Internet Engineering Task Force 105, 23 Jul 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: IETF105-IPWAVE-20190723-1130

Description

IPWAVE meeting session at IETF105
2019/07/23 1130

https://datatracker.ietf.org/meeting/105/proceedings/

A

B

Okay, let's go ahead and get started, welcome.

B

We have only 30 minutes.

B

Some things have happened in the past couple days, so we're juggling the agenda quickly, we'll see if we have time to get to everything, so we put it in priority order.

B

So just reminder on the intellectual property- you you must say if there's any IPR associated with your drafts and your contributions and in terms of anything you say if the might as well.

B

This is the no well you've been seeing it all day yesterday. So well, but if you haven't read it, please pay attention to it.

B

Blue sheets are going around.

B

We need a new ticker.

B

Someone please volunteer it's going to easy today: Thank You robber. Can we have a joker star.

B

This is just take to go to the mic if anyone who is remote and wants to share something with the room.

B

Okay, I'll, do it let you run the meeting all right.

B

We have two documents of progress: you'll see in a few minutes that the number has gone up since we made these slides, but we're going to talk about them in this order. The first one is with the is G and the second one has just been in last call this is the agenda we're going to deal with the amount of time necessary in this order, so I hope we get to everything, but we'll see.

B

Okay and that's it tradition lesser than the agenda bashes.

B

All right, Billy you're up.

C

Okay, good morning, everybody, my name is Naveen amar I'm, the main Otto, the the first author of the the district after Alex left. The draft in vision, 43, so I will give an update about what have been done in this draft since version 44 until 15. Currently we are in 50, but we are expecting to to to post a new version 51 after receiving some comments from a romanian ID, so from 44 45. So we have in 44.

C

We have updated the author's list and in 45 we have updated the abstracts so that we can give more focus on the the main scope. So we have redefined the scope, so that's to animate any ambiguity on what we want to do and we want to achieve in this draft, as suggested by the main reviewer at that time, pascal figure, so we have added in ran of one another of I simply a to do so.

C

We are dealing with ranch and that any other optimization are out of scope in in this draft from 44 to 46, we have a new abstract, which is more concise, so that we don't have much details in the abstract and we keep them to the main body of the text, and we have introduced that these drafts doesn't make challenge in existing stack so with minimal challenge to existing stack and also, we focus once once out of time that the more complex scenarios related to vehicular communication will be covered in future. Work.

C

Also in from 45 to 46, we have specifying the scope also in the introduction, so we have now in 46 this new text. The document describes the leading of ipv6 networking on top of the I Triple E standard, mock layer or I Triple, E, 8 or 2.3 ma clear, with a frame translation underneath, and we have added these centers- that the results in stacks operates over 80 2.11 ocv and provides at least peer to peer connectivity.

C

So the term peer to peer was not in previous versions of the draft using ipv6 in d and link local addresses in the extension and IP way of optimization for vehicular. Communication are not in scope because we have been dealing with Pascal about what kind of Indy what are the the limits of Indy when it comes to vehicle communication, etc.

C

So we made this sentence so to clarify things better.

C

In also we have in security consideration, we have added some some ideas here. So, for example, we added that previous work at say: V working group present some threats while sent presented in these children. These two RFC's is the solution against breast eft, but it's complex had not deployed.

C

These comment was sent also by from from Pascal, but we had some other comments from the 80s when it goes to the iesg that why talking about mentioning sand, if it's complex had not deployed- and we replied that this is the only solution that we have at that time. So this is what why we mention it in this draft from 46 to 47. We have changed the title so now we have this new title of the the draft basic support for ipv6 over X cetera and yesterday I asked the RFC editor if they can help.

C

Even if it's not the time, it's not an FCS, but if we can gain some time to if we can receive some feedback, so they told me that the title is a little bit long, so maybe we can, we can think about it in next version. We have also included in the introduction that the resulting stuck in here it's from ipv6 over Ethernet, so these lenses also some debates in the mailing list and with the ideas what we mean by in heretic cetera.

C

We explained that we do the same as in Ethernet as as ipv6 defined in RFC 24 64. So there is no charge and.

C

That's it yes for for this version, and then yes- and this was one of the hot topics, hot debates, which is about the the length of the address or the the interface identifier. So we specify clearly that independently of the the method used to provide the even identifier, the left is 64 bits. So if we need any other, if there is any comments that should be discussed in other working group, not in IP wave from 47 to 48 Mohammed Boogedy thanks to Mohammed, he suggested he proofreads the whole document, and he suggests me in private.

C

He sent me corrections proofreading and such some ameliorations to clarify some text.

C

So many correction were made in this version: 48 in from 48 to 14 9. We had some issues about the references which are normative and which are informative. I think that can be I can happen in many stage and hopefully we have solved this problem and we fix it some remaining typos. In every time we have a lot of typos from 49 to 50, so it was the the ballot. So we got 10 okay. So it's it's an okay for the moment, but we still have some three discuss.

C

We reflected the comments received from the IDS we removed to the following, not in so in the new version, which is 50 compliance with standard and the regulation set in different countries when using the 5.9 gigai frequency band is required. We remove this part because we don't need to deal with this detail in in IETF. No specific reason why this needs to be said here. So this is why we remove this part. I think this is what the so. This one was one of the main comment of the of one of the discuss.

C

So hopefully this discuss will turn on okay, the map pin to the exit. So we have this. Must that became a shirt because one of the ad is say that must it's strong here and is in this? This idea deals with I Triple E, so why we need to force something like that. So we agreed that this must be. Should I need to remind the participant here that this part of the draft was sent us as a review from I Triple E 802 dogeleven working group based on their review, a long review?

C

We we made this this sentence here, so we just changed the must into shoot. We also corrected the normative use of may and should in the text the normal way of writing this. These words also, we reformulated the text in subnet structure, as pointed out by Roman one of the 80s as follows. So.

C

So he suggested to to to rewrite, but with the help of Pascal, we rewrite this this part, so we introduce the some specific passivity is related to neighbor discovery, to reflect what we want to say here and why nd with current ocv can so I can lead to other works that can be done in the future.

C

We also reformist the Security section, because one of the ATS suggested that an attacker may not heed to legal limit for radio power and can use a very sensitive directional antenna, so it was okay, so I added this sentence. If attackers wish to attack a given exchange, they don't necessarily need to be in close physical proximity, so this detail was not present in previous version.

C

What remains to do is to answer one last comment in the discuss that we have now of Roman, which deals with pseudonym handling, so I think we have now with the help of Erik. We have now I think a good text that Roman will I I hope will will agree on. So that's it for for these versions so in equation. Our comments are welcome. Thank you say: what are you going to say to addresses?

C

So we are, we have just sent few lines to Roman and we can die and we are still waiting for his feedback just to fix the the this short paragraph about what is the the the the difference between what is the impact of confidentiality, because we used the term confidentiality in the text. It was on big use for him and we changed it into the identity of the destination and making the trust, because identity is not as confidentiality its to term different. So we have proposed to to change these two terms.

C

That's it that's the main change that we have in this section.

D

Suresh krisshnan, so Russ like in charge. It tastes like the upper layer protocol should take care of it, like that's the gist of this Texas like propose to roman okay. Thank.

B

D

E

So, basically, when you get the okay from from any, you will be submitting version 51. Yes,.

C

E

Then we will ask the the three scums, because there are three discussions to check if they are fine and 2/3 of that. Okay, thank.

C

F

Hello, everybody, this is poor, John from SK, kill you so basically, this IP waive promise statement document liberated by Charlie pockys and three and basically three kundo, barely certified. With on this television and Charlie Kaiser, he looks like didn't finished and Libya yet, but I believe most of his comments are addressed.

F

So after last the prom meeting revised it two times it so right now the version is 11, so so I will explain to Orly be jealous, so the first revision from 9 to 10 so from a chali. So the question is the mortar linker, so Burnette models. So this is the problem statement. So I remove a multi-link, a sudden, a discussion. Just we specify characteristic of a virtual or linker model. Is it symmetric and so mobile attack and that took property.

F

So we just addressed and also he mentioned, the DNS motivation, the pqy Deanna's important, so I explained it especially in beaker has a lot of issue unit, so they can be lester their own domain, name or so remotely automotive services enter can remote a monitor regulator, so in the case dns laceration. Also, these are resolution is important. So that's why I added a some text so in appendix I added that detail changes. So you can refer to that so timing, issues, Andy.

F

So the you can see the national highway traffic agencies as a product service as appetizers. Thank you. So they mentioned the 0.25, the second very important timing to avoid accident. So in the case the neighbor discovery message: if used for the safety, a message, 0.25 exchanges required so depending on tentative vehicle, we can adjust the interval of neighbor discovery message, so I added a tap text.

F

Next, the security consideration case, expanding the considered, neighbor discovery, part and the mobility management part, so I added a some text the. Secondly, this you can see documentation predicts Charlie, maybe some confused. Why? What is it is so this is? You can see the documentation predicts right for the example. So definitely any allowed rubber IP version 6 address. It can be loud trouble in Bennett and V killer, Edie vehicle Annette on including infrastructure, loader, RS use. Also the third one.

F

In the piggy run case, the separate vehicle on a dog netted panel can be merged to using infrastructure node. Over newly a joining of vehicles, so you can see two separate: a banach can link with the infrastructure loader, all the other arriving not I mentioned, and that part and suggestion are how a good a nose depress the prom, a better nose, or so we I added a text, so certificate can be used to authenticate the beaker. So I mentioned the high lever.

F

So our a document again is for problem Stateville letter then giving the solution so and Burton ten case I reflected the chalice. The editorial comment, so many editorial comments are addressed. Some several questions case in the appendix that I address the. What change are ton and three hakuna volley case the article she a fifty eighty five, zero five reference: this is for the six all of em the address alliterations optimization. So in this text that we can lead up them just may be the solution development, the page we can consider. So we tell it either.

F

Next step is after confirmation of a charlie and then Carlos will leave you and then we can call last call and then we can submit is G. Okay, that's it.

E

G

I miss the reason why you remove tags denied her for five. Is that because you're not talking solution.

F

G

Did not understand why you removed that text. Okay,.

F

Three, his comment, the years that and the optimization is the possible a good solution, but our problem statement- the case we don't imagine it may be how we can discuss that's.

G

Because you don't enter solution space, that's because you're in promised ice. That's very.

F

Good yeah, yes, yes, yeah, okay,.

E

So yeah, basically, we will wait until suddenly confirms that this is happy with the review. Yeah.

H

Just closing on the last comment, exactly I think it's not touching on the solution. That's one thing second thing is this was the last minute you know the comments you know, but that you are addressing were mostly based on the reviews from Charlie and others right there. As to some discussion randomly, we cannot insert some new sections without any discussion in the mailing list right, so there's a reason why.

G

So it was scary, yeah I don't buy the last argument. If you say, if you say: okay, that's not a subjective document. I completely buy it. If you tell me it's because we don't discuss enough well, let's discuss right now that you can't just remove text. That's odd, because it's like, if you say, okay, we have to ship now. So whatever is discussion is pending. We don't take it. I don't buy!

G

That I mean you want a good document, and if you want a good document, then you want to make sure you have settled on the issues right. So if there was a good reason to mention it then, and you remove it just because you decided at the last minute- that's not fair, either right so so the group must are the consensus either way we want to talk about it. We don't want to talk about it. It's the right document, wrong document, but you can't just make a decision like that in a work group document.

G

Just at the moment, you do last call to snip some document effect without the group saying. Yes, you can right.

H

So the text that showed up in that version there was no priority discussion. It's just a randomly showed up that intersection. Point of our point number two again, this spec is about requirements right, so we are not touching I. Think those for those reasons. I said you know they should be okay,.

E

So I mean we we can have discussion if somebody thinks that it has to be the earth text. But I read that it's not about solution. The document, okay, so then I was instead we're saying we'll wait for Charlie to confirm and then once at least happy with the review I will I will.

E

F

So we moved to the hackathon, propose just one minute it enough, because yeah- and so basically this has only the second time hackathon. We so I think that first card, the solution of our solution is the proto-tool heated previous. The work also the hackathon project case.

F

We approved the OCB and also one possible solution on the big hill neighbor discovery, the multi hub at dat, so we proved concept and so the basically we using this multi hub at the ad to lead years at the ad arm delay or so we can allow picker can start the TCP connection, Egyptian and possible, using the multi hub, adding network and the impact on network. So we extended the connectivity, so we probably using the simulation and okay, so the appendices are explained and this website can show the video clip and github. That's it.

F

Thank you yeah. So if you have a question fully, you can discuss so Paschal. You have yeah short time.

G

Yes, if you want one, are you talking that? Are you saying that you're flooding the dead of are multiple Hopf, so you're actually doing a.

C

G

Flood for that, well, that's the one thing we just did not want it absolutely to avoid in the type of networks that.

C

G

Been working on right because, because of the bandwidth that the flooding cost you know the flooding well, each broadcast is that at the slowest speed possible- and if you have to flood it, that means you're, basically consuming all the bandwidth for everybody time you want to form or just check in your dress and well, many environments is just not acceptable. They can't see that it would be acceptable. The CD I, don't just don't see why it would be acceptable because not you get it. Of course.

G

The only thing you will see is that on your network, there is nobody else or anything else, not not quite a good idea. So the way six load has that of our multiple hearts is unicast, so you need to have for a given subnet a registry, and you just unicast check that registry. That's how you avoid the flood, because the flood with just one call it looks great. Then you have one thousand cars. You know nothing works. Okay,.

F

So, inter pastor the adding load mark are he suggested we need to consider a multi up the ad with the unicast. So that's why we implemented some possible solution, but once we, the penises promise statement document can discuss it not just example.

E

Do you plan to do some real OCB.

E

F

Definitely yes, my group already have the Kota Wireless our Cu and ODU, so we will test paper protocol. Okay, thank you, yeah, so yeah.

C

Yes, thank you for pointing out to the hackathon, so we made the two hackathons in the past, one in Dakar and second one in Kampala last month during the Africa internet summit. So we got two good results because we are using not simulation but real cards and to test if there is possibility to to connect and to send pings and we use in Wireshark to capture and to see the the frame and all fields and all details, that's what we want also to do next time. Okay,.

E

Thanks, thank you I guess we don't have time for for the presentation if they I mean for.

C

Those are interested on.

E

That Piscotty within six month I mean six long, six robot a like a standard version right.

I

E

You can take the video.

J

E

I

Sure, yeah, okay, okay,.

F

E

We are done Thanks. Where are the blue seats.

F

Okay, thank you.