[c-nsp] Shaping and dialer ints 12.4(24)T vs. 15T8

Dale Shaw dale.shaw+cisco-nsp at gmail.com
Fri Jun 19 21:52:37 EDT 2009


Hi Charles,

On Sat, Jun 20, 2009 at 11:22 AM, Church, Charles<cchurc05 at harris.com> wrote:
> Can anyone confirm for me if some shaping and/or NBAR bugs were fixed
> between 24T and older 15T7 or T8?

Hmm, it doesn't directly match your scenario, but there were some new
QoS features introduced in 12.4(20)T -- most notably "Hierarchical
Queuing Framework (HQF)" -- that may have had an effect on your
configuration.

http://www.cisco.com/en/US/docs/ios/qos/configuration/guide/qos_frhqf_support.html
http://www.cisco.com/en/US/prod/collateral/iosswrel/ps6537/ps6558/white_paper_c11-481499.html

I have a similar configuration to yours on my home router, but I use a
top-level shaper to introduce backpressure. Partial config follows:

policy-map aardvark-queue
 class voice-sip
  bandwidth 20
 class voice-packet
  priority 96
 class class-default
  fair-queue
  random-detect
!
policy-map aardvark-shape
 class class-default
  shape average 1177000 11770 0
  service-policy aardvark-queue
!
interface FastEthernet0/1
 pppoe enable group global
 pppoe-client dial-pool-number 1
 service-policy output aardvark-shape
!
interface Dialer1
 bandwidth 1000
 bandwidth receive 12000
 ip address negotiated
 ip mtu 1492
 dialer pool 1
 dialer idle-timeout 0
 dialer persistent
 dialer-group 1

NB: I am running 12.4(15)T9.

cheers,
Dale


More information about the cisco-nsp mailing list