Don’t exactly know what the problem is, but it seems that healthcheck curl commands to https://example.com/ currently fail, probably because of a ssl failure. This causes container that rely on this to fail. Just wanted to let you guys know.

  • WhyJiffie@sh.itjust.works
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    18 hours ago

    the healthcheck URL should point to some HTTP API that the container makes available, so it should point to the container.

    in place of localhost should be the container’s name, and port should be the port the container exposes as the web server. some services, like Jellyfin, have a specific webpage path for this purpose: https://jellyfin.org/docs/general/post-install/networking/advanced/monitoring/

    and others, like gitea, hide the fact well that they have a health check endpoint, because its not mentioned in documentation: https://github.com/go-gitea/gitea/pull/18465

    but check if docker’s way of doing healthchecks produces a lot of spam in the system log, in which case you could choose to just disable health checking, because it would push out useful logs

    • Pete@feddit.orgOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      19 hours ago

      Thanks for explaining it to me in a simple manner, I will update the url accordingly!