[c-nsp] Mac address flapping..

[James Ashton]

On both 4506s I see that one of the upl;ink ports is in blocking mode.

On 4506-a is it port g1/2
On 4506-b is it port g1/1

There is no blocking going on for this vlan on the 6509s.  But I wouldnt expect that with the 4506s blocking.


As got the timers for HSRP matching the log frequency..  I agree... But my confusion is that HSRP isnt flipping between routers..   So are the spanning tree Hello packets causing this??   If so, Why on this one vlan out of over 120 configured exactily the same???


This is the physical port config of the 4506s..    All 4506 uplink ports ahve the same config:

interface GigabitEthernet1/1
 description dotiq  trunk to fi1/8 core switch
 switchport trunk encapsulation dot1q
 switchport trunk native vlan 999
 switchport trunk allowed vlan 2-999
 switchport mode trunk
end





This is the same for one of the 6509 ports:

interface GigabitEthernet1/7
 description dot1q trunk to int gig 1/1 on as1
 switchport
 switchport access vlan 999
 switchport trunk encapsulation dot1q
 switchport trunk native vlan 999
 switchport trunk allowed vlan 2-995,998,999
 switchport mode trunk
 no ip address
 hold-queue 3000 in
end



Thanks for the help.  This doesnt appear to be effecting the customer on this vlan..  But it is truly vexing me.



James


________________________________________
From: Jonathan Lassoff [jof at thejof.com]
Sent: Saturday, July 11, 2009 5:12 AM
To: James Ashton
Cc: cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] Mac address flapping..

Excerpts from James Ashton's message of Fri Jul 10 08:56:40 -0700 2009:
>
>        %MAC_MOVE-SP-4-NOTIF: Host 00d0.009e.2400 in vlan 42 is flapping between
> port Po1 and port Gi1/7
>
> I see about 20 of these for this one vlan each minute.
> Spanning tree is not reconverging.  It hasn't had a topology change in over 48
> hours.
> HSRP has not changed state.
>
> 6509-b:
> VLAN0042
>   Spanning tree enabled protocol ieee
>   Root ID    Priority    24618
>              Address     00d0.00a7.f000
>              Cost        3
>              Port        1665 (Port-channel1)
>              Hello Time   2 sec  Max Age 20 sec  Forward Delay 15 sec
>
>   Bridge ID  Priority    28714  (priority 28672 sys-id-ext 42)
>              Address     00d0.009e.2400

This is telling -- notice that one of the burned-in addresses on 6509-b
is the one from your log message.

20 times a minute?
HSRP's default hold timer is every 3 seconds -- 20 times a minute.

You also described Gi1/7 as going to 4506-b, right? I would investigate
why spanning tree isn't blocking one of the uplink ports, as it's
causing what sounds like a loop.

Perhaps check that there's something listed from a "show spanning-tree
blockedports".

Cheers,
jonathan