[c-nsp] Rate-Limit Problem

Paolo Lucente pl+list at pmacct.net
Thu Dec 20 16:44:56 EST 2007


Hi Paul,

also consider the access rate (100Mbps) is much higher than your CIR (some
4Mbps); this means that TCP tests will perform badly unless you introduce
some sort of egress shaping instead of that brutal rate-limiting which just
drops non-conforming packets. Rate-limiting and policing are generally
suitable egress only for UDP and multicast traffic. They are also suitable
ingress to avoid customers pumping too much traffic at the edges (in which
case, again if the access rate of the medium is higher than the CIR, they
should be shaping outbound).

Cheers,
Paolo


On Thu, Dec 20, 2007 at 03:46:05PM -0500, Paul Stewart wrote:
> Can anyone tell me why this doesn't work?  We have this implemented in
> dozens of other locations and it works fine...
> 
> Cisco 2621 router
> 
> interface FastEthernet0/0.10
>  description Office LAN
>  encapsulation dot1Q 10
>  ip address 216.168.114.57 255.255.255.248
>  rate-limit input 4200000 2100 2100 conform-action transmit exceed-action
> drop
>  rate-limit output 4200000 2100 2100 conform-action transmit exceed-action
> drop
> 
> 
> When we run tests, we can barely get 200Kb/s up or down....??
> 
> Thanks,
> 
> Paul
> 
> 
> No virus found in this outgoing message.
> Checked by AVG Free Edition. 
> Version: 7.5.503 / Virus Database: 269.17.5/1190 - Release Date: 12/19/2007
> 7:37 PM
>  
> 
> _______________________________________________
> 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