Uploaded image for project: 'Flink'
  1. Flink
  2. FLINK-6664 Extend Restart Strategy for Fine-grained Recovery and missing Resources
  3. FLINK-6667

Pass a callback type to the RestartStrategy, rather than the full ExecutionGraph

    XMLWordPrintableJSON

Details

    • Sub-task
    • Status: Closed
    • Major
    • Resolution: Fixed
    • 1.3.0
    • 1.3.2, 1.4.0
    • Runtime / Coordination
    • None

    Description

      To let the ResourceStrategy work across multiple FailoverStrategy implementations, it needs to be passed a "callback" to call to trigger the restart of tasks/regions/etc.

      Such a "callback" would be a nice abstraction to use for global restarts as well, to not expose the full execution graph.

      Ideally, the callback is one-shot, so it cannot accidentally be used to call restart() multiple times.

      This would also make the testing of RestartStrategies much easier.

      Attachments

        Issue Links

          blocks

          Bug - A problem which impairs or prevents the functions of the product. FLINK-7216 ExecutionGraph can perform concurrent global restarts to scheduling

          • Blocker - Blocks development and/or testing work, production could not run
          • Closed
          links to

          Web Link GitHub Pull Request #4277

          Activity

            People

              zjureel Fang Yong
              sewen Stephan Ewen
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: