Uploaded image for project: 'Flink'
  1. Flink
  2. FLINK-4911 Non-disruptive JobManager Failures via Reconciliation
  3. FLINK-5862

When JobManager fails, TaskManagers do not cancel tasks, but attempt to reconnect to the JobManager, and report states of tasks deployed in

    XMLWordPrintableJSON

Details

    • Sub-task
    • Status: Closed
    • Major
    • Resolution: Won't Fix
    • None
    • None
    • Runtime / Coordination
    • None

    Description

      When TaskManager detects failure of JobManager, it will not cancel the tasks, but attempt to reconnect to JobManager for a duration. If reconnecting successfully, TaskManager will report the states of Tasks deployed in it. If reconnecting is failed because timeout happened or some other reasons, TaskManager will cancel the tasks and mark the slots inactive.

      Attachments

        Issue Links

          is blocked by

          Sub-task - The sub-task of the issue FLINK-5703 ExecutionGraph recovery based on reconciliation with TaskManager reports

          • Major - Major loss of function.
          • Closed

          Sub-task - The sub-task of the issue FLINK-5861 TaskManager's components support updating JobManagerConnection

          • Major - Major loss of function.
          • Closed

          Activity

            People

              SleePy Biao Liu
              SleePy Biao Liu
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: