10.7 Configuring the Start, Failover, and Failback Modes for Cluster Resources

You can configure the start, failover, and failback of cluster resources to happen manually or automatically.

IMPORTANT:Cluster Services works with NCP user connections so that the user data sessions are resumed after failover. However, non-NCP users might experience service interruption and need to reconnect to the server after the failover. Applications using server based storage must be restarted on the client even with NCP unless they are NCP reconnect aware.

10.7.1 Understanding Cluster Resource Modes

With the resource Start mode set to AUTO, the resource automatically starts on a server when the cluster is first brought up. If the resource Start mode is set to MANUAL, you can manually start the resource on a server when you want, instead of having it automatically start when servers in the cluster are brought up.

With the resource Failover mode set to AUTO, the resource automatically starts on the next server in the Assigned Nodes list in the event of a hardware or software failure. If the resource Failover mode is set to MANUAL, you can intervene after a failure occurs and before the resource is moved to another node.

With the resource Failback mode set to DISABLE, the resource does not fail back to its most preferred node when the most preferred node rejoins the cluster. If the resource Failback mode is set to AUTO, the resource automatically fails back to its most preferred node when the most preferred node rejoins the cluster. Set the resource Failback mode to MANUAL to prevent the resource from moving back to its preferred node when that node is brought back online, until you are ready to allow it to happen.

The preferred node is the first server in the Assigned Nodes list for the resource.

IMPORTANT:Resources fail back only to the first node in their Assigned Nodes list. For example, if a resource has failed over to three servers since it originally ran on its preferred node, and the second server the resource was running on comes back up, the resource does not fail back to that second server.

Resources do not automatically move from node to node just because a node higher in the Assigned Nodes list rejoins the cluster, unless the Failback mode is set to AUTO and the first node in the Assigned Nodes list rejoins the cluster.

10.7.2 Setting the Start, Failover, and Failback Modes for a Resource

If you are creating a new cluster resource, the Resource Policies page should already be displayed. You can start with Step 5.

  1. In iManager, click Clusters and then click Cluster Options.

  2. Browse to locate and select the Cluster object of the cluster you want to manage.

  3. Select the box next to the resource whose Start, Failover, or Failback modes you want to view or edit, then click the Details link.

  4. Click the Policies tab.

  5. (Optional) Select the Resource Follows Master check box if you want to ensure that the resource runs only on the master node in the cluster.

    If the master node in the cluster fails, the resource fails over to whichever node becomes the master.

  6. (Optional) Select the Ignore Quorum check box if you don't want the cluster-wide timeout period and node number limit enforced.

    The quorum default values were set when you installed Novell Cluster Services. You can change the quorum default values by accessing the properties page for the Cluster object.

    Selecting this box ensures that the resource is launched immediately on any server in the Assigned Nodes list as soon as any server in the list is brought online.

  7. Specify the Start, Failover, and Failback modes for this resource.

    The default for both Start and Failover modes is AUTO, and the default for Failback mode is DISABLE.

  8. Do one of the following:

    • If you are configuring a new resource, click Next, then continue with Section 10.8, Configuring Preferred Nodes for a Resource.

    • Click Apply to save your changes.

      Changes for a resource’s properties are not applied while the resource is loaded or running on a server. You must offline, then online the resource to activate the changes for the resource.