Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-20483

Mesos Coarse mode may starve other Mesos frameworks if max cores is not a multiple of executor cores

    XMLWordPrintableJSON

    Details

    • Type: Bug
    • Status: Resolved
    • Priority: Minor
    • Resolution: Fixed
    • Affects Version/s: 2.0.2, 2.1.0, 2.1.1
    • Fix Version/s: 2.2.0
    • Component/s: Mesos
    • Labels:
      None

      Description

      if spark.cores.max = 10 for example and spark.executor.cores = 4, 2 executors will get launched thus totalCoresAcquired = 8. All future Mesos offers will not get tasks launched because sc.conf.getInt("spark.executor.cores", ...) + totalCoresAcquired <= maxCores will always evaluate to false. However, in handleMatchedOffers we check if totalCoresAcquired >= maxCores to determine if we should decline the offer "for a configurable amount of time to avoid starving other frameworks", and this will always evaluate to false in the above scenario. This leaves the framework in a state of limbo where it will never launch any new executors, but only decline offers for the Mesos default of 5 seconds, thus starving other frameworks of offers.

        Attachments

          Issue Links

            Activity

              People

              • Assignee:
                dgshep Davis Shepherd
                Reporter:
                dgshep Davis Shepherd
                Shepherd:
                DB Tsai
              • Votes:
                0 Vote for this issue
                Watchers:
                4 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: