From: Gavin Lawson (GavinL@titan.net.au)
Date: Thu May 31 2007 - 02:19:08 ART
Hi mihai
The reason for this is the Hardware Queues.
When ever you apply the software queue's (i.e the LLQ) the router will
shorten the hardware queue to 2 packets.
This hardware queue will act in a FIFO fashion.
Since your background traffic is using 1500 byte packets (12000 bits)
there will most likely be 2 of these packets in the hardware queue in
front of each "priority icmp" packet.
To put the 2 1500 byte packets on the line, it will take about 360 ms,
then your "priority icmp" packet will be put on the line.
The same will apply in the reverse direction. i.e another 360 ms - then
your "priority icmp" packet.
All up about 720ms :-)
All this even though the priority queue is configured for the interface.
The way you would alleviate this delay in a voice network would be to
use LFI (Link fragmentation and Interleaving)
The protocols to do LFI, is either Frame Relay or Multilink PPP.
I hope this helps.
GL
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
mihai
Sent: Thursday, 31 May 2007 8:58 AM
To: ccielab@groupstudy.com
Subject: Re: QoS test
Going further I added two more routers R1 - R5 - R3 - R6 - R4 Traffic
generator on R5, test ping from R1 to R4, link between R3 and R6 100%
usage.
Just plain and simple priority command for icmp traffic and I would like
to see same RTT time from ping with that serial link flooded.
Is this how u guys would practice QoS commands?
Bandwidth command seems to be working ... shape average command seems to
be working ... I got stuck with this priority :))
I mean all it looks nice in the workbooks but I would like to see it
working live on my 26xx routers.
Let me know even if this question of mine is the most stupid question
asked :))) ... I'll take the heat!
Thx for your time,
Mihai
mihai wrote:
> Here is what I can see on R3 and is similar to all the routers
> involved in this small test.
>
> r3#sh access-lists
> Extended IP access list 100
> 10 permit icmp any any echo (65 matches)
> 20 permit icmp any any echo-reply (40 matches) r3#
>
> Before flooding the s1/3:
>
> r5#ping 6.6.6.6
>
> Type escape sequence to abort.
> Sending 5, 100-byte ICMP Echos to 6.6.6.6, timeout is 2 seconds:
> !!!!!
> Success rate is 100 percent (5/5), round-trip min/avg/max = 28/31/32
> ms r5#
>
> Flooding started:
>
> r3#sh int s1/3
> Serial1/3 is up, line protocol is up
> Hardware is CD2430 in sync mode
> Internet address is 155.1.36.3/24
> MTU 1500 bytes, BW 64 Kbit, DLY 20000 usec,
> reliability 255/255, txload 255/255, rxload 239/255
> Encapsulation HDLC, loopback not set
> Keepalive set (10 sec)
> Last input 00:00:00, output 00:00:00, output hang never
> Last clearing of "show interface" counters 00:01:26
> Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops:
0
> Queueing strategy: Class-based queueing
> Output queue: 7/1000/64/0 (size/max total/threshold/drops)
> Conversations 3/8/16 (active/max active/max total)
> Reserved Conversations 0/0 (allocated/max allocated)
> Available Bandwidth 28 kilobits/sec
> 30 second input rate 60000 bits/sec, 15 packets/sec
> 30 second output rate 64000 bits/sec, 16 packets/sec
> 1288 packets input, 661412 bytes, 0 no buffer
> Received 11 broadcasts, 0 runts, 0 giants, 0 throttles
> 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
> 1406 packets output, 696106 bytes, 0 underruns
> 0 output errors, 0 collisions, 0 interface resets
> 0 output buffer failures, 0 output buffers swapped out
> 0 carrier transitions
> DCD=up DSR=up DTR=up RTS=up CTS=up r3#
>
>
> r3#sh policy-map int s1/3
> Serial1/3
>
> Service-policy output: p-icmp
>
> Class-map: icmp (match-all)
> 59 packets, 6136 bytes
> 30 second offered rate 2000 bps, drop rate 0 bps
> Match: access-group 100
> Queueing
> Strict Priority
> Output Queue: Conversation 24
> Bandwidth 20 (kbps) Burst 500 (Bytes)
> (pkts matched/bytes matched) 59/6136
> (total drops/bytes drops) 0/0
>
> Class-map: class-default (match-any)
> 3040 packets, 1510159 bytes
> 30 second offered rate 69000 bps, drop rate 3000 bps
> Match: any
> r3#sh policy-map int s1/3
> Serial1/3
>
> Service-policy output: p-icmp
>
> Class-map: icmp (match-all)
> 74 packets, 7696 bytes
> 30 second offered rate 2000 bps, drop rate 0 bps
> Match: access-group 100
> Queueing
> Strict Priority
> Output Queue: Conversation 24
> Bandwidth 20 (kbps) Burst 500 (Bytes)
> (pkts matched/bytes matched) 74/7696
> (total drops/bytes drops) 0/0
>
> Class-map: class-default (match-any)
> 3098 packets, 1539943 bytes
> 30 second offered rate 69000 bps, drop rate 3000 bps
> Match: any
> r3#
>
> r3#sh access-lists
> Extended IP access list 100
> 10 permit icmp any any echo (185 matches)
> 20 permit icmp any any echo-reply (175 matches) r3#
>
>
> Tarun Pahuja wrote:
>> Mihai,
>> What does the show output of Policy-Map say, do you see any
>> packets being matched and being put in the prority queue?
>>
>> Thanks,
>> Tarun
>>
>>
>>
>>
>> On 5/29/07, mihai <mihai@ngnpath.com> wrote:
>>> Hi all,
>>>
>>> Not sure what's going on ...
>>> Here is what I have:
>>>
>>> R5 --Eth--R3--Serial with clock rate 64k--R6
>>>
>>> IP connectivity ok and ping from R5 to R6 ~ 6x ms Start rtr on R5 to
>>> bring the serial between R3 and R6 to 100% usage.
>>> Ping from R5 to R6 ~ 800 ms
>>>
>>> I would like to test priority command with MQC for ping traffic
>>> (icmp) My expectation is to get RTT around 6x ms with priority
>>> command in place.
>>>
>>> This is not happening ... and I can't figure out what I'm doing
wrong.
>>>
>>> MQC related config:
>>>
>>> class-map match-all t1
>>> match access-group 100
>>> !
>>> policy-map p-t1
>>> class t1
>>> priority 16
>>>
>>> access-list 100 permit icmp any any echo access-list 100 permit icmp
>>> any any echo-reply
>>>
>>> Service policy output p-t1 on R5,R3 and R6.
>>>
>>> Any idea what's wrong with this test of mine?
>>> Thx.
>>>
>>> Mihai
>>>
>>> ____________________________________________________________________
>>> ___ Subscription information may be found at:
>>> http://www.groupstudy.com/list/CCIELab.html
>>
>> _____________________________________________________________________
>> __ Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>
> ______________________________________________________________________
> _ Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Fri Jun 01 2007 - 06:55:23 ART