Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR, but I have a question I want to throw out there: Do you think IPv6 geolocatoin (GeoIP) will ever be viable? If so, when do you think this will happen? If not, what's the superseding solution? (The W3C location technology fails miserably for me 100% of the time even on IPv4). Two of the "big four" GeoIP providers don't even catalog IPv6, and the other two's IPv6 database is unremarkable and usually only has the country. (Or, in my case, a block that's clearly in the United States is deemed as simply "(somewhere in) Asia".) What I'm getting at is: IPv6 geolocation is presently rather hopeless and useless. Eager to hear thoughts from my fellow network thinkers! - Blair
On 2013-10-22, at 15:16, Blair Trosper <blair.trosper@gmail.com> wrote:
Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR,
Not sure what that means, but...
but I have a question I want to throw out there:
Do you think IPv6 geolocatoin (GeoIP) will ever be viable?
To me it seems like an easier problem to solve than IPv4. There's no historical assignment swamp. Subnets are of fixed size. Many/most organisations who receive a direct assignment will never need a second.
If so, when do you think this will happen?
As soon as enough people using geo-located services start doing so over v6. Joe
I meant that PTR isn't a priority for ISPs. A la Comcast's rollout of IPv6 lacks PTR, as does Google in general for v4 and v6 (even though they have it internally). On Tue, Oct 22, 2013 at 2:21 PM, Joe Abley <jabley@hopcount.ca> wrote:
On 2013-10-22, at 15:16, Blair Trosper <blair.trosper@gmail.com> wrote:
Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR,
Not sure what that means, but...
but I have a question I want to throw out there:
Do you think IPv6 geolocatoin (GeoIP) will ever be viable?
To me it seems like an easier problem to solve than IPv4. There's no historical assignment swamp. Subnets are of fixed size. Many/most organisations who receive a direct assignment will never need a second.
If so, when do you think this will happen?
As soon as enough people using geo-located services start doing so over v6.
Joe
it seems like solving your first complaint is the same work as solving your second: 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.8.b.d.0.1.0.0.2.ip6.arpa. 1h IN PTR host.example.com. 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.8.b.d.0.1.0.0.2.ip6.arpa. 1h IN LOC 40 45 33 N 73 59 07 W 100m ________________________________________ From: Blair Trosper [blair.trosper@gmail.com] Sent: Tuesday, October 22, 2013 3:16 PM To: nanog@nanog.org Subject: ipv6 and geolocation Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR, but I have a question I want to throw out there: Do you think IPv6 geolocatoin (GeoIP) will ever be viable? If so, when do you think this will happen? If not, what's the superseding solution? (The W3C location technology fails miserably for me 100% of the time even on IPv4). Two of the "big four" GeoIP providers don't even catalog IPv6, and the other two's IPv6 database is unremarkable and usually only has the country. (Or, in my case, a block that's clearly in the United States is deemed as simply "(somewhere in) Asia".) What I'm getting at is: IPv6 geolocation is presently rather hopeless and useless. Eager to hear thoughts from my fellow network thinkers! - Blair
On 2013-10-22 21:16, Blair Trosper wrote:
Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR, but I have a question I want to throw out there:
Do you think IPv6 geolocatoin (GeoIP) will ever be viable?
Yes, in the same way as it happens for IPv4: customer types their address into the database for a geo-provider when they type it in to get stuff shipped out to them... Most consumer/hard-line ISPs typically have their users in the same country/region as they operate, hence geo-location up to city level will be 'easy'. For VPN providers or more specifically IPv6-tunnel providers that is not the case, the user might be in a completely different country than the PoP is, or where the address space for that PoP comes from. As such, for SixXS we are using the "Self-published IP Geolocation Data" specification as defined in this draft by Google folks: http://tools.ietf.org/html/draft-google-self-published-geofeeds-02 This resolves this problem for us. More details about this and where to find the feed etc can be found at: http://www.sixxs.net/faq/misc/?faq=geolocation As mentioned there, as a content-provider, please use that data, and if you want do also please send a notification so that we can either list you on the above page and/or at least notify you in case things change. Note that most VPN providers actually are more 'geo-location changers' and thus likely will not want to do this, or will want to "lie" in their data, for them I don't think that providers will be accepting their feeds.
What I'm getting at is: IPv6 geolocation is presently rather hopeless and useless.
One very simple approach is to take RIR data, you then end up with a reasonably accurate location, unless, like in the above detail the traffic actually is tunneled from somewhere else. If wanted I can make a geo-feed available containing the data from GRH, as that has all these little details already anyway. If somebody finds it useful, give a yell, and I'll kick the system to produce one (separately from the SixXS specific prefixes of course, thus under a different URL). Greets, Jeroen
On Tue, Oct 22, 2013 at 9:16 PM, Blair Trosper <blair.trosper@gmail.com> wrote:
Everyone loves IPv6, and it's a fantastic technology. However, I've been pondering a few quirks of v6, including the low priority of PTR, but I have a question I want to throw out there:
Do you think IPv6 geolocatoin (GeoIP) will ever be viable?
You might find this interesting: https://ripe67.ripe.net/presentations/324-self-published-geo_RIPE67.pdf http://tools.ietf.org/html/draft-google-self-published-geofeeds-02
participants (5)
-
Blair Trosper -
Ian Smith -
Jen Linkova -
Jeroen Massar -
Joe Abley