Cabling a fabric-attached MetroCluster configuration with disks

You cable the fabric-attached MetroCluster configuration so that the controller and the disk shelves at each site are connected to Brocade or Cisco switches. In turn, the switches at one site are connected through inter-switch links to the switches at the other site.

  • To cable a fabric-attached MetroCluster configuration, you must be familiar with HA configuration, the Brocade or Cisco command-line interface, and synchronous mirroring.
  • You must also be familiar with the characteristics of fabric-attached MetroCluster configurations.
  • You must also have the following information:
    • Correct Brocade or Cisco licenses for each switch
    • Unique domain IDs for each of the switches
      Note: You can use the switch numbers (1, 2, 3, and 4) as the switch Domain ID.
    • Ethernet IP address for both the switches and nodes
      Note: The switches ship with a default IP address (, which you can use if the switches are not attached to a network.
    • Ethernet subnetmask
    • Gateway address
A fabric-attached MetroCluster configuration involves two nodes at physically separated sites. To differentiate between the nodes, the guide refers to the two nodes as Node A and Node B.
Note: If you are using SAS disk shelves, the disk shelves connect to FibreBridge 6500N bridges. To see an example of a cabled fabric-attached MetroCluster system with FibreBridges and SAS disk shelves, see the Configuring a MetroCluster system with SAS disk shelves and FibreBridge 6500N bridges at the NetApp Support Site at

The following illustration shows a fabric-attached MetroCluster configuration:

This graphic is described by the surrounding text