search

intarchboard / draft-use-it-or-lose-it

Long-term Viability of Protocol Extension Mechanisms
Other
2 stars 4 forks source link

Appendex A.1: BGP #72

Closed elear closed 3 years ago

elear commented 3 years ago

There's not enough information to understand what happened and how this draft pertains to addressing the BGP failure. For one thing, the reference [RIPE-99] has a reference to a Cisco URL that apparently no longer exists; IOS-XR at that time was brand spanking new. And it's not like BGP extensions weren't a regular thing at that time.

However, there is a different use case that might be interesting to pursue: partial deployment of ROAs and S*BGP led to some unintended consequences. See Lychev, Goldberg, Shapira. This gets into multiparty aspects.

martinthomson commented 3 years ago

As an example, I think that this is sufficient. Bad link notwithstanding... (see https://www.w3.org/Provider/Style/URI.html).

This isn't saying that the problem is constant or that BGP can't change, but that BGP, like many other protocols, has experienced the same sorts of issues too.

I read the paper; it's good work, but I'm not seeing the relevance. It is primarily about deployment incentives and how various amounts of deployment might contribute to security goals.

elear commented 3 years ago

The issue with BGP is that changes to one router's routing table can impact an entire system. And so protocol options used between two routers can have an impact beyond those two routers. I am suspecting what you are seeing is a bug of that variety.

martinthomson commented 3 years ago

I've dropped the BGP example. As you noted privately, it doesn't really support any particular advice in the document, it just existed and that's bloat.