[
https://issues.apache.org/jira/browse/YARN-7876?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16350762#comment-16350762
]
Miklos Szegedi edited comment on YARN-7876 at 2/2/18 6:19 PM:
--------------------------------------------------------------
Thank you [~grepas] for the review and [~jlowe] for updating the title. I
refined the title a little bit since what happens is that the localized and
extracted files should be there. The patch does not change that. What might be
truncated is the jar that is left around for compatibility reasons. If the job
tries to extract this with zip for example, it could run into an issue.
was (Author: miklos.szeg...@cloudera.com):
Thank you [~grepas] for the review and [~jlowe] for updating the title. I
refined the title a little bit since what happens is that the localized and
extracted files should be there. The patch does not change that. What might be
truncated is the jar that is left around for compatibility reasons. If the job
tries to extract this with zip for example, it might run into an issue.
> Localized jars that are expanded after localization are not fully copied
> ------------------------------------------------------------------------
>
> Key: YARN-7876
> URL: https://issues.apache.org/jira/browse/YARN-7876
> Project: Hadoop YARN
> Issue Type: Bug
> Affects Versions: 3.1.0
> Reporter: Miklos Szegedi
> Assignee: Miklos Szegedi
> Priority: Blocker
> Attachments: YARN-7876.000.patch, YARN-7876.001.patch
>
>
> YARN-2185 added the ability to localize jar files as a stream instead of
> copying to local disk and then extracting. ZipInputStream does not need the
> end of the file. Let's read it out. This helps with an additional
> TeeInputStream on the input.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
