On a docker compose install, with a non default port, docker health check always fails

[Nifle-CGE]

Description

On a docker compose install with this config

services:
  glance:
    container_name: glance
    image: glanceapp/glance
    restart: always
    volumes:
      - ./config:/app/config
      - ./assets:/app/assets
      # Optionally, also mount docker socket if you want to use the docker containers widget
      - /var/run/docker.sock:/var/run/docker.sock:ro
      # To get storage stats
      - /mnt/pool:/mnt/pool:ro
    network_mode: host
    env_file: .env

with this in glance.yml

server:
  port: 60010
  assets-path: /app/assets

theme:
  # Note: assets are cached by the browser, changes to the CSS file
  # will not be reflected until the browser cache is cleared (Ctrl+F5)
  custom-css-file: /assets/user.css

  # Teal City
  background-color: 225 14 15
  primary-color: 157 47 65
  contrast-multiplier: 1.1

pages:
  # It's not necessary to create a new file for each page and include it, you can simply
  # put its contents here, though multiple pages are easier to manage when separated
  !include: home.yml

Docker health check always fails (see unhealthy) even though glance works perfectly

0e257e99b242    glanceapp/glance    "/app/glance    --confi…"    6 hours ago    Up 6 hours (unhealthy)

Edit: oops, didn't see there already was a pull request for this issue, see #591

[svilenmarkov]

I'm considering removing the health check since it's not really representative of anything meaningful. The only time it would show up as unhealthy is if the check is broken as is in this case, or if the server is actually down, in which case it would exit and the container would restart anyways. I don't see the point in having it, and those who really want it can add it in their docker-compose fairly easily.

[Nifle-CGE]

I agree for deleting the health check.