[IMPALA-4604] Hung query for several hours during end-to-end testing on ubuntu1404 cluster

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Duplicate
• Affects Version/s: Impala 2.7.0
• Component/s: Backend
• Labels: crash
• Target Version: Impala 2.8.0
• Assignee: Sailesh Mukil
• Reporter: David Knupp
• Created: 06/Dec/16 23:33
• Updated: 07/Dec/16 01:56
• Resolved: 07/Dec/16 00:58

Description

From the impalad log, the last entry is:

09:51:25.582063 25017 coordinator.cc:1410] CancelFragmentInstances() query_id=db409a1bfe8222f4:6afdcdc400000000, tried to cancel 0 fragment instances

The time stamp corresponds to the last time stamp in the jenkins log where this test was being run:

09:51:25 [gw3] PASSED query_test/test_tpch_queries.py::TestTpchQuery::test_tpch[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-TPC-H: Q18]

From the Query Details page on the CM host, the actual query is:

select id, cnt
from functional_parquet.complextypestbl t,
  (select count(item) cnt from t.int_array) v
order by id
limit 10

Attaching impalads log, stack traces, and minidump.