Details

    • Type: Bug
    • Status: Resolved
    • Priority: Blocker
    • Resolution: Fixed
    • Affects Version/s: Impala 2.8.0
    • Fix Version/s: Impala 2.8.0
    • Component/s: Backend
    • Labels:

      Description

      Failed build:
      http://sandbox.jenkins.cloudera.com/view/Impala/view/Evergreen-asf-master/job/impala-asf-master-core/393/

      From the BE test logs:

      [==========] Running 4 tests from 1 test case.
      [----------] Global test environment set-up.
      [----------] 4 tests from KuduScanNodeTest
      [ RUN      ] KuduScanNodeTest.TestScanNode
      [       OK ] KuduScanNodeTest.TestScanNode (1800 ms)
      [ RUN      ] KuduScanNodeTest.TestScanNullColValues
      [       OK ] KuduScanNodeTest.TestScanNullColValues (1037 ms)
      [ RUN      ] KuduScanNodeTest.TestScanEmptyString
      Wrote minidump to /data/jenkins/workspace/impala-umbrella-build-and-test/repos/Impala/logs/be_tests/minidumps/kudu-scan-node-test/59cf0ffe-3076-3690-3cba5646-0275528f.dmp
      Wrote minidump to /data/jenkins/workspace/impala-umbrella-build-and-test/repos/Impala/logs/be_tests/minidumps/kudu-scan-node-test/59cf0ffe-3076-3690-3cba5646-0275528f.dmp
      #
      # A fatal error has been detected by the Java Runtime Environment:
      #
      #  SIGSEGV (0xb) at pc=0x000000000124a9a8, pid=19751, tid=140208655132800
      #
      # JRE version: Java(TM) SE Runtime Environment (7.0_75-b13) (build 1.7.0_75-b13)
      # Java VM: Java HotSpot(TM) 64-Bit Server VM (24.75-b04 mixed mode linux-amd64 compressed oops)
      # Problematic frame:
      # C  [kudu-scan-node-test+0xe4a9a8]  std::vector<impala::TupleDescriptor*, std::allocator<impala::TupleDescriptor*> >::size() const+0xc
      #
      # Core dump written. Default location: /data/jenkins/workspace/impala-umbrella-build-and-test/repos/Impala/be/src/exec/core or core.19751
      #
      # An error report file with more information is saved as:
      # /data/jenkins/workspace/impala-umbrella-build-and-test/repos/Impala/be/src/exec/hs_err_pid19751.log
      #
      # If you would like to submit a bug report, please visit:
      #   http://bugreport.sun.com/bugreport/crash.jsp
      # The crash happened outside the Java Virtual Machine in native code.
      # See problematic frame for where to report the bug.
      #
      <end of output>
      Test time =   7.45 sec
      ----------------------------------------------------------
      Test Failed.
      "kudu-scan-node-test" end time: Sep 07 23:42 PDT
      "kudu-scan-node-test" time elapsed: 00:00:07
      

      Unfortunately, I could not dig up any more information since the build slave is gone. You should be able to find the core in the S3 archive.

        Activity

        Show
        tarmstrong Tim Armstrong added a comment - Also hit it here: http://sandbox.jenkins.cloudera.com/job/impala-external-gerrit-verify-merge-ASF/179/
        Hide
        mjacobs Matthew Jacobs added a comment -
        Show
        mjacobs Matthew Jacobs added a comment - waiting for a review: https://gerrit.cloudera.org/#/c/4337/
        Hide
        mjacobs Matthew Jacobs added a comment -

        commit 7b4a6fac1acf16fb84b38a005f69c228620ddb3f
        Author: Matthew Jacobs <mj@cloudera.com>
        Date: Thu Sep 8 13:19:47 2016 -0700

        IMPALA-4097: Crash in kudu-scan-node-test

        The kudu-scan-node-test was calling GetNext() with a NULL
        row batch, which isn't valid. This wasn't failing until a
        recent code change, and only occasionally due to the timing
        of scanner threads producing row batches. One batch is empty
        and the other has 1 row. This test worked when the empty row
        batch was added to the batch queue first (which usually
        happened), but the test code couldn't handle the other
        ordering properly. This fixes the test to be more careful
        about what is being exercised.

        Change-Id: I0e30964232ed137fce7a99bf7e9557293b5905c8
        Reviewed-on: http://gerrit.cloudera.org:8080/4337
        Reviewed-by: Dan Hecht <dhecht@cloudera.com>
        Tested-by: Internal Jenkins

        Show
        mjacobs Matthew Jacobs added a comment - commit 7b4a6fac1acf16fb84b38a005f69c228620ddb3f Author: Matthew Jacobs <mj@cloudera.com> Date: Thu Sep 8 13:19:47 2016 -0700 IMPALA-4097 : Crash in kudu-scan-node-test The kudu-scan-node-test was calling GetNext() with a NULL row batch, which isn't valid. This wasn't failing until a recent code change, and only occasionally due to the timing of scanner threads producing row batches. One batch is empty and the other has 1 row. This test worked when the empty row batch was added to the batch queue first (which usually happened), but the test code couldn't handle the other ordering properly. This fixes the test to be more careful about what is being exercised. Change-Id: I0e30964232ed137fce7a99bf7e9557293b5905c8 Reviewed-on: http://gerrit.cloudera.org:8080/4337 Reviewed-by: Dan Hecht <dhecht@cloudera.com> Tested-by: Internal Jenkins

          People

          • Assignee:
            mjacobs Matthew Jacobs
            Reporter:
            alex.behm Alexander Behm
          • Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development