You're viewing an archived page. It is no longer being updated.
24 October 2017, 11:00-12:30
WG co-Chairs: Remco van Mook and Florence Lavroff
Scribe: Rumy Spratley-Kanis
Status: Draft
Remco van Mook
The presentation is available at:
https://ripe75.ripe.net/archives/video/123
Vasileios Giotsas
The presentation is available at:
https://ripe75.ripe.net/archives/video/124
There were no questions
Barry O’Donovan, INEX
The presentation is available at:
https://ripe75.ripe.net/archives/video/125
There were no questions
Falk von Bornstaedt, Deutsche Telekom
The presentation is available at:
https://ripe75.ripe.net/archives/video/126
Aaron Hughes, ARIN, asked if they tried to use a known attribute to adjust their analytics to see the difference between performance. Falk replied that, yes, they have done, but it didn’t give enough insight. He advised Aaron to talk to Ruediger Volk to get a more in-depth answer.
Christoph Dietzel, DE-CIX/TU Berling
The presentation is available at:
https://ripe75.ripe.net/archives/video/127
Gert Döring asked if they tried to reach the most heavy hitters to get more insight on what they are doing and why for such a long time
Christoph replied that this was tricky. They did approach some people but they were reluctant to cooperate and wondering why he was interested in how they manage their networks – even if the data is publicly available. One person did acknowledge that they used blackholing, just to play around with it, and then forgot about it.
Gert replied that that was what he expected, and why he was so curious about it, to which Christoph answered that they didn’t do it on a large scale because it’s hard to get answers. Some mentioned that they did it to avert DDoS attacks but at the end it’s hard to pinpoint why their announcements are so long.
Martin Levy, Cloudflare, mentioned he saw value in this, and wanted to know whether blackholing should be the only community that is built into a registry, or whether it could be generalised. According to him, the Internet community should use community control within their connections to various backbones, and either documenting it or building consistency could be useful. He then asked whether this effort would be capable of expanding or bringing in other players in order to increase that registry.
Christoph replied that he was not sure about the registry part, but that he was of the same opinion. According to him it could make life a lot easier if it was possible to define and standardise more communites across each user case.
Martin agreed 100%, since using automation tools helps people getting around some of this complexity. As a closing comment he added that if something like this were to go ahead, it would be good to expand it over and above just a blackhole community, since it’s an industry-wide backbone-wide requirement. He offered his help.
Rudiger Volk commented that the place to really discuss this is the IETF Working Group. He mentioned that he has been looking forward to get more standardised functions there, since the set of potentially standardised functions is limited. The interactions signalled with communities between networks are fairly specific and it does not make a lot of sense, even if they are very similar. However, with the large communities, we could have code space where we could do interesting things like any party that gets an AS number can do their own registry of standardised communities, and they can be used between any two other parties that like the semantics.
Christoph replied that while it is possible with large communities to each do it the way they please, it still might make sense to use it the same way across different parties, to make life for the users easier.
Rudiger commented that if you are talking and thinking about standardised communites you are essentially pointing to communities that are registered by IANA. Christoph agreed. Finally Rudiger asked what would keep people from opening the AS registry for communites that are part of the Euro-IX repertoire.
Florence Lavroff closed the microphone and asked the room if they could take further questions directly with Christoph.
Florence Lavroff
The presentation is available at:
https://ripe75.ripe.net/archives/video/128
Martin Levy commended the co-chairs for the lack of IXP updates within the session.