From: Ivan Walker (ivan@itpro.co.nz)
Date: Sat Dec 13 2008 - 22:40:51 ARST
Hi Pavel,
Here is the output from a Cisco 1841 running 12.4.13b configured as per
question 2 in my original email. I believe that previously 10ms was the
lowest Tc available but now on some platforms 4ms is possible - I can't
tell if the 1841 is honoring the configuration or not but the output
below seems to suggest it may be.
1841#show policy-map interface fa0/0
FastEthernet0/0
Service-policy output: SHAPE
Class-map: class-default (match-any)
29 packets, 5806 bytes
30 second offered rate 0 bps, drop rate 0 bps
Match: any
Traffic Shaping
Target/Average Byte Sustain Excess Interval Increment
Rate Limit bits/int bits/int (ms) (bytes)
1000000/1000000 500 4000 0 4 500
Adapt Queue Packets Bytes Packets Bytes Shaping
Active Depth Delayed Delayed Active
- 0 29 5806 2 460 no
As a possible answer to question 2 I have found a post from 2003 from
Brian Dennis -
http://www.groupstudy.com/archives/ccielab/200310/msg01353.html.
"As per the engineer spec on FRTS, if the size of the packet is larger
than the max Bc bucket size, the Bc bucket will still transmit the
packet but the packet will cause the Bc bucket to go into a debt
situation. I normally do not mention this as it just creates confusion
as to the fact that shaping is allowing a debt situation to occur."
The discussion is in relation to FRTS but it is the only place I have
seen any mention of a Bc deficit in order to send packets larger than
Bc. Hopefully someone can confirm if this is the same for traffic
shaping using MQC for a non frame-relay interface. I would be
interested to learn the details of the deficit operation if this is the
case.
Cheers
Ivan
Pavel Bykov wrote:
> I think it's not possible to configure Tc lower than 10ms.
> As far as I can remember, with such configuration "show policy-map
> interface" showed bogus results.
> Can you post show policy-map with such config? In #4 output you sent
> the tc is 10ms.
>
> On Sat, Dec 13, 2008 at 3:01 AM, Ivan Walker <ivan@itpro.co.nz
> <mailto:ivan@itpro.co.nz>> wrote:
>
> 1. When shaping using MQC on a interface that also passes VoIP
> traffic (priority queuing of voice traffic via child policy) what
> are the pros and cons of using a Tc of 4ms (minimum possible as
> far as I understand) compared with a Tc of 10ms (Cisco recommended
> value for voice in some documents I have seen). For example the
> impact on CPU or possible delay.
>
> 2. If shaping is configured as follows how are packets larger than
> Bc accommodated? (no interleaving configured).
>
> CIR = 1000000bps
> Bc = 4000bits (500bytes)
> Be = 0
> Tc = 4ms
>
> 3. Again with the above shaping configuration, if only packets of
> size 400bytes were sent does that mean it will never be possible
> to utilise the full 1000000bps but rather only 400/500 * 1000000 =
> 800000bps?
>
> 4. Using IOS 12.4.13b it was possible to see the stats for delayed
> packets and other values such as Tc.
>
> router#show policy-map interface gi0/0
> GigabitEthernet0/0
>
> Service-policy output: SHAPE
>
> Class-map: class-default (match-any)
> 7522946 packets, 1961480754 bytes
> 30 second offered rate 66000 bps, drop rate 0 bps
> Match: any
> Traffic Shaping
> Target/Average Byte Sustain Excess Interval
> Increment
> Rate Limit bits/int bits/int (ms)
> (bytes) 5000000/5000000 6250 50000 0 10
> 6250
> Adapt Queue Packets Bytes Packets Bytes Shaping
> Active Depth Delayed Delayed Active
> - 0 7518711 1956530048 118002 129484844 no
>
> I can't seem to get the same stats from 12.4.22T
>
> router#show policy-map interface fa0/0 output
> FastEthernet0/0
>
> Service-policy output: SHAPE
>
> Class-map: class-default (match-any)
> 13 packets, 2184 bytes
> 30 second offered rate 0 bps, drop rate 0 bps
> Match: any
> Queueing
> queue limit 64 packets
> (queue depth/total drops/no-buffer drops) 0/0/0
> (pkts output/bytes output) 13/2366
> shape (average) cir 3000000, bc 12000, be 12000
> target shape rate 3000000
>
> Any ideas?
>
> Thanks
>
> Ivan
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>
>
>
> --
> Pavel Bykov
> ----------------
> Don't forget to help stopping the braindumps, use of which reduces
> value of your certifications. Sign the petition at
> http://www.stopbraindumps.com/
Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Thu Jan 01 2009 - 12:53:08 ARST