To increase the availability of Orchestrator, you can configure a cluster of Orchestrator server instances. In the cluster, multiple Orchestrator server instances (Orchestrator server nodes) work together. To achieve this, the nodes must share one database and have identical configuration of the Orchestrator server and plug-ins.

The active Orchestrator server nodes respond to client requests and run workflows. If an active Orchestrator server node fails to send heartbeats to indicate it is up and running, it is considered as non-responsive and an inactive Orchestrator node becomes active to take control and resume all of the workflows from the point they were interrupted.

After you configure an Orchestrator server instance in cluster mode, you can create the rest of the Orchestrator cluster nodes by exporting the configuration of the main Orchestrator server instance and importing it to the newly installed Orchestrator server instances.

Note:

All Orchestrator server nodes of a cluster must have identical server and plug-ins' configuration and contents. If you want to make changes on the Orchestrator content, for example to edit a workflow or an action, you must stop all Orchestrator server nodes except one and cancel all running tasks that refer to the content you want to change. You can then make changes to the only server node that is active, and restart the other Orchestrator server instances in the cluster.

The following use case describes how to build an Orchestrator cluster by installing and configuring the main Orchestrator server instance (Orchestrator server 1) and importing its configuration to a newly installed Orchestrator server instance (Orchestrator server 2).

  1. Install Orchestrator server 1 or download and deploy the Orchestrator Appliance.

    For information about installing Orchestrator standalone, see Install Orchestrator Standalone. For information about downloading and deploying the Orchestrator Appliance, see Download and Deploy the Orchestrator Appliance.

  2. Configure a database instance.

    Important:

    Configure the database to accept multiple connections so that it can accept connections from the different Orchestrator instances. To prevent possible transactional deadlocks when the database is Microsoft SQL Server database, you must set the ALLOW_SNAPSHOT_ISOLATION and READ_COMMITTED_SNAPSHOT database options to on.

    You must use an external database.

  3. Log in to the Orchestrator configuration interface as vmware, and configure Orchestrator server 1 to work with the database you configured.

    See Configuring the Orchestrator Database Connection.

  4. Configure Licensing.

    See Importing the vCenter Server License

  5. Click the Reset current version link on the Troubleshooting tab to reinstall previously installed Orchestrator plug-ins with the newly configured database.

    See Configure the Orchestrator Plug-Ins.

  6. Configure an authentication provider.

    See Selecting the Authentication Type.

  7. Configure Orchestrator server 1 to work in cluster mode.

    See Configure Cluster Mode.

  8. (Optional) Install and configure additional Orchestrator plug-ins.

    See Installing a New Plug-In

  9. Start Orchestrator server 1 and wait until it starts successfully.

  10. Restart the configuration interface and export the Orchestrator server 1 configuration.

    See Export the Orchestrator Configuration.

  11. Install Orchestrator server 2 or download and deploy the Orchestrator Appliance.

    For information about installing Orchestrator standalone, see Install Orchestrator Standalone. For information about downloading and deploying the Orchestrator Appliance, see Download and Deploy the Orchestrator Appliance.

  12. On Orchestrator server 2, install the plug-ins that you have installed on Orchestrator server 1.

  13. Import the Orchestrator configuration of Orchestrator server 1 to Orchestrator server 2.

    By importing the Orchestrator configuration, you make both configurations identical. For more information about importing the Orchestrator configuration, see Import the Orchestrator Configuration.

  14. Verify that both Orchestrator server instances have identical configurations and configure the plug-ins on Orchestrator server 2 identically with the plug-ins on Orchestrator server 1.

  15. Modify the network settings on both Orchestrator server instances to reflect your environment, if necessary.

    For instructions about configuring the Orchestrator network settings, see Configure the Network Connection.

  16. Synchronize the the clock of the Orchestrator server 2 machine with the clock of the Orchestrator server 1 machine.

  17. Start Orchestrator server 2.

    To verify that the server started successfully, click the Server Availability tab of the Orchestrator configuration interface and wait until the name of the Orchestrator server appears under Started cluster nodes with a Running or StandBy status.

You can add more Orchestrator server active nodes to the cluster by changing the value of the Number of active nodes field in the Server Availability tab.