From: Alexei Monastyrnyi (alexeim@orcsoftware.com)
Date: Fri Mar 31 2006 - 09:43:34 GMT-3
Some interesting math behind the dampening ...
Hope such a reverse engineering does not break NDA :-)
Cisco says "The maximum value of the accumulated penalty is calculated,
based on the maximum suppress time, reuse threshold, and half-life period."
see "Maximum Suppress Time" here
http://www.cisco.com/univercd/cc/td/doc/product/software/ios122s/122snwft/release/122s14/fsipevdp.htm#wp1048361
Borrowing variables definition from the URL below, we have
HL = half life
RT = reuse threshold
ST = supress threshold
MS = mas supress time
MP = F(HL, RT, MS)
After gathering some statistics from "dampening HL RT ST MS" output on
WS-C3750G-24T (sorry, my 2500 routers are too old for this stuff), one can
draw the following expression for F.
F(HL, RT, MS) = RT*2^(MS/HL).
There are some minor issues with rounding to integer.
Catalyst switch gives me
sw1(config)#in gi 1/0/1
sw1(config-if)#dam 15 1482 2000 181
% Maximum penalty (6357517) is more than allowed maximum (20000). Reduce
maximum suppress time.
Windows calculator gives me 6357360.1895226113880692603777501
When there is no fraction in (MS/HL) power, the result is 100% as accurate
as from Catalyst.
A.
----- Original Message -----
From: "David Vasek" <dvasek@satx.rr.com>
To: "'Scott Smith'" <hioctane@gmail.com>; "'Nick Griffin'"
<ngriffin@sio.midco.net>
Cc: <ccielab@groupstudy.com>
Sent: Friday, March 31, 2006 6:09 AM
Subject: RE: Flapping Interface
> Thank you, this is exactly what I was looking for.
> -DV
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> Scott Smith
> Sent: Thursday, March 30, 2006 9:15 PM
> To: Nick Griffin
> Cc: ccielab@groupstudy.com
> Subject: Re: Flapping Interface
>
> Check out this thread:
>
> http://www.groupstudy.com/archives/ccielab/200510/msg01553.html
>
> -Scott
>
>
> On 3/30/06, Nick Griffin <ngriffin@sio.midco.net> wrote:
>> If your asked to provide a configuration to remove an interface from
>> participating in exchanging routing information, after a certain
>> number of interface flaps, what configuration options are available for
> this?
>> I'm not aware of any options.
>>
>> TIA
>>
>> ______________________________________________________________________
>> _ 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 : Sat Apr 01 2006 - 10:07:40 GMT-3