cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1812
Views
0
Helpful
8
Replies

STP issue 3750X

adrien.brun
Level 1
Level 1

Hi Everybody,

We have an issue on a Catalyst 3750X with a VPLS link.

There is a shematic

We have a main site with two access on a VPLS, and some extended sites with one access to the VPLS.

Interconnecting is made by some layer 3 vlans between our 6506 and Layer 3  switchs on extended sites.

On the main site, we made redundancy with spanning tree (for many reasons...).

The first link have a priority for vlan stronger than the secondary link. L3 Vlan are blocked on the secondary link.

When the first link goes down vlan are automatically unbloked on the secondary link and it works well.

We made a new site recently and it has a C3750X connected to the VPLS. Before we had C3750G.

But now when the first link, in the main site, goes down, the Layer 3 vlan for the extended site with a 3750X is not unbloked automatically on the second link in the main site.

In fact when when the first link goes down, i don't receive bpdu from extended site with 3750X. And it works well with 3750G.

Configurations are equal.

Do you have any idea ?

Best Regards

Adrien

8 Replies 8

adrien.brun
Level 1
Level 1

Nobody have an idea ?

I'm alone ?

andtoth
Level 4
Level 4

Hi Adrien,

I would recommend to verify that your configuration is correct and equal with the old site where this is working.

Additionally, if you are not seeing BPDU ingressing, check with your VPLS provider to make sure they are not blocking or filtering BPDU packets on any of the links (maybe on one link only due to a misconfig on their side).

Also, if you could share the relevant configuration and the output of 'sh span' from the devices in a working and a non-working scenario, that might help as well.

What do you see in sh span for the second link on the new site when the first link is down? Is it in BLK state? Or is it in blocking state on the other end?

Best regards,

Andras

Hi Andras,

Configuration are equal and my provider doesn't filtering bpdu.

I have the result of sh span in my different case :

- First case Vlan deleted on Master Link / Active on Second Link / Vlan is ok on 3750G

Interface           Role Sts Cost      Prio.Nbr Type

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

Gi1/0/24            Root FWD 19        128.24   P2p

- Second Case Vlan deleted on Master Link / Active on Second Link / Vlan is nok on  3750X

Interface           Role Sts Cost      Prio.Nbr Type

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

Gi1/0/48            Root BKN*19        128.48   P2p *LOOP_Inc

Best regards

Hi,

seems you're in a loop inconsistent state so you have loopguard enabled on this link and it is doing its job.

Next step is to find why there is a loop. Are you using BPDU filter  on this port ? Can you post sh run int g1/0/48

Regards.

Alain.

Don't forget to rate helpful posts.

Hi,

here it is :

interface GigabitEthernet1/0/48

description Interco_PFS

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 463

switchport mode trunk

speed 100

duplex full

priority-queue out

mls qos trust dscp

end

Alain,

I'm not sure that there is a loop. Because I don't have thie problem with my 3750G.

Maybe there is some thing in the new 3750X that didn't work well wirh VPLS like that.

Regards,

Adrien

Hi Adrien,

Probably the STP BPDU packets are not getting through the VPLS. Please see the following link for more information about STP Loopguard and explanation how it's working:

http://www.cisco.com/en/US/tech/tk389/tk621/technologies_tech_note09186a0080094640.shtml

Loop Guard kicks in when there's a loss of hearing BPDU packets from the other side.

Best regards,

Andras