[
https://issues.apache.org/jira/browse/SPARK-1297?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14089499#comment-14089499
]
Sean Owen commented on SPARK-1297:
----------------------------------
Yes, the problem though is that at least one of the profiles has to be selected
for the project to build, then, and I don't think we can ask everyone who
builds it to always choose one. Now, let's say hbase-hadoop1 were marked as
active by default. Then I think a caller could either specify nothing and
compile vs Hadoop 1, or, specify "-Phbase-hadoop2,-hbase-hadoop1" to disable
the one and enable the other.
That's closer to workable. Although I think then every time someone chooses a
Hadoop 2 profile they have to know to add this invocation to make HBase work.
Still kind of lame.
Of course, the HBase Hadoop 2 deps can be copied into the various hadoop22,
hadoop23, etc. profiles that already exist. The problem is there is no existing
"hadoop1" profile. If there were, the Hadoop 1 deps could go there and the
problem is solved.
This is why I wonder if this is best solved by introducing a "hadoop1" profile
that is active by default, to solve this issue for all components. I *think* it
means that people then have to write things like "-Phadoop23,-hadoop1", which
is still lame.
Am I missing a much easier solution, given this problem? It's tricky!
> Upgrade HBase dependency to 0.98.0
> ----------------------------------
>
> Key: SPARK-1297
> URL: https://issues.apache.org/jira/browse/SPARK-1297
> Project: Spark
> Issue Type: Task
> Reporter: Ted Yu
> Priority: Minor
>
> HBase 0.94.6 was released 11 months ago.
> Upgrade HBase dependency to 0.98.0
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
