> On Jul 28, 2018, at 11:24 PM, Mikael Abrahamsson <[email protected]> wrote: > > On Sat, 28 Jul 2018, Joe Touch wrote: > >> because DPI and NAT devices don’t reassemble. And they don’t because it’s >> cheaper to sell devices that say they run at 1 Gbps (e.g.) that don’t bother >> to reassemble. > > Keeping lots of state is always more expensive than not keeping state, and > customers like lower cost devices.
Yes, but they need to be told that their device is “hobbled”. >> So pushing this to another layer will never solve it. What will solve it >> will only be a compliance requirement for #6 - which could be done right >> now, and has to be done for ANY solution to work. > > Where is that Internet Protocol Police when you need it? I appreciate your > struggle, but I don't see how you will succeed in your struggle, in reality. > > So I prefer to recommend not to rely on IP level fragmentation, and fragment > at higher layers. It works better in reality. Until it doesn’t, for exactly the same reason it isn’t working at IP. You’re engaging in a game of escalation - whatever layer you add fragmentation will end up being a layer that a vendor puts a device that does DPI that fails. Joe _______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
