drill-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "abdelhakim deneche" <adene...@gmail.com>
Subject Re: Review Request 34829: DRILL-3190: Invalid FragmentState transition from CANCELLATION_REQUESTED in QueryManager
Date Fri, 29 May 2015 21:49:07 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/34829/#review85822
-----------------------------------------------------------



exec/java-exec/src/main/java/org/apache/drill/exec/work/foreman/FragmentData.java
<https://reviews.apache.org/r/34829/#comment137643>

    Foreman state management is already split between the Foreman and the QueryManager. This
will make the FragmentData handle part of that logic too!
    
    Shouldn't we move this to the QueryManager instead ? at least we would have one less class
to check involved


- abdelhakim deneche


On May 29, 2015, 8:44 p.m., Sudheesh Katkam wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/34829/
> -----------------------------------------------------------
> 
> (Updated May 29, 2015, 8:44 p.m.)
> 
> 
> Review request for drill, abdelhakim deneche, Chris Westin, and Jacques Nadeau.
> 
> 
> Bugs: DRILL-3190
>     https://issues.apache.org/jira/browse/DRILL-3190
> 
> 
> Repository: drill-git
> 
> 
> Description
> -------
> 
> DRILL-3190: Check for transition from CANCELLATION_REQUESTED to non-terminal state in
FragmentData
> 
> 
> Diffs
> -----
> 
>   exec/java-exec/src/main/java/org/apache/drill/exec/work/foreman/FragmentData.java ceb77f0

> 
> Diff: https://reviews.apache.org/r/34829/diff/
> 
> 
> Testing
> -------
> 
> Running unit and regresssion tests.
> 
> 
> Thanks,
> 
> Sudheesh Katkam
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message