The Cisco Nexus device can be deployed as a Fibre Channel Forwarder (FCF) as shown in the following figure.

Figure 1. Directly Connected Fibre Channel Forwarder

The following rules are used to process FIP frames to avoid the FCF being used as a transit between an FCoE node (ENode) and another FCF. These rules also prevent login sessions between ENodes and FCFs in different fabrics.

• FIP solicitation and login frames received from the CNAs are processed by the FCF and are not forwarded.

• If an FCF receives solicitations and advertisements from other FCFs over an interface, the following occurs:

  • The frames are ignored and discarded if the FC-MAP value in the frame matches the value of the FCF (the FCF is in the same fabric).

  • The interface is placed in the "FCoE Isolated" state if the FC-MAP value in the FIP frame does not match that of the FCF (the FCF is in a different fabric).

CNAs cannot discover or log in to FCFs that are reachable only through a transit Cisco Nexus FCF. The Cisco Nexus device cannot perform the FCoE transit function between a CNA and another FCF due to hardware limitations.

Because the Cisco Nexus FCF cannot perform the transit FCoE function, you must design your network topology so that the active Spanning Tree Protocol (STP) path of FCoE VLANs is always over the directly connected links between the CNA and the FCF. Make sure that you configure the FCoE VLAN on the directly connected links only.

The Cisco Nexus device can be deployed as a Fibre Channel Forwarder (FCF) for remotely connected CNAs, but not as a FIP snooping bridge, as shown in the following figure.

Figure 2. Remotely Connected Fibre Channel Forwarder

The following rules are used to process FIP frames to avoid the FCF being used as a transit between an ENode and another FCF. These rules also prevent login sessions between ENodes and FCFs in different fabrics.

• FIP solicitation and login frames received from the CNAs are processed by the FCF and are not forwarded.

• If an FCF receives solicitations and advertisements from other FCFs over an interface, the following occurs:

  • The frames are ignored and discarded if the FC-MAP value in the frame matches the value of the FCF (the FCF is in the same fabric).

  • The interface is placed in the "FCoE Isolated" state if the FC-MAP value in the FIP frame does not match that of the FCF (the FCF is in a different fabric).

Because the Cisco Nexus FCF cannot perform the transit FCoE function, you must design your network topology so that the active STP path of FCoE VLANs is always over the directly connected links between the CNA and the FCF. Make sure that you configure the FCoE VLAN on the directly connected links only.

The following figure shows a best practices topology for an access network that is using directly connected CNAs with Cisco Nexus devices.

Figure 3. Directly Connected CNA

Follow these configuration best practices for the deployment topology in the preceding figure:

1. You must configure a unique dedicated VLAN at every converged access switch to carry traffic for each Virtual Fabric (VSAN) in the SAN (for example, VLAN 1002 for VSAN 1, VLAN 1003 for VSAN 2, and so on). If you enable Multiple Spanning Tree (MST), you must use a separate MST instance for FCoE VLANs.

2. You must configure the unified fabric (UF) links as trunk ports. Do not configure the FCoE VLAN as a native VLAN. You must configure all FCoE VLANs as members of the UF links to allow extensions for VF_Port trunking and VSAN management for the virtual Fibre Channel interfaces.

   Note	A unified wire carries both Ethernet and FCoE traffic.

3. You must configure the UF links as spanning-tree edge ports.

4. You must not configure the FCoE VLANs as members of Ethernet links that are not designated to carry FCoE traffic because you want to ensure that the scope of the STP for the FCoE VLANs is limited to UF links only.

5. If the converged access switches (in the same SAN fabric or in another) need to be connected to each other over Ethernet links for a LAN alternate path, you must explicitly configure such links to exclude all FCoE VLANs from membership. This action ensures that the scope of the STP for the FCoE VLANs is limited to UF links only.

6. You must use separate FCoE VLANs for FCoE in SAN-A and SAN-B.

Note	All Gen-1 (pre-FIP) and Gen-2 (FIP) CNAs are supported in a directly connected topology.

The following figure shows a best practices topology for an access network using remotely connected CNAs with Cisco Nexus devices.

Figure 4. Remotely Connected CNAs

Follow these configuration best practices for the deployment topology in the preceding figure:

1. You must configure a unique dedicated VLAN at every converged access switch to carry traffic for each Virtual Fabric (VSAN) in the SAN (for example, VLAN 1002 for VSAN 1, VLAN 1003 for VSAN 2, and so on). If you enable MST, you must use a separate MST instance for FCoE VLANs.

2. You must configure the unified fabric (UF) links as trunk ports. Do not configure the FCoE VLAN as a native VLAN. You must configure all FCoE VLANs as members of the UF links to allow extensions for VF_Port trunking and VSAN management for the virtual Fibre Channel interfaces.

   Note	A unified fabric link carries both Ethernet and FCoE traffic.

3. You must configure the CNAs and the blade switches as spanning-tree edge ports.

4. A blade switch must connect to exactly one Cisco Nexus device converged access switch, preferably over an EtherChannel, to avoid disruption due to STP reconvergence on events such as provisioning new links or blade switches.

5. You must configure the Cisco Nexus device converged access switch with a better STP priority than the blade switches that are connected to it. This requirement allows you to create an island of FCoE VLANs where the converged access switch is the spanning-tree root and all the blade switches connected to it become downstream nodes.

6. Do not configure the FCoE VLANs as members of Ethernet links that are not designated to carry FCoE traffic because you want to ensure that the scope of the STP for the FCoE VLANs is limited to UF links only.

7. If the converged access switches and/or the blade switches need to be connected to each over Ethernet links for the purposes of LAN alternate pathing, you must explicitly configure such links to exclude all FCoE VLANs from membership. This action ensures the scope of the STP for FCoE VLANs is limited to UF links only.

8. You must use separate FCoE VLANs for FCoE in SAN-A and SAN-B.

Note	A remotely connected topology is supported only with Gen-2 (FIP) CNAs.

FCoE has the following guidelines and limitations:

• FCoE on Cisco Nexus devices support the Gen-1 (pre-FIP) and Gen-2 (FIP) CNAs. FCoE on the Cisco Nexus 2232PP fabric extender (FEX) supports Gen-2 CNAs only.

• Enabling FCoE on VLAN 1 is not supported.

• A combination of straight-through and active-active topologies is not supported on the same FEX.

• Direct connect FCoE (that is, a direct connect to CNAs through a bind interface) is not supported on a port channel of a Cisco Nexus device or FEX interface if it is configured to have more than one interface. Direct connect FCoE is supported on port channels with a single link to allow for FCoE from a CNA connected through a vPC with one 10 GB link to each upstream switch/FEX.

Note

For a description of the default quality of service (QoS) policies for FCoE, see the Quality of Service guide for your device. for the Nexus software release that you are using. The available versions of this document can be found at the following URL: http:/​/​www.cisco.com/​en/​US/​products/​ps9670/​products_​installation_​and_​configuration_​guides_​list.html..

This example shows how to configure the default Ethernet system class to support the jumbo MTU:

switch(config)# policy-map type network-qos jumbo
switch(config-pmap-nq)# class type network-qos class-fcoe
switch(config-pmap-c-nq)# pause no-drop
switch(config-pmap-c-nq)# mtu 2158
switch(config-pmap-nq)# class type network-qos class-default
switch(config-pmap-c-nq)# mtu 9216
switch(config-pmap-c-nq)# exit
switch(config-pmap-nq)# exit
switch(config)# system qos
switch(config-sys-qos)# service-policy type qos input fcoe-default-in-policy
switch(config-sys-qos)# service-policy type queuing input fcoe-default-in-policy
switch(config-sys-qos)# service-policy type queuing output fcoe-default-out-policy
switch(config-sys-qos)# service-policy type network-qos jumbo