crunch-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ryan Brush (JIRA)" <>
Subject [jira] [Commented] (CRUNCH-481) Support independent output committers for multiple outputs
Date Fri, 06 Feb 2015 20:46:35 GMT


Ryan Brush commented on CRUNCH-481:

Hey Tom,

You're right that the output committer is constructed with a task context using a different
Job instance. I handled this in the patch by "decorating" the job instance passed to the task
context as well.  (And it did indeed fail before I handled that additional Job instance, as
one would expect with your comment.) So I think the logic here is okay. In any case, the tests
for Kite, Crunch, and our local usage all pass. I can't say this is the most elegant code
I've ever written, but these hoops seem to be necessary to simulate multiple output formats
in a job.

Thanks for the clarification on the committer logic in Hadoop 1. That explains what I saw.

> Support independent output committers for multiple outputs
> ----------------------------------------------------------
>                 Key: CRUNCH-481
>                 URL:
>             Project: Crunch
>          Issue Type: Bug
>          Components: Core
>            Reporter: Aniket Kulkarni
>            Assignee: Josh Wills
>            Priority: Minor
>             Fix For: 0.12.0
>         Attachments: CRUNCH-481-hadoop-2-compat.patch, CRUNCH-481.patch, CRUNCH-481.patch,
CRUNCH-481.patch, CRUNCH-481c.patch
> I faced this issue while trying to write to Kite and HDFS in the same pipeline. A similar
issue was logged for Kite[1][2]. 
> I was attempting to write a PCollection to Kite and a different PTable to HDFS as a text
file. The write to Kite succeeded, however the write to HDFS only produced a _SUCCESS file
with no text file.
> Commenting out the write to Kite seems to solve the issue and I can see the text file
being written.
> [1] -
> [2] -

This message was sent by Atlassian JIRA

View raw message