Post-Upgrade Steps for Drill

For Drill under Warden, complete the following steps:

1. If you made any changes to configuration files in the previous version, compare and restore your previous configurations in the /opt/mapr/drill/drill-<version>/conf directory. Configuration files from the previous installation now reside in /opt/mapr/drill/OLD_DRILL_VERSIONS. Also, copy over any UDF or custom storage or format plugin JAR files that you added to the previous Drill directory.
   NOTE: The drill-override.conf file contains your ZooKeeper configuration and any other options specified in the file. The drill-env.sh file contains any options that you modified, such as Drill memory allocation. The logback.xml file contains changes you may have made to use Lilith.
2. Run configure.sh to refresh the node configuration.

   $ /opt/mapr/server/configure.sh -R

For Drill under YARN, complete the following steps:

1. Issue the following command to start the Application Master:

   $DRILL_HOME/bin/drill-on-yarn.sh start --site <path/to/the/site/directory/for/Drill/cluster>

   NOTE: Note the provided URL at start-up.
2. Optionally, issue the following command to add or remove drillbits to the cluster:

   $DRILL_HOME/bin/drill-on-yarn.sh resize <value>

   NOTE: Value is the number of drillbits that you want started in the cluster. The default value is 1. You can also configure the resize value of the drill cluster in the Application Master Web UI. You can change the number of default drillbits in the drill-on-yarn.conf file.
3. If you have multiple Drill clusters running under YARN, repeat steps 2 and 3 for each cluster.

Once you have completed the steps above for Drill under Warden or YARN, Drill should be configured and running on the node(s). The storage plugin configurations should have been preserved from the previous Drill installation, however you should verify that the Drillbit service is running on each node and also verify that the storage plugin configurations were preserved. Review the Drill logs files if you encounter any issues.

Verify that the Drillbit service is running.

You can use the following methods to verify that the Drillbit service is running on the node(s):

• For Drill under Warden or YARN, issue the following command to verify the status of the Drillbit service from the command line:

  jps

• For Drill under Warden, log in to the MCS at https://<host name>:8443 to verify the status of the Drillbit service.
• For Drill under YARN, access the Application Master Web UI, using the URL provided at start-up, to see the active drillbits.

Verify that the storage plugin configurations were preserved.

If your storage plugin configurations were not preserved, you can restore them using the backup taken before the upgrade. You can use the following methods to verify that the storage plugin configurations were preserved during the upgrade:

• For Drill under Warden or YARN, enter the following URL in a web browser to access the Drill Web Console for an individual drillbit:

  http://<IP address or host name of drillbit>:8047/storage

• For Drill under YARN, access the Application Master Web UI, using the URL provided at start-up, to see the active drillbits.

Access the Drill log files.

You can use the following methods to access the Drill log files:

• For Drill under Warden or YARN, you can access the Drill log files in the Drill Web Console.
• For Drill under YARN, you can access the log files in the Drill container.

Start, stop, or restart drillbits.

You can use the following methods to start, stop, or restart drillbits, as needed:

• For Drill under Warden, you can start, stop, or restart the Drillbit service on one or more nodes using the MCS or the following command:

  $ maprcli node services -name drill-bits -action start|restart|stop -nodes <node host names separated by a space>

  Use the host name if possible. Using host names instead of IP addresses is a best practice.
• For Drill under YARN, you can start or stop the Application Master using the following command:

  DRILL_HOME/bin/drill-on-yarn.sh stop|start --site <path/to/the/site/directory/for/Drill/cluster>

  NOTE: Site is the site directory where the site specific files are stored. For example, opt/mapr/drill/site.