Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-5559

Reconsidering the policy of ignoring the blacklist after reaching the threshold

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Open
    • Major
    • Resolution: Unresolved
    • 2.1.1-beta
    • None
    • None
    • None

    Description

      Nowadays, when MR AM find the number of blacklisted nodes reaches one threshold, the blacklist will be totally ignored. The newly assigned containers on the blacklisted nodes will be allocated. This may be not the best practice. We need to reconsider of it.

      Attachments

        Issue Links

          relates to

          Bug - A problem which impairs or prevents the functions of the product. MAPREDUCE-3339 Job is getting hanged indefinitely,if the child processes are killed on the NM. KILL_CONTAINER eventtype is continuosly sent to the containers that are not existing

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

          Bug - A problem which impairs or prevents the functions of the product. MAPREDUCE-5489 MR jobs hangs as it does not use the node-blacklisting feature in RM requests

          • Critical - Crashes, loss of data, severe memory leak.
          • Closed

          Activity

            People

              zjshen Zhijie Shen
              zjshen Zhijie Shen
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated: