Hi James, Thank you for your valuable feedback, and see below for responses:
> First my understanding of this solution is that is BGP > underlay, BGP Overlay.. Is this correct? It is a BGP overlay over an underlying connected wide area network such as the Internet. So, yes, it is BGP in both the overlay and underlay. >If so, what is the ask in Section 6 last paragraph? If it is to create >multiple instances >of BGP would that be using Local-AS, Dual-AS etc… or to actually to create two >BGP >Instances each with their own AS? We have had that discussion at IETF and it >is similar >to an ask I have where I am using BGP 3107 as underlay and 2547, VPN etc… as >overlay.. The idea is that the overlay BGP instance will carry only Mobile Network Prefix (MNP) routes, while the underlay BGP instance carries all of the routes in the Internet. If we join the two instances, the overlay portion would only advertise one or a few aggregated prefixes into the underlay portion – it would not advertise all of the more-specific MNPs. >There is no discussion of the actual topology in terms of s-ASBRs and >C-ASBRs.. Is >there some notion of geography, nation, continent etc… s-ASBRs will be regionally distributed, e.g., some in Europe, some in the US, some in Asia, etc. And, end systems choose which s-ASBRs to associate with based on a map that tells which ones are closest. So, the end system chooses a nearby s-ASBR and if it moves far from the s-ASBR it hands over to a different s-ASBR. >Sect 3 ( Top of Pg 8 ) > >“Since ATN/IPS end systems …” > >You indicate that airplanes remain in the stub area for an extended period of >time >and therefore intra-domain mobility events are handled in the stub area only. >How >is this possible? There must come a time where the MNP for said aircraft will >not >be available via a given cell tower or cellular provider. What am I missing >here.. The s-ASBR keeps track of the airplane’s current data link addresses and associates these with the airplane’s IPv6 MNP. To get packets to the airplane, the s-ASBR uses encapsulation with MNP addresses in the encapsulated packet header and data link addresses in the encapsulation header. The airplane may have multiple data links, and it may be handing over between cell towers and/or service providers as it moves. But, the s-ASBR never reports these kinds of mobility events to the c-ASBRs via BGP updates, and so mobility-related churn is kept out of the BGP routing system. >Sect 3 ( Middle of Pg 8 ) > >It seems that you are going to apply route filters on the c-asbrs so that a >given set >of them will service a given set of MSPs. No, the c-ASBRs have full topology knowledge of all MNPs in the entire system. The s-ASBRs only have knowledge of the MNPs they are currently serving. So I think yes when I set up the implementation I used route filters. But, what they did was allow the s-ASBRs to tell the c-ASBRs everything and did not allow the c-ASBRs to tell the s-ASBRs anything. > a) Are these filters applied inbound/outbound >b) Is this generally static or will operations be mucking with these often. >More > hands equals more problems is my experience. >c)What happens if mis-config? What is the blast radius of getting this wrong. In terms of filters, we have it that the c-ASBRs originate “default” but do not advertise any MNPs to any of the s-ASBRs. And, the s-ASBRs tell their peer c-ASBRs about every MNP they are currently servicing. In terms of static, c-ASBRs and s-ASBRs only need to be configured once, and then the configuration would only be changed if new c/s-ASBRS are added or old c/s-ASBRs are decommissioned. So, I think the configuration footprint would be similar to the way the global Internet works but at a much smaller scale – for example, there would probably only be about 100 or so s-ASBRs and 1o or so c-ASBRs. In terms of misconfigurations, if a peering between an s-ASBR and c-ASBR is not set up correctly all of the end systems that associate with the (defunct) s-ASBR will not receive service. But, in that case, the end systems can simply choose a different s-ASBR. I hope this addresses your questions. If not, I would be happy to iterate. Thanks for your time. Fred From: UTTARO, JAMES [mailto:[email protected]] Sent: Wednesday, March 14, 2018 12:10 PM To: Templin, Fred L <[email protected]>; Christopher Morrow <[email protected]> Cc: [email protected]; Saccone, Gregory T <[email protected]>; Gaurav Dawra <[email protected]> Subject: RE: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network Fred, Took a read on the draft.. A few comments below.. First my understanding of this solution is that is BGP underlay, BGP Overlay.. Is this correct? If so, what is the ask in Section 6 last paragraph? If it is to create multiple instances of BGP would that be using Local-AS, Dual-AS etc… or to actually to create two BGP Instances each with their own AS? We have had that discussion at IETF and it is similar to an ask I have where I am using BGP 3107 as underlay and 2547, VPN etc… as overlay.. There is no discussion of the actual topology in terms of s-ASBRs and C-ASBRs.. Is there some notion of geography, nation, continent etc… Sect 3 ( Top of Pg 8 ) “Since ATN/IPS end systems …” You indicate that airplanes remain in the stub area for an extended period of time and therefore intra-domain mobility events are handled in the stub area only. How is this possible? There must come a time where the MNP for said aircraft will not be available via a given cell tower or cellular provider. What am I missing here.. Sect 3 ( Middle of Pg 8 ) It seems that you are going to apply route filters on the c-asbrs so that a given set of them will service a given set of MSPs. a) Are these filters applied inbound/outbound b) Is this generally static or will operations be mucking with these often. More hands equals more problems is my experience. c) What happens if mis-config? What is the blast radius of getting this wrong. Thanks, Jim Uttaro From: GROW [mailto:[email protected]] On Behalf Of Templin, Fred L Sent: Wednesday, March 14, 2018 2:19 PM To: Christopher Morrow <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; Saccone, Gregory T <[email protected]<mailto:[email protected]>>; Gaurav Dawra <[email protected]<mailto:[email protected]>> Subject: Re: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network Chris, I forgot to mention that one of the key requirements is that there be no dynamic routing protocol running over the air-to-ground data links. The data links we are talking about have data rates as low as 1Mbps and even lower, and the civil aviation community has declared that the control message overhead must be kept to a minimum. So, placing a BGP speaker on-board the airplane would not be acceptable. Is there interest in having a presentation about this in London next week? Thanks - Fred From: GROW [mailto:[email protected]] On Behalf Of Templin, Fred L Sent: Tuesday, March 13, 2018 8:57 AM To: Christopher Morrow <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; Saccone, Gregory T <[email protected]<mailto:[email protected]>>; Gaurav Dawra <[email protected]<mailto:[email protected]>> Subject: Re: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network Hi Chris, >it's bad for bgp on the global scale, but in a VPN scenario you're talking >about ~10k routes? (number of planes concurrently in the air) and transitions >at a rate of 100/second? 500/second? (what rate is >expected at 10k planes? at >100k planes?) The model is that each airplane gets one or more IPv6 prefixes and acts as a mobile network. So, it has a mobile router on board, and uses the IPv6 prefixes to number its downstream-attached devices and networks – an airborne Internet of Things. The IPv6 prefixes stay the same wherever the plane roams to (more on that below). But, the plane’s underlying data link connections can be changing very dynamically, e.g., switch from SATCOM to cellular, update QoS due to signal fading, etc. >For quick/dirty numbers: >https://www.telegraph.co.uk/travel/travel-truths/how-many-planes-are-there-in-the-world/<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.telegraph.co.uk_travel_travel-2Dtruths_how-2Dmany-2Dplanes-2Dare-2Dthere-2Din-2Dthe-2Dworld_&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=3qhKphE8RnwJQ6u8MrAGeA&m=1in3Oz4sy5FNjimSBTapMQePTjFvBG-cVosjFEOCeoc&s=WQSPIejYgOPpbdGwqVI7BM6J4EqZPOSmpKsBMoFyyfo&e=> > >says there are 25k planes (round numbers) planes that I think qualify in your >pool. You are very correct to check on the current numbers of planes. For civil aviation, we currently see tens of thousands. But, the system should be flexible to support several orders of magnitude more than that with the multitudes of unmanned aircraft expected to be coming into the airspace in the near future. >why would you change ip addressing on the plane? having them keep their >addressing seems simpler and more conducive to stability, no? Right, the airplane’s on-board IPv6 prefixes used for downstream IoT addressing never change. It is the plane’s upstream data link addresses that can change dynamically, i.e., in the same way that a cellphone’s WiFi and/or 4G addresses can change. Again, the design is to keep mobility-related churn out of BGP in the core of the network and to keep the churn out in the edges of the network. Thanks - Fred From: Christopher Morrow [mailto:[email protected]] Sent: Tuesday, March 13, 2018 8:24 AM To: Templin, Fred L <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; Saccone, Gregory T <[email protected]<mailto:[email protected]>>; Gaurav Dawra <[email protected]<mailto:[email protected]>>; Acee Lindem (acee) <[email protected]<mailto:[email protected]>> Subject: Re: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network On Tue, Mar 13, 2018 at 10:58 AM, Templin, Fred L <[email protected]<mailto:[email protected]>> wrote: Hi Chris, >yup, sure what I was proposing is that they DO participate. >I could see a world where the plane has a simple BGP instance, and some >orchestration does the equivalent of the mobile cell hand-off for hand-devices: > "about to leave AS1, start peering with AS2, ... drop peering with AS1" With the Connexion By Boeing experience, we have proof that frequent injections and withdrawals of prefixes due to mobility is bad for BGP. Plus, aircraft are it's bad for bgp on the global scale, but in a VPN scenario you're talking about ~10k routes? (number of planes concurrently in the air) and transitions at a rate of 100/second? 500/second? (what rate is expected at 10k planes? at 100k planes?) For quick/dirty numbers: https://www.telegraph.co.uk/travel/travel-truths/how-many-planes-are-there-in-the-world/<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.telegraph.co.uk_travel_travel-2Dtruths_how-2Dmany-2Dplanes-2Dare-2Dthere-2Din-2Dthe-2Dworld_&d=DwMGaQ&c=LFYZ-o9_HUMeMTSQicvjIg&r=3qhKphE8RnwJQ6u8MrAGeA&m=1in3Oz4sy5FNjimSBTapMQePTjFvBG-cVosjFEOCeoc&s=WQSPIejYgOPpbdGwqVI7BM6J4EqZPOSmpKsBMoFyyfo&e=> says there are 25k planes (round numbers) planes that I think qualify in your pool. multi-link entities that often have multiple data links operational simultaneously, where each data link connects to a data link service provider network that may know nothing about BGP internally. And, we want to avoid tunneling over the low data rate wireless data links themselves. >I imagine each plane could even maintain more than one live BGP session with >the ground stations, even. It's good to hear that the expected churn is low, >that makes 'plane based bgp' even more >attractive (to me anyway). No, the aircraft could be moving into and out of service range dynamically, changing their IP addresses frequently, switching between available data links, and updating why would you change ip addressing on the plane? having them keep their addressing seems simpler and more conducive to stability, no? their QoS conditions, e.g., based on signal strength changes. So, there is a lot going on from a mobility standpoint, but the architecture in our doc prevents that from percolating up into BGP. >Again this still sounds like /2547 mpls vpn/ sorts of stuff, not something >super related to grow's 'global routing (internet focused) operations' area, >is it? Again, this is a simple BGP arrangement – no RFC2547, no mpls, etc. About whether or not it is related to grow, that’s what we’re here to find out. ok, other folk ought to chime in then. Thanks - Fred >in a 2547 sort of scenario (any of the vpn overlays really) the carrier >network doesn't have to know anything at all about the vpn content or routes. From: Christopher Morrow [mailto:[email protected]<mailto:[email protected]>] Sent: Monday, March 12, 2018 7:16 PM To: Templin, Fred L <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; Saccone, Gregory T <[email protected]<mailto:[email protected]>>; Gaurav Dawra <[email protected]<mailto:[email protected]>> Subject: Re: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network On Mon, Mar 12, 2018 at 4:59 PM, Templin, Fred L <[email protected]<mailto:[email protected]>> wrote: Hi Chris, Thanks for the comments, but no the planes (as Clients) do not do BGP; only the ground-domain Servers and Relays do BGP. Servers are ASBRs for stub ASes and connect to Relays that are ASBRs for a core AS in a hub-and-spokes fashion. When a plane contacts a Server, it becomes part of that Server’s stub AS. And, because planes do not move rapidly from Server to Server, the amount of mobility-related BGP update churn as seen by the core AS is dampened. But, the planes themselves do not participate in BGP, and are therefore not mobile ASes. yup, sure what I was proposing is that they DO participate. I could see a world where the plane has a simple BGP instance, and some orchestration does the equivalent of the mobile cell hand-off for hand-devices: "about to leave AS1, start peering with AS2, ... drop peering with AS1" I imagine each plane could even maintain more than one live BGP session with the ground stations, even. It's good to hear that the expected churn is low, that makes 'plane based bgp' even more attractive (to me anyway). Again this still sounds like /2547 mpls vpn/ sorts of stuff, not something super related to grow's 'global routing (internet focused) operations' area, is it? in a 2547 sort of scenario (any of the vpn overlays really) the carrier network doesn't have to know anything at all about the vpn content or routes. Thanks - Fred From: Christopher Morrow [mailto:[email protected]<mailto:[email protected]>] Sent: Monday, March 12, 2018 12:31 PM To: Templin, Fred L <[email protected]<mailto:[email protected]>> Cc: [email protected]<mailto:[email protected]>; Saccone, Gregory T <[email protected]<mailto:[email protected]>>; Gaurav Dawra <[email protected]<mailto:[email protected]>> Subject: Re: [GROW] A Simple BGP-based Mobile Routing System for the Aeronautical Telecommunications Network (as a normal participant) On Thu, Mar 8, 2018 at 3:14 PM, Templin, Fred L <[email protected]<mailto:[email protected]>> wrote: Hello, We have published a document that proposes BGP as the core of a mobile routing service for worldwide civil aviation in the Aeronautical Telecommunications Network with Internet Protocol Services (ATN/IPS). This would be an overlay network deployment of standard BGP with ASes arranged in such a way as to mitigate the mobility-related instability that was inherent in past approaches. The system also leverages an adjunct route optimization service known as AERO. The ATN/IPS is planned to eventually replace existing air traffic management services with an IPv6-based service as part of a long-term evolution. The choice of mobile routing services is being made now, with this approach, LISP and Mobile IPv6 as candidates. Although the decision is being considered in the International Civil Aviation Organization (ICAO), we feel the time is right to socialize the effort in the IETF. Hey, much of this document reads like: "hey, the global internet is messy, and slowish, we think making our own bgp domain will make that problem go away" Followed by what smells a lot like any old RFC2547 MPLS VPN deployment. I'm not sure I buy the need for 'ip mobility' in a world where the plane COULD be a BGP speaker and just negotiate upstream connectivity 'in real time'... but overall this just sounds like any other 2547 deployment to me? You'd have to convince your constituent parts that depending upon various providers 2547 interconnection agreements to work out properly is sane/useful/cost-effective/not-prone-to-explosion... but ... sure, make a 2547 network, make the planes do bgp, and orchestrate the add/remove peerings part across the network as planes move around.
_______________________________________________ GROW mailing list [email protected] https://www.ietf.org/mailman/listinfo/grow
