When a switch stack forms a port-channel, the active stack member’s MAC Address is used for the LACP ID. The active stack member is the Master. If you look below, you will see the “Switch/Stack Mac Address” is the Master’s.
TPA-SWA#sh switch
Switch/Stack Mac Address : 1c6a.7ad3.0c80
H/W Current
Switch# Role Mac Address Priority Version State
----------------------------------------------------------
*1 Master 1c6a.7ad3.0c80 15 1 Ready
2 Member 1cde.a722.0900 10 1 Ready
Now if you look at the LACP details, you will see the Master switches Mac Address being used on both uplinks for the System ID:
TPA-SWA#sh lacp internal detail
Flags: S - Device is requesting Slow LACPDUs
F - Device is requesting Fast LACPDUs
A - Device is in Active mode P - Device is in Passive mode
Channel group 1
Actor (internal) information:
Actor Actor Actor
Port System ID Port Number Age Flags
Te1/1/1 32768,1c6a.7ad3.0c80 0x136 27s SA
LACP Actor Actor Actor
Port Priority Oper Key Port State
32768 0x1 0x3D
Port State Flags Decode:
Activity: Timeout: Aggregation: Synchronization:
Active Long Yes Yes
Collecting: Distributing: Defaulted: Expired:
Yes Yes No No
Actor Actor Actor
Port System ID Port Number Age Flags
Te2/1/1 32768,1c6a.7ad3.0c80 0x236 26s SA
LACP Actor Actor Actor
Port Priority Oper Key Port State
32768 0x1 0x3D
Port State Flags Decode:
Activity: Timeout: Aggregation: Synchronization:
Active Long Yes Yes
Collecting: Distributing: Defaulted: Expired:
Yes Yes No No
This will be a problem if the Master switch goes offline. This LACP System ID will change causing the Etherchannel to flap. So if you have any servers teamed with LACP interfaces, they will go down. Usually it would stay down until the Master switch is reloaded.
So basically when you use LACP, the System ID uses the stack MAC Address from the stack master. If the stack Master goes down, the LACP System ID can change resulting in the Etherchannel flapping and STP reconverging. This will bring down your LACP Etherchannels.
The fix for this is to use “stack-mac persistent timer 0” command. This will keep the Mac Address of the previous Master until you enter the “no stack-mac persistent timer”.
You will get the warning below when you enter the command:
WARNING: Stack MAC persistency timer value of 0 means that, after a
WARNING: master switchover, the current stack-mac will continue
WARNING: to be used indefinitely.
WARNING: The Network Administrators must make sure that the old
WARNING: stack-mac does not appear elsewhere in this network
WARNING: domain. If it does, user traffic may be blackholed.
After adding the command you will see a new line when you do the “show switch” command:
TPA-SWA#sh switch
Switch/Stack Mac Address : 1c6a.7ad3.0c80
Mac persistency wait time: Indefinite
H/W Current
Switch# Role Mac Address Priority Version State
----------------------------------------------------------
*1 Master 1c6a.7ad3.0c80 15 1 Ready
2 Member 1cde.a722.0900 10 1 Ready
One last thing and this is stated in the warning message above. When and if you do this persistent mac address config, you want to make sure you don’t reuse the Master switch somewhere else on your network. If you do, you will end up with duplicate Mac Addresses and lost traffic.