Details
-
Bug
-
Status: Resolved
-
Blocker
-
Resolution: Duplicate
-
Impala 2.1, Impala 2.2
-
None
Description
In very low mem situations the parquet-scanner will fail with the wrong error. That causes the mem_usage_scaling test to fail sometimes, example below:
FAILED TESTS ================================================ 1 tests failed. REGRESSION: query_test.test_mem_usage_scaling.TestTpchMemLimitError.test_low_mem_limit_q21[mem_limit: 100 | exec_option: {'disable_codegen': False, 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0, 'batch_size': 0, 'num_nodes': 0} | table_format: parquet/none] Error Message: test failure Stack Trace: query_test/test_mem_usage_scaling.py:155: in test_low_mem_limit_q21 self.low_memory_limit_test(vector, 'tpch-q21', self.MIN_MEM_FOR_TPCH['Q21']); query_test/test_mem_usage_scaling.py:119: in low_memory_limit_test assert TestTpchMemLimitError.EXPECTED_ERROR_MSG in str(e) E assert 'Memory limit exceeded' in 'ImpalaBeeswaxException:\n Query aborted:\nMetadata states that in group hdfs://localhost:20500/test-warehouse/tpch.or...parquet/a345e61c42eae84f-e2322ebea5a08dbe_467083517_data.0.parq(0) there are 318497 rows, but only 0 rows were read.\n' E + where 'Memory limit exceeded' = TestTpchMemLimitError.EXPECTED_ERROR_MSG E + and 'ImpalaBeeswaxException:\n Query aborted:\nMetadata states that in group hdfs://localhost:20500/test-warehouse/tpch.or...parquet/a345e61c42eae84f-e2322ebea5a08dbe_467083517_data.0.parq(0) there are 318497 rows, but only 0 rows were read.\n' = str(ImpalaBeeswaxException())
When run with ASAN enabled, it may cause impalad to crash.
Attachments
Issue Links
- duplicates
-
IMPALA-2327 Query completed without error but incorrectly returned no results
- Resolved
- relates to
-
IMPALA-2327 Query completed without error but incorrectly returned no results
- Resolved