. Slim/alpine/distroless images often ship neither, and multi-stage builds drop them if they're only added in the build stage.

    # Debian/Ubuntu base
RUN apt-get update && apt-get install -y curl && rm -rf /var/lib/apt/lists/*
# Alpine base
RUN apk add --no-cache wget

  3. Add a lightweight health route if your app has no cheap 200 endpoint, then point the healthcheck Path at it.

    // e.g. Express
app.get('/health', (_req, res) => res.sendStatus(200))

  4. For Dockerfile / Docker Compose deploys, define the healthcheck in the Dockerfile or compose file; per the docs it takes precedence over the UI settings. wget is often more reliable than curl in minimal images.

    HEALTHCHECK --interval=10s --timeout=5s --retries=5 \
  CMD wget -qO- http://127.0.0.1:3000/health || exit 1

  5. If you just need it live now (a one-off job, or a container with no shell tools and no HTTP route), toggle the healthcheck off under Configuration > Healthchecks, redeploy, then re-enable once you've added curl/wget and a proper check.

Verify it worked

Coolify marks the container Healthy and the deploy completes/swaps in the new container; docker inspect reports health status healthy.

docker inspect --format '{{.State.Health.Status}}' <container-name>

Related Coolify errors

Sources

Configuration reference: Umami. Ports, environment variables, volumes and the release timeline, each row linked to its upstream source.

Still stuck? Send it to Emsden Studio and get a straight answer on what it is and what fixing it costs.

All Coolify deploy errors