Product Administration › Choose Another Host in a Cluster When Selected Host Fails

Choose Another Host in a Cluster When Selected Host Fails

If the database host that is specified during Data Aggregator installation fails at runtime, Data Aggregator shuts down automatically. If you installed Data Repository in a cluster, point database connections to another host in the cluster before you restart Data Aggregator.

Follow these steps:

1. Open the Data Aggregator installation directory/apache-karaf-2.3.0/etc/dbconnection.cfg file on the Data Aggregator host.
2. Modify the following line in the dbconnection.cfg file. Modify the line to reference a hostname or IP address of one of the Data Repository cluster hosts that is still up and running:

   dbUrl=jdbc:vertica://database server hostname:database server port/databasename?use35CopyFormat=true&BinaryDataTransfer=false
   

   database server hostname:database server port

   Indicates the hostname or IP address of Data Repository and the Data Repository port number that you entered during the Data Aggregator installation.

   Default port number: 5433

   Example:

   If host2 is up and running in the cluster and you choose database connections to point to host2, your updated dbUrl entry could look like the following line:

   dbUrl=jdbc:vertica://host2:5433/mydatabasename?use35CopyFormat=true&BinaryDataTransfer=false
   

3. Save the dbconnection.cfg file.
4. To restart Data Aggregator, type the following command:

   /etc/init.d/dadaemon start
   

5. To help ensure that Data Aggregator is not still running, type the following command:

   Ps -ef | grep java | grep -v grep 
   

   Data Aggregator processes are not returned when Data Aggregator is not running.

   Database connections point to the specified host in the cluster going forward.

If more than one host in the Data Repository cluster fails, Data Repository and Data Aggregator shuts down automatically. The Data Repository cluster is only capable of losing one host.

If a single host in the cluster that is not specified during the Data Aggregator installation disconnects from the network (for example, because a firewall was put in place, or the Ethernet cable was removed), Data Aggregator shuts down. Data Aggregator restarts automatically if you set up the automatic recovery of the Data Aggregator process during the Data Aggregator installation. Once the host that is offline becomes available, return that host to the cluster. Select the “Restart Vertica on Host” option on the main menu of the admintools utility and follow the prompts.

Note: For information about setting up the automatic recovery of the Data Aggregator process, see the Data Aggregator Installation Guide.

If a single host in the cluster that is not specified during the Data Aggregator installation is stopped through the “Kill Vertica Process on Host” option on the Advanced Menu of the admintools utility, Data Aggregator continues to function. Once the host that is offline becomes available, return that host to the cluster. Select the “Restart Vertica on Host” option on the main menu of the admintools utility and follow the prompts.