Brocade Communications Systems IP250 user manual 428 53-1003056-01

14 Logical chassis cluster operations

•If the cluster is configured with a virtual IP address after it is discovered by the Management application, the virtual IP address is collected and saved in the database for the next lazy polling or next adaptive collection.

•If another switch becomes the principal switch, the Management application sets the cluster IP address to that of the new principal switch at the next lazy polling or next adaptive collection.

•The State column in the Discover Setup - IP dialog applies only to nodes that are in logical chassis mode. Refer to Figure 156. Possible states are:

-Online—A node that is currently connected and operational.

-On discovery, only online members are considered active cluster members. The Management application server collects the device, port, and LAG information of active cluster members. The the Management application client displays the member node as a cluster member in the Ethernet Fabrics topology.

-Offline—A cluster member node that cannot be reached by the primary cluster node.

-On refresh, if the member was an active member of a cluster and is now offline, the member is marked as missing. If the member is not online after three consecutive short ticks, auto-discovery gets initiated. If auto-chasing fails, the member remains missing.

-Rejoining—A node that is in the process of rejoining its cluster.

-On refresh, if the member was an active member of the cluster and is now rejoining, then the member is marked as missing. If the member is not online after three consecutive short ticks, auto discovery gets initiated. If auto chasing fails, the member remains missing.

-Replacing—A node that is being replaced.

-On refresh, if a member node is in the Replacing state, the member is shown as missing.

-If the member is in the Replacing state for more than three consecutive short ticks, auto-discovery gets initiated. If auto-chasing fails, the member remains missing.

Administratively removing a node from a logical chassis cluster

You can remove a node from a logical chassis cluster by using the Network OS command line interface. For instructions, refer to the Network OS Administrator’s Guide and the Network OS Command Reference, versions 4.0 or later.

Once the node is removed, all configurations corresponding to that node are removed from the cluster configuration database.

The deleted node gets rebooted automatically and boots in VCS-disabled mode. The deleted node also gets marked as missing in the cluster.

The Management application initiates auto-discovery immediately, and the deleted node gets rediscovered in its current state.

As an example, Figure 157 shows the Discover Setup - IP dialog box before the administrator removes a node from the cluster.

428	Brocade Network Advisor IP User Manual
	53-1003056-01