Good talk by Emile on visualisation of traceroutes and the various sources of geo-loc that can be munged and derived to try and give more accurate info. While there was talk of commercial geo-loc such as Maxmind, what about mining the DNS for LOC records? Has there been any investigation of that? Or am I just an ossified dinosaur, talking about LOC records? :) Cheers, Mike
Hi, On Thu, Oct 17, 2013 at 03:14:18PM +0100, Mike Hughes wrote:
Or am I just an ossified dinosaur, talking about LOC records? :)
I can still remember them. Which might make me an ossified dinosaur as well :-) - we have never very actively deployed them, as it's work, and the benefit was small. It might be time to re-think that - and one of the question would be "where to put them"? In the reverse zone (so mapping "ip-address in traceroute -> LOC" would be straightforward), or in the forward zone - where we usually don't have forward records for the "interface specific" PTRs... ("router-X-interface-Y" is not there, while "router-X" would be)... Gert Doering -- dinosaur -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
On 17/10/2013 17:14, Mike Hughes wrote:
Good talk by Emile on visualisation of traceroutes and the various sources of geo-loc that can be munged and derived to try and give more accurate info.
Thanks!
While there was talk of commercial geo-loc such as Maxmind, what about mining the DNS for LOC records? Has there been any investigation of that?
We haven't looked into that, but I remember people mentioning that it never really took off. If nobody else has done work on this recently, we could certainly take a stab at looking into that as an additional source of data and/or as a means of having people providing their geoloc info.
Or am I just an ossified dinosaur, talking about LOC records? :)
maybe ;) best regards, Emile Aben RIPE NCC
* Mike Hughes
Good talk by Emile on visualisation of traceroutes and the various sources of geo-loc that can be munged and derived to try and give more accurate info.
+1 I was wondering if there's any tricks available to figure out the actual hops (and in turn geolocate them) from those transit networks that are using MPLS with core hiding enabled, like AS3549? Tore
On 2013/10/17 16:43 , Tore Anderson wrote:
I was wondering if there's any tricks available to figure out the actual hops (and in turn geolocate them) from those transit networks that are using MPLS with core hiding enabled, like AS3549?
Typically, traceroute will reveal the number of MPLS hops. The MPLS labels in error ICMPs do not contain any kind of router identifier, so pin pointing them will be difficult. By creating a full mesh through the MPLS core it may possible to statistically show a minimum number of routers needed to support the measurements. But I have no idea how tricky that measurement and analysis will be. Philip
participants (5)
-
Emile Aben -
Gert Doering -
Mike Hughes -
Philip Homburg -
Tore Anderson