Uploaded image for project: 'Infrastructure'
  1. Infrastructure
  2. INFRA-2312

Meaningful health check

    XMLWordPrintable

    Details

    • Similar Issues:

      Description

      I recall Olivier Vernin saying that the health check for the server was pretty basic: just that it was responding to general HTTP requests. We should consider instituting a more demanding health check that would verify that the system as a whole is really working and serving CI needs. For example, every hour, run a Pipeline build which (in parallel) tries to get a node block from each of the key labels—Linux/Docker VM, Windows VM, Linux ACI, Windows ACI—and run some hello-world sh / powershell step. If this build fails to complete within, say, 30m, then we can say the system is, if not broken, then certainly degraded or overloaded.

        Attachments

          Activity

          There are no comments yet on this issue.

            People

            Assignee:
            markewaite Mark Waite
            Reporter:
            jglick Jesse Glick
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Dates

              Created:
              Updated: