[
https://issues.apache.org/jira/browse/IMPALA-9775?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17114423#comment-17114423
]
Wenzhe Zhou edited comment on IMPALA-9775 at 5/25/20, 4:05 AM:
---------------------------------------------------------------
This issue was caused by
I[MPALA-9764|http://issues.apache.org/jira/browse/IMPALA-9764]
([https://gerrit.cloudera.org/#/c/15956/).|https://gerrit.cloudera.org/#/c/15956/]
The sleep interval between heartbeats was changed from 60 seconds to a value
deduced from the Hive configuration parameter "hive.txn.timeout". If there is
no setting for "hive.txn.timeout", the sleep interval will be set as 100
seconds by default. The test case TestAcid.test_acid_heartbeats will be failed
if there is no heartbeat detected in 200 seconds. To carry through this test
case, we should not allow the sleep time beyond 60 seconds, or set
hive.txn.timeout as 180 seconds. In our test environment, there is no
hive-site.xml so hive.txn.timeout is not set.
The Issue could be reproduced on local machine by running following command. It
may failed for different table format in each run.
./bin/impala-py.test
tests/query_test/test_acid.py::TestAcid::test_acid_heartbeats
--workload_exploration_strategy=functional-query:exhaustive
Verified that the issue would not happen by setting the maximum sleep interval
as 60 sec.
was (Author: wzhou):
This issue was caused by
[MPALA-9764|http://issues.apache.org/jira/browse/IMPALA-9764]
([https://gerrit.cloudera.org/#/c/15956/).|https://gerrit.cloudera.org/#/c/15956/]
The sleep interval between heartbeats was changed from 60 seconds to a value
deduced from the Hive configuration parameter "hive.txn.timeout". If there is
no setting for "hive.txn.timeout", the sleep interval will be set as 100
seconds by default. The test case TestAcid.test_acid_heartbeats will be failed
if there is no heartbeat detected in 200 seconds. To carry through this test
case, we should not allow the sleep time beyond 60 seconds, or set
hive.txn.timeout as 180 seconds. In our test environment, there is no
hive-site.xml so hive.txn.timeout is not set.
The Issue could be reproduced on local machine by running following command. It
may failed for different table format in each run.
./bin/impala-py.test
tests/query_test/test_acid.py::TestAcid::test_acid_heartbeats
--workload_exploration_strategy=functional-query:exhaustive
Verified that the issue would not happen by setting the maximum sleep interval
as 60 sec.
> Failure in TestAcid.test_acid_heartbeats
> ----------------------------------------
>
> Key: IMPALA-9775
> URL: https://issues.apache.org/jira/browse/IMPALA-9775
> Project: IMPALA
> Issue Type: Bug
> Components: Backend, Frontend
> Affects Versions: Impala 4.0
> Reporter: Wenzhe Zhou
> Assignee: Wenzhe Zhou
> Priority: Blocker
> Labels: broken-build
>
> Got following failures when running exhaustive pre-review test.
> *
> [query_test.test_queries.TestQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestQueries/]
> *
> [query_test.test_queries.TestHdfsQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestHdfsQueries/]
> *
> [query_test.test_queries.TestQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestQueries_2/]
> *
> [query_test.test_queries.TestHdfsQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestHdfsQueries_2/]
> * [query_test.test_acid.TestAcid.test_acid_heartbeats[protocol: beeswax |
> exec_option: \{'batch_size': 0, 'num_nodes': 0,
> 'disable_codegen_rows_threshold': 5000, 'disable_codegen': False,
> 'abort_on_error': 1, 'exec_single_node_rows_threshold': 0} | table_format:
> text/snap/block]|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test.test_acid/TestAcid/test_acid_heartbeats_protocol__beeswax___exec_option____batch_size___0___num_nodes___0___disable_codegen_rows_threshold___5000___disable_codegen___False___abort_on_error___1___exec_single_node_rows_threshold___0____table_format__text_snap_block_/]
> *
> [query_test.test_queries.TestQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestQueries_3/]
> *
> [query_test.test_queries.TestHdfsQueries|https://jenkins.impala.io/job/ubuntu-16.04-from-scratch/10724/testReport/junit/query_test/test_queries/TestHdfsQueries_3/]
> h3. Stacktrace
> query_test/test_queries.py:46: in add_test_dimensions
> extend_exec_option_dimension(cls, "exec_single_node_rows_threshold", "100") E
> NameError: global name 'extend_exec_option_dimension' is not defined
>
> query_test/test_acid.py:201: in test_acid_heartbeats
> assert success E assert False
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
