cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1162
Views
15
Helpful
5
Replies

Mac inconsistency on N5010 vPC pair

Matt Craig
Level 1
Level 1

I have two Nexus 5010s in a vPC pair with a bunch of FEXs vPC dual-homed between them.  I am encountering a problem where, from the point of view of the vPC Primary switch, a MAC address would move from the switch port its connecting to (Eth119/1/25) to the VPL port-channel (Po1).  However from the point of view of the vPC Secondary switch the MAC address would remain on the switch port.  I lose connectivy to the server from outside the datacenter when this happens.  This is happening sporadically to only a small handful for MACs in the datacenter... most other MACs are fine.

Anyone have any clues?

switch1 -----------------

   |  |                         |

   |  | VPL              FEX ------- server

   |  |                         |

switch2 -----------------

State 1:connectivity is fine, ping works

switch1# sh mac address-table address 0050.56a5.0002

   VLAN     MAC Address      Type      age     Secure NTFY    Ports         

---------+-----------------+--------+---------+------+----+------------------

* 1        0050.56a5.0002    dynamic   0          F    F  Eth119/1/25

switch2# sh mac address-table address 0050.56a5.0002

   VLAN     MAC Address      Type      age     Secure NTFY    Ports         

---------+-----------------+--------+---------+------+----+------------------

* 1        0050.56a5.0002    dynamic   0          F    F  Eth119/1/25

State 2: Connectivty lost, ping fails

switch1# sh mac address-table address 0050.56a5.0002

   VLAN     MAC Address      Type      age     Secure NTFY    Ports         

---------+-----------------+--------+---------+------+----+------------------

* 1        0050.56a5.0002    dynamic   0          F    F  Po1

switch2# sh mac address-table address 0050.56a5.0002

   VLAN     MAC Address      Type      age     Secure NTFY    Ports         

---------+-----------------+--------+---------+------+----+------------------

* 1        0050.56a5.0002    dynamic   0          F    F  Eth119/1/25

5 Replies 5

Chad Peterson
Cisco Employee
Cisco Employee

This is odd.

When it gets out of sync like this and one of the switches shows the mac learned on po1...how long does it stay in that condition?

You may want to open a TAC case about this.

Chad

It jumps in and out of that condition for incosisten time periods and at incosistent intervals... ie. mac learned on po1 and stays for couple minutes, then its learned on e119/1/25 on both switches for several minutes, then goes back to po1 after several mintues... whackamole

This is a software defect and a upgrade to 5.0(3)n2(2b) or 5.1(3) is the recommended path to resolve this issue.

Thanks Carlo.  Is there a bug-id for this?  Just want to have it as a reference for any issues I do see like this in the future.  It felt like a bug, but didn't know if one written already.

The Bug ID is: CSCti29571

Regards,

Carlo

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: