New measurement target suggestion: 6to4 anycast
Hello, I'd like to suggest adding to the list of monitored addresses a new one: 192.88.99.1 - 6to4 anycast IPv4 to IPv6 gateway; https://en.wikipedia.org/wiki/6to4 https://tools.ietf.org/html/rfc3068 This could provide interesting data on what percentage of networks have this address blocked/inaccessible (resulting in broken operation of 6to4), and also what is the average latency to the closest 6to4 gateway across various countries/regions. The next step would possibly be the collection of statistics which ASes 'see' whose 6to4 gateways, this can be inferred from the AS of the traceroute hop that is right before the 192.88.99.1 on the trace. -- With respect, Roman ~~~~~~~~~~~~~~~~~~~~~~~~~~~ "Stallman had a printer, with code he could not see. So he began to tinker, and set the software free."
Dear All, 6to4 should not used any longer. However it would be nice to see how 6to4 usage decreasing over the time.... Janos Mohacsi Head of HBONE+ project Network Engineer, Deputy Director of Network Planning and Projects NIIF/HUNGARNET, HUNGARY Key 70EF9882: DEC2 C685 1ED4 C95A 145F 4300 6F64 7B00 70EF 9882 On Wed, 21 Dec 2011, Roman Mamedov wrote:
Hello,
I'd like to suggest adding to the list of monitored addresses a new one:
192.88.99.1 - 6to4 anycast IPv4 to IPv6 gateway;
https://en.wikipedia.org/wiki/6to4 https://tools.ietf.org/html/rfc3068
This could provide interesting data on what percentage of networks have this address blocked/inaccessible (resulting in broken operation of 6to4), and also what is the average latency to the closest 6to4 gateway across various countries/regions.
The next step would possibly be the collection of statistics which ASes 'see' whose 6to4 gateways, this can be inferred from the AS of the traceroute hop that is right before the 192.88.99.1 on the trace.
-- With respect, Roman
~~~~~~~~~~~~~~~~~~~~~~~~~~~ "Stallman had a printer, with code he could not see. So he began to tinker, and set the software free."
On Wed, Dec 21, 2011 at 11:57 AM, Mohacsi Janos <mohacsi@niif.hu> wrote:
Dear All, 6to4 should not used any longer. However it would be nice to see how 6to4 usage decreasing over the time....
I agree. It may be interesting to see its eventual sunset, which is already showing up according to Google's stats: http://www.google.com/intl/en/ipv6/statistics/ Also, quoting Robert Kisteleki: " We are aware that in some cases the lookup has a surprising result. One example is 6to4: for IPv6 related ASNs the prefix is 2002::/16 but the ASN can be misleading. We'll most likely stop looking up ASNs for 6to4 (and Teredo) prefixes. If you see anomalies *not* related to this, you can let us know at atlas-dev@ripe.net " it is unclear (to me) how much interest and resources should be put into these measurements.
Janos Mohacsi Head of HBONE+ project Network Engineer, Deputy Director of Network Planning and Projects NIIF/HUNGARNET, HUNGARY Key 70EF9882: DEC2 C685 1ED4 C95A 145F 4300 6F64 7B00 70EF 9882
Best,
On Wed, 21 Dec 2011, Roman Mamedov wrote:
Hello,
I'd like to suggest adding to the list of monitored addresses a new one:
192.88.99.1 - 6to4 anycast IPv4 to IPv6 gateway;
https://en.wikipedia.org/wiki/6to4 https://tools.ietf.org/html/rfc3068
This could provide interesting data on what percentage of networks have this address blocked/inaccessible (resulting in broken operation of 6to4), and also what is the average latency to the closest 6to4 gateway across various countries/regions.
The next step would possibly be the collection of statistics which ASes 'see' whose 6to4 gateways, this can be inferred from the AS of the traceroute hop that is right before the 192.88.99.1 on the trace.
-- With respect, Roman
~~~~~~~~~~~~~~~~~~~~~~~~~~~ "Stallman had a printer, with code he could not see. So he began to tinker, and set the software free."
On 2011.12.21. 12:21, Iñigo Ortiz de Urbina wrote:
On Wed, Dec 21, 2011 at 11:57 AM, Mohacsi Janos <mohacsi@niif.hu> wrote:
Dear All, 6to4 should not used any longer. However it would be nice to see how 6to4 usage decreasing over the time....
I agree. It may be interesting to see its eventual sunset, which is already showing up according to Google's stats:
It seems that there is interest for doing this measurement, so we'll add it to the list -- most likely in January, after the holiday break.
Also, quoting Robert Kisteleki:
" We are aware that in some cases the lookup has a surprising result. One example is 6to4: for IPv6 related ASNs the prefix is 2002::/16 but the ASN can be misleading. We'll most likely stop looking up ASNs for 6to4 (and Teredo) prefixes. If you see anomalies *not* related to this, you can let us know at atlas-dev@ripe.net "
it is unclear (to me) how much interest and resources should be put into these measurements.
With the current setup we have enough control that setting up a new system-wide measurement is not difficult. However, we have to balance the usefulness of such measurements with the measurement capacity they consume. Regards, Robert
participants (4)
-
Iñigo Ortiz de Urbina -
Mohacsi Janos -
Robert Kisteleki -
Roman Mamedov