[
https://issues.apache.org/jira/browse/HIVE-7349?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14717860#comment-14717860
]
ASF GitHub Bot commented on HIVE-7349:
--------------------------------------
Github user jamescao commented on a diff in the pull request:
https://github.com/apache/flink/pull/1064#discussion_r38163609
--- Diff: flink-staging/flink-hcatalog/pom.xml ---
@@ -34,17 +34,64 @@ under the License.
<packaging>jar</packaging>
+ <repositories>
+ <repository>
+ <id>cloudera</id>
+
<url>https://repository.cloudera.com/artifactory/cloudera-repos/</url>
+ </repository>
+ </repositories>
+
<dependencies>
<dependency>
<groupId>org.apache.flink</groupId>
<artifactId>flink-java</artifactId>
<version>${project.version}</version>
</dependency>
-
+ <dependency>
+ <groupId>org.apache.flink</groupId>
+ <artifactId>flink-scala</artifactId>
+ <version>${project.version}</version>
+ </dependency>
+ <dependency>
+ <groupId>org.apache.hive.hcatalog</groupId>
+ <artifactId>hive-hcatalog-core</artifactId>
+ <version>1.1.0-cdh5.4.0</version>
--- End diff --
This is a known issue with hcatalog, the maven artifact is compiled against
hadoop1 and blocks unit testing.
https://issues.apache.org/jira/browse/HIVE-7349
@fhueske
I found there is no pre-exist test when I begin to work on this issue. How
did hcatalog get tested before? If we stick to the vanilla hcatalog, I guess
one way is to move it to hadoop1 profile(flink compiled against the maven
hcatalog jar can't be used in a hadoop2 env anyway), which will limit its usage
to a large extend since almost all hive production servers on run on hadoop2
now.
> Consuming published Hive HCatalog artificats in a Hadoop 2 build environment
> fails
> ----------------------------------------------------------------------------------
>
> Key: HIVE-7349
> URL: https://issues.apache.org/jira/browse/HIVE-7349
> Project: Hive
> Issue Type: Bug
> Components: HCatalog
> Affects Versions: 0.13.0
> Reporter: Venkat Ranganathan
>
> The published Hive artifacts are built with Hadoop 1 profile. Even though
> Hive has Hadoop 1 and Hadoop 2 shims, some of the HCatalog Mapreduce classes
> are still dependent on the compiled environment.
> For example, using Hive artifacts published in a Sqoop Hcatalog Hadoop 2
> build environment results in the following failure
> Found interface org.apache.hadoop.mapreduce.JobContext, but class was expected
> java.lang.IncompatibleClassChangeError: Found interface
> org.apache.hadoop.mapreduce.JobContext, but class was expected
> at
> org.apache.hive.hcatalog.mapreduce.HCatBaseOutputFormat.getJobInfo(HCatBaseOutputFormat.java:104)
> at
> org.apache.hive.hcatalog.mapreduce.HCatBaseOutputFormat.getOutputFormat(HCatBaseOutputFormat.java:84)
> at
> org.apache.hive.hcatalog.mapreduce.HCatBaseOutputFormat.checkOutputSpecs(HCatBaseOutputFormat.java:73)
> at
> org.apache.hadoop.mapreduce.JobSubmitter.checkSpecs(JobSubmitter.java:418)
> at
> org.apache.hadoop.mapreduce.JobSubmitter.submitJobInternal(JobSubmitter.java:333)
> at org.apache.hadoop.mapreduce.Job$11.run(Job.java:1218)
> at org.apache.hadoop.mapreduce.Job$11.run(Job.java:1215)
> at java.security.AccessController.doPrivileged(Native Method)
> at javax.security.auth.Subject.doAs(Subject.java:396)
> at
> org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1478)
> at org.apache.hadoop.mapreduce.Job.submit(Job.java:1215)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
