11-30-2022 06:39 AM
Hello Cisco community.
I have been doing some research on running BFD with BFD Dampening and IP Event Dampening together as a method to ensure that BFD is detecting these layer 3 type failures and correctly notifying upper layer protocols of failures to reroute traffic accordingly. But in the other hand possibly using IP Event Dampening to assist with layer1 type flapping causing instability in upper layer protocols. I can't seem to find good documentation of running these two protocols together and best practices around this. Anyone doing this currently? share any tips for using these on high speed long haul fiber links?
Solved! Go to Solution.
12-01-2022 07:59 AM
Yeah, I reviewed the presentation and am familiar with the concepts outlined there. Was hoping someone in the community here that is operating a service provider network could chime in on BFD dampening and Cisco IP event dampening being used at the same time for this scenario. I think we would lean more towards IP event dampening rather than carrier delay up an interface.
11-30-2022 11:26 AM
Take a look at this presentation:
https://www.ciscolive.com/c/dam/r/ciscolive/emea/docs/2019/pdf/BRKRST-2042.pdf
Now to answer your question it really depends on what you want to detect and protect against, there is no right answer, for example with carrier-delay some companies want to bring down the interface immediately, some want to wait in case it was a transport switchover event. BFD is useful as well but if you get a spike in latency it can cause your session to bounce, and you need to remember qos for echo mode.
Sam
11-30-2022 12:15 PM - edited 11-30-2022 12:16 PM
Currently we use BFD for that microsecond level detection of link failure at layer 3 in which BFD dampening takes care of any subsequent layer 3 flapping by going through the process of the backoff counter and exponential increase of that hold down for subsequent layer 3 flaps. But in the event of layer 1 (optical loss) through a 3rd party carrier that is flapping due to some bad fiber splice, fiber cut , fiber damage etc... BFD is able to correctly detect these "failures" and bring upper level protocols/routing protocols up and down but it cannot distinguish that this particular link has been flapping up and down due to layer 1 type events. Since layer 1 events cause a BFD session removal/deletion event on IOS-XR causing the BFD dampening timers to essentially "start over" and never exponentially grow, I would think that coupling BFD dampening with Cisco IP Event Dampening could help with saying yeah...your layer 1 on this has been flapping up and down for a long period of time. Maybe we should hold that down for an X amount of preconfigured time before attempting to bring layer 1 up, followed by upper layers...BFD/routing protocols/convergence etc..
Does that make sense?
11-30-2022 12:46 PM
dampening could help you in that case. Have you had a chance to review the presentation, you might also want carrier-delay. There is nothing wrong with a combination of all these technologies.
Sam
12-01-2022 07:59 AM
Yeah, I reviewed the presentation and am familiar with the concepts outlined there. Was hoping someone in the community here that is operating a service provider network could chime in on BFD dampening and Cisco IP event dampening being used at the same time for this scenario. I think we would lean more towards IP event dampening rather than carrier delay up an interface.
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