Re: OSPF route filtering teaser

From: Carlos G Mendioroz (tron@xxxxxxxxxxx)
Date: Sat Jun 29 2002 - 16:08:29 GMT-3


   
May I politelly disagree...
the original request clearly states config could only be done on R2 &
R3.
Changing area 1 requires config in all routers in area 1, including but
no limited to, R1.

P729 wrote:
>
> David,
>
> I think you've got it. Making area 1 an NSSA would cause the external RIP
> domain to be advertised via type 7 LSAs throughout area 1. Using
> 'summary-address not-advertise' on R2 would allow you to control which type
> 7s are translated into type 5s to be flooded throughout the rest of the OSPF
> AS.
>
> Regards,
>
> Mas Kato
> https://ecardfile.com/id/mkato
> ----- Original Message -----
> From: "David Ham" <ccieau@hotmail.com>
> To: <franjime@cisco.com>; <peter@whittle-systems.demon.co.uk>;
> <ccielab@groupstudy.com>
> Sent: Saturday, June 29, 2002 8:18 AM
> Subject: RE: OSPF route filtering teaser
>
> > Frank,
> >
> > Could you pls bit more detail. I used nssa with summary-address
> > not-advertise, it works great. But not sure about 2 ospf processes.
> >
> > Regards,
> >
> > David Ham
> >
> >
> > >From: "Frank Jimenez" <franjime@cisco.com>
> > >Reply-To: "Frank Jimenez" <franjime@cisco.com>
> > >To: "'Peter Whittle'" <peter@whittle-systems.demon.co.uk>,
> "'CCIELab
> > > Studygroup'" <ccielab@groupstudy.com>
> > >Subject: RE: OSPF route filtering teaser
> > >Date: Fri, 28 Jun 2002 17:30:07 -0500
> > >
> > >Are we allowed to have two different OSPF processes running? If so, we
> > >can run two different OSPF processes and redistribute between the R2/R3
> > >process and the R3/R4 process.
> > >
> > >Frank Jimenez, CCIE #5738
> > >franjime@cisco.com
> > >
> > >-----Original Message-----
> > >From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
> > >Peter Whittle
> > >Sent: Friday, June 28, 2002 3:11 PM
> > >To: CCIELab Studygroup
> > >Subject: OSPF route filtering teaser
> > >
> > >
> > >Selectively blocking OSPF routes between areas
> > >----------------------------------------------
> > >
> > >I would like to pose a simple scenario to the group.
> > >
> > >There are 4 routers R1 .. R4
> > >
> > >R1 is injecting routes into RIPv2 say 10.1.0.0/16
> > >
> > >R2 is running RIPv2 on e0, it is also connected by e1 to OSPF Area 1 and
> > >must inject the routes learnt from RIP into AREA 1. There are other
> > >routers also connected to the ethernet segment in Area 1 again running
> > >OSPF. The routes injected by R2 into area 1 must be visible to these
> > >routers via OSPF.
> > >
> > >R3 has 2 ethernet interfaces e0 in OSPF Area 0, and e1 in OSPF Area 1.
> > >It must see the RIP routes injected into OSPF by R2 (ie 10.1.0.0/16).
> > >
> > >R4 in connected to the ethernet in Area 0 and is also running OSPF and
> > >is outside of your control.
> > >
> > >STOP the RIP routes that were injected by R2 from being seen in Area 0.
> > >(i.e. block the 10.1.0.0/16 route)
> > >
> > >You may only program routers R2, & R3 to achieve this.
> > >
> > >
> > >Any thoughts, ideas, solutions?
> > >
> > >
> > >I have one solution in mind but it is not very elegant. I will share
> > >this next week when you have had time to think about the problem.
> > >
> > >
> > >------------
> > >
> > >A distribute-list applied to R3, an ABR, will of course not work. (When
> > >the 10.1 route reaches R3 it is in an LSA. If we apply the distribute-
> > >list x in, it will only block the route going into R3's routing table,
> > >it will not prevent the LSA from being sent on to R4. We are not
> > >permitted to change the other routers in Area 0 so we can not use the
> > >conventional approach of applying the distribute-list x in to each of
> > >the routers in Area 0.
> > >
> > >If we apply a distribute-list x out to the ABR it will again have no
> > >impact on the LSA advertising the 10.1 route into Area 0.)
> > >
> > >==========================
> > >
> > >May enlightenment be yours.
> > >
> > >Peter
> > >
> > >--
> > >Peter Whittle



This archive was generated by hypermail 2.1.4 : Tue Jul 02 2002 - 08:12:44 GMT-3