[c-nsp] CoPP class-default policer

[Rodney Dunn]

I just happened to see this asked by someone else.

http://www.cisco.com/univercd/cc/td/doc/product/lan/cat6000/122sx/swcg/qos.htm#wp1171634

class-default isn't yet supported on the 65xx.

You can create a class with an ACL and match all traffic
to simulate the default class.  It will just be all
IP traffic though.

In future code this restriction should be removed.
I don't have any exact timeframes for you.

Rodney

On Wed, Dec 08, 2004 at 03:08:46PM -0600, John Kristoff wrote:
> I've got a 6509 with a 720 and 12.2(18)SXD2 and am playing around with
> the CoPP stuff.  I've used this page as a rough guide and my test config
> is not too much different than it:
> 
>   <http://www.cisco.com/en/US/products/sw/iosswrel/ps1838/products_white_paper09186a0080211f39.shtml>
> 
> However, I run into a problem when I try to apply a specific policer
> to the class-default in the policy-map.  As an example, this is what
> I get:
> 
>   Router(config-pmap-c)#police 32000 6000 12000 conform-action transmit exceed-action drop
>   Info: pir/max-burst ignored
>   Router(config-pmap-c)#exit
>   Router(config-pmap)#exit
>   Warning (QoS): Control Plane Interface: Problem installing policy [control-plane-policy] with class class-default. Class can only have
>   'match access-group' OR
>   'match ip dscp' OR 'match ip precedence' OR
>   'match dscp' OR 'match precedence' OR 'match mpls exp' OR
>   'match dscp' OR 'match precedence' with one of'match protocol ip' OR
>   'match protocol ipv6'
> 
> I suspect I'm doing something wrong, but this CoPP stuff is new to me so
> I'm a bit confused as to what that might be.  Anyone recognize this problem
> and know what to look for to get around it?
> 
> FYI... apparently I'm also not calculating the max-burst correctly.  The
> calculation for those values have apparently changed from earlier uses of
> the rate limiter commands I'm used to.  I've tried it with other values
> that don't get the 'Info: pir/max-burst ignored' message and it doesn't
> seem to make a difference.
> 
> Thanks for any insight.
> 
> John
> _______________________________________________
> 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/