UCS | MAC Address Ageing

[ptaneja12]

We have migrated NLB cluster from HP Servers to UCS Servers, and since migration to the Cisco UCS environment, customer have been complaining about issues with the NLB (Network Load Balancing) service on their systems.  These machines were previously residing inside of a traditional vSphere 5.0 cluster running on HP BL465c G7 blades, and were recently migrated to the UCS platform on blades running vSphere 5.1. 

One further researching came to an option > MAC Address Table Ageing (under Global Policies)

Last section on the below link > > 

http://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/sw/gui/config/guide/2-0/b_UCSM_GUI_Configuration_Guide_2_0/b_UCSM_GUI_Configuration_Guide_2_0_chapter_01100.html

Before we go ahead and change the MAC Address Table Ageing setting to "Never" - so that entries are not washed away at set interval of time, need to know if changing this can cause any issues globally to overall Infra? Maybe something like buffer overflow wherein the table is overwhelmed by MAC entries.  Need to know the suggestions !!  

[Walter Dey]

Please read

http://www.cisco.com/c/en/us/support/docs/servers-unified-computing/ucs-b-series-blade-servers/118262-config-nlb-00.html

in particular

Caution: NLB in unicast mode relies on unknown unicast flooding for delivery of cluster bound packets. Unicast mode will not work on UCS B-Series servers when the FI is in End-Host Mode since unknown unicast frames are not flooded as required by this mode. For more details on the L2 forwarding behavior of UCS in End-Host mode, see Cisco Unified Computing System Ethernet Switching Modes.