You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues-all@impala.apache.org by "Michael Ho (JIRA)" <ji...@apache.org> on 2019/04/22 18:35:00 UTC

[jira] [Commented] (IMPALA-4631) plan-fragment-executor.cc:518] Check failed: other_time <= total_time (25999394 vs. 25999393)

    [ https://issues.apache.org/jira/browse/IMPALA-4631?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16823331#comment-16823331 ] 

Michael Ho commented on IMPALA-4631:
------------------------------------

The problem still occurs after loosening the DCHECK:

{noformat}
F0420 03:50:03.056339  2817 fragment-instance-state.cc:330] Check failed: other_time <= total_time + 3 (646003282 vs. 646003278)
{noformat}

Makes me wonder if there is a genuine bug here.

> plan-fragment-executor.cc:518] Check failed: other_time <= total_time (25999394 vs. 25999393)
> ---------------------------------------------------------------------------------------------
>
>                 Key: IMPALA-4631
>                 URL: https://issues.apache.org/jira/browse/IMPALA-4631
>             Project: IMPALA
>          Issue Type: Bug
>          Components: Backend
>    Affects Versions: Impala 2.8.0
>            Reporter: Dan Hecht
>            Assignee: Dan Hecht
>            Priority: Major
>              Labels: flaky
>
> This dcheck occasionally fires:
> {code}
> impalad.FATAL:F1201 22:35:58.617157 30293 plan-fragment-executor.cc:518] Check failed: other_time <= total_time (25999394 vs. 25999393)
> {code}
> I suspect the problem is with using floating point operations in places like this:
> {code}
>    timespec ts;
>     clock_gettime(OsInfo::fast_clock(), &ts);
>     return ts.tv_sec * 1e9 + ts.tv_nsec;
> {code}
> and because floating point doesn't distribute, and we can end up with {noformat} c * (a + b) < c * a + c * b {noformat} which is effectively what the dcheck does.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscribe@impala.apache.org
For additional commands, e-mail: issues-all-help@impala.apache.org