You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-issues@hadoop.apache.org by "Bharat Viswanadham (JIRA)" <ji...@apache.org> on 2017/09/11 20:39:00 UTC

[jira] [Comment Edited] (HADOOP-14089) Shaded Hadoop client runtime includes non-shaded classes

    [ https://issues.apache.org/jira/browse/HADOOP-14089?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16161946#comment-16161946 ] 

Bharat Viswanadham edited comment on HADOOP-14089 at 9/11/17 8:38 PM:
----------------------------------------------------------------------

[~busbey] Tested this patch, checked the correctness of jar and pre-integration test for jar correctness are passing and every thing looks good.

One general question, why there is a seperate pre-integration-test for hadoop-client-minicluster jar for jar correctness check in hadoop-client-check-test-invariants, and for client-api and client-runtime in hadoop-client-check-invariants.
Can they be combined and have it in one place or is there is any reason for doing this separately?


was (Author: bharatviswa):
[~busbey] Tested this patch, and every thing looks good.

One general question, why there is a seperate pre-integration-test for hadoop-client-minicluster jar for jar correctness check in hadoop-client-check-test-invariants, and for client-api and client-runtime in hadoop-client-check-invariants.
Can they be combined and have it in one place or is there is any reason for doing this separately?

> Shaded Hadoop client runtime includes non-shaded classes
> --------------------------------------------------------
>
>                 Key: HADOOP-14089
>                 URL: https://issues.apache.org/jira/browse/HADOOP-14089
>             Project: Hadoop Common
>          Issue Type: Sub-task
>    Affects Versions: 3.0.0-alpha2
>            Reporter: David Phillips
>            Assignee: Sean Busbey
>            Priority: Critical
>         Attachments: HADOOP-14089.WIP.0.patch, HADOOP-14089.WIP.1.patch
>
>
> The jar includes things like {{assets}}, {{okio}}, {{javax/annotation}}, {{javax/ws}}, {{mozilla}}, etc.
> An easy way to verify this is to look at the contents of the jar:
> {code}
> jar tf hadoop-client-runtime-xxx.jar | sort | grep -v '^org/apache/hadoop'
> {code}
> For standard dependencies, such as the JSR 305 {{javax.annotation}} or JAX-RS {{javax.ws}}, it makes sense for those to be normal dependencies in the POM -- they are standard, so version conflicts shouldn't be a problem. The JSR 305 annotations can be {{<optional>true</optional>}} since they aren't needed at runtime (this is what Guava does).



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org