This is an automated email from the ASF dual-hosted git repository. stigahuang pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/impala.git
commit 49bffd236ca6caa4d86f988c3f7290b7dca0b2ff Author: Zoltan Borok-Nagy <[email protected]> AuthorDate: Fri Sep 23 15:41:02 2022 +0200 IMPALA-11572: deflake test_mt_dop_skew_lpt part 2 I suspect the test is still flaky because of parallel execution of other tests that have high CPU usage, hence marking the test for serial execution to make it more stable. Also printing out the query profile to stdout when we observe a skew. Change-Id: I521685d683d51a52e54ad138f8466dd41c844f72 Reviewed-on: http://gerrit.cloudera.org:8080/19035 Reviewed-by: Impala Public Jenkins <[email protected]> Tested-by: Impala Public Jenkins <[email protected]> --- tests/query_test/test_scanners.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/query_test/test_scanners.py b/tests/query_test/test_scanners.py index f49ebbd53..29979929a 100644 --- a/tests/query_test/test_scanners.py +++ b/tests/query_test/test_scanners.py @@ -379,6 +379,7 @@ class TestHdfsScannerSkew(ImpalaTestSuite): v.get_value('table_format').compression_codec == 'none') @SkipIfLocal.multiple_impalad + @pytest.mark.execute_serially def test_mt_dop_skew_lpt(self, vector, unique_database): """IMPALA-11539: Sanity check for MT scan nodes to make sure that the intra-node skew is mitigated. For intra-node scan range assignment we are using dynamic @@ -410,6 +411,7 @@ class TestHdfsScannerSkew(ImpalaTestSuite): print "Intra-node bytes read ratio:", ratio if ratio < SKEW_THRESHOLD: count_skew += 1 + print "Profile of skewed execution: ", profile return count_skew tbl_name = unique_database + ".lineitem_skew"
