Is this an STP problem or a Cisco bug?

[Marc Abaya]

I'm constantly seeing this error on my 2960 Catalyst switch:

SW#show spanning-tree detail | inc ieee|occurr|from|is exec

 VLAN0010 is executing the ieee compatible Spanning Tree protocol
 Number of topology changes 58334 last change occurred 00:26:48 ago
 from GigabitEthernet1/0/41

This is an access port, why will it received such Topology Change Notifications (TCN)?

This is a remote site so I can't visually check what's connected to it. When I check the "show interface summary" it's showing that this port is down (not connected). I still shut it down but after a few minutes, if I check the STP status again, I saw another TCN message from the same switch but different port. Is this normal? I don't see this happening on my other switches. Here's how the port is cofigured and  show ver of the switch:

interface GigabitEthernet1/0/41

switchport access vlan 10

switchport mode access

no logging event link-status

storm-control broadcast level 1.00

storm-control multicast level 1.00

spanning-tree bpduguard enable

===========================

cisco WS-C2960S-48TS-L (PowerPC) processor (revision F0) with 131072K bytes of memory.

Processor board ID FOC1547Y0XE

Last reset from power-on

2 Virtual Ethernet interfaces

1 FastEthernet interface

52 Gigabit Ethernet interfaces

The password-recovery mechanism is enabled.

512K bytes of flash-simulated non-volatile configuration memory.

Base ethernet MAC Address : 0C:85:25:50:AA:00

Motherboard assembly number : 73-11909-06

Power supply part number : 341-0327-04

Motherboard serial number : FOC154648RW

Power supply serial number : LIT15420HA7

Model revision number : F0

Motherboard revision number : A0

Model number : WS-C2960S-48TS-L

Daughterboard assembly number : 73-11933-04

Daughterboard serial number : FOC15464AKU

System serial number : FOC1547Y0XE

Top Assembly Part Number : 800-30950-04

Top Assembly Revision Number : B0

Version ID : V04

CLEI Code Number : COMGF00ARD

Daughterboard revision number : A0

Hardware Board Revision Number : 0x01

Switch Ports Model SW Version SW Image

------ ----- ----- ---------- ----------

* 1 52 WS-C2960S-48TS-L 12.2(55)SE3 C2960S-UNIVERSALK9-M

Configuration register is 0xF

[julijime · Accepted solution]

Hi Marc,

Even if Gi1/0/41 is an access port it can generate a TCN if there is change in its forwarding state. It would be important to check what is connected to the port, perhaps it is just an end host disconnecting its pc from the network. 

If these ports are destined only for your end users, you can enable the "portfast" feature on them in order to prevent the TCNs to be constantly generated when a user connects/disonnects its pc.

Here is a very useful link which explains in more detail this behavior:

http://www.cisco.com/c/en/us/support/docs/lan-switching/spanning-tree-protocol/12013-17.html#anc12

HTH

Julio

[Marc Abaya]

Thanks Julio. Will take a look.

I just wonder why it's showing only for this switch? We have 4 other Catalyst 2960 switches  in there similarly configured and we do not get this TCNs?

[Mark Malone]

If you do not have portfast enabled on an access port you will see tcns generated whne there is stp changes as its part of the process, set port fast it will remove it from stp calculation and no more tcns will be seen on the port

[Marc Abaya]

My mistake. I thought I saw "spanning-tree portfast" configured on this switch but it is not. Adding it fixes the problem. Thanks.

[Jeroen Huysmans]

Hi,

sorry to bump this old thread...

Even if you see "spanning-tree portfast" configured on an interface it does not mean portfast is operational...

There is a difference between the portfast configured state and operational state.

An interface configured with portfast (trunk) will immediately move to the FWD state when it becomes active, but once a BPDU is received on this interface the switch will remove the operational portfast state from the interface and revert to a traditional STP enabled port.

regards,

Jeroen