Yesterday, Job Snijders presented the NLNOG RING at the RIPE 65 Plenary meeting. In his presentation he illustrated how fast the RING has grown (137 hosts in 123 networks and 27 countries at this moment) and, more importantly, how powerful the RING has become. Various tools make debugging network related problems a lot easier. In his presentation Job also demonstrated the new (currently beta) AMP tool, which can be used to detect various problems like packetloss, MTU issues and routability problems.
People visiting RIPE65 please visit the NLNOG RING BoF drink on Thursday at 18.00h. Location is tentatively Grand Ballroom IV-V but may change, please check the RIPE BoF page for confirmation.
One of the nice features of the ring is the ability to view routing information from a large (and growing!) number of sources all around te world. We can combine this information to create cool things, like graphs which show the similarities and differences in traceroutes towards a common destination. This is exactly what
ring-trace does. More detailed information on how to use
ring-trace can be found on the Toolbox page.
Here are a few examples of graphs generated by ring-trace which show some interesting things:
The traces towards
www.apple.comshow indications on how Apple implemented global loadbalancing. The trace towards
www.telstra.netis a nice example of a host “far away”, which gives us some indications of the transit providers used by various networks. The traces towards
www.arin.netshow a somewhat similar picture. The
www.ripe.netshow some IPv6 traces with ‘broken’ hops in it.
ring-trace can be downloaded here. It requires
python-dnspython packages on Ubuntu. Bug reports, fixes and suggestions for additional code are welcome of course, as well as graphs of the most awesome traces around the world.