Working vulnerability? (Cisco exploit)

>I'd estimate than less than a tenth of a percent (that's 0.1%) of
>edge paths use RPF, even though BCP38 states the case clearly and the
>technology makes it easy

"Makes it easy" if you live in an Internet with a number of routes
significantly less than the limit imposed for having stable RPF
enabled on your devices, or have devices without bugs in RPF checking
when said limit is spotted vaguely across the horizon.

since those are two very common beliefs about BCP38 deployment, i'm going
to leave it to barry to repost his standard answers to them, once he's gotten
some sleep.

I dont seem to be in either of those places. (Although I have not
sacrificed a router in the last upgrade version or two to see if things
have improved.)

please do, and please post your results, as an example to others.