You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Cheng Lian (JIRA)" <ji...@apache.org> on 2014/11/21 05:41:33 UTC
[jira] [Comment Edited] (SPARK-4499) Problems building and running
1.2 with Hive support
[ https://issues.apache.org/jira/browse/SPARK-4499?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14220513#comment-14220513 ]
Cheng Lian edited comment on SPARK-4499 at 11/21/14 4:40 AM:
-------------------------------------------------------------
Hey [~GKyriacou], thanks for reporting, but both consequences you mentioned are expected.
The {{hive-thriftserver}} profile was once removed but added back recently because of dependency issues related to Scala 2.11. So for 1.2.x, you need {{-Phive-thriftserver}} to enable the Thrift server. Please refer to [this notice mail|http://apache-spark-developers-list.1001551.n3.nabble.com/NOTICE-BUILD-Minor-changes-to-Spark-s-build-td9299.html] by Patrick for more details.
For the second issue, we're working with Simba to enables future versions of Simba ODBC driver to inspect Spark SQL and underlying Hive versions precisely via the {{SQLGetInfo}} ODBC API. I believe updated version of Simba ODBC driver will be delivered soon. The reason why Hortonworks driver works, I guess, should be it doesn't rely on {{SQLGetInfo}} to inspect Spark/Hive version. You may refer to [PR #2843|https://github.com/apache/spark/pull/2843] for details.
was (Author: lian cheng):
Hey [~GKyriacou], thanks for reporting, but both consequences you mentioned are expected.
The {{hive-thriftserver}} profile was once removed but added back recently because of dependency issues related to Scala 2.11. So for 1.2.x, you need {{-Phive-thriftserver}} to enable the Thrift server. Please refer to [this notice mail|http://apache-spark-developers-list.1001551.n3.nabble.com/NOTICE-BUILD-Minor-changes-to-Spark-s-build-td9299.html] by Patrick for more details.
For the second issue, we're working with Simba to enables future versions of Simba ODBC driver to inspect Spark SQL and underlying Hive versions precisely via the {{SQLGetInfo}} ODBC API. I believe updated version of Simba ODBC driver will be delivered soon. They reason why Hortonworks driver works, I guess, should be it doesn't rely on {{SQLGetInfo}} to inspect Spark/Hive version. You may refer to [PR #2843|https://github.com/apache/spark/pull/2843] for details.
> Problems building and running 1.2 with Hive support
> ---------------------------------------------------
>
> Key: SPARK-4499
> URL: https://issues.apache.org/jira/browse/SPARK-4499
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.2.0
> Environment: Hadoop 2.4.1, Hive 0.12, Ubuntu Server 12.04 LTS x64
> Reporter: George Kyriacou
> Labels: build, hive, hive-thiftserver, maven
>
> Building using Maven with the '-Phive' option causes an error during start-up of the Thrift Server. It complains that the project was not built using the '-Phive' and '-Phive-thriftserver' options and exits abnormally.
> Compiling with both the '-Phive' and '-Phive-thiftserver' allows the Thrift Server to start correctly however it seems that now it is running the same kind of Thrift server that Hive would be running. Connecting using the Simba Spark ODBC driver to the port that the Spark Thrift Server is running on displays an error that I am attempting to connect to an incorrect server type. Switching to a Hive ODBC driver from Hortonworks and pointing it to the same port connects successfully.
> This did not happen with previous Spark 1.2 builds.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org