[
https://issues.apache.org/jira/browse/HIVE-2757?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13262283#comment-13262283
]
Edward Capriolo commented on HIVE-2757:
---------------------------------------
We are missing the larger issue here. Bigtop only cares about two entry points
does not mean there are only two we should care about. We have hive web
interface, hive-thrift-service, lineage tool. Oozie should probably interface
with the Hive-thrift-service I did that here:
https://github.com/edwardcapriolo/m6d_oozie/blob/master/src/main/java/com/m6d/oozie/HiveServiceBAction.java.
Thrift is the best programmatic way to interface with hive.
We have a long standing ticket open to standardize the environment for all
these processes so they all have a common entry point. (I can not find the
ticket ATM).
We are always going to need something like HADOOP_HOME because hive assumes
hadoop jars are "provided" (like in maven terms) and hadoop configuration is
"provided". Unless we copy all the hadoop jars to hive/lib and copy all the
hadoop configuration to hive. With the patch about we still use HADOOP_HOME to
build the classpath to start hive.
Also HADOOP_HOME is deprecated but in the setup docs
"http://hadoop.apache.org/common/docs/r0.23.1/hadoop-yarn/hadoop-yarn-site/SingleCluster.html";
"Assuming that the environment variables $HADOOP_COMMON_HOME,
$HADOOP_HDFS_HOME, $HADOO_MAPRED_HOME, $YARN_HOME, $JAVA_HOME and
$HADOOP_CONF_DIR have been set appropriately. Set $$YARN_CONF_DIR the same as
$HADOOP_CONF_DIR"
I still feel like hive is going to end up using environment variables to start
up.
> hive can't find hadoop executor scripts without HADOOP_HOME set
> ---------------------------------------------------------------
>
> Key: HIVE-2757
> URL: https://issues.apache.org/jira/browse/HIVE-2757
> Project: Hive
> Issue Type: Bug
> Components: CLI
> Affects Versions: 0.8.0
> Reporter: Roman Shaposhnik
> Attachments: HIVE-2757.patch.txt
>
>
> The trouble is that in Hadoop 0.23 HADOOP_HOME has been deprecated. I think
> it would be really nice if bin/hive can be modified to capture the which
> hadoop
> and pass that as a property into the JVM.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
