[
https://issues.apache.org/jira/browse/IMPALA-7990?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16724344#comment-16724344
]
ASF subversion and git services commented on IMPALA-7990:
---------------------------------------------------------
Commit 01f436979cc7ed26208ffb81a9218e090192d205 in impala's branch
refs/heads/master from Thomas Tauber-Marshall
[ https://git-wip-us.apache.org/repos/asf?p=impala.git;h=01f4369 ]
IMPALA-7990: make test_lifecycle_failures more permissive
test_lifecycle_failures runs a query with a debug action set to cause
fragment thread creation to fail 50% of the time and checks that the
query either succeeded or failed with an expected error.
Generally, the failure happens early enough that the query hasn't yet
reached FINISHED, but its also correct under Impala's query lifecycle
model for the query to reach FINISHED before the error is hit, in
which case it will be returned by a call to fetch().
Previously, the expected error in the test only handled the first
case. This patch modifies it to be more permissive in order to handle
both cases.
Testing:
- Ran test_lifecycle_failures in a loop 200 times without a failure.
Previously it repro-ed for me ~10% of the time.
Change-Id: I2da4ca95692b445e05796105d3b7d14e51fece88
Reviewed-on: http://gerrit.cloudera.org:8080/12103
Reviewed-by: Tim Armstrong <[email protected]>
Tested-by: Impala Public Jenkins <[email protected]>
> Failing assert in TestFailpoints .test_lifecycle_failures
> ---------------------------------------------------------
>
> Key: IMPALA-7990
> URL: https://issues.apache.org/jira/browse/IMPALA-7990
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 3.1.0
> Reporter: bharath v
> Assignee: Thomas Tauber-Marshall
> Priority: Blocker
> Labels: broken-build
>
> This test is hitting the assert intermittently and I'm not able to repro it
> locally even after looping the test for a while.
> Error Message
> {noformat}
> failure/test_failpoints.py:176: in test_lifecycle_failures assert 'Query
> aborted:Debug Action: FIS_FAIL_THREAD_CREATION:[email protected]' \ E
> AssertionError: ImpalaBeeswaxException: E INNER EXCEPTION: <class
> 'beeswaxd.ttypes.BeeswaxException'> E MESSAGE: Debug Action:
> FIS_FAIL_THREAD_CREATION:[email protected] E E assert 'Query aborted:Debug
> Action: FIS_FAIL_THREAD_CREATION:[email protected]' in "ImpalaBeeswaxException:\n
> INNER EXCEPTION: <class 'beeswaxd.ttypes.BeeswaxException'>\n MESSAGE: Debug
> Action: FIS_FAIL_THREAD_CREATION:[email protected]\n" E + where
> "ImpalaBeeswaxException:\n INNER EXCEPTION: <class
> 'beeswaxd.ttypes.BeeswaxException'>\n MESSAGE: Debug Action:
> FIS_FAIL_THREAD_CREATION:[email protected]\n" = str(ImpalaBeeswaxException())
> {noformat}
> Stacktrace
> {noformat}
> failure/test_failpoints.py:176: in test_lifecycle_failures
> assert 'Query aborted:Debug Action: FIS_FAIL_THREAD_CREATION:[email protected]' \
> E AssertionError: ImpalaBeeswaxException:
> E INNER EXCEPTION: <class 'beeswaxd.ttypes.BeeswaxException'>
> E MESSAGE: Debug Action: FIS_FAIL_THREAD_CREATION:[email protected]
> E
> E assert 'Query aborted:Debug Action: FIS_FAIL_THREAD_CREATION:[email protected]' in
> "ImpalaBeeswaxException:\n INNER EXCEPTION: <class
> 'beeswaxd.ttypes.BeeswaxException'>\n MESSAGE: Debug Action:
> FIS_FAIL_THREAD_CREATION:[email protected]\n"
> E + where "ImpalaBeeswaxException:\n INNER EXCEPTION: <class
> 'beeswaxd.ttypes.BeeswaxException'>\n MESSAGE: Debug Action:
> FIS_FAIL_THREAD_CREATION:[email protected]\n" = str(ImpalaBeeswaxException())
> {noformat}
> Standard Error
> {noformat}
> -- connecting to: localhost:21000
> -- connecting to localhost:21050 with impyla
> Conn <impala.hiveserver2.HiveServer2Connection object at 0x654a450>
> -- 2018-12-16 17:59:19,399 INFO MainThread: Closing active operation
> SET debug_action=FIS_IN_PREPARE:[email protected];
> -- executing against localhost:21000
> select * from tpch.lineitem limit 10000;
> -- 2018-12-16 17:59:22,779 INFO MainThread: Started query
> 5e4209c709db77df:4c9baa7800000000
> SET debug_action=FIS_IN_OPEN:[email protected];
> -- executing against localhost:21000
> select * from tpch.lineitem limit 10000;
> -- 2018-12-16 17:59:22,891 INFO MainThread: Started query
> b341dcc7d97f3d56:1855e49100000000
> SET debug_action=FIS_IN_EXEC_INTERNAL:[email protected];
> -- executing against localhost:21000
> select * from tpch.lineitem limit 10000;
> -- 2018-12-16 17:59:22,999 INFO MainThread: Started query
> f74b482fd8f7bb5f:3515880600000000
> SET debug_action=FIS_FAIL_THREAD_CREATION:[email protected];
> -- executing against localhost:21000
> select * from tpch.lineitem limit 10000;
> -- 2018-12-16 17:59:23,110 INFO MainThread: Started query
> fa4ae2eff60d340b:a074c6f600000000
> {noformat}
> Commit:
> {noformat}
> Checking out Revision b9377d3fdebb8f13fe4cc56ce16b666897b35822 (origin/master)
> 16:21:43 > git checkout -f b9377d3fdebb8f13fe4cc56ce16b666897b35822
> 16:21:46 Commit message: "IMPALA-7989: Revert "Remove Python 2.4 workarounds
> in start-impala-cluster.py""
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]