Uploaded image for project: 'Core Server'
  1. Core Server
  2. SERVER-61869

Race in DuplicateReturnAfterReachingTimestamp test

    • Fully Compatible
    • ALL
    • v5.2, v5.1
    • 46

      This test expects two threads to simultaneously update the tenant migration recipient state doc with the same value for the rejectReadsBeforeTimestamp field. This is the SERVER-60969 scenario, and the test checks that the server detected and logged it.

      If the first thread writes to the state doc and calls logOp() it will advance lastApplied, and the second thread will choose max(returnAfterReachingTimestamp, lastApplied) as the rejectReadsBeforeTimestamp. Thus it will update the state doc with a different rejectReadsBeforeTimestamp and its write will NOT be a no-op. Make logOp() hang to force the test scenario.

            Assignee:
            jesse@mongodb.com A. Jesse Jiryu Davis
            Reporter:
            jesse@mongodb.com A. Jesse Jiryu Davis
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

              Created:
              Updated:
              Resolved: