Details
-
Bug
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
None
-
None
-
None
Description
After the initial install of the cluster and trying to start up Zeppelin service the following comes up in Ambari
Traceback (most recent call last): File "/var/lib/ambari-agent/cache/common-services/ZEPPELIN/0.6.0.2.5/package/scripts/master.py", line 448, in <module> Master().execute() File "/usr/lib/python2.6/site-packages/resource_management/libraries/script/script.py", line 329, in execute method(env) File "/var/lib/ambari-agent/cache/common-services/ZEPPELIN/0.6.0.2.5/package/scripts/master.py", line 225, in start self.update_kerberos_properties() File "/var/lib/ambari-agent/cache/common-services/ZEPPELIN/0.6.0.2.5/package/scripts/master.py", line 300, in update_kerberos_properties and params.zookeeper_znode_parent not in interpreter['properties']['phoenix.url']: KeyError: 'phoenix.url'
WORKAROUND:
It is likely there are no phoenix properties in /etc/zeppelin/conf/interpreter.json file. In that situation:
- edit this file
- search for "jdbc"
- in that section go to "properties" where you can see definition for hive, default, etc.
- add the below:
"phoenix.user": "phoenixuser",
"phoenix.driver": "org.apache.phoenix.jdbc.PhoenixDriver",
"phoenix.url": "jdbc:phoenix:<hbase.zookeeper.quorum>:<zookeeper.znode.parent>",
"phoenix.hbase.client.retries.number": "1", - save the changes
- restart Zeppelin
OR
Remove /etc/zeppelin/conf/interpreter.json file and restart Zeppelin service
Steps to reproduce; (this happens only on a Kerberos enabled cluster)
- On a fresh cluster install just Zeppelin
- Once the above is done, then installed hbase/phoenix.
- Now restart Zeppelin
Ambari will fail to start Zeppelin with the above-mentioned error.
Attachments
Attachments
Issue Links
- links to