From: Keith Kruepke (lister@xxxxxxxxxxx)
Date: Fri Jul 28 2000 - 01:48:25 GMT-3
Brian,
I disagree--I think you may be misunderstanding the question. Rereading Dave's
original message, he is asking how to lower a hop count, not the maximum hops.
In his first paragraph, he asks about lowering the hop count value, and he ev
en relates it to redistribution (which would have nothing to do with the maximu
m).
In the second paragraph, he talks about changing the maximum hop count (but NOT
lowering it) as an alternate solution. And he's right--if you lower your indi
vidual route hop counts to make them more acceptable, or if you increase your m
aximum, the net result is the same--more routes are accepted.
Of course, Dave could probably clarify this a lot better than anyone else here,
so what exactly did you mean, Dave?
Keith
----- Original Message -----
From: "Brian Hescock" <bhescock@cisco.com>
To: <dhbrown@Pipeline.com>
Cc: "William Swedberg" <swedbergwp@yahoo.com>; "Dave Gingrich" <Dave@dcg.org>;
<ccielab@groupstudy.com>
Sent: Thursday, July 27, 2000 10:17 PM
Subject: Re: Re: Changing IPX RIP Hop Count
Guys, think about what I'm saying. The max hop count for ipx rip is 15,
correct? What if we want to make the MAXIMUM hop count for ANY ipx route
10 instead? Not just make it 10 or less by using tunnels, etc, but make
it the MAXIMUM it could EVER be, not due to any configuration magic you do
by using tunnels but a complete limit. Pretend the RFC now says the max
hop is 10, make your code only accept a max of 10. That's the point we've
been trying to make. Yes, I could make a router 100 hops aways look like
1 ipx hop by putting in a tunnel, that's not the issue here. You can
increase the max hop count with ipx maximum-hop, which will increase the
limit from 15 to a value between 16-254. But how can you lower the
absolute limit the software will only accept a max hop count of 10? We're
not talking about ways to get around the max hop count but a way to set
the absolute value IN THE CODE to a value lower than 15. Hopefully this
makes sense now, I'm going to go have a beer... ;-) Yes, it's somewhat
confusing because why would you want do it?
Why you would want to do that I have no idea but that's what Dave inferred
and he has yet to correct me after numerous e-mails. I don't see any way
it can be done but then again, new commands are being added all the
time...
B.
On Thu, 27 Jul 2000 dhbrown@Pipeline.com wrote:
> On the contrary, a tunnel is the correct solution. Consider:
>
> R1 - R2 - R3 - R4 - R5 - R6
>
> The hop count from R1 to R6 would be 5. If you had a tunnel from R1 to R6, t
hat would be 1 hop...
>
> David
> (RTP Lab 8/6)
>
> Brian Hescock <bhescock@cisco.com> wrote:
> > William,
> But he want's to reduce the max hop count, not increase it. Tunnels
> would allow you to increase it.
>
> B.
>
> On Thu, 27 Jul 2000, William Swedberg wrote:
>
> > Remember this concept....
> >
> > "tunnel"
> >
> >
> > William Swedberg CCNP CCDP
> >
> >
> > --- Dave Gingrich wrote:
> > > At 09:30 7/27/00 -0400, Brian Hescock wrote:
> > > >You should be able to set the metric when you
> > > redistribute or set the
> > > >default metric when you redistribute. You could
> > > say the metric is 5 so
> > > >then the max hop count from that point would be 10.
> > > Another option may be
> > > >to use a route-map to change the metric.
> > >
> > > Unfortunately niether of those are available with
> > > IPX. The metric used to
> > > choose a path is a delay value measured in "ticks."
> > > You can fiddle with
> > > that a bit, but if the hop count is over 15, you are
> > > done as far as IPX RIP
> > > is concerned.
> > >
> > >
> > > =========================
> > > David C. Gingrich, K9DC
> > > Indianapolis, Indiana
> > > Dave@dcg.org
> > > =========================
> > >
> > >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:23:59 GMT-3