[
https://issues.apache.org/jira/browse/YARN-4086?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Robert Kanter updated YARN-4086:
--------------------------------
Attachment: YARN-4086.001.patch
The YARN-4086.001.patch allows the yarn CLI and web UIs to read aggregated logs
from HAR files. It's mostly the same as the prelim patch in MAPREDUCE-6415,
with some minor changes and unit tests. The patches for this and
MAPREDUCE-6415 can be applied independently.
*Important:* For the unit tests, I had to include some HAR files, which are
basically folders with a few files in them. One of the files is a binary file,
which makes generating and applying the patch tricky. I got it to work by
generating it with {{git diff --binary > FILE}} and to apply with {{git apply
<FILE>}}. The regular {{patch}} command won't work and it has to be {{-p1}}
and not {{-p0}}. I'm not sure if Jenkins will be able to handle this.
> Allow Aggregated Log readers to handle HAR files
> ------------------------------------------------
>
> Key: YARN-4086
> URL: https://issues.apache.org/jira/browse/YARN-4086
> Project: Hadoop YARN
> Issue Type: Improvement
> Affects Versions: 2.8.0
> Reporter: Robert Kanter
> Assignee: Robert Kanter
> Attachments: YARN-4086.001.patch
>
>
> This is for the YARN changes for MAPREDUCE-6415. It allows the yarn CLI and
> web UIs to read aggregated logs from HAR files.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
