cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1105
Views
0
Helpful
2
Replies

Packet Shaping question on 4510

CSCO12052693
Level 1
Level 1

4510 info:

License Information for 'WS-X45-SUP7-E'

License Level: entservices

Version 03.03.01.SG RELEASE SOFTWARE (fc2)

I have a packet shaper on one of my ports that connects to our service provider for all our remote branches. 13.5 mpbs is dedicated to real-time traffic and 37.5 is left for everything else. I know we are over utilizing the link for best effort traffic and are working on getting more bandwidth. We are seeing dropped packets (for best effort traffic, real time traffic is fine). I believe this is caused by our queue filling up on our shaper and the queue limit is just the default 1688 packets.  My question is should I increase the queue? Is that recommended? or is just resending traffic that gets dropped better?

Service policy:

policy-map VOIP

class voice

    priority

class class-default

    shape average 39321600

Viewing the policy-map over one morning.

#show policy-map interface gi3/31

#show policy-map interface gi3/31

GigabitEthernet3/31

Service-policy output: VOIP

    queue stats for all priority classes:

      Queueing

      queue limit 400 packets

      (queue depth/total drops)  0/10

      (bytes output) 4647901573

    Class-map: voice (match-any)

      29172429 packets

      Match: ip dscp ef (46)

        29172429 packets

      Priority: Strict, b/w exceed drops: 10

    Class-map: class-default (match-any)

      1151446829 packets

      Match: any

        1151446829 packets

      Queueing

      queue limit 1688 packets

      (queue depth/total drops)  197/43009   <-----   queue depth fluctuates and total drops slowly increases over days

      (bytes output) 1174686093970

      shape (average) cir 39321600, bc 157287, be 157287

      target shape rate 39321600

1 Accepted Solution

Accepted Solutions

Joseph W. Doherty
Hall of Fame
Hall of Fame

Disclaimer

The   Author of this posting offers the information contained within this   posting without consideration and with the reader's understanding that   there's no implied or expressed suitability or fitness for any purpose.   Information provided is for informational purposes only and should not   be construed as rendering professional advice of any kind. Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In   no event shall Author be liable for any damages whatsoever (including,   without limitation, damages for loss of use, data or profit) arising  out  of the use or inability to use the posting's information even if  Author  has been advised of the possibility of such damage.

Posting

Answer depends on the service needs of your traffic.

Realize TCP was originally designed to use up to host's interface's full bandwidth.  TCP "discovers" available bandwidth by increasing its transmission rate until it "sees" an issue (like packet drops).  I.e. this means, if you have TCP traffic, some drops are perfectly normal.

What you want to avoid is dropping packets from non-rate adaptive traffic or excessively dropping ordinary TCP traffic.

PS:

BTW, creating extremely large queues may decrease drop rate, but then creates queuing latency, which can also detrimental to some traffic types.

View solution in original post

2 Replies 2

Joseph W. Doherty
Hall of Fame
Hall of Fame

Disclaimer

The   Author of this posting offers the information contained within this   posting without consideration and with the reader's understanding that   there's no implied or expressed suitability or fitness for any purpose.   Information provided is for informational purposes only and should not   be construed as rendering professional advice of any kind. Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In   no event shall Author be liable for any damages whatsoever (including,   without limitation, damages for loss of use, data or profit) arising  out  of the use or inability to use the posting's information even if  Author  has been advised of the possibility of such damage.

Posting

Answer depends on the service needs of your traffic.

Realize TCP was originally designed to use up to host's interface's full bandwidth.  TCP "discovers" available bandwidth by increasing its transmission rate until it "sees" an issue (like packet drops).  I.e. this means, if you have TCP traffic, some drops are perfectly normal.

What you want to avoid is dropping packets from non-rate adaptive traffic or excessively dropping ordinary TCP traffic.

PS:

BTW, creating extremely large queues may decrease drop rate, but then creates queuing latency, which can also detrimental to some traffic types.

Thanks for the reply. I'll probably leave the queue limit alone till we  get more bandwidth.

Jeremy

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco