From: Peter van Oene (pvo@usermail.com)
Date: Sun Mar 02 2003 - 02:32:55 GMT-3
At 04:29 AM 3/2/2003 +0000, Sage Vadi wrote:
>All of these costs are unidirectional in nature. In
>your A to B example, none of these costs as set on A
>will change B's perspective of the cost of the B to A
>link. Said another way, there are two links here,
>BtoA and AtoB. The AtoB link is assigned a cost on A
>and the BtoA link on B. You can't set the cost of
>BtoA from A. Furthermore, it has a lot to do with the
>use of an spf-like algorithm which is a typical
>characteristic of link state protocols.
>
>NB: I do not agree with your above quote. Example:
Your example isn't really the same as mine. If R1 is A, and R2 B, B is not
using the AtoB link to get to networkB.
>NetworkB---R1-----R2---NetworkA
>
>Using OSPF we can manipulate the cost of NetworkB as
>R2 receives it, by doing it locally on R1 itself.
By setting the cost on R1' s like toward netB, you will influence R2's cost
to reach it.
>People get confused as they try to manipulate metrics
>on directly connected interfaces, which is generally
>pointless.
It is only pointless on the router itself. The rest of the network will
see and react to these local costs.
>Can the same thing be done in ISIS?
Yes
>As you said to accurately see this I would have to
>recreate it, which I will... when I have time...
>
>I believe it can but possibly the interfaces cannot be
>a pure L1, if the router that is receiving those
>routes is a pure L2, in my situation: I have tested
>this and the pure L2 router does not receive the
>metric.
Your pure L2 won't become adjacent with your pure L1 :-) I'm sure you mean
you have an L1L2 in between. In either case, the metric should flow
properly. Essentially to calculate a metric to a prefix, one sums the cost
of the prefix as seen in the LSP with the cost of reaching the router that
announced it (for intra area). In inter-area, you just add the cost(s)
toward the L1L2 routers to that set. In your case, whether intra area, or
inter area, you should be properly setting and seeing the
metrics. Sections 3.1 and 3.2 in rfc 1195 (about a page an a half of
text) cover metric setting in ISIS. A great deal of real world ISIS differs
from the rather vague spec, but I believe basic metric setting is
accurate. Wide metrics should also work in this fashion.
Assuming you have the basic stuff setup, why not post the config/routing
table output.
>...
>Sage
>
>
>
>
>
>
>__________________________________________________
>Do You Yahoo!?
>Everything you'll ever need on one web page
>from News and Sport to Email and Music Charts
>http://uk.my.yahoo.com
This archive was generated by hypermail 2.1.4 : Sat Apr 05 2003 - 08:51:30 GMT-3