MCT Configuration Considerations
- Configuring flow-based MAC address learning and MCT on the same device is not supported.
- For MCT, Layer 2 protocols such as STP and RSTP are not supported on cluster VLANs and session VLANs. As a result, STP must be disabled explicitly on cluster VLANs and session VLANs. STP is automatically disabled in the router image.
- One ICL can be configured per device or stack, and a device or stack can be in only one cluster.
- An ICL port must not be an untagged member of any VLAN.
- It is recommended that you set up ICL as a static LAG with at least two ports. This provides port-level redundancy and higher bandwidth for cluster communication.
- ICL ports must be part of MCT VLANs and session VLANs.
- An ICL cannot be a regular port link or an LACP trunk. It must be a single- or multiple-port static LAG.
- MDUP synchronizes all MAC entries for VLANs served by an ICL link.
- In any MCT configuration, there are two different cluster-related IDs, the Cluster ID and the Cluster RBridge ID. The Cluster ID uniquely identifies a cluster. All cluster devices in the same MCT cluster must have the same Cluster ID. The Cluster RBridge ID uniquely identifies a cluster device within the cluster. To avoid conflicts, ensure that the Cluster ID and the Cluster RBridge IDs are unique within an MCT configuration and cannot be confused with each other.
- The Cluster RBridge ID must not conflict with any Client RBridge ID or with the peer RBridge ID.
- The Client RBridge ID is unique and must be the same on cluster devices.
- RUCKUS recommends keeping only ICL ports in the session VLAN during operation.
- MCT is not supported on the ICX 8100 devices.
- MCT supports all the ports configured in a LAG. The maximum number of ports supported per LAG is 16 on all ICX devices that support MCT.
- An ICL interface cannot be configured as the CCEP in any client.
- BPDU guard and root guard configurations must be identical on both cluster devices.
- RUCKUS recommends that you configure a keep-alive VLAN as a separate link (not ICL). The keep-alive VLAN provides a backup control path when CCP goes down.
- From the perspective of MCT, a cluster can consist of a pair of standalones or a pair of stacks. RUCKUS does not recommend pairing a standalone and a stack in the same cluster.
- ICX 7550 and ICX 7850 tacks can be configured as a cluster. Devices from other platforms can be configured as standalone cluster devices.
- All devices in an MCT cluster must be from the same platform family (for example, all ICX 7850 devices).
- From FastIron release 09.0.10a, the maximum number of LACP LAGs that can be configured for MCT cluster increases from 256 to 300.
- Layer 2 multicast traffic may drop for a longer time (approximately 2 minutes and 27 seconds) after an active controller unit is added to an MCT peer stack.
- In an MCT configuration with stacking rings, after a standby controller goes down and comes back up, traffic loss may occur, including on ports from stack units other than the standby controller.
- In an MCT configuration with ICX 7850 peer stacks that include a static LAG connection through a copper GBIC, traffic loss of up to 12 seconds may occur when any unit rejoins the stack following a failover or unit reload.
- In a scaled MCT configuration, Layer 2 traffic spikes may occur when one of the MCT peers rejoins after a reload.
- When a core link flap or peer reload occurs and uplink traffic is being received via a cluster, a significant delay may occur before CR MAC entries are changed to CL MAC entries.