Re: Trying to understand OSPF Point-to-Multipoint Non-Broadcast

From: Narbik Kocharians <narbikk_at_gmail.com>
Date: Sat, 4 Jun 2011 21:34:45 -0700

Andrew,

Just to add to what Bob nicely explained.

Some times to know how to manipulate routes and how to take advantage of a
given feature within a protocol, you have to know whats available to you in
that protocol and what options (Once configured) can give you certain
behavior that you are looking to achieve.

Lets say R1 is the hub router and R2 and R3 are configured as the spokes.

R2 and R3 are both advertising a network, lets say 1.0.0.0/8.

R2 and R3 are advertising the same cost to R1 for network 1.0.0.0/8.

R1 adds its cost to this multi-access network, and it performs a load
balancing of some type for network 1.0.0.0/8, through R2 and R3.

This is all good, but what if I told you that R2 has a connection rate of
64Kbps and R3 has a T1 connection to the same multi-access network, I guess
you would NOT want to perform equal cost load balancing and the interesting
part is that none of the spokes advertise their cost to this NMBA network to
R1. So how do you tell OSPF which spoke R1 should go through to reach
network 1.0.0.0/8.

In this case, if the network type is configured as a Point-to-Multipoint
non-broadcast, when you configure the neighbors on R1 you will also see a
cost option, and this cost option only shows up under this network type.
NOW..manipulating this cost option you can tell R1 to go through the
spoke that has a better total cost.

Maybe for what you are asking or wanting to do (If i understand you
correctly), an OSPF "Non-broadcast" network type may also be an option.

I hope this helped.

On Sat, Jun 4, 2011 at 6:30 PM, Abdullah Al-Malki
<a.almalki1402_at_gmail.com>wrote:

> Hi Andrew,
> Hub-spoke topology can better illustrate the point-to-multipoint
> non-broadcast which is cisco proprietary.
>
> This mode of operation seems strange at first glance because it uses static
> neighbor peering using the neighbor command.
>
> However, think of the use of the neighbor command that we can use in the
> hub
> to solve the asymmetric cost to spokes.
>
> Neighbor command can set the cost to the neighbor which allows better
> metric
> control.
>
> Regards,
> Abdullah
>
> On Sun, Jun 5, 2011 at 3:59 AM, Andrew LaPorte <andy_at_cloud9.net> wrote:
>
> > Hi,
> >
> >
> >
> > I m hoping that someone can explain to me the way OSPF works on a
> > Point-to-Multipoint Non-Broadcast.
> >
> >
> >
> > Let me explain the situation I m running across. I need to connect three
> > routers using OSPF. The routers are:
> >
> >
> >
> > R1 fa0/0 (10.1.1.1/24) _-` fa0/0 (10.1.1.2/24) R2 fa0/1 (10.2.2.1/24) _
> `
> > fa0/1 (10.2.2.2/24)R3
> >
> >
> >
> > R1
> >
> > Interface fa0/0
> >
> > Ip add 10.1.1.1 255.255.255.0
> >
> > Router ospf 1
> >
> > Network 10.1.1.1 0.0.0.0 area 0
> >
> >
> >
> > R2
> >
> > Interface fa 0/0
> >
> > Ip address 10.1.1.2 255.255.255.0
> >
> > Interface fa 0/1
> >
> > Ip address 10.2.2.1 255.255.255.0
> >
> > Ip ospf network point-to-multipoint non-broadcast
> >
> > Router ospf 1
> >
> > Network 10.1.1.2 0.0.0.0 area 0
> >
> > Network 10.2.2.1 0.0.0.0 area 1
> >
> > Neighbor 10.2.2.2
> >
> >
> >
> >
> >
> > R3
> >
> > Inter fa 0/1
> >
> > Ip address 10.2.2.2 255.255.255.0
> >
> > Ip ospf network point-to-multipoint non-broadcast
> >
> > Router ospf 1
> >
> > Network 10.2.2.2 0.0.0.0 area 1
> >
> > Neighbor 10.2.2.1
> >
> >
> >
> > All connections are Ethernet. The restrictions given are that R2 to R3
> > cannot form a DR/BDR relationship and they are not allowed to exchange
> > multicast.
> >
> >
> >
> > My configuration is to use Point-to-Multipoint Non-Broadcast on the R2 to
> > R3
> > interfaces. When I do that I get a surprising result.
> >
> >
> >
> > I get 2 routes on R1 that are both /32 for 10.2.2.1 and 10.2.2.2 I m
> also
> > not getting the /24 for the 10.2.2.0/24 that I m expecting.
> >
> >
> >
> > R1 show ip route
> >
> > O IA 10.2.2.2/32 [110/2] via 10.1.1.2, 00:00:34, FastEthernet0/0
> >
> > C 10.1.1.0/24 is directly connected, FastEthernet0/0
> >
> > O IA 10.2.2.1/32 [110/1] via 10.1.1.2, 00:01:07, FastEthernet0/0
> >
> >
> >
> > I m trying to understand why the /32 are showing up. My guess would be
> > that
> > the point-to-multipoint non-broadcast is causing the issue but I m just
> not
> > sure.
> >
> >
> >
> > The second problem that I m trying to understand is why does the route
> for
> > 10.2.2.0/24 not appear on R1.
> >
> >
> >
> > Any insight would greatly appreciated.
> >
> >
> >
> > Thanks,
> >
> >
> >
> > 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
>
>
>
>
>
>
>
>

--
*Narbik Kocharians
*CCSI#30832, CCIE# 12410 (R&S, SP, Security)
www.MicronicsTraining.com <http://www.micronicstraining.com/>
Sr. Technical Instructor
*Ask about our FREE Lab Voucher with our Boot Camps*
YES! We take Cisco Learning Credits!
Training & Remote Racks available
Blogs and organic groups at http://www.ccie.net
Received on Sat Jun 04 2011 - 21:34:45 ART

This archive was generated by hypermail 2.2.0 : Fri Jul 01 2011 - 06:24:27 ART