[
https://issues.apache.org/jira/browse/FLINK-18256?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17133168#comment-17133168
]
Yang Wang commented on FLINK-18256:
-----------------------------------
[~lzljs3620320] Could you assign this ticket to me? I will attach a PR to fix
this.
> Hadoop is bundled in flink-orc exceptionally
> --------------------------------------------
>
> Key: FLINK-18256
> URL: https://issues.apache.org/jira/browse/FLINK-18256
> Project: Flink
> Issue Type: Bug
> Components: Formats (JSON, Avro, Parquet, ORC, SequenceFile)
> Affects Versions: 1.11.0, 1.12.0
> Reporter: Yang Wang
> Priority: Blocker
> Fix For: 1.11.0, 1.12.0
>
>
> The {{hadoop-common}} and {{hadoop-hdfs}} have been bundled into the
> {{flink-orc_2.11-1.11-SNAPSHOT-jar-with-dependencies.jar}}. Since we suggest
> our users to put this jar into lib directory or add dependency in their pom
> to build fat jar, it might be incorrect to have hadoop here. Flink now is not
> hadoop free, we have some specific hadoop version classes in the released
> jars.
>
> > How to fix?
> Let flink-orc depends on {{hadoop-common}} and {{hadoop-hdfs}} directly and
> make them provided.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
