From: yuen me (yuen_me@xxxxxxxxxxx)
Date: Sat Feb 09 2002 - 00:13:54 GMT-3
I agree that Banner's approach is the most logical, not because he is from
cisco. Pls say you are not the procter. All you want to do is to create a
summary route in routing table so that it can be redistributed into eigrp.
Remember the redistribution is based on routing table, not the database area
of the redistributed routing process. area x range will only generate
summary update to other area, not creating summary route. "summary" will
create desired summary route, but depending on the redistributed external
specific routes. So you create 2 ospf process let say 1 and 2. ospf 1 is the
normal one that governs all ospf interface. ospf 2 is the additional one
which has ospf 1 route redistributed into it. After that those routes are
"external routes" and qualified to be summarized by "summary" statement and
then a summary route can be created in the routing table and then it is
qualkified to be redistributed into eigrp.
Think of it as a factory production line.
I think why don't you use loopback or secondary address ? Because if all
specific routes die, you want the summary route should die too. That's the
only reason I can think of.
Yuenme
>From: "Lupi, Guy" <Guy.Lupi@eurekaggn.com>
>Reply-To: "Lupi, Guy" <Guy.Lupi@eurekaggn.com>
>To: "'DAN DORTON'" <DHSTS68@dhs.state.il.us>, parry.chua@compaq.com,
>ssowell@gate.net, ccielab@groupstudy.com
>Subject: RE: RE: RE: RE: solie skinet lab - and the old topic ospf-igrp
>Date: Fri, 8 Feb 2002 18:49:24 -0500
>
>Ok, so lets say now that the requirement is to get the 165.10.185.0 and
>165.10.12.0 subnets into the routing table on the IGRP Router B. I put
>secondary addresses on Router A's frame interface for the 165.10.185.0 and
>165.10.12.0 subnets and redistribute them, works like a charm. But in the
>event that Router C goes down, and the frame interface on Router A stays
>up,
>Routers A and B still think that they have a route to these subnets.
>Question is, do I pass or fail this section? The routes are in Router B's
>routing table, and connectivity is there, but in the event that Router C
>went down and Router A's interface stays up the routes will still appear in
>the tables even though there is no reachability. What do you think?
>Additionally, what way is there other than secondary ospf process, tunnels,
>and secondary addresses on the serial interface? I tried putting
>secondary's on the interface between Router A and Router B that matched the
>major net and mask of these networks, but it doesn't seem to work, maybe I
>am missing something there?
>
> OSPF Area 0
> 165.10.11.0/25
> |
> |
> RouterA------IGRP 165.10.26.0/24-------RouterB
> |
> |
> Primary address 165.10.7.2/24
> Secondary address 165.10.185.74/24
> Secondary address 165.10.12.254/24
> |
> |
> |
> OSPF Area 0
> 165.10.7.0/24
> |
> | OSPF Area 0
> RouterC--------165.10.185.0/26
> 165.10.12.0/30
>
>
>~-----Original Message-----
>~From: DAN DORTON [mailto:DHSTS68@dhs.state.il.us]
>~Sent: Friday, February 08, 2002 3:51 PM
>~To: parry.chua@compaq.com; ssowell@gate.net; ccielab@groupstudy.com
>~Subject: RE: RE: RE: RE: solie skinet lab - and the old topic ospf-igrp
>~
>~
>~Well, you got one on me anyways...
>~
>~Actually 4 on me.
>~
>~The 7317 after your name.
>~
>~All I managed to come back from San Jose with was a sore bum...
>~
>~>>> "ssowell@gate.net" <ssowell@gate.net> 02/08/02 02:25PM >>>
>~Aye Dan, you caught my poor wording. Since I'm 4 hours outta my time
>~zone, can I use the jet lag excuse?
>~I was just trying to steer him away from "area x range" command. Good
>~point, and thanks for the heads up!
>~Steven Sowell
>~CCIE#7317
>~
>~Original Message:
>~-----------------
>~From: DAN DORTON DHSTS68@dhs.state.il.us
>~Date: Fri, 08 Feb 2002 12:41:27 -0600
>~To: parry.chua@compaq.com, ssowell@gate.net, ccielab@groupstudy.com
>~Subject: RE: RE: RE: solie skinet lab - and the old topic ospf-igrp
>~
>~
>~Why MUST you configure a second OSPF process?
>~
>~Seems to me you can create some loopbacks & redistribute connected &
>~accomplish the same thing. IE 165.10.11.129/25 Summary-Address
>~165.10.11.9/24.
>~
>~Also seems you could build some tunnels, or add some secondaries to
>~IGRP to do the job.
>~
>~Things that make you go Hmmmmm.
>~
>~>>> "ssowell@gate.net" <ssowell@gate.net> 02/08/02 12:20PM >>>
>~While it is true that area range will work on ABR, there is no ABR in
>~this scenario. All routers are in area 0, except the "router B". This
>~is
>~why you must create a second OSPF process to pull off the
>~summarization
>~with summary-address command.
>~Cheers,
>~Steven Sowell
>~CCIE#7317
>~
>~Original Message:
>~-----------------
>~From: Chua, Parry Parry.Chua@compaq.com
>~Date: Fri, 8 Feb 2002 11:29:51 +0800
>~To: ssowell@gate.net, ccielab@groupstudy.com
>~Subject: RE: RE: solie skinet lab - and the old topic ospf-igrp
>~
>~
>~Hello Lupi,
>~
>~Area range will work on ABR.
>~
>~Parry Chua
>~
>~-----Original Message-----
>~From: ssowell@gate.net [mailto:ssowell@gate.net]
>~Sent: Friday, February 08, 2002 10:59 AM
>~To: ccielab@groupstudy.com
>~Subject: RE: RE: solie skinet lab - and the old topic ospf-igrp
>~
>~
>~I recreated your scenario on my lab rack with similar results. The
>~difference you seem to be noticing is that the ".11" network is
>~directly
>~connected, while the ".185" network is learned about via OSPF.
>~If you'd like to see the configs of my working solution, lemme know
>~and
>~I'll send 'em to you.
>~Regards,
>~Steven Sowell
>~CCIE#7137
>~
>~
>~Original Message:
>~-----------------
>~From: Lupi, Guy Guy.Lupi@eurekaggn.com
>~Date: Thu, 7 Feb 2002 19:52:00 -0500
>~To: jbanner@cisco.com, rodrigo.silva@cpm.com.br, ccielab@groupstudy.com
>~
>~
>~Subject: RE: solie skinet lab - and the old topic ospf-igrp
>~
>~
>~Ok, I see that this works, but I am having trouble understanding why.
>~Could
>~someone explain this to me with a little detail. Lets say I have the
>~following:
>~
>~ OSPF Area 0
>~ 165.10.11.0/25
>~ |
>~ |
>~ RouterA------IGRP 165.10.26.0/24-------RouterB
>~ |
>~ |
>~ |
>~ OSPF Area 0
>~ 165.10.7.0/24
>~ |
>~ | OSPF Area 0
>~ RouterC--------165.10.185.0/26
>~
>~
>~All interfaces are in OSPF area 0 except for the one to Router B, so
>~Router
>~A gets the routes for everything, Router B is restricted to /24's
>~within
>~the
>~165.10.0.0 network. I can do a summary address 165.10.11.0/24 under
>~the
>~original OSPF process and Router B gets it, but it will not get
>~165.10.185.0/24 until I do a second routing process, like OSPF 10,
>~redistribute the original process into this one, do a summary address
>~for
>~165.10.185.0/24 in the second process, and redistribute both processes
>~into
>~IGRP. I am having trouble understanding why, if both the 165.10.11.0
>~and
>~165.10.185.0 are OSPF internal routes, what is the difference, why
>~does
>~redistributing into the other process work? Sorry for the long email,
>~but I
>~really want to understand this.
>~
>~
>~~-----Original Message-----
>~~From: S. John Banner [mailto:jbanner@cisco.com]
>~~Sent: Thursday, February 07, 2002 6:08 PM
>~~To: Rodrigo Espinha T Da Silva; ccielab@groupstudy.com
>~~Subject: RE: solie skinet lab - and the old topic ospf-igrp
>~~
>~~
>~~ To get this to work (I did it yesterday), redistribute all
>~~OSPF routes
>~~into the second OSPF process, which makes them all External
>~~routes, and as
>~~such not redistributable (the original OSPF routes are
>~redistributable
>~~because they are internal and thus prefered over the external
>~~routes and
>~~thus appear in the routing table). Then use "summary-address"
>~~to summarize
>~~in the second OSPF process. These summarized routes show up
>~~in the routing
>~~table (because that is the only place they appear) and are thus
>~~redistributable. Finally, redistribute *BOTH* OSPF processes
>~~into IGRP (and
>~~make sure that IGRP has a higher admin distance than OSPF or
>~~everything will
>~~break).
>~~ Also, make sure that you don't redistribute the summaries
>~~back into OSPF
>~~as that will cause nasty problems...
>~~
>~~ sjb.
>~~
>~~> -----Original Message-----
>~~> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On
>~~Behalf Of
>~~> Rodrigo Espinha T Da Silva
>~~> Sent: Thursday, February 07, 2002 9:29 AM
>~~> To: ccielab@groupstudy.com
>~~> Subject: solie skinet lab - and the old topic ospf-igrp
>~~>
>~~>
>~~> Hi Guys,
>~~>
>~~> I did the skynet and still have the problem with redistribution
>~~> between ospf
>~~> and igrp :
>~~>
>~~> regardind the mask /28 ospf to /24 igrp , the solution
>~~discussed in email
>~~> before ( 25/jan untilll 29/jan ) regarding create another
>~~ospf process and
>~~> use the summary adrress comand works perfect, I tried to
>~~create in the R3
>~~> router and ABR configuring the loopback in another area , works
>~~> for the ospf
>~~> database , i can see an summarized network geneated for the
>~~area 0 range
>~~> command in R1 , but does'nt solve the redistr problem ( help ).
>~~>
>~~>
>~~> I can't find a solution for /28 ospf to /24 igrp ( reverse
>~~> summarization ),
>~~> someone can send me an numeric example.
>~~> help!
>~~>
>~~> challenge day : 27/fev
>~~>
>~~> thanks in advance
>~~>
>~~> Rodrigo E. Teixeira da Silva
>~~>
>~~>
>~~> Rodrigo E. Teixeira da Silva
>~~> CPM Comunicacoes
>~~> Analista de Suporte Tecnico
>~~> CCNP Lan & Wan Certified.
>~~> Phone: + 55-11-4196-0793
>~~> Fax: + 55-11-4196-0900
>~~> Call Dispach: 0800-117239
This archive was generated by hypermail 2.1.4 : Thu Jun 20 2002 - 13:46:16 GMT-3