This article, along with counterpart articles for Cloudera and MapR, describes the simple 3-step process to connect the IRI Voracity big data management platform to any Ambari distribution through the VGrid Gateway.
After connecting, data can be conveniently moved between HDFS and other systems. Furthermore, that data can be manipulated and masked in Hadoop via MR2, Spark, Spark Stream, Storm or Tez using the jobs created in Voracity’s Eclipse IDE, IRI Workbench.
Step 1 – Collect Information from Ambari Dashboard
1. Log into the Ambari Dashboard.
2. Click HDFS. Then, click Configs.
3. Make a note of the NameNode Web UI Port (dfs.namenode.http-address) value (in this case: 50070)
4. Make a note of the NameNode Port (fs.default) value (in this case: 8020)
5. Click on YARN.
6. Make a note of the ResourceManager Web Application HTTP Port (yarn.resourcemanager.webapp.address) value (in this case: 8088)
7. Make note of the Resource Manager Address (yarn.resourcemanager.address) value (in this case: 8050)
8. Click on MapReduce2.
9. Make a note of the MapReduce JobHistory Web Application HTTP Port (mapreduce.jobhistory.webapp.address) value (in this case: 19888)
10. Click on Oozie.
11. Make a note of the Oozie HTTP Port (oozie.base.url) value (in this case: 11000)
Step 2 – Enter configuration details in VGrid Dashboard
1. Log into the VGrid Gateway.
2. Click User > Add User and enter the user information.
3. Click the X in the success banner to refresh the screen.
4. Click Detail in the Action section of the new user.
5. Make note of the generated API key shown. It will be needed in the VGrid Gateway setup in the Workbench preferences screen.
6. Click HadoopConfig and Add Hadoop Config.
- Cluster = Cluster nmae
- User = User name that will be used as the user in the Hadoop file system when working in Workbench
- HDFS = NameNode Web UI Port
- Namenod = NameNode Port
- Proxy = ResourceManager Web Application HTTP Port
- History = MapReduce JobHistory Web Application HTTP Port
- Jobtracker = Resource Manager Address
- Oozie = Oozie HTTP Port
7. Click the X in the success banner to refresh the screen.
8. Click HadoopConfig and click inactive to activate that configuration.
9. Multiple configurations can be associated with each user; however, only one can be active at any given time.
Step 3 – Enter configuration details in IRI Workbench
1. Open IRI Workbench. On the IRI > VGrid Gateway preferences screen, enter the details of the connection.
2. Click Test Connection to confirm that a successful connection is made. Included engines can be narrowed down here if only certain engines are being used in the Hadoop environment. Also, a default engine can be selected for Hadoop run configurations.