Add Health Check Example for Self-Hosted Directus in Docker

[podlomar]

Describe the Request

When self-hosting a Directus instance using Docker, it's common to have another container running on top of Directus as part of a larger application stack. This app container often needs to wait until Directus is fully up and ready before starting.

A working health check that I use looks like this:

healthcheck:
      test: ["CMD", "wget", "--spider", "-q", "http://directus:8055/server/health"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_interval: 5s
      start_period: 30s

I suggest adding this example to the Directus documentation to help others avoid the trial-and-error process of figuring out a proper health check setup.

Maintenance Strategy

Unlikely to require frequent maintenance as long as the health API endpoint stays the same and the directus container contains wget by default.

[linear[bot]]

GEN-1225 Add Health Check Example for Self-Hosted Directus in Docker

[podlomar]

I would like to be assgined to this issue. It's my first, so not sure if this comment is enough...

[linear[bot]]

EDU-375 Add Health Check Example for Self-Hosted Directus in Docker

[Nitwel]

@podlomar feel free to create a PR to the docs repo and thank you very much for wanting to improve our docs. ❤

[hichamelkaddioui]

Directus may report as unhealthy if, for instance, the EMAIL_CONNECTION is missing. In this case the response is a 503. In my case I only need to check whether the HTTP server is accepting requests to be considered healthy, hence the /server/ping endpoint suits me better.

services:
  directus:
    healthcheck:
      test:
        [
          "CMD-SHELL",
          "[ $$(wget -q -O - http://directus:8055/server/ping) = 'pong' ]",
        ]
      interval: 30s
      timeout: 5s
      retries: 5
      start_interval: 5s
      start_period: 30s