Hi Wei, > Hi Carles, > > I will give some comments as we have been doing both CoAP over Websocket > and > CoAP over TCP by extending Califorium.
Thanks a lot! > I agree Carsten that CoAP over TCP will be used in the cloud and may also > be > in less-constrained networks. > It is hard to say that CoAP over TCP will be used in Constrained > Networks. There exist relevant examples such as: https://www.ietf.org/proceedings/87/slides/slides-87-lwig-6.pdf It could be that, within the constrained-node network space, TCP is used rather in class 2 devices. However, we would need to collect more data to be able to confirm this. On the other hand, in our opinion, the CoAP over TCP activity may increase the number of constrained devices using TCP (probably, in the 'constrained device' to 'cloud' scenario). Thanks again! Best regards, Carles > Rgards, > > Gengyu WEI > Network Technology Center > School of Computer > Beijing University of Posts and Telecommunications > -----åå§é®ä»¶----- > From: Carles Gomez Montenegro > Sent: Friday, June 10, 2016 11:10 PM > To: Carsten Bormann > Cc: lwip@ietf.org ; t...@ietf.org Extensions ; jon.crowcr...@cl.cam.ac.uk > ; > c...@ietf.org WG > Subject: Re: [core] [tcpm] [Lwip] [Fwd: New Version Notification for > draft-gomez-core-tcp-constrained-node-networks-00.txt] > > Hi Carsten, > > Thanks a lot for your comments. > > While we work to address those, it would be really helpful if folks that > have faced 'bad constrained TCP implementations', and/or have struggled > with middlebox traversal can share their experience. > > Cheers, > > Carles > > >> Carles, >> >> thanks for submitting this. >> >> I think that this draft is truly best handled in LWIG. >> >> We don't *have* to profile TCP for CoAP-over-TCP; people are free to use >> whatever parts of TCP they think are useful. (And, of course, there are >> applications for CoAP-over-TCP that are in the backend.) >> >> On the other hand, it is useful to >> -- manage expectations: >> what can I expect that the *other* side will offer in TCP >> functionality >> -- give advice to implementers: >> what is useful to implement, what not >> -- collect implementation experience that is relevant for these two >> >> (One interesting effect I'm seeing is that people know how good TCP can >> be, which shapes their expectations, but then they are hurt by using >> really bad constrained TCP implementations... We certainly should be >> paying attention to this on the CoRE WG side.) >> >> My biggest comment is probably that for device-to-cloud, the level of >> TCP functions implemented will be asymmetric (full TCP on cloud side, >> possibly more limited on the device side) -- what is the effect of this >> asymmetry? >> >> Maybe there also needs to be more discussion on the role of the >> middlebox (after all, we are doing CoAP-over-TCP to devices for the sole >> reason to climb over middleboxes). >> >> Grüße, Carsten >> >> >> Scharf, Michael (Nokia - DE) wrote: >>> Heads-up >>> >>> Michael >>> >>> >>> -----Original Message----- >>> From: Lwip [mailto:lwip-boun...@ietf.org] On Behalf Of Carles Gomez >>> Montenegro >>> Sent: Friday, June 10, 2016 11:36 AM >>> To: lwip@ietf.org >>> Cc: jon.crowcr...@cl.cam.ac.uk >>> Subject: [Lwip] [Fwd: New Version Notification for >>> draft-gomez-core-tcp-constrained-node-networks-00.txt] >>> >>> Dear LWIG WG, >>> >>> /** Apologies for possibly multiple similar e-mails... **/ >>> >>> We have just submitted the draft entitled 'TCP over Constrained-Node >>> Networks', which we believe may be of interest to the members of this >>> group. >>> >>> We would like to kindly ask for feedback, specially on the basis of >>> implementation experience. >>> >>> Thank you very much! >>> >>> Kind regards, >>> >>> The authors >>> >>> >>> ---------------------------- Original Message >>> ---------------------------- >>> Subject: New Version Notification for >>> draft-gomez-core-tcp-constrained-node-networks-00.txt >>> From: internet-dra...@ietf.org >>> Date: Fri, June 10, 2016 10:38 am >>> To: "Jon Crowcroft" <jon.crowcr...@cl.cam.ac.uk> >>> "Carles Gomez" <carle...@entel.upc.edu> >>> -------------------------------------------------------------------------- >>> >>> >>> A new version of I-D, >>> draft-gomez-core-tcp-constrained-node-networks-00.txt >>> has been successfully submitted by Carles Gomez and posted to the IETF >>> repository. >>> >>> Name: draft-gomez-core-tcp-constrained-node-networks >>> Revision: 00 >>> Title: TCP over Constrained-Node Networks >>> Document date: 2016-06-10 >>> Group: Individual Submission >>> Pages: 9 >>> URL: >>> https://www.ietf.org/internet-drafts/draft-gomez-core-tcp-constrained-node-networks-00.txt >>> Status: >>> https://datatracker.ietf.org/doc/draft-gomez-core-tcp-constrained-node-networks/ >>> Htmlized: >>> https://tools.ietf.org/html/draft-gomez-core-tcp-constrained-node-networks-00 >>> >>> >>> Abstract: >>> This document provides a profile for the Transmission Control >>> Protocol (TCP) over Constrained-Node Networks (CNNs). The >>> overarching goal is to offer simple measures to allow for >>> lightweight >>> TCP implementation and suitable operation in such environments. >>> >>> >>> >>> >>> Please note that it may take a couple of minutes from the time of >>> submission until the htmlized version and diff are available at >>> tools.ietf.org. >>> >>> The IETF Secretariat >>> >>> >>> >>> _______________________________________________ >>> Lwip mailing list >>> Lwip@ietf.org >>> https://www.ietf.org/mailman/listinfo/lwip >>> >>> _______________________________________________ >>> tcpm mailing list >>> t...@ietf.org >>> https://www.ietf.org/mailman/listinfo/tcpm >>> >> > > > _______________________________________________ > core mailing list > c...@ietf.org > https://www.ietf.org/mailman/listinfo/core > > > _______________________________________________ Lwip mailing list Lwip@ietf.org https://www.ietf.org/mailman/listinfo/lwip
