Re: EIGRP offset-lists...... have i got them all wrong?

I dont know if it will mess up other parts of your scenario, but another way
to do it without thinking to much is take bandwidth out of the calculation
using ""metric weight 0 0 0 1 0 0". Then you basically just have to put
delay 5 on one interface and delay 1 on another (and use the variance
command) to get 5:1 traffic sharing. Probably wouldn't want to do this in
the real world, but it works in a jam :)

Chris

On Sat, Jul 4, 2009 at 9:10 AM, andrew <andrew.coates_at_internode.on.net>wrote:

> Yes offset lists changes the AD of a route, use it to make a route a FS.
> if you want to unequal cost load balance then your best bet is delay. It
> really isn't that hard to figure out.
>
> Asumming route is a FS,
>
> Take the F route and times its metric by the desired balancing ratio , so
> 5:1 would be FD x 5 .
>
> Now minus the FD of the FS from the FD x 5 value
>
> Now just divide that value by 256. what ever your answer is apply that to
> the incomming interface of the FS as delay.
>
> Easy as pie.
>
> So an example with real numbers
>
> F FD = 10501
> FS FD = 15123
> Balance value 7:1
>
> F FD X 7 = 73507
> F FDX7 - FD FS = 58384
>
> 58384 \ 256 = 228 (rounded)
>
> So on the incoming interfaces of the FS route add a delay of 228
>
> Remember to set the varience.
>
> cheers
>
> -----Original Message-----
> From: nobody_at_groupstudy.com [mailto:nobody_at_groupstudy.com] On Behalf Of
> andy
> Chisholm
> Sent: Saturday, July 04, 2009 10:53 PM
> To: ccielab_at_groupstudy.com
> Subject: EIGRP offset-lists...... have i got them all wrong?
>
> Hi
>
> I've been playing with offset-lists in eigrp thinking that i had read you
> could use them with variance to help load-balance, as a direct alternative
> to doing maths and altering the delay on one of the interfaces i wish to
> loadbalance through. I thought, for example if i were asked to LB at a
> ratio
> of 5 to 1 it may be quicker in my lab to use offset lists.
>
> However, when i started playing with them, it seems that they alter the
> AD/RD of a route rather than just adding to the composite metric. So when
> applied to the route i wish to make less preferable (ratio 1) it actually
> made the AD so high it couldn't be a feasible successor.
>
> I guess my question is: have i got offset lists all wong? are they used in
> this instance to increase the 'composite metric' of a preferred route
> higher
> than the secondary routes' AD/RD? In which case you've still got play with
> the delay....
>
> Can anyone explain how they are used properly?
>
> cheers
> Andy
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html

Blogs and organic groups at http://www.ccie.net
Received on Sat Jul 04 2009 - 15:02:54 ART