If there is a communication error, the grid manager assumes that the process has gone down. The grid manager makes one of the secondary controller instances the primary and then attempts to deploy the object in the new controller instance.

If there is a deployment error, the grid manager assumes that the deployment cannot proceed and starts the rollback process for undeploying the deployed objects. As part of the rollback process, it proceeds to undeploy all the objects that it had deployed previously on the controller instances. After the rollback is complete, the server returns an error back to the user interface to notify about the failed deployment.

If the active grid manager process fails while deploying the objects, the passive grid manager process takes over from where the active grid manager process stopped during the deployment process and rolls back the changes.

If the run time is down during deployment or undeployment, an error message appears, stating that the grid manager cannot be reached.