SPIKE: investigate improving CI reliability

ExportXMLWordPrintableJSON

    • Type: Task
    • Resolution: Done
    • Priority: Unknown
    • None
    • Affects Version/s: None
    • Component/s: None
    • None
    • None
    • PHP Drivers
    • None
    • None
    • None
    • None
    • None
    • None

      Scope
      Time-box investigating reliability improvements to GitHub Actions and Evergreen.

      Background
      Quoting TeamRetro on 2026-08-26:

      Evergreen seems to be doing a lot of random system failures, which is painful on top of the fact it also just takes longer to report status, etc.

      At the granularity of an Evergreen task, there is a retry_on_failure field. This may be too-coarse of a setting.

      A transient failure in a GitHub Action may cancel other GitHub Actions.

      See Google Doc PHPLIB-1938: CI failures for tracking known CI failures.

            Assignee:
            Kevin Albertson
            Reporter:
            Kevin Albertson
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated:
              Resolved: