Spanning Tree Update

[jdeisenm]

We are looking to update Spanning tree from PVST. The network has 2 cisco 6500's in the middle and a bunch of cisco 4500's in the wire closets. We want to stay with a per vlan spanning tree. Any thoughts?

 

[KiscoKid]

If you want to stay with a single STP instance per VLAN, then I'd continue to use PVST+.

If you want to optimise your switched environment and reduce the number of STP instances and BPDU traffic, you could consider migrating your current environment to a Multiple Spanning Tree (MST) setup.

Here's a link that explains further about it with some configuration guidelines:

http://www.cisco.com/en/US/products...figuration_guide_chapter09186a0080518f21.html

 

[jdeisenm]

What about Rapid-PVST? Has anyone used that?

 

[vipergg]

why do you want to change from pvst ? If its convergence time you can use pvst+ along with other features like uplink fast or backbone fast to greatly reduce convergence times . Rapid pvst is basically like regular pvst running with like uplink fast , failover is in the range of like 3-5 seconds from what we have seen . Works but we have seen some anomalies with it , don't think it is as stable as pvst+ , but just my opinion. they also use diferent terminolgy for rapid pvst so you have to get used to that also.

 

[jdeisenm]

I my test. 2 2950's back to back with a trunk on fa0/24, convergence for rpvst was 1-2 seconds vs 4-5 sec for pvst w uplink vs 1-2 sec for rstp

 

[Ru55ell]

What convergence are you measuring between the two switches? Updates to your VTP domain or link outages and table changes? Please enlighten me as I’m unfamiliar with the tests.

 

[jdeisenm]

I connected a pc to the non-spanning tree root switch and set a constant ping to the virtual interface of the vlan 1 ip on the spanning tree root switch. I disconnected and reconnected the xover cable and 1-2 supsequent pings would fail after the reconnect with rpvst and 4-5 would fail with pvst.

 

[Ru55ell]

Confirm
It's not clearly stated,

Both switches are Cisco switches.

The xover cable is the only one between the two switches.

You


found on the web..

CSCdy00143
With Multiple Spanning Tree (MST) or Rapid PVST (RPVST) enabled, CSM failover time is excessive. When MST or RPVST is enabled and the spanning tree protocol is configured on a CSM client or server VLAN, it may take up to one minute for traffic flow through a CSM to resume after a CSM failover. The delay is caused by reconvergence of the Spanning Tree Protocol (STP).

Workaround: Ensure that MST and RPVST are not enabled on the Catalyst 6500 series switch containing the CSM, or ensure that the spanning tree protocol is disabled on the CSM client and server VLANs.

 

[jdeisenm]

For my "test bed" I have 2 Cisco 2950's connected on Fa0/24 with an xover cable.

 

[Ru55ell]

Rpvst was developed to make convergence faster. I’m sticking with pvst and have optimized my Spanning tree configuration to my liking.