1. Installation Guide
  2. Installation Quick Reference

A Installation Quick Reference

This condensed description of the installation steps may be useful if you have installed the release before, or, feel that you do not need a detailed explanation of the steps. You can refer back to the full instructions in the main body of the guide for more information.

About Restarts

Hadoop Side:

  • Cloudera Configuration Manager (or Ambari) may be restarted by the installation. This in itself does not interrupt any services.

  • Hive, YARN , and any other services that have a dependency on Hive or YARN (such as Impala) are restarted.

    The Hive libraries parameter is updated in order to include Oracle Big Data SQL JARs. On Cloudera installations, if the YARN Resource Manager is enabled, then it is restarted in order to set cgroup memory limit for Oracle Big Data SQL and the other Hadoop services. On Oracle Big Data Appliance, the YARN Resource Manager is always enabled and therefore always restarted.

Database Side:

  • In some cases where Grid infrastructure is present, it must be restarted. If the system uses Grid then you should have the Grid user credentials on hand in case a restart is required

  • The Linux users grid and oracle (or other database owner) must both in the same group (usually oinstall).

Table A-1 Checking Prerequisites and Patches

Step For More Information

1. Check for the prerequisites on both the Hadoop or Oracle Database systems. (The installation will alert you if any are missing.)

2. Ensure that the latest Database Proactive Bundle Patch is installed on database as well as any one-off patches required for this product.

NOTE: The Database Proactive Bundle Patch is different than the Patch Set Update. Refer to the Master Compatibility Matrix for complete details on required patches.

Oracle Big Data SQL Master Compatibility Matrix (Doc ID 2119369.1 in My Oracle Support

3. On the database server, check that opatch is included in PATH . Then run opatch to see what patches are installed on the database: $ ./opatch lspatches

  

4. Copy patch 26170659 to the database server if not already installed.

 

5. Make sure database and listener are down.

Managing Oracle Software and Applying Patches (Database 2 Day + Real Application Clusters Guide)

6. Unzip the patch archive on the database server. Navigate to the patch and apply it: $ ./opatch apply

If the database server has both a database and Grid home, then you must apply the one-off patch to BOTH the grid home and database home.

  

7. Navigate to $ORACLE_HOME/Opatch and run the datapatch command: $ ./datapatch

  

Table A-2 Installing the Hadoop Side of Oracle Big Data SQL

Step For More Information

8. Download the Oracle Big Data SQL installation bundle from OTN or eDelivery (Oracle Software Delivery Cloud).

9. Copy this file on to master node of the Hadoop cluster (where Ambari or CM is running). Unzip it and navigate to extracted directory.

 

10. Create a configuration file: 

{
"cluster": {
  "name": <Your cluster name>
  }
}
Some important implicit default settings you may want to modify are database_compatibility and database_auth_enabled. See the description of these parameters in the Jaguar Configuration Parameter and Command Reference in this guide.
{
"cluster": {
  "name": <Your cluster name>
  "database_compatibility" : [ "12.2" ]
  },
"security": {
  "database_auth_enabled": "false"
  }
}

Other configuration options are available.

11. As root, run : ./jaguar install <config file>

If you want to generate a key for Database Authentication, then include the --requestdb parameter and list the databases that should accept the key: 

./jaguar --requestdb orcl,testdb,proddb install <config file>
  

12. Copy the database-side installation bundle generated in <Big Data SQL Install directory>/BDSjaguar-3.2.1.2/dbundles to the database server. If you used --requestdb to generate a request key, find the .reqkey file under /BDSjaguar-3.2.1.2/dbkeys and copy it to the database server as well.

 

Table A-3 Installing the Database Side of Oracle Big Data SQL

Step For More Information
13. Log on as the database owner, locate the database-side installation bundle, and unzip it. Steps for Installing on Oracle Database Nodes
14. Execute the .run file extracted from the ZIP file.  
15. Navigate to $ORACLE_HOME/BDSJaguar-3.2.1.2 and find the installation directory created by the .run file . The directory name will indicate the cluster, node, and domain where the installation bundle originated. Cd into the directory and run the installer: 

$ cd cdh510-6-node1.my.domain.com
$ ./bds-database-install.sh

You may need to include some parameters with this command.

 Command Line Parameter Reference for bds-database-install.sh

Perform the steps in the table below if database_auth_enabled and/or impersonation_enabled was set to “true” in the configuration.

Table A-4 Extra Steps to Complete the Configuration for Database Authentication and/or Multi-User Authorization

Step For More Information

16. If Database Authentication and/or Multi-User Authorization are enabled, the database-side installation generates a ZIP file. Find this file in the installation directory and copy it back to the path indicated below on the Hadoop cluster management server: 

$ ls $ORACLE_HOME/BDSJaguar-3.2.1.2/cdh510-6-node1.my.domain.com/*.zip
  mycluster1-18-mycluster1node03.mydomain.com-myoradb1.mydomain.com.zip
$ scp mycluster1-18-mycluster1node03.mydomain.com-myoradb1.mydomain.com.zip root@node1:/opt/oracle/DM/databases/conf
 Steps for Installing on Oracle Database Nodes
17. Log on to the Hadoop cluster management server as root and run the Database Acknowledge operation:
# cd <Big Data SQL Install Directory>/BDSJaguar-3.2.1.2
# ./jaguar databaseack <config file>
  

Table A-5 Creating Users and Setting Up Mappings for Multi-User Authorization

Action For More Information

18. To grant a user access:

  • Grant the BDSQL_USER role.

  • Grant read privileges on the BigDataSQL configuration directory object. 

SQL> grant BDSQL_USER to user1;
SQL> grant read on directory ORACLE_BIGDATA_CONFIG to user1;
  

19. To grant users Hadoop access through the Multi-User Authorization feature, use the procedures of the DBMS_BDSQL PL/SQL Package:

  • Grant the BDSQL_ADMIN role to designated administrators in order to allow them to invoke these procedures.

  • Use the ADD_USER_MAP procedure to define the rules for a user. This creates a row in the BDSQL_USER_MAP table.

  • Use REMOVE_USER_MAP to remove a row from the BDSQL_USER_MAP table.

DBMS_BDSQL PL/SQL Package

See Also

These workflow diagrams provide a visual overview of the installation process as a whole and a close-up view of the steps needed to complete the setup of Database Authentication:

The next two sections of this guide provide useful code examples: