Jumbo MTU on Nexus 5000

Setting a per interface MTU (maximum transmission unit) is not supported on the Nexus 5000/2000 series switches.
If a Jumbo packet is required to traverse a Nexus 5000 series switch , the jumbo MTU must be set in a policy-map and applied to the ‘Sytem QOS’.

Configuration:

Configuration, PRE NX-OS 4.1:
policy-map JUMBO
 class class-default
  mtu 9216
system qos
 service-policy JUMBO

Configuration with POST NX-OS 4.1:
policy-map type network-qos JUMBO
 class type network-qos class-default
  mtu 9216
system qos
 service-policy type network-qos JUMBO

Verification:
First useful command is “sh queuing”. For clarity to understand the output of the command, by default QOS-Group 0 is for IP traffic and QOS-Group 1 is for FCOE traffic.  Thus look at the MTU setting for QOS-Group 0. As you can see it is set to 9216.

Nexus5k-lab2# sh queuing int e1/1
Ethernet1/1 queuing information:
TX Queuing
qos-group  sched-type  oper-bandwidth
0       WRR             50
1       WRR             50

RX Queuing
qos-group 0
q-size: 243200, HW MTU: 9280 (9216 configured)
drop-type: drop, xon: 0, xoff: 1520
Statistics:
Pkts received over the port             : 86
Ucast pkts sent to the cross-bar        : 77
Mcast pkts sent to the cross-bar        : 9
Ucast pkts received from the cross-bar  : 0
Pkts sent to the port                   : 0
Pkts discarded on ingress               : 0
Per-priority-pause status               : Rx (Inactive), Tx (Inactive)

qos-group 1
q-size: 76800, HW MTU: 2240 (2158 configured)
drop-type: no-drop, xon: 128, xoff: 240
Statistics:
Pkts received over the port             : 0
Ucast pkts sent to the cross-bar        : 0
Mcast pkts sent to the cross-bar        : 0
Ucast pkts received from the cross-bar  : 0
Pkts sent to the port                   : 0
Pkts discarded on ingress               : 0
Per-priority-pause status               : Rx (Inactive), Tx (Inactive)

Total Multicast crossbar statistics:
Mcast pkts received from the cross-bar      : 23

Assuming you’ve had jumbo traffic,  the jumbo counters with “show int” and “show interface {int} counter detail” will be non-zero.

Nexus-5k-SW1# sh int e1/2 | i jumbo
    31 jumbo packets  0 storm suppression packets
    10 jumbo packets

Nexus5k-lab1# sh int counters detailed | i Jumbo
 Rx Jumbo Packets:                                    10
 Tx Jumbo Packets:                                    51
 Rx Jumbo Packets:                                    10
 Tx Jumbo Packets:                                    51

It is important to note that the command “show interface” does not display the correct MTU value. This is a cosmetic bug which has been logged:

CSCsl21529, Symptom: An incorrect MTU value is displayed in the show interface command output. The Cisco Nexus 5000 Series switch only supports class-based MTU. Per-interface level MTU configuration is not supported. The switch supports jumbo frames by default. However, the show interface command output currently displays an incorrect MTU value of 1500 bytes.

Advertisement

4 thoughts on “Jumbo MTU on Nexus 5000

  1. Excellent post. I’ve had many customers ask why they can’t see MTU under show interface and I’ve always referenced the bugID. Now I will reference your blog instead :)

  2. oho man you are lucky , getting experience in next generation nexus switches , thanks for sharing the knowledge

  3. Great article and insight on class based MTU for Jumbo packets on the FEX (2K) and 5K Nexus switches. Bravo Sir!

Please leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.