[
https://issues.apache.org/jira/browse/PHOENIX-1132?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14079315#comment-14079315
]
Jesse Yates commented on PHOENIX-1132:
--------------------------------------
I don't think this is really a big issue; the overhead of adding 1 jar vs. 2 is
really pretty minor. And then if you are using flume, or pig you need to add
that as well. Oh and know which generated phoenix jar you use.
I could see having just a single phoenix-hadoop<version>.jar that includes the
compat layer and the hadoop<version> (e.g. Hadoop1 or Hadoop2) code.
What we really need to do is take a look at all the jars created an generate a
cohesive story (and docs!) around usage + naming.
bq. What's exactly in the phoenix-hadoop2-compat.jar
Our own code for using hadoop2 under the hadoop-compat layer interfaces
bq. and phoenix-hadoop-compat.jar
Yes, this is the shim layer
> Merge server side phoenix-hadoop2-compat.jar, phoenix-hadoop-compat.jar and
> phoenix-<version>.jar
> -------------------------------------------------------------------------------------------------
>
> Key: PHOENIX-1132
> URL: https://issues.apache.org/jira/browse/PHOENIX-1132
> Project: Phoenix
> Issue Type: Sub-task
> Reporter: James Taylor
>
> With tracing added to Phoenix 4 branch
> https://git-wip-us.apache.org/repos/asf?p=phoenix.git;a=commitdiff;h=e8def027a9c60e978ac79acbc6c71979e7478a43
> requires 3 server side jars. Need to reduce it to a single jar for ease of
> deployment in our user community.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
