i have a perl script which takes in the form
133.10.0.0
133.11.0.0
155.162.0.0/15
155.164.0.0/14
and tells me that i can aggregate the first two into a /15.
unfortunately, the actual data as paths can differ
133.10.0.0 2914 2907
133.11.0.0 2914 2907 2501
155.162.0.0/15 2914
155.164.0.0/14 2914
so, what i would like is a tool which takes something like the latter syntax
and tells me what the origin ass could aggregate. i.e. i am specifically
not inclined to proxy aggregation at this point.
Maybe CIDRadvisor does this, but I have been unable to make it do so.
randy
I once started a project that would do this, which would
allow realtime aggregation by speaking ebgp to a router. I've not had
time to play with it too much, but it does bring up a bgp session and
receive routes properly.
If I ever finish it, I'll run it and put up a telnet/http interface
to it.
I would also check out Tony Bates CIDR report, as you can do
interesting things with that data, it will tell you if you can
aggregatte better.
- jared
Randy Bush (randy@psg.com) on November 28:
i have a perl script which takes in the form
133.10.0.0
133.11.0.0
155.162.0.0/15
155.164.0.0/14
and tells me that i can aggregate the first two into a /15.
unfortunately, the actual data as paths can differ
133.10.0.0 2914 2907
133.11.0.0 2914 2907 2501
155.162.0.0/15 2914
155.164.0.0/14 2914
so, what i would like is a tool which takes something like the latter syntax
and tells me what the origin ass could aggregate. i.e. i am specifically
not inclined to proxy aggregation at this point.
Maybe CIDRadvisor does this, but I have been unable to make it do so.
CIDRadvisor requires aut-num's to be populated to figure out differing
as paths. Given the state of aut-num's in IRR, it is not a very useful
tool.
One could use as paths from a BGP route table dump. But that is only
an estimate, as two routes flap the as paths may diverge differently.
randy
Cengiz