01-19-2012 04:03 PM - edited 03-07-2019 04:27 AM
We have two L3 3750 switches running HSRP and routing for various Vlans on our corporate network. Rapid PVST is running across our entire WAN. We are introducing a third party solution for remote communications over MPLS. When interconnecting this equipment to our core network, we have experienced less than desirable failover times of 32 seconds and recovery times of 60 seconds. The vendors engineers are telling is that the interoperability between IEEE spanning tree and Rapid PVST is the culprit. They are suggesting two scenarios - either converting our corporate network to MST (which I prefer not to do but provides us the fastest fail/recovery times) or adjust the timers on our Rapid PVST forwarding timer to 4 seconds. What would the implications be to change these timers on our network, and if I choose to do this, do I only do it at the core (on the two devices that would become root bridge)? I have been doing some reading that says to not mess with the timers, but nowhere do I see reasons why. Can anyone shed some light on this for me?
Many thanks...
Alice
01-19-2012 04:50 PM
What protocol are running between your 3750 switches and the MPLS provider? You can simply use a couple of routed ports on your 3750s, so there is no spanning tree involved and test to see if there is improvement.
HTH
01-20-2012 09:29 AM
We are packing a few of our vlans across trunk ports to be distributed over the MPLS network.
The MPLS device is running RSTP.
Alice
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide