Orchestrator requires a database to store workflows and actions.
The Orchestrator server is preconfigured to use an embedded database, which is suitable for testing purposes only. You must configure Orchestrator to use a separate database by using the Orchestrator configuration interface. When the database is embedded, you cannot set up Orchestrator to work in cluster mode, or change the license and the server certificate from the Orchestrator configuration interface. To change the server certificates without changing the database settings, you must run the configuration workflows by using either the Orchestrator client or the REST API. For more information about running the configuration workflows by using the Orchestrator client, see Using the VMware vRealize OrchestratorPlug-Ins. For detailed instructions about running the configuration workflows by using the REST API, see Configuring Orchestrator by Using the Configuration Plug-In and the REST API.
To use Orchestrator in a production environment, you must configure the Orchestrator server to use a dedicated Orchestrator database.
If you download and deploy the Orchestrator Appliance, the Orchestrator server is preconfigured to work with the PostgreSQL database distributed with the appliance. The default Orchestrator Appliance database configuration is suitable for small- or medium-scale environment. To use Orchestrator in a production environment, you must set up a database and configure Orchestrator to work with it.
Orchestrator server supports Oracle, Microsoft SQL Server, and PostgreSQL databases. Orchestrator can work with Microsoft SQL Server Express in small-scale environments consisting of up to 5 hosts and 50 virtual machines. For details about using SQL Server Express with Orchestrator, see Configure SQL Server Express to Use with Orchestrator.
The common workflow for setting up the Orchestrator database consists of the following steps:
Create a new database. For more information about creating a new database, see the documentation of your database provider.
Enable the database for remote connection. For an example, see Configure SQL Server Express to Use with Orchestrator.
Configure the database connection parameters. For more information, see Configuring the Orchestrator Database Connection.
If you plan to set up an Orchestrator cluster, you must configure the database to accept multiple connections so that it can accept connections from the different Orchestrator server instances in the cluster.
The database setup can affect Orchestrator performance. Install the database on a machine other than the one on which the Orchestrator server is installed. This approach ensures that the JVM and database server do not share CPU, RAM, and I/O.
The location of the database is important because almost every activity on the Orchestrator server triggers operations on the database. To avoid latency in the database connection, connect to the database server that is geographically closest to your Orchestrator server and that is on the network with the highest available bandwidth.
The size of the Orchestrator database varies depending on the setup and how workflow tokens are handled. Allow for approximately 50 KB for each vCenter Server object and 4 KB for each workflow run.
Verify that at least 1 GB of disk space is available on the machine where the Orchestrator database is installed and on the machine where the Orchestrator server is installed.
Insufficient disk storage space might cause the Orchestrator server and client to not function correctly.