You must have the Hive JDBC standalone jar compatible with your HDP version and the
latest Hive JDBC standalone jar compatible with Cloudera on
the edge node between the HDP and Cloudera cluster.
-
On the HDP cluster, navigate to /usr/hdp/current/hive-client/lib check that the
Apache Hive standalone JAR was installed by HDP on your cluster.
For example, on HDP 2.6.5 you find a JAR having a name something like
hive-jdbc-1.2.1000.2.6.5.5000-33-standalone.jar.
On HDP
3.1.5, you find a JAR having a name something like
hive-jdbc-3.1.0-315.133-2-standalone.jar.
-
If the JAR is not on the cluster, using the version number of the JAR in
/usr/hdp/current/hive-client/lib, search for and download the Apache Hive JDBC
driver hive-jdbc from the driver archive.
-
On the Cloudera cluster, navigate to
/opt/cloudera/parcels/<version>/jars/hive-jdbc-3<version>,
and check that the Apache Hive standalone JAR is on the cluster.
-
If the JAR is not on the cluster, download the latest Hive JDBC driver from
the Cloudera Downloads page.
-
Make a note of the path to the JAR to connect the HDP and Cloudera clusters later.
