Update healthcheck endpoint in docker-compose.yml from /api/v1/trending to /api/v1/stats for better reliability

[SolarisEclipse]

Description

The current healthcheck in docker-compose.yml for the invidious service uses /api/v1/trending (wget -nv --tries=1 --spider http://127.0.0.1:3000/api/v1/trending || exit 1), which can fail in certain setups (e.g., with Traefik as a reverse proxy), preventing the container from being exposed even when the app is otherwise functional. Switching to /api/v1/stats resolves this without issues.
This was discussed in a community chat (screenshot attached for reference), where maintainer @unixfox requested an issue as a reminder to make the change.

Steps to Reproduce

Deploy Invidious via docker-compose with Traefik labels enabled.
Observe Traefik logs showing healthcheck failures on /trending.
Remove or change the healthcheck to /stats; the setup works.

Proposed Solution

Update the healthcheck test in docker-compose.yml to:

  test: wget -nv --tries=1 --spider http://127.0.0.1:3000/api/v1/stats || exit 1
  interval: 30s
  timeout: 5s
  retries: 2

This endpoint is lighter and more consistent, per API docs.

Additional Context

/trending may fail due to load or optional param dependencies, while /stats is always quick.
No damage from the change, as noted in the discussion.
Attach screenshot of chat for full context.

Labels: enhancement, docker

Sources Used (with Dates):

Invidious API Documentation (accessed 2025, last updated ~2023): Confirmed /trending and /stats endpoints.
Invidious Installation Guide (accessed 2025, last updated ~2024): Verified current healthcheck config.
GitHub Issue #2982 (March 2022): Historical note on /trending param bugs, though not directly related.

[unixfox]

Thanks