On 2013-05-27 13:30, Chris Bagnall wrote:
On 27/5/13 6:18 pm, Zach Underwood wrote:
network 216.105.159.0/24
network 216.105.158.0/24
network 216.105.157.0/24
network 216.105.128.0/24
network 216.105.135.0/24
network 216.105.136.0/23
network 216.105.141.0/24
network 216.105.143.0/24
network 216.105.144.0/22
network 216.105.153.0/24
network 216.105.155.0/24
network 216.105.156.0/24
network 216.105.151.0/24
Not related to the issue you're experiencing, but why don't you
aggregate some of those?
That's an excellent point... your upstreams may aggregate for you, but
the bigger the advertised netblock, the better.
Oh. What maniac gave you *that* allocation? Ah, I see, you're only
advertising some of your routes on this router; I assume you are
advertising the rest of your address space from another router
somewhere?
Still, there is one optimization you can do: replace 216.105.156.0/24
through 216.105.159.0/24 with 216.105.156.0/22. That shrinks your
prefix list by 3. If everyone on the 'net did that, we'd all need less
RAM to hold routing tables.
If you aren't advertising the rest of your address space elsewhere, you
have to choose between local or global optimization. To optimize
locally, you're doing it right. To optimize globally, you should be
advertising the largest prefixes possible, e.g. 216.105.128.0/23 instead
of /24. That causes a lot of traffic to reach you that you'll
immediately drop, however.
-Adam Thompson
[email protected]
_______________________________________________
List mailing list
[email protected]
http://lists.pfsense.org/mailman/listinfo/list
