Re: TCP proxys in mobile networks

On 7/28/2016 1:29 PM, Theodore V Faber wrote: > On 7/28/16 13:18, Joe Touch wrote: > > > > On 7/28/2016 1:11 PM, Theodore V Faber wrote: > >> On 7/28/16 13:04, Joe Touch wrote: > >> > >>> I.e., we should NEVER use these boxes to govern how we build > >>> TCP for the masses. > >> > >> To say that

Re: TCP proxys in mobile networks

> On 28 Jul 2016, at 22:17, Joe Touch wrote: > > > > On 7/28/2016 1:11 PM, Theodore V Faber wrote: >> On 7/28/16 13:04, Joe Touch wrote: >> >>> I.e., we should NEVER use these boxes to govern how we build TCP >>> for the masses. >> >> To say that another way: vendors who

Re: TCP proxys in mobile networks

-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 7/28/16 13:18, Joe Touch wrote: > > > On 7/28/2016 1:11 PM, Theodore V Faber wrote: >> On 7/28/16 13:04, Joe Touch wrote: >> >>> I.e., we should NEVER use these boxes to govern how we build >>> TCP for the masses. >> >> To say that another

Re: TCP proxys in mobile networks

On 7/28/2016 1:11 PM, Theodore V Faber wrote: > On 7/28/16 13:04, Joe Touch wrote: > > > I.e., we should NEVER use these boxes to govern how we build TCP > > for the masses. > > To say that another way: vendors who produce such devices are failing > to follow the Postel principle. The Postel

Re: TCP proxys in mobile networks

On 7/28/2016 1:07 PM, Olle E. Johansson wrote: >> I.e., we should NEVER use these boxes to govern how we build TCP for the >> > masses. > That’s right - but we do need to find out what happens. Right now our > support is getting a lot of issues that is hard to explain and we’re not even >

Re: TCP proxys in mobile networks

> On 28 Jul 2016, at 22:03, Joe Touch wrote: > > > > On 7/28/2016 12:57 PM, Olle E. Johansson wrote: >>> All you can do is cause them visibly break so you can detect and eradicate them. >> If you check the paper I referred to they have detected the presence >> of TCP

Re: TCP proxys in mobile networks

On 7/28/2016 12:57 PM, Olle E. Johansson wrote: >> All you can do is cause them visibly break so you can detect and >> > eradicate them. > If you check the paper I referred to they have detected the presence > of TCP proxys, which may help us with setting protocol options > right in order to

Re: TCP proxys in mobile networks

On 7/28/2016 12:33 AM, Toerless Eckert wrote: > You may want to move this discussion to the spud mailing list. Thats IMHO the > "improve STUN" solution. If this were a situation SPUD or STUN could improve, I might agree. I don't think that's the goal of either system, though - there is no

Re: TCP proxys in mobile networks

On 7/28/2016 12:20 AM, Olle E. Johansson wrote: >> > The challenge with STUN has always been that many middleboxes *do not >> > want to be found*. > Which is one reason to improve STUN - right? You can't fix something that doesn't *want* to be found. So-called "transparent" middleboxes (I call

Draft IETF 96 TSVAREA Minutes now available for review

Draft minutes are available at https://www.ietf.org/proceedings/96/minutes/minutes-96-tsvarea. Please let Mirja and I know about any corrections. Thank you, Mat Ford, for producing these minutes and reviewing them against the MeetEcho recording for the meeting. Spencer

Re: TCP proxys in mobile networks

On 27 Jul 2016, at 20:51 , Joe Touch > wrote: On 7/27/2016 8:27 AM, Spencer Dawkins at IETF wrote: Hi, Joe, On Wed, Jul 27, 2016 at 10:18 AM, Joe Touch > wrote: Olle, On 7/27/2016 5:41 AM, Olle E. Johansson wrote: > ... >

Re: TCP proxys in mobile networks

You may want to move this discussion to the spud mailing list. Thats IMHO the "improve STUN" solution. On Thu, Jul 28, 2016 at 09:20:32AM +0200, Olle E. Johansson wrote: > > > On 27 Jul 2016, at 17:18, Joe Touch wrote: > > > > Olle, > > > > On 7/27/2016 5:41 AM, Olle E.

Re: TCP proxys in mobile networks

> On 27 Jul 2016, at 17:18, Joe Touch wrote: > > Olle, > > On 7/27/2016 5:41 AM, Olle E. Johansson wrote: >> ... >> >> This mess caused me sadly to suggest that we need to discuss breaking the >> assumption that TCP delivery is always reliable >> and implement retransmits even