This documentation is for a version that is out of support.

You may want to view the same page for the latest version, or for any of the versions available in the drop-down menu in the top-right corner.

Cluster administration

For Enterprise and Performance editions only.

Discover how to administer, monitor and manage a Bonita cluster.

Backing up a cluster

Back up your cluster in the same as you would back up a standalone Bonita system, by backing up the database. You also need to back up the load balancer (see your load balancer documentation for details of how to do this).

Monitoring a process

To monitor a process in a cluster, use the Bonita Portal as usual. The information presented applies to process activity on all nodes in the cluster.

Deploying a process in a Bonita cluster

To deploy a process in a cluster, configure the process in Bonita Studio, and then create a .bar file using the Build option of the Server menu. Import the process bar file into Bonita Portal on one of the nodes. The process is now available to Bonita Portal on every node in the cluster. The process must be started on one node. Subsequently, instances can be created on all nodes. This means that all nodes have the possibility of running instances of tasks of all processes.

You can also deploy a process using the API, exactly as you would in a standalone system.

Managing your organization in a cluster

The organization data used by Bonita is stored in the database, so only has to be configured once. It can be maintained from any Bonita Portal in the cluster.

Troubleshooting

Each Bonita Engine maintains its own log file on the node

You can find it under <BUNDLE_HOME>/logs/bonita.`date.log`, where date is the date the file is created. In addition, there is an incident file <BUNDLE_HOME>/logs/tenants/[TENANT_ID]/incident.log (where [TENANT_ID] is the ID of the tenant on which the error occurred) that is used as a last resort to record an error that cannot be handled, such as failure to access the database.