[c-nsp] ASR9K: BPDU stats on non-MSTP enabled interfaces

[Pavel Stefanov]

In case anyone is interested, I managed to find the right set of 
commands to get this information. The "0x08000a60" in the log below is 
the interface handle, which can be captured as follows.

RP/0/RSP0/CPU0:xxxx#show netio idb all brief location 0/3/CPU0 | in 
"Bundle-Ether2000”
Mon Apr 14 13:24:39.086 UTC
0x08000a60 Bundle-Ether2000 608 0x00000000 F 2 4096 30

RP/0/RSP0/CPU0:ASRPE2#sh spanning-tree trace io | in "Received a BPDU"
Mon Apr 14 13:40:03.417 UTC
Dec 12 17:18:08.819 l2protocols/ethernet/stp/io/unique 0/RSP0/CPU0 228# 
t1 080009e0/--: 0090 Packet :- Received a BPDU on a port that does not 
have MSTP configured on it
Apr 11 13:11:31.374 l2protocols/ethernet/stp/io/unique 0/RSP0/CPU0 1168# 
t1 08000a60/--: 0090 Packet :- Received a BPDU on a port that does not 
have MSTP configured on it


On 11/04/2014 14:21, Pavel Stefanov wrote:
> Hi all,
>
> I need to enable MSTP with BPDU guard on several hundred ports on an 
> ASR9K. Unfortunately, due to the high number of ports it is not 
> feasible to identify whether the remote end runs xSTP and since MSTP 
> is not enabled on the ports on the ASR9K, I cannot see the BPDU stats 
> for themt. I tried looking in the STP trace but all I get is the below 
> which gives no readable reference to the actual port number.
>
> Is anyone aware of a command that can show whether BPDUs are actually 
> being received and dropped on the port (due to the protocol not being 
> enabled)?
>
> RP/0/RSP0/CPU0:xxxx#show spanning-tree trace io | in configured
>
> Fri Apr 11 12:26:12.911 UTC
>
> Dec 12 17:18:08.819 l2protocols/ethernet/stp/io/unique 0/RSP0/CPU0 
> 228# t1 080009e0/--: 0090 Packet :- Received a BPDU on a port that 
> does not have MSTP configured on it
>
> Nov 26 12:29:04.138 l2protocols/ethernet/stp/io/fast 0/RSP0/CPU0 t1 
> --------/--: 0010 Packet :- Constructing BPDUs for 1 ports. 0 MSTIs 
> configured, max 0 Rl2GP MSTIs.
>
> I could potentially enable MSTP without BPDU guard first, check the 
> stats and then enable BPDU guard on the port but that might cause a 
> reconvergence event and drops for traffic on other ports so I have 
> ruled this out as an option.
>
> Thanks,
> Pavel