[
https://issues.apache.org/jira/browse/IMPALA-12633?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17796967#comment-17796967
]
Michael Smith commented on IMPALA-12633:
----------------------------------------
I believe this happened because the original query failed and was retried
before it was set to inflight. The HS2 and Beeswax servers start execution,
then SetInflight. Execution runs asynchronously, and it appears possible for
execution to fail and retry before the SetInflight call (I can reproduce it
with a debug action).
I think I'll restore the previous behavior, even though I'm not convinced it
makes sense to skip cleanup if the query is not yet inflight.
> DCHECK failure on query_handle->is_inflight()
> ---------------------------------------------
>
> Key: IMPALA-12633
> URL: https://issues.apache.org/jira/browse/IMPALA-12633
> Project: IMPALA
> Issue Type: Bug
> Components: Backend
> Affects Versions: Impala 1.4.2
> Reporter: Andrew Sherman
> Assignee: Michael Smith
> Priority: Critical
> Attachments: 951ee391-85bd-4fdb-b43550a1-7564a485.dmp_dumped
>
>
> This is a new DCHECK introduced in IMPALA-12493
> {code}
> 12 impalad!google::LogMessageFatal::~LogMessageFatal() + 0x9
> rsp = 0x00007fd0d4e5d2b0 rip = 0x00000000038f8ad9
> Found by: stack scanning
> 13 impalad!impala::QueryDriver::RetryQueryFromThread(impala::Status
> const&, std::shared_ptr<impala::QueryDriver>) [query-driver.cc :
> 376 + 0xc]
> rsp = 0x00007fd0d4e5d2c0 rip = 0x0000000002295372
> Found by: stack scanning
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
