Fails to build / Any development? / Licences
tim.streater at dante.org.uk
Mon Nov 7 10:06:35 UTC 2005
At 11:10 06/11/2005, Marco d'Itri wrote:
>On Nov 06, Francesco Ferreri <f.ferreri at namex.it> wrote:
>> But I think it's worth considering that IRRToolSet is quite an ambitious
>> software project, with a rich set of features and tasks which may
>This is an important point: how many people are actually using
>IRRToolSet to build complete configurations using the IRR as the
>authoritative source of data about *their own* network?
>Even ignoring objects like peering-set and inet-rtr, how many people
>build their configurations by taking the list of their peers from their
>aut-num object instead of from a local database?
>I think that a more realistic goal would be to use the IRR to gather
>data about other people's networks, which is more practical to deploy
>and much easier to support by tools since the only relevant objects are
>route, inetnum, as-set and rs-set.
We expect our customers (European national research networks) to keep their route: object entries up to date in RIPE, and we generate prefix lists for the routes we accept from them, based on these entries. Our commercial upstreams do the same for us. Of course, this is only a portion of our configuration, but its a well-defined portion.
More information about the irrtoolset