Fair call Rick. Well I daresay they've sorted the issue in the later
12.2S versions. Still good to know about the issue nonetheless.

I hear you on 12.2T also. Yesterday I was working on a lab with a
122-15T router. It was a PE and exchanging labels using BGP - however
CEF wasn't picking these up properly. Workaround was to redistribute
the Inter-AS PE route into the IGP to get things working.

Oh well - that's the game we're in I suppose. Making things work with
what you've got.


On Sun, Aug 30, 2009 at 11:33 PM, Rick Mur<[email protected]> wrote:
> I had this constantly on a 12.2S 7200 on a PE-CE connection and when
> advertising an EIGRP summary to a CE so the interface being in a VRF.
> I can't recall the exact IOS version, but I did most the labs on the
> ProctorLabs ATM pod.
> --
>
> Regards,
>
> Rick Mur
> CCIE2 #21946 (R&S / Service Provider)
> Sr. Support Engineer – IPexpert, Inc.
> URL: http://www.IPexpert.com
>
>
> On Sun, Aug 30, 2009 at 11:17 AM, Con Spathas <[email protected]>
> wrote:
>>
>> I agree with Rick that this is a bug - except that I could not
>> re-produce the issue.
>>
>> I tested using the following "SP" blueprint IOS's.
>>
>> 12.2(11)T (note: VRF aware EIGRP not available until 12.2(15)T but
>> it's not available on CCO any longer)
>> 12.3(8)T11
>> 12.2(18)S13 <--- closest I could get to your IOS Bryan
>>
>> I performed a whole heap of tests creating eigrp adjacencies between
>> all these versions.
>>
>> I used the following interfaces types:
>>
>> FastEthernet Dot1q SubIF
>> ATM P2P SubIf
>>
>> and they were placed either the following routing tables (except for
>> 12.2(11)T for VRF see above note):
>>
>> Global <-> Global
>> Global <-> VRF
>> VRF <-> VRF
>>
>> After a lengthy amount of testing with the 0/0 summary, I could not
>> re-create Bryans issue with any combination of the above interface
>> types, tables or IOSs.
>> Any chance Bryan you could re-try with the 12.2(18)S13 IOS I've listed
>> above?
>>
>> Cheers,
>> Con.
>>
>> On Sat, Aug 29, 2009 at 9:58 PM, Rick Mur<[email protected]> wrote:
>> > Ah true, that's the advantage of a distance vector protocol :-)  Still a
>> > huge bug in IOS though
>> >
>> > --
>> > Rick Mur
>> > CCIE2 #21946 (R&S, Service provider)
>> > Sr. Support Engineer -- IPexpert, Inc.
>> > Sent from my iPhone
>> > On 29 aug 2009, at 22:43, Bryan Bartik <[email protected]> wrote:
>> >
>> > Yep, still does. The AD is just for the null route,
>> >
>> > On Sat, Aug 29, 2009 at 2:43 PM, Rick Mur <[email protected]> wrote:
>> >>
>> >> But does the summary get advertised then? I remember not.
>> >>
>> >> --
>> >> Rick Mur
>> >> CCIE2 #21946 (R&S, Service provider)
>> >> Sr. Support Engineer -- IPexpert, Inc.
>> >> Sent from my iPhone
>> >> On 29 aug 2009, at 22:17, Bryan Bartik <[email protected]> wrote:
>> >>
>> >> Thanks for the replies guys. Yep, It's c7200-p-mz.122-18.S2.bin.
>> >> The AD 255 thing was a shot in the dark and it worked :)
>> >>
>> >> On Sat, Aug 29, 2009 at 2:13 PM, Con Spathas <[email protected]>
>> >> wrote:
>> >>>
>> >>> Gday Bryan,
>> >>>
>> >>> A couple of questions:
>> >>>
>> >>> 1. Can you let me know IOS version/platform you're using (I want to
>> >>> try it) - At a guess I'd say 7200 if you're playing with ATM.
>> >>> 2. Is the summary you're configuring under a global or VRF interface?
>> >>> 3. Any other kookie stuff going on where R7 might be receiving or
>> >>> advertising a 0/0 from another protocol? (again to try and recreate
>> >>> issue in my lab)
>> >>>
>> >>> I'll play with this first thing in the AM - it's telly (read: wind
>> >>> down) time now! :-)
>> >>>
>> >>> Cheers,
>> >>> Con.
>> >>>
>> >>> On Sat, Aug 29, 2009 at 7:12 PM, Bryan Bartik<[email protected]>
>> >>> wrote:
>> >>> > Have you guys ever seen this?
>> >>> >
>> >>> > R7(config)#interface ATM1/0.711 point-to-point
>> >>> > R7(config-subif)#ip summary-address eigrp 117 0.0.0.0 0.0.0.0
>> >>> > R7(config-subif)#^Z
>> >>> > R7#
>> >>> > 04:19:34: %DUAL-5-NBRCHANGE: IP-EIGRP(2) 117: Neighbor
>> >>> > 117.117.117.12
>> >>> > (ATM1/0.711) is down: summary configured
>> >>> >
>> >>> > The neighbor never comes back up! Remove the summary
>> >>> >
>> >>> > R7(config-subif)#no  ip summary-address eigrp 117 0.0.0.0 0.0.0.0 5
>> >>> > R7(config-subif)#
>> >>> > 04:21:06: %DUAL-5-NBRCHANGE: IP-EIGRP(2) 117: Neighbor
>> >>> > 117.117.117.12
>> >>> > (ATM1/0.711) is up: new adjacency
>> >>> > R7(config-subif)#
>> >>> >
>> >>> > Comes back up!
>> >>> >
>> >>> > On a hunch I decided to use AD 255:
>> >>> >
>> >>> > R7(config-subif)#ip summary-address eigrp 117 0.0.0.0 0.0.0.0 255
>> >>> > R7(config-subif)#
>> >>> > 04:23:55: %DUAL-5-NBRCHANGE: IP-EIGRP(2) 117: Neighbor
>> >>> > 117.117.117.12
>> >>> > (ATM1/0.711) is up: new adjacency
>> >>> >
>> >>> > It's up! Is this correct behavior?
>> >>> >
>> >>> > --
>> >>> > Bryan Bartik
>> >>> > CCIE #23707 (R&S), CCNP
>> >>> > Sr. Support Engineer - IPexpert, Inc.
>> >>> > URL: http://www.IPexpert.com
>> >>> >
>> >>> > _______________________________________________
>> >>> > For more information regarding industry leading CCIE Lab training,
>> >>> > please
>> >>> > visit www.ipexpert.com
>> >>> >
>> >>> >
>> >>
>> >>
>> >>
>> >> --
>> >> Bryan Bartik
>> >> CCIE #23707 (R&S), CCNP
>> >> Sr. Support Engineer - IPexpert, Inc.
>> >> URL: http://www.IPexpert.com
>> >>
>> >> _______________________________________________
>> >> For more information regarding industry leading CCIE Lab training,
>> >> please
>> >> visit www.ipexpert.com
>> >
>> >
>> >
>> > --
>> > Bryan Bartik
>> > CCIE #23707 (R&S), CCNP
>> > Sr. Support Engineer - IPexpert, Inc.
>> > URL: http://www.IPexpert.com
>> >
>
>
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Reply via email to