You are here: Home / Blog / Should I use BFD in my Network ?

Should I Use BFD in My Network ?

19th October 2011 by 6 Comments

Q. Should I use BFD in my Network Design ?

A. Yes. Always.

More correctly, BFD is such a great idea for improving routing protocol convergence by improving neighbour failure detection independent of physical circuit that connects those neighbours, that your ONLY consideration is why you WOULD NOT use BFD.

Subnote: although [Loop Free Alternates](http://etherealmind.com/loop-free-alternate-routes/) is even more awesome but it hasn’t arrived yet for OSPF and IS-IS – but it’s coming.

Filed Under: Blog Tagged With: ,
About Greg Ferro

Greg Ferro is a Network Engineer/Architect, mostly focussed on Data Centre, Security Infrastructure, and recently Virtualization. He has over 20 years in IT, in wide range of employers working as a freelance consultant including Finance, Service Providers and Online Companies. He is CCIE#6920 and has a few ideas about the world, but not enough to really count.

He is a host on the Packet Pushers Podcast, blogger at EtherealMind.com and on Twitter @etherealmind and Google Plus

  • Dave Noonan

    Just in case anyone else wondered what “Big F**king Deal” had to do with routing…

    BFD = Bidirectional Forwarding Detection

    (Which I’m familiar with but thought the acronym was BDF or some variation.)

  • kevin

    Right On! but you also have to consider the application and platform.  E.g. there are certain versions of NX-OS that require direct  (same subnet) neighbor relationships to support BFD for things like BGP.. then you must weight the importance of the loopback in the equation for your peer and design implications.  I’ve also seen memory space conflicts for certain functions on a particular platform… not bad, just another factor to consider.  Bottom line.. use it … but make sure you research the limitations of the platform before you do… and interoperability between vendors (e.g. timer limitations, etc.). 

  • http://www.facebook.com/profile.php?id=1340236845 Tristan Rhodes

    Sadly, this is not supported on 3750 routers (except the metro versions).  I can’t find a confirmation on 4500s, which is what we are replacing the 3750s with.  Anyone know the answer?

  • Blah

    LFA as been in Junos since 10.0

  • mivs

    LFA for IS-IS has been supported in IOS-XR since 3.5.0 and for OSPF since 3.9.0.

  • Aliou

    in some implementation we can set fast convergence at sub-second level for IGPs, in that case how can we benefit from BFD?