08-09-2006 08:44 AM - edited 03-03-2019 01:36 PM
From Routing TCP/IP Volume II P212
Ex 3-79
The book highlights 3 next hops for 10.20.0.0/16 but I only find two:
Network Next Hop
* 10.20.0.0/16 10.200.60.1 0 200 50 ?
*> 10.50.250.1 0 50 ?
The thing that throws me is why isis level-2 is redistributed into bgp 30 when it origionates from th esame area - shouldn't it be level-1?
I don't suppose anyone else has tried this lab and has saved the configs??
08-09-2006 12:24 PM
There is indeed 3 NH for 10.20.0.0/16. 1 via Meribel, 1 via Cervinia and finally 1 via Zermatt. Bear in mind that there is a NAP at the top of the diagram, where Innsbruck can connect directly to Meribel and Cervinia.
As for your other question, it is very common to see large service provider running a flat level-2 network. Therefore, there is no level-1 routers at all, just level-2.
Hope this helps,
08-10-2006 03:52 AM
So are you saying I should configure the ISIS cloud so that both Zermatt & Moritz should detect these routes as L2? I did this my manipulating my R5 area.
I manage to TX & Rx the RIP & ISIS networks round the ring in both directions but I still can't get Innsbruck to learn the RIP networks through Moritz-Zermatt for some reason.
I only have one hub so maybe my topology is wrong?? Or, I have a missconfiguration??
I have attached my drawing and results - if you have time :-)
Cheers!!
08-10-2006 04:23 AM
1- You should in principle configure Zermatt and Moritz (and R5 in you case) to only establish a L2 adjacency between themselves.
This can be done by adding the following statement under router isis:
is-type level-2-only
2- "I still can't get Innsbruck to learn the RIP networks through Moritz-Zermatt for some reason."
Could you please attach a "show ip route" from Zermatt to help us determine why you are not learning these routes on Innsbruck.
Thanks,
08-10-2006 05:00 AM
08-10-2006 06:54 AM
According to a previous "sh ip bgp" taken from Innsbruck, 10.20.0.0/16 was learnt from 3 different NHs.
* 10.20.0.0/16 172.30.255.254 10 0 30 ?
* 10.200.60.1 0 200 50 ?
*> 10.50.250.1 1 0 50 ?
And in a subsequent output from the same router, you had 2. This is actually a timing issue.
Let me explain. If Zermatt receives 10.20.0.0/16 from via BGP from Innsbruck before it receives it from ISIS, it is selected as the best path and installes in the RIB. Since the BGP best path is from Innsbruck, you don't announce it back to that same router, which leaves Innsbruck with two BGP NH for 10.20.0.0/16.
On the other hand, if Zermatt receives 10.20.0.0/16 from ISIS first, it redistributes it into BGP and Innsbuck see 3 BGP NHs.
Hope this helps,
08-10-2006 07:46 AM
This does help!
I noticed that when I clear ip bgp * at Zermatt and Moritz (different times) the learned routes toggle between being BGP & ISIS at these two routers. So, if 10.20.0.0/16 is learned by Zermatt as L2 ISIS from Moritz, then at the same time Moritz shows this network as learned through BGP from Cervinia.
It really does prove your point!!
One last thing...
If Zermatt first learns 10.20.0.0/16 through ISIS 10.20.0.0/16 [115/10] why doesn't it change when it later receives the same network from Innsbruck with a lower AD 10.20.0.0/16 [20/0] via 10.200.60.1 ??
Thanks for your help once again.
08-11-2006 05:57 AM
"If Zermatt first learns 10.20.0.0/16 through ISIS 10.20.0.0/16 [115/10] why doesn't it change when it later receives the same network from Innsbruck with a lower AD 10.20.0.0/16 [20/0] via 10.200.60.1 ??"
This is because the locally sourced BGP path(redistributed from ISIS) will be preferred over the eBGP learnt from Innsbruck. This is because sourced BGP paths have a weight of 32768 and will therefore win over just about any other BGP paths.
The AD will not come into play on this selection.
Hope this helps,
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: