Robot Schedule Enterprise and Clustering

 

Windows Server Failover Clustering (WSFC) allows you to configure two or more server computers to work together in a cluster to provide failover protection and increased availability of applications. Companies may have one or more clusters with failover ability running on each server.

The agents you create in Robot Schedule Enterprise can be configured to work in a failover cluster environment. To do so:

• Create shared Robot Schedule Enterprise agents on each of the SQL Servers in the cluster. This includes the main server and all of the standby servers.

• Create a Generic Application resource on each server. It's used to start the servicing agent when needed.

Important: You must be running Robot Schedule Enterprise V01R34 or higher in order to use its agents in a cluster environment.

Creating Shared Agents

To create shared agents, add identical agent definitions on each machine in the cluster. Instructions are below. For a detailed description on installing the agent manager and creating agents, see "Installing Robot Schedule Enterprise Agents."

1. Install (or update) the Robot Schedule Enterprise Agent Manager on the first machine in the cluster. To do so:

   Download and run the Robot Schedule Enterprise Agent Manager for Windows installer (setupAgent.exe). You can download it from My Downloads on our website.

2. Open the Robot Schedule Enterprise Agent Manager.

3. Click Add. Create the agent definition. Some things to keep in mind:

   • The name should match the SQL Server name.

   • Select the option to manually start the agent.

   • The shared agent's listening port must be the same on each of the servers in the cluster.

   Note: After you add the agent, do not start it. In the next section, you'll use the Failover Cluster Manager to bring the agent online.

4. Repeat step 3 to add more agents. You'll need to add a shared agent for each of the servers in the cluster.

5. Repeat steps 1-4 on each server in the cluster.

Creating a Generic Application Resource

The steps in this section are done in the Failover Cluster Manager that's part of Windows Server Failover Clustering. For additional information, see the documentation for the Failover Cluster Manager.

1. Start the Failover Cluster Manager.

2. Under your cluster definition, expand Services and Applications.

3. Right-click the first SQL Server installation and select Add a resource, then Generic Application. A Wizard starts.

4. Under Command Line, enter the path to the cluster program installed with the Robot Schedule Enterprise Agent software. The default path is: C:\Program Files (x86)\Help Systems\Robot Schedule Enterprise Agent\Robot Schedule Cluster Agent Process.exe.

5. Under Parameters, enter an XML file name for the agent. The file/agent name should match the SQL Server installation you selected in step 3. Use quotation marks around the name, for example: "sqlsvr1.xml".

6. Click Next.

7. Accept the rest of the default settings by clicking Next on each page, and then Finish.

8. Right-click the resource you just added and select Properties.

9. On the Dependencies tab, select the main SQL Server job from the list and click Insert. This creates a dependency between the job and the resource.

10. Click OK to save the properties.

11. Right-click the resource again, and select Properties.

12. On the General tab, select Use Network Name for Computer Name.

13. Click OK to save the properties.

14. Repeat steps 3-13 for each of the other SQL Server installations in the cluster.

15. Right-click each of the resources you just added and select Bring this resource online. This starts the Robot Schedule Enterprise agents.

After You Are Done

After you bring the agents online in the Failover Cluster Manager, they will appear in the list of agents under Enterprise Server in the Robot Schedule Explorer.

Agent jobs that were created using the SQL Server Job function type in Robot Schedule will run normally. However, if you've used another method of creating SQL Server jobs in Robot Schedule (for example, using tsql commands in the job definition), you should change the agent assigned to those jobs to one of the new agents you created in the above procedure. This ensures that the jobs will run in both a normal situation and also in a failover, without you having to manually switch the agent at the time of the failure.