You can run the connector server on the same host as the identity resource server, or you can run it on a different host, for example, that’s dedicated to hosting only connectors.
Preparing to implement the RCS
Before you start implementing the RCS, you should consider the following recommendations:
- Setting up the RCS,
connectors, an d mappings should be the first thing you do during implementation to ensure you have user data flowing into the Identity Cloud.
- You should implement the RCS in all three Identity Cloud environments (Development,
Staging, and Production).
Always use dedicated RCS instances and external identity servers for each Identity Cloud environment because each environment has different needs. For example, Development will only contain test data whereas Production will hold real data. Separating the RCS and external identity servers for each environment means you will not impact Production data during testing.
Physically locate the RCS as close to the external identity servers as possible to reduce latency.
- Cluster the RCS to achieve high availability. Although this step is considered optional, it is strongly recommended to prevent the RCS from being a single point of failure.
- You have three working Identity Cloud environments (Development, Staging, and Production).
- You have corresponding external identity servers set up for each environment.
Java 11 is installed on the server where you want to install the RCS.
- You have set the JAVA_HOME environment variable to point to your Java 11 install.
Implementing the RCS
The documentation covers the steps needed to implement the RCS: ForgeRock Identity Cloud Docs › Before you begin. This article references the steps in the documentation, but goes further to include additional checks along the way to help ensure a successful implementation. It also reinforces the need to implement the RCS in all three environments (Development, Staging, and Production).
The steps to implement the RCS are as follows:
- Navigate to the Admin UI in your Identity Cloud Development environment.
- Register a remote server.
Reset the client secret (
unless you have already done this and know the password).
- Download a remote server.
- Configure a remote server to connect to Identity Cloud.
Check that the remote server connection is working before you continue by navigating to Identities > Connect in the Admin UI. It should show as connected.
Install and configure a connector:
Navigate to Native Consoles > Identity Management > Configure > Connectors to configure your connector.
- If you want to use a scripted connector or configure a connector that is not supported in the Admin UI, you can configure it via REST: ForgeRock Identity Cloud Docs › Create a connector configuration over REST.
- Check that the connector is successfully connected to the external identity server by navigating to the Data tab in the connector. You should see user data from your external identity server.
Create a mapping between identities in Identity Cloud and identities in your identity server:
- You can create correlation queries and/or scripts to control what is synchronized in different situations: Synchronization Guide › Correlating Source Objects With Existing Target Objects.
- You can transform attributes during synchronization if required: Synchronization Guide › Transform Attributes in a Mapping.
- Perform an initial reconciliation based on your mapping.
- Register a server cluster if you plan to set up load balancing or failover.
- Navigate to the Admin UI in your Identity Cloud Staging environment.
- Repeat steps 2 to 11 in your Staging environment.
- Navigate to the Admin UI in your Identity Cloud Production environment.
- Repeat steps 2 to 11 in your Production environment.
Once you have successfully implemented the RCS, you should:
- Tune the RCS appropriately according to your needs. The default maximum JVM heap size is 512m, but you can increase this if needed as described below.
- Configure an effective log rotation strategy to manage your logs.
Changing the JVM heap size
Changing the JVM heap size can improve performance and reduce the time it takes to run reconciliations. You should try different heap sizes to see what impact it has to determine the best heap size for your setup.
You can set the JVM heap size via the OPENICF_OPTS environment variable. For example, to set the maximum heap size to 1GB, you would enter the following prior to starting the RCS:
On Unix® and Linux® systems:$ cd /path/to/openicf/bin $ export OPENICF_OPTS="-Xmx1024m" $ ./ConnectorServer.sh /run
On Microsoft® Windows® systems:C:\> cd \path\to\openicf\bin C:\path\to\openicf\bin> set OPENICF_OPTS=-Xmx1024m C:\path\to\openicf\bin> ConnectorServer.bat /run
- Check if the remote server connection is working by navigating to Identities > Connect in the Admin UI. It should show as connected.
- Check that the connector is successfully connected to the external identity server by navigating to Native Consoles > Identity Management > Configure > [Connector] > Data. You should see user data from your external identity server if it's receiving data.
If the RCS is not connected, you should:
- Verify the settings in your ConnectorServer.properties file (located in the /path/to/openicf/conf directory) are correct for your environment and update if needed.
Reset the client secret to the known password if there is a chance someone else could have
Increase the logging level as follows:
- Create a logging.properties file in the /path/to/openicf/conf directory.
- Add the following content to this file:.level=WARNING org.forgerock.openidm.provisioner.openicf.level = FINEST org.forgerock.openicf.connectors.level = FINEST # Monitor Websocket traffic as well (this can be verbose) org.identityconnectors.framework.level=FINEST org.forgerock.openicf.framework.level=FINEST # Troubleshoot Grizzly's WebSocket Implementation (this can be verbose) org.glassfish.grizzly.websockets.BaseWebSocketFilter.level = FINEST
- Restart the RCS.
- Check the resulting logs (located in the /path/to/openicf/logs directory) for clues to see if you can resolve your issues yourself. A common
error message to look out for is:
[name of rcs] 0 active WebSocket(s), 3 remaining permits
This error message means the RCS cannot obtain a connection to the Identity Cloud. You should check the following settings in the ConnectorServer.properties file and update as needed:
If the connector is not receiving data, you should verify the connector settings are correct and update them if needed.
If you're still experiencing issues after taking these troubleshooting steps,
Raising a ticket
To help us troubleshoot issues with the RCS,
- Increase the logging level for the RCS logs if you've not already done this.
- Collect the logs from the RCS server (located in the /path/to/openicf/logs directory) and indicate what timezone is used for logging.
- Collect all the idm- prefixed logs from Identity Cloud as explained in ForgeRock Identity Cloud Docs › View Audit Logs.
Collect the ConnectorServer.properties file (located in the /path/to/openicf/conf directory).
- Confirm the versions of the RCS and connectors you are experiencing issues with.
- Provide details about the environment where you are running the RCS, including whether the
y are clustered?
- Provide details of any configuration changes you made prior to experiencing issues.
- Provide any scripts you use in your connectors.