[
https://issues.apache.org/jira/browse/TEZ-2192?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14392898#comment-14392898
]
Hitesh Shah commented on TEZ-2192:
----------------------------------
[~zjffdu] The sample job was just mainly for manual testing. I logged the file
names and the file size ( not the md5 ) just to eyeball the info. In any case,
it was not something that I planned to commit to the codebase. Do either you or
[~sseth] see a need for it to be part of the tez-tests ( as compared to just
having the unit test )?
> Relocalization does not check for source
> ----------------------------------------
>
> Key: TEZ-2192
> URL: https://issues.apache.org/jira/browse/TEZ-2192
> Project: Apache Tez
> Issue Type: Bug
> Affects Versions: 0.6.0, 0.5.2
> Reporter: Rohini Palaniswamy
> Assignee: Hitesh Shah
> Priority: Blocker
> Attachments: TEZ-2192.1.patch, TEZ-2192.2.patch, TEZ-2192.3.patch,
> test-job-2192.patch
>
>
> PIG-4443 spills the input splits to disk if serialized split size is greater
> than some threshold. It faces issues with relocalization when more than one
> vertex has job.split file. If a job.split file is already there on container
> reuse, it is reused causing wrong data to be read.
> Either need a way to turn off relocalization or check the source+timestamp
> and redownload the file during relocalization.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
