10 messages in net.nether.puck.cisco-nsp[c-nsp] how to validate cell rates?
FromSent OnAttachments
Gert DoeringJan 20, 2005 5:46 am 
Hudson Delbert J Contr 61 CS/SCBNJan 20, 2005 10:58 am 
MADMANJan 20, 2005 11:29 am 
Gert DoeringJan 20, 2005 11:37 am 
Clinton WorkJan 20, 2005 11:49 am 
MADMANJan 20, 2005 12:33 pm 
Gert DoeringJan 20, 2005 12:50 pm 
Gert DoeringJan 20, 2005 1:01 pm 
Clinton WorkJan 20, 2005 1:44 pm 
MADMANJan 20, 2005 2:48 pm 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:[c-nsp] how to validate cell rates?Actions...
From:MADMAN (davi@qwest.com)
Date:Jan 20, 2005 12:33:15 pm
List:net.nether.puck.cisco-nsp

Gert Doering wrote:

Hi,

On Thu, Jan 20, 2005 at 10:29:30AM -0600, MADMAN wrote:

Try shaping down a little more, roughly 10%. I do know that earlier IOS did not always take into account the overhead in the ATM cell when shaping. I have seen this several times in the past though admittedly not in 12.3.

The values listed for cells/seconds in "show atm vc <name>" are correctly calculated for 53 bytes/cell (full size). Like this:

Cisco-M-XIV>sh atm vc sp1-936 ATM2/0.129: VCD: 46, VPI: 15, VCI: 129, Connection Name: sp1-936 UBR, PeakRate: 2150 (5071 cps) ^^^^ (PVC configured to "ubr 2150" instead of the 'documented' value of "ubr 2303", so that the customers won't suffer while we argue with the Telco)

So are you telling me that earlier IOSes sometimes used 48 bytes for calculating cells/s., or that the displayed values are wrong?

gert

In a nutshell, yes to your first question. A couple of us here had a couple a cases open but could never got anywhere. knowing how to resolve the problem and realizing we were not loosing any bandwidth it wasn't worth battling.

Dave

"Emotion should reflect reason not guide it"