You can using varinace command to achieve load balance.
"Roger Dellaca" ¼¶¼g©ó¤å³¹ ...
>I'd like to answer this to the group as well because it's an important
distinction. From CCO,
>http://www.cisco.com/univercd/cc/td/doc/cisintwk/idg4/nd2003.htm#xtocid1338
311
>
>a few paragraphs in:
><Begin copy>
>
>Figure 3-6: DUAL feasible successor.
>
>[graphic on page - check it out, it's important to the example]
>
>If Router B becomes unavailable, Router A will go through the following
three-step process to find a feasible successor for Network 7:
>
>Step 1 Determining which neighbors have an advertised distance to Network 7
that is less than Router A's feasible distance (FD) to Network 7. The FD is
31 and Router H meets this condition. Therefore, Router H is a member of V1.
>
>Step 2 Calculating the minimum computed cost to Network 7. Router H
provides a cost of 40, and Router D provides a cost of 230. Dmin is,
therefore, 40.
>
>Step 3 Determining the set of neighbors that are in V1 whose computed cost
to Network 7 equals Dmin (40). Router H meets this condition.
><End copy>
>
>This is an explanation in the section on the topology database & what
happens when a route fails - if equal-cost load balancing & the route isn't
the same cost, it wouldn't be used, BUT it would be in the topology database
as a feasible successor - that allows it to be used immediately upon failure
of the best route without a new calculation.
>
>And the feasible successors are the possible routes for unequal cost, as
long as their cost divided by the cost of the best route is less than or
equal to the variance.
>
>In step 1, feasible distance is total cost from current router to
destination (is Arizona to Florida); advertised distance is cost from next
possible hop to destination (is Kansas to Florida).
>
>It is the same as your explanation below, however when they say current
router, that is your starting point in my analogy - even if the packet
started somewhere before that, once it's in the router, as far as that
router is concerned it's the starting point. The router doesn't care where
the packet has been, only where it's going (of course policy routing is an
exception to that, but wasn't part of this discussion!!).
>
>
>>>> <[EMAIL PROTECTED]> 09/12 5:08 PM >>>
>
>Hi Roger,
>
>Thanks so much for answering my email :o)
>
>I'm rather confused about the first condition as the BSCN course notes
says: "A
>route is feasible if the next router in the path is closer to the
destination
>than the current router", using your analogy, isn't that saying that you
>consider Kansas only when it's closer to Florida than Texas is, not when
you
>can't use Texas?
>
>Thanks
>Helena
>
>
>
>
>
>
>
>The first condition is used in either unequal OR equal cost load balancing.
It
>is to prevent routing loops. Think of it like this: Let's say I'm going
on
>vacation. I'm in Arizona. I want to get to Florida. I can go thru Texas,
which
>is my best route, and there is a certain distance from California to
Florida.
>If I can't get thru Texas becauuse all the roads are closed. My options
are to
>go thru Kansas or California. Kansas is closer to Florida than Arizona, so
it
>fits the rules & I will consider it. California is further from Florida
than
>Arizona is, so I won't consider it. If I went thru California, once I got
there
>the gas station attendant would tell me to go thru Arizona, and I would
drive in
>a loop forever.
>
>Notice this is comparing total distance of best route (from source to
>destination) with advertised distance of alternate routes (from next hop to
>destination).
>
>The second condition is how much difference in total distances should I
>tolerate. Same analogy as above - If I am doing equal cost load balancing,
The
>total distance from Arizona - Texas - Florida is shorter than from
Arizona -
>Kansas - Florida, so I and my buddy will both drive thru texas. If I am
doing
>unequal cost, the variance tells me how much different the alternatives can
be
>from the best (as a multiple of the best).
>
>Notice here we are comparing total distance of the best router to total
distance
>of alternate routes.
>
>Now the only thing beyond that is to remember that unlike in driving, the
>hop-by-hop metrics in EIGRP are not calculated ina totally additive
fashion -
>rather it's lowest bandwidth on a link in the path (which is not additive)
plus
>total delay (which is additive).
>
>>>> Helena <[EMAIL PROTECTED]> 09/12 12:46 AM >>>
>Hi everyone,
>
>CAn someone please explain the two feasibility conditions in EIGRP
>unequal-cost load balancing?
>
>They are:
>- local best metric > best metric learned from the next router
>
>- the "multiplier" x Local bestmetric for the destination > metric through
>the next router
>
>Thanks in advance
>Helen
>
>**NOTE: New CCNA/CCDA List has been formed. For more information go to
>http://www.groupstudy.com/list/Associates.html
>_________________________________
>UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
>FAQ, list archives, and subscription info: http://www.groupstudy.com
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
>
>
>
>
>
>======================================================================
>This email message has been swept by MIMEsweeper.
>
>**NOTE: New CCNA/CCDA List has been formed. For more information go to
>http://www.groupstudy.com/list/Associates.html
>_________________________________
>UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
>FAQ, list archives, and subscription info: http://www.groupstudy.com
>Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
**NOTE: New CCNA/CCDA List has been formed. For more information go to
http://www.groupstudy.com/list/Associates.html
_________________________________
UPDATED Posting Guidelines: http://www.groupstudy.com/list/guide.html
FAQ, list archives, and subscription info: http://www.groupstudy.com
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
