Details

    • Type: Improvement Improvement
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: 1.4.0
    • Fix Version/s: None
    • Component/s: tserver
    • Labels:

      Description

      After running randomwalk test for several days, tablet servers began to run out of memory. Heap dump analysis revealed a large number of HashMap entries. Inspection of those entries found they were referenced by zookeeper watches. A quick search finds this to be a known zookeeper issue: ZOOKEEPER-442

        Activity

        Mike Drob made changes -
        Description After running randomwalk test for several days, tablet servers began to run out of memory. Heap dump analysis revealed a large number of HashMap entries. Inspection of those entries found they were referenced by zookeeper watches. A quick search finds this to be a known zookeeper issue: https://issues.apache.org/jira/browse/ZOOKEEPER-442 After running randomwalk test for several days, tablet servers began to run out of memory. Heap dump analysis revealed a large number of HashMap entries. Inspection of those entries found they were referenced by zookeeper watches. A quick search finds this to be a known zookeeper issue: ZOOKEEPER-442
        Eric Newton made changes -
        Assignee Eric Newton [ ecn ]
        Gavin made changes -
        Workflow no-reopen-closed, patch-avail [ 12654206 ] patch-available, re-open possible [ 12671746 ]
        Eric Newton made changes -
        Summary tablet servers running out of memory cannot unwatch zookeeper nodes
        Keith Turner made changes -
        Field Original Value New Value
        Labels 14_qa_bug
        Eric Newton created issue -

          People

          • Assignee:
            Unassigned
            Reporter:
            Eric Newton
          • Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

            • Created:
              Updated:

              Development