Re: [sidr] Burstiness of BGP updates

[Russ White <[e-mail]>]

>> Is it really that complex?
> (not really aimed at russ)
> 
> is the never-ending rathole of 'what are we trying to protect' really
> required on-list? I think the most simple case we care about is: "Is
> the routing system telling us what it is supposed to?" Or rephrased
> some: "Did the route injected at the source get faithfully reproduced
> down the line to the receiver?"

But SIDR is currently saying that as long as the route was injected
correctly a week or two ago, "it's all good." Sorry, but I disagree.
It's not "all good."

Security compares what the state currently looks like to what the state
should look like. If "what the state should look like" could be a week
old, and you've ruled out "intentions" (which really rules out what the
system should look like), then you've ruled out "security."

Russ