[c-nsp] VLSM

Stephen J. Wilcox steve at telecomplete.co.uk
Tue Jan 11 13:30:42 EST 2005


Hi Mark,
 that isnt specifically the result of enabling classless behaviour, in classful 
mode you are still using the last resort just that it assumes any 172.16.0.0 
Class B traffic stays internal so if theres no route its dropped.

Just route 172.16.0.0/16 to null0 and it will drop the packets...

Alternatively route 0.0.0.0/0 to null0 and only configure routes on your last 
resort gateway for known active destinations (whichever of these produces the 
tidiest config)

Of course, you've not mentioned why if you have the .1 and .2 /24s in use but 
not the .3 is traffic being sent to it in such large quantities as to be 
noticable?

Steve

On Tue, 11 Jan 2005, Mark Persiko wrote:

> If you have a hierarchical network topology with the gateway of last
> resort facing  one egress point at the core, then I've noticed that "ip
> classless" has the unfortunate side effect of sending all traffic out
> that egress pipe, whose destination is for subnets that aren't used
> within the network.  That is my condition right now and I am burning
> bandwidth, and seeking a solution!
> 
> In other words, if you are using 172.16.1/24 and 172.16.2/24, and you
> have packets headed for 172.16.3/24, "ip classless" will make them head
> to the gateway of last resort instead of just dropping them.
> 
> I am using EIGRP in my network.  I have it in mind to turn off "ip
> classless."   This is one advantage of "classful routing;" traffic bound
> for any non-defined subnets are dropped.  However, in my network, I have
> other subnets of 172.16 that are beyond the EIGRP cloud, so I need to be
> able to get to them.
> 
> One solution I've considered is explicit static routing for all valid
> subnets of 172.16 and then null routing anything else for 172.16/16
> itself.  What do you think?
> 
> Thanks,
>  Mark P.
> 
> - Mark C. Persiko, Network Engineer
> - IT Division, Boulder Valley School District
> 
> 
> -----Original Message-----
> From: cisco-nsp-bounces at puck.nether.net
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Brant I. Stevens
> Sent: Monday, January 10, 2005 9:26 PM
> To: lists at hojmark.org; 'matthew zeier'; 'Gert Doering'; 'Shaun'
> Cc: cisco-nsp at puck.nether.net
> Subject: Re: [c-nsp] VLSM
> 
> While I agree that having no class (when talking about networking :B) is
> a
> good thing, "classfulness" is not completely dead...  Sometimes you
> still
> have to use RIP v1..  There's also EIGRP and BGP auto-summary using
> classful
> boundaries.
> 
> 
> On 01/10/2005 06:59 PM, "lists at hojmark.org" <lists at hojmark.org> wrote:
> 
> >> However, I can't get people (sales) to stop calling it "class c".
> > 
> > Even worse is when they (and some 'techs') call everything /24 a
> "class C"
> > and a every /16 a "class B", even when it's 10.10.10/24 and 10.10/16,
> for
> > example.
> > 
> > Yuck.
> > 
> > -A
> > 
> > _______________________________________________
> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
> 
> 
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
> 



More information about the cisco-nsp mailing list