Robot SCHEDULE Enterprise R01M34 and higher support failover cluster configurations.
An agent should be installed on every server that can run in a cluster. Once installed, open the Robot SCHEDULE Agent Manager. Select Add, and create an agent with the same name as each instance in the cluster.
For example, we have instances named SBSQLSVR and SBSQLSVR2. So two agents were created on each of the two server in the cluster. The agents were named SBSQLSVR and SBSQLSVR2.
Once the agents are created, set them to inactive. Also go into Edit, and check the box marked "No, I'll manually start it when needed". That way the agent can be started by the cluster instead of the local server.
The configuration requires a Generic Resource be added to each instance. To add a generic resource, right click on the instance from the Microsoft Failover Cluster Manager, select Add a resource, and Generic Application.
There will be two fields.
Once the command line and parameters are set, click next, and finish adding the resource.
A resource will appear under "Other Resources". Right click on that resource, and go to Properties. Select the Dependencies tab at the top, and insert a dependency with SQL Server as the resource.
Then go back to the General tab, and select the box next to "Use Network Name for computer name" Select Okay when finished.
Repeat these steps for every instance in the cluster.
Agents will be started from the Failover Cluster Manager. Select the instance on the left, and find the generic resource created above. Right click and select 'Bring this resource online".
Now when you go to Agents in the Robot SCHEDULE GUI, agents with the same name as the instances in the cluster will appear. When a job is run on an agent in the cluster, the job will first go to the cluster, and the cluster will direct the job towards whatever server the agent process is active on. When the server running the agent goes down, the failover cluster should start the agent process on another server. When the job runs again, it still goes to the same cluster, and the cluster directs it towards the new server instead of the failed server.
Still have questions? We can help. Submit a case to Technical Support.