From: ccie candidate (ccie1@xxxxxxxxx)
Date: Sat Jul 27 2002 - 00:41:53 GMT-3
there are 3 other methods to solve this problem , however all of them should in
troduce something new ( like IP addressing )which are not particularly on the l
ab
1- create loopbacks inside your ospf domain on the redistribution router , thos
e loopbacks are all of the same mask as the IGRP , put those loopbacks in the s
ame subnet as your OSPF subnets which of different mask .
for example assume you have 172.3.10.0/28 somewhere on your ospf domain ..creat
e loopback with 172.3.10.0/24 on the redistribution router , this network will
propagate to the IGRP domain , the redistribution router will have two subnets
now , the more specific network will work .
2-create secondary addresses on the IGRP domain redistribution router ( this to
allow the IGRP routers to accept differnt subnet masks) to the downstream rout
ers .
3-create tunnels instead of secondary addresses to do the same like 2
the easist way to do this is also to create another ospf process on the redistr
ibutionn router , summarize ospf1 to ospf2 and redistribute both into IGRP
however one of the guys on the list claim that the last method should be unacce
ptable .
if anyone has different opinion ,can post please
--On Fri, 26 Jul 2002 22:38:55 Cade Wagner wrote: > I am curious how these other two methods work. (tunnel and secondary >addressing) Could someone explain these? I have some ideas, but they are >untested: > >Tunnel: > >1. Use addressing in the same subnet with the same mask as what needs to be >distributed. >2. Use addressing in an entirely different subnet so that you get the >summarization effect. > >Secondary: > >1. Not sure here. > >Any help is greatly appreciated. > >Cade > >-----Original Message----- >From: ccie candidate [mailto:ccie1@lycos.com] >Sent: Thursday, July 25, 2002 3:42 PM >To: Donny MATEO; Anthony Pace >Cc: ccielab@groupstudy.com >Subject: Re: Redistrbuting from OSPF to RIP/IGRP > > on previous post by one CCIE guy >he said this technique is not allowed on the lab ?? >however techniques like tunnel and secodary ip addresses is acceptable . >can anyone confirm this ? and why ?? > > > > > >-- > >On Thu, 25 Jul 2002 18:48:57 > Anthony Pace wrote: >>Donny, >> >>THis sounds correct. It sounds like the same principle which causes you >>to have to do "full mesh", 3 way redistribution on a router with 3 >>routing protocols to be redistributed. I have noticed that in this >>scenario the same thing happens. >> >>Anthony PAce >> >> >>On Thu, 25 Jul 2002 11:43:04 +0800, "Donny MATEO" >><donny.mateo@sg.ca-indosuez.com> said: >>> >>> I'm not sure but perhaps >>> >>> ospf 1 is distributed to ospf 2. >>> then ospf 2 is distribute to igrp. >>> All this is done under one router. >>> >>> The question is why the route of ospf 1 does not appear in the routing >>> table of igrp. >>> I'm not sure but perhaps it has something to do with the fact that the >>> route that is distributed to >>> other routing protocol has to appear in the routing table ( this is >>> where I might be wrong... ) >>> If this happens in a single router, the routing table would be that of >>> the ospf 1 process (as in >>> ospf 2 it would be external). So when you redistribute to ospf 2 to >>> igrp, only the "summarized" >>> route appears cause that one is in the routing table and known from >>> ospf 2. While the rest of the >>> route osfp 2 knows are external and are know in ospf 1 as internal, >>> which is prefered and listed in >>> the routing table. >>> I will have to test this to verify, but I'm sure someone in the list >>> would have the answer by now. >>> Search the archive, I believe this had been discussed before. >>> >>> Donny >>> >>> >>> > >>> "Anthony Pace" >>> <anthonypace@fast To: "ccie >>> candidate" <ccie1@lycos.com>, >>> ccielab@groupstudy.com, "jin" >>> mail.fm> >>> <jin10101010@hotmail.com> >>> Sent by: cc: >>> nobody@groupstudy Subject: Re: >>> Redistrbuting from OSPF to RIP/IGRP >>> .com >>> > >>> > >>> 25-07-2002 01:18 >>> Please respond to >>> "Anthony Pace" >>> > >>> > >>> >>> >>> >>> >>> I had a question earlier in this thread: >>> >>> I have also used this 2 process method but still am curious as to why >>> both OSPF processes need to be REDISTRIBUTED into IGRP. I have found >>> that this is needed; but it seems like the second process would contain >>> a full set of the OSPF routes and I would think it would be the only >>> thing that would need to be RED into IGRP. DOes anyone know why both >>> need to go into IGRP? >>> >>> The answer seemed to "the requirements of the lab asked for the first >>> process to be redistributed". Setting the requiremments of the lab >>> aside, why won't this work (it won't work for me): >>> >>> OSPF1 => OSPF2 => IGRP >>> >>> This works: >>> >>> OSPF1 => OSPF2 => IGRP >>> OSPF1 => IGRP >>> >>> >>> >>> >>> >>> >>> On Wed, 24 Jul 2002 03:08:55 -0700, "ccie candidate" <ccie1@lycos.com> >>> said: >>> > well i didnt get all your points ..however the two ospf processes is >>> > just working as perfect solution for the summary problem . >>> > the question is to redistribute the ospf running on the interfaces into >>> > IGRP , so you SHOULD fulfill this requirement , the other process is >>> > your own way to solve the summarization issue ..so you end up >>> > redistibuting both .. >>> > >>> > >>> > good luck >>> > -- >>> > >>> > On Wed, 24 Jul 2002 13:37:52 >>> > jin wrote: >>> > >Right, >>> > >ospf and igrp should be redistributed mutually. >>> > >but he told us 'redistributed' , only about 'redistributed'. >>> > >If we already made static route or default route, we can use the >static and default route >>> origination. >>> > >but if we not make that already, we can't use anything. >>> > >Should Be only Redistributed. >>> > > >>> > >I think. >>> > >Only way for that problem is Understanding how to use of Summary >address command on the ospf. >>> > >The important thing is that summary address command can summarize the >any routes that isn't exist >>> on the routing table Tagging OSPF. >>> > >If you can understand this, You can redistrubute the ospf into igrp >and rip. >>> > >And I already make a success on that situation. >>> > > >>> > >Thanks. >>> > > >>> > >----- Original Message ----- >>> > >From: "ccie candidate" <ccie1@lycos.com> >>> > >To: "kym blair" <kymblair@hotmail.com>; <ccie1@lycos.com>; ><fangloma@pacific.net.hk>; >>> <Darryl.Munro@computerland.co.nz>; "Anthony Pace" >>> <anthonypace@fastmail.fm> >>> > >Cc: <ccielab@groupstudy.com> >>> > >Sent: Tuesday, July 23, 2002 5:03 AM >>> > >Subject: Re: Redistrbuting from OSPF to RIP/IGRP >>> > > >>> > > >>> > >> probably because the question is asking you to redistribute the ospf >(ospf1) into IGRP on that >>> router .:)))) >>> > >> >>> > >> good point ..HAH >>> > >> >>> > >> >>> > >> -- >>> > >> >>> > >> On Mon, 22 Jul 2002 18:28:40 >>> > >> Anthony Pace wrote: >>> > >> >I have also used this 2 process method but still am curious as to >why >>> > >> >both OSPF processes need to be REDISTRIBUTED into IGRP. I have >found >>> > >> >that this is needed; but it seems like the second process would >contain >>> > >> >a full set of the OSPF routes and I would think it would be the >only >>> > >> >thing that would need to be RED into IGRP. DOes anyone know why >both >>> > >> >need to go into IGRP? >>> > >> > >>> > >> >Anthony Pace >>> > >> > >>> > >> > >>> > >> > >>> > >> >On Sat, 20 Jul 2002 23:28:26 +0000, "kym blair" ><kymblair@hotmail.com> >>> > >> >said: >>> > >> >> C, >>> > >> >> >>> > >> >> Example OSPF1 area, you have: >>> > >> >> >>> > >> >> 192.168.1.0/24 >>> > >> >> 192.168.2.0/24 >>> > >> >> 192.168.3.0/26 >>> > >> >> >>> > >> >> redistribute ospf1 into IGRP, but IGRP only receives .1 and .2 >>> > >> >> networks. >>> > >> >> Solution: >>> > >> >> >>> > >> >> router ospf 2 >>> > >> >> redistribute ospf 1 metric-type 1 subnets >>> > >> >> summary-address 192.168.3.0 255.255.255.0 >>> > >> >> >>> > >> >> router igrp 100 >>> > >> >> redistribute ospf 1 metric 1000 100 255 1 1500 >>> > >> >> redistribute ospf 2 metric 1000 100 255 1 1500 >>> > >> >> >>> > >> >> Of course add appropriate filtering and passive-interfaces. >>> > >> >> >>> > >> >> HTH, Kym >>> > >> >> >>> > >> >> >>> > >> >> >From: "ccie candidate" <ccie1@lycos.com> >>> > >> >> >Reply-To: "ccie candidate" <ccie1@lycos.com> >>> > >> >> >To: fangloma@pacific.net.hk, Darryl.Munro@computerland.co.nz, >"kym >>> > >> >> >blair" <kymblair@hotmail.com> >>> > >> >> >CC: ccielab@groupstudy.com >>> > >> >> >Subject: Re: Redistrbuting from OSPF to RIP/IGRP >>> > >> >> >Date: Sat, 20 Jul 2002 14:44:23 -0700 >>> > >> >> > >>> > >> >> > guys ; >>> > >> >> >im still having confusing about this method . >>> > >> >> > >>> > >> >> >if you create an OSPF2 process , and you want to summarize the >OSPF1 into >>> > >> >> >it , again you are using the summary command into the wrong >direction !!! >>> > >> >> >,summary address is supposed to summarize external routes into >OSPF1 and >>> > >> >> >not OSPF1 internal non-classful routes into OSPF2 ...am i right >or im >>> > >> >> >missing something here . >>> > >> >> > >>> > >> >> >this subject has been killed on this mailing list hundered of >times >>> > >> >> >..however i didnt find any clue for it . >>> > >> >> > >>> > >> >> >can any folk post the right dierctions to solve this problem ..i >would >>> > >> >> >appreciate if anyone correct my concepts. >>> > >> >> > >>> > >> >> >candidate >>> > >> >> > >>> > >> >> > >>> > >> >> > >>> > >> >> > >>> > >> >> >-- >>> > >> >> > >>> > >> >> >On Sat, 20 Jul 2002 13:44:32 >>> > >> >> > kym blair wrote: >>> > >> >> > >Darryl, >>> > >> >> > > >>> > >> >> > >There are a couple methods. The one many people like is to >create a >>> > >> >> >second >>> > >> >> > >OSPF process, redistribute the first ospf process into the >second, >>> > >> >> >summarize >>> > >> >> > >each non-classful network under the second ospf process, then >>> > >> >> >redistribute >>> > >> >> > >both ospf processes into RIP/IGRP. >>> > >> >> > > >>> > >> >> > >HTH, Kym >>> > >> >> > > >>> > >> >> > > >>> > >> >> > > >>> > >> >> > > >>> > >> >> > >>From: Fanglo MA <fangloma@pacific.net.hk> >>> > >> >> > >>Reply-To: Fanglo MA <fangloma@pacific.net.hk> >>> > >> >> > >>To: Darryl Munro <Darryl.Munro@computerland.co.nz> >>> > >> >> > >>CC: Group Study <ccielab@groupstudy.com> >>> > >> >> > >>Subject: Re: Redistrbuting from OSPF to RIP/IGRP >>> > >> >> > >>Date: Sat, 20 Jul 2002 15:59:03 +0800 (HKT) >>> > >> >> > >> >>> > >> >> > >>Would you consider using route-map to direct summary address >point to >>> > >> >> > >>null0 to replace the static route functionality? >>> > >> >> > >> >>> > >> >> > >>Regards, >>> > >> >> > >>Fanglo >>> > >> >> > >> >>> > >> >> > >>On Sat, 20 Jul 2002, Darryl Munro wrote: >>> > >> >> > >> >>> > >> >> > >> > How is it possible to redistribute from OSPF to IGRP/RIP >without >>> > >> >> >using >>> > >> >> > >> > statics to Null0? I know that the mask needs to be the >same as the >>> > >> >> > >>IGRP/RIP >>> > >> >> > >> > domain, however is it achievable to do this with area >range commands >>> > >> >> >and >>> > >> >> > >> > summary-address's positioned at the right the places in >your OSPF >>> > >> >> > >>domain? >>> > >> >> > >> > Area range should take care of all of the OSPF inter area >routes and >>> > >> >> > >>summary >>> > >> >> > >> > address the external addresses from other routing >protocols. I just >>> > >> >> > >>can't >>> > >> >> > >> > seem to work this one out in my lab. Any suggestions would >be >>> > >> >> > >>appreciated. >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > TIA >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > Darryl Munro >>> > >> >> > >> > >>> > >> >> > >> > CNE, MCSE, CCNP, CCDP, CCEA >>> > >> >> > >> > >>> > >> >> > >> > Systems Consultant >>> > >> >> > >> > >>> > >> >> > >> > Computerland NZ >>> > >> >> > >> > >>> > >> >> > >> > 104-106 Customs St West >>> > >> >> > >> > >>> > >> >> > >> > PO Box 3631, Auckland >>> > >> >> > >> > >>> > >> >> > >> > Phone: 09 306 8700 >>> > >> >> > >> > >>> > >> >> > >> > Cell Phone 027 2897786 >>> > >> >> > >> > >>> > >> >> > >> > Darryl <mailto:darryl.munro@computerland.co.nz> Munro >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > >>> > >> >> > >> > CAUTION: This e-mail message and accompanying data may >contain >>> > >> >> > >>information >>> > >> >> > >> > that is confidential and subject to privilege. If you are >not the >>> > >> >> > >>intended >>> > >> >> > >> > recipient, you are notified that any use, dissemination, >distribution >>> > >> >> >or >>> > >> >> > >> > copying of this message or data is prohibited. If you >have received >>> > >> >> > >>this >>> > >> >> > >> > e-mail in error, please notify me immediately and delete >all material >>> > >> >> > >> > pertaining to this e-mail. Ceritas / Computerland will not >accept >>> > >> >> > >>liability >>> > >> >> > >> > for any loss or damage caused by using any material or >attachments >>> > >> >> > >>contained >>> > >> >> > >> > in this message. While every best practice has been taken >to, no >>> > >> >> > >>warranty is >>> > >> >> > >> > made that this material is free from computer virus or >other defect. >>> > >> >> > >> > Ceritas/Computerland's entire liability will be limited to >>> > >> >> >resupplying >>> > >> >> > >>the >>> > >> >> > >> > material. Thank you >>> > >> >> > >> >
This archive was generated by hypermail 2.1.4 : Sat Sep 07 2002 - 19:36:46 GMT-3