You can enable the various capabilities of the Ethernet interfaces on a per-interface basis using the interface command. When you enter the interface command, you specify the following information:

• Interface type—All physical Ethernet interfaces use the ethernet keyword.
• Slot number
  • Slot 1 includes all the fixed ports.
  • Slot 2 includes the ports on the upper expansion module (if populated).
  • Slot 3 includes the ports on the lower expansion module (if populated).
  • Slot 4 includes the ports on the lower expansion module (if populated).
• Port number— Port number within the group.

The interface numbering convention is extended to support use with a Cisco Nexus Fabric Extender as follows:

switch(config)# interface ethernet [chassis/]slot/port

• Chassis ID is an optional entry to address the ports of a connected Fabric Extender. The chassis ID is configured on a physical Ethernet or EtherChannel interface on the switch to identify the Fabric Extender discovered via the interface. The chassis ID ranges from 100 to 199.

The Cisco-proprietary Unidirectional Link Detection (UDLD) protocol allows ports that are connected through fiber optics or copper (for example, Category 5 cabling) Ethernet cables to monitor the physical configuration of the cables and detect when a unidirectional link exists. When the switch detects a unidirectional link, UDLD shuts down the affected LAN port and alerts the user. Unidirectional links can cause a variety of problems, including spanning tree topology loops.

UDLD is a Layer 2 protocol that works with the Layer 1 protocols to determine the physical status of a link. At Layer 1, autonegotiation takes care of physical signaling and fault detection. UDLD performs tasks that autonegotiation cannot perform, such as detecting the identities of neighbors and shutting down misconnected LAN ports. When you enable both autonegotiation and UDLD, Layer 1 and Layer 2 detections work together to prevent physical and logical unidirectional connections and the malfunctioning of other protocols.

A unidirectional link occurs whenever traffic transmitted by the local device over a link is received by the neighbor but traffic transmitted from the neighbor is not received by the local device. If one of the fiber strands in a pair is disconnected, as long as autonegotiation is active, the link does not stay up. In this case, the logical link is undetermined, and UDLD does not take any action. If both fibers are working normally at Layer 1, then UDLD at Layer 2 determines whether those fibers are connected correctly and whether traffic is flowing bidirectionally between the correct neighbors. This check cannot be performed by autonegotiation, because autonegotiation operates at Layer 1.

A Cisco Nexus device periodically transmits UDLD frames to neighbor devices on LAN ports with UDLD enabled. If the frames are echoed back within a specific time frame and they lack a specific acknowledgment (echo), the link is flagged as unidirectional and the LAN port is shut down. Devices on both ends of the link must support UDLD in order for the protocol to successfully identify and disable unidirectional links.

Note	By default, UDLD is locally disabled on copper LAN ports to avoid sending unnecessary control traffic on this type of media.

The following figure shows an example of a unidirectional link condition. Device B successfully receives traffic from Device A on the port. However, Device A does not receive traffic from Device B on the same port. UDLD detects the problem and disables the port.

Figure 1. Unidirectional Link

The Cisco Discovery Protocol (CDP) is a device discovery protocol that runs over Layer 2 (the data link layer) on all Cisco-manufactured devices (routers, bridges, access servers, and switches) and allows network management applications to discover Cisco devices that are neighbors of already known devices. With CDP, network management applications can learn the device type and the Simple Network Management Protocol (SNMP) agent address of neighboring devices running lower-layer, transparent protocols. This feature enables applications to send SNMP queries to neighboring devices.

CDP runs on all media that support Subnetwork Access Protocol (SNAP). Because CDP runs over the data-link layer only, two systems that support different network-layer protocols can learn about each other.

Each CDP-configured device sends periodic messages to a multicast address, advertising at least one address at which it can receive SNMP messages. The advertisements also contain time-to-live, or holdtime information, which is the length of time a receiving device holds CDP information before discarding it. Each device also listens to the messages sent by other devices to learn about neighboring devices.

The switch supports both CDP Version 1 and Version 2.

The Cisco Nexus device switch does not fragment frames. As a result, the switch cannot have two ports in the same Layer 2 domain with different maximum transmission units (MTUs). A per-physical Ethernet interface MTU is not supported. Instead, the MTU is set according to the QoS classes. You modify the MTU by setting class and policy maps.

Note	When you show the interface settings, a default MTU of 1500 is displayed for physical Ethernet interfaces.

You can configure normal or aggressive unidirectional link detection (UDLD) modes for Ethernet interfaces on devices configured to run UDLD. Before you can enable a UDLD mode for an interface, you must make sure that UDLD is already enabled on the device that includes the interface. UDLD must also be enabled on the other linked interface and its device.

To use the normal UDLD mode, you must configure one of the ports for normal mode and configure the other port for the normal or aggressive mode. To use the aggressive UDLD mode, you must configure both ports for the aggressive mode.

Note	Before you begin, UDLD must be enabled for the other linked port and its device.

You can configure a Quad small form-factor pluggable (QSFP+) port by using the hardware profile portmode command. To restore the defaults, use the no form of this command.

You can disable link negotiation using the no negotiate auto command. By default, auto-negotiation is enabled on 1-Gigabit ports and disabled on 10-Gigabit ports. By default, auto-negotiation is enabled on the Cisco Nexus 3064 and 3064-X switches and disabled on the Cisco Nexus 3048 switch.

This command is equivalent to the Cisco IOS speed non-negotiate command.

Note	Auto negotiation configuration is not applicable on 10-Gigabit ports. When auto-negotiation is configured on a 10-Gigabit port the following error message is displayed: ERROR: Ethernet1/40: Configuration does not match the port capability

You can configure a SVI to remain active even if no interfaces are up in the corresponding VLAN. This enhancement is called Autostate Disable.

When you enable or disable autostate behavior it is applied to all the SVIs in the switch unless you configure autostate per SVI .

Note	Autostate behavior is enabled by default.

You can configure the frequency of Cisco Discovery Protocol (CDP) updates, the amount of time to hold the information before discarding it, and whether or not to send Version-2 advertisements.

You can enable or disable CDP for Ethernet interfaces. This protocol works only when you have it enabled on both interfaces on the same link.

You can enable error-disable (err-disabled) detection in an application. As a result, when a cause is detected on an interface, the interface is placed in an err-disabled state, which is an operational state that is similar to the link-down state.

You can specify the application to bring the interface out of the error-disabled (err-disabled) state and retry coming up. It retries after 300 seconds, unless you configure the recovery timer (see the errdisable recovery interval command).

You can use this procedure to configure the err-disabled recovery timer value. The range is from 30 to 65535 seconds. The default is 300 seconds.

You can provide textual interface descriptions for the Ethernet ports.

You can shut down and restart an Ethernet interface. This action disables all of the interface functions and marks the interface as being down on all monitoring displays. This information is communicated to other network servers through all dynamic routing protocols. When shut down, the interface is not included in any routing updates.