[f-nsp] Question about MLX/XMR capabilities

[Allan Eising]

Right, thank you all very much for your explanations. It has been most helpful.

Regarding rate-limiting versus shaping, I'm interested in both
limiting customer IP traffic as well as adding various Quality of
Service profiles in order to do stuff with voice traffic.

I've contacted a SE guy, so I'm waiting to get their official reply to
these questions, but at the moment, I have the impression that the
only way to solve the overlapping vlan/QinQ issue would be one of the
following:

 - a pair CES2000 that does vlan translation based on ESI
 - a pair of MPLS PE routers that takes the vlan trunks and converts
them into VPLS's for the core routers to route.

None of the solutions are particularly elegant, and at least the first
one would double the price...

Anyway, thanks for your help!

Allan

On Wed, Jan 28, 2009 at 10:18 AM, Colin Horsington <colin at horsington.com> wrote:
> Tomek,
>
> That is true, but my assumption was that he Allan wanted to simply limit
> customer IP traffic.  You can apply the shaping per queue as well.
>
> Colin
>
>
> On 28/01/2009, at 6:37 PM, Tomasz Szewczyk wrote:
>
>> Colin,
>>
>> Rate-limit is different feature than traffic shaping. You can apply VALN
>> based RL, but as far as I know on XMRs it is not possible for traffic
>> shaping. Traffic shaping can be applied only per-port.
>>
>> Tomek
>>>
>>> If you then wish to rate-limit all traffic associated with this
>>> sub-interface, in this case VLAN 100 traffic you can perform the following
>>> on interface 4/5.
>>>
>>> The below example rate-limits VLAN 100 to 100Kbps with a 10Kbps burst.
>>>
>>> interface ethernet 4/5
>>>  rate-limit input vlan-id 100 100000 10000
>>>  rate-limit output vlan-id 100 100000 10000
>>>
>>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>