Eigrp and auto-summarize

[ycae]

Dear all,

I am preparing my ccna exam and was reading this evening through the EIGRP protocol. At a point the autor writes that if you have a discontiguous network, you have to use the "no auto-summarize" command to make it possible that the routing updates are done correctly between the routers in the AS.

I tried a small netowrk on Boson NetSim where i had 2 routers A and B connected with a serial line and a network of 10.3.1.0/24 and on one router (A) i had a network like 172.16.10.0/24 on the fa interface and on the other one (B) a network called 172.16.20.0/24.

I configured than the eigrp and i didn't use the "no auto-summarize" command. So theoreticaly if i understood everything right i wouldn't be able to ping the 172.16.10.0/24 from router B.....but it worked....i could ping it. And if i checked the routing table everything was just fine.

So my question now....Is it possible that there is a bug in the Boson NetSim software or did i understand something wrong?

Could somebody please give me some light in this matter.

Thanks a lot for your help.

cu Yves

[m-geisler]

Hello Yves,

As I understand the problem, you connect two routers, and their common network is 10.3.1.0/24.

Now each router has exactly one network 172.16.?.? on the other interface? What should happen is that e.g. one router summarizes his 172.16.20.0/24 network and advertises it as 172.16.0.0/ 16 (ClassB) to his neighbour.

The other router now knows about his 172.16.20.0/24 network (as C = directly connected) and 172.16.0.0/16 on the other side. He should be able to reach each network (longest match!).

To break the network try at least 3 routers. Same setup

Mike

[ycae]

the point is that i took basicaly the example they gave in the course. I even found the same example in a cisco press article where they explained the same thing.

i will of course also try the 3 router approch but i am still wondering why they give this example in the course book even if it is not needed to have the "no auto-summarize" command.

What they say in the book is that both routers will start sending updates like 172.16.0.0/16 and that both routers will refuse these updates since they think they know about this network and thus you will not be able to reach the net work on the other side.

here is the article i've found from cisco press....just search for "discontiguous" in the document and you will find the example i am talking about.

http://www.ciscopress.com/articles/article.asp?p=27839&seqNum=3&rl=1

Yves

[vladrac-ccna]

hello Yves,

how are you doing?

not sure how this Boson software works, if it indeed implement a full IOS, but anyway:

If you want to see what routes EIGRP is advertising to its neighbor try :

debug ip eigrp AS (dont forget terminal monitor in case you're not in console).

Now, check for the routing tables of both routers.(you dont need 3 routers to check this issue).

You'll probably see that on router 1 you have a 172.16.0.0/16 pointing to router 2 , and in router 2 you have the same 172.16.0.0 pointing to router 1.

So, you have routes to reach the network you're trying to reach.

I'm not sure this set up you have is the best to check problems auto-summary will cause, but again you'll clearly see on the debug what he's not adverstising out.

1d18h: IP-EIGRP(Default-IP-Routing-Table:100): 172.16.10.0/24 - don't advertise out FastEthernet0/1

1d18h: IP-EIGRP(Default-IP-Routing-Table:100): 10.3.1.0/24 - do advertise out FastEthernet0/1

1d18h: IP-EIGRP(Default-IP-Routing-Table:100): 172.16.0.0/16 - do advertise out FastEthernet0/1

HtH,

If it does, I'd appreciate if you rate this post.

Vlad

[ycae]

Hi Vlad,

I've set up the example of cisco press with 2 1603 routers to have the same lab as in the book....and if i follow the configuration, i cannot ping the ethernet interface of the routers. here is the output of the ip table of Lab_A router

Lab_A#sh ip route

Codes: C - connected, S - static, I - IGRP, R - RIP, M - mobile, B - BGP

D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area

N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2

E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP

i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, * - candidate default

U - per-user static route, o - ODR

Gateway of last resort is not set

172.16.0.0/16 is variably subnetted, 2 subnets, 2 masks

C 172.16.10.0/24 is directly connected, Ethernet0

D 172.16.0.0/16 is a summary, 00:03:14, Null0

10.0.0.0/8 is variably subnetted, 2 subnets, 2 masks

C 10.1.3.0/24 is directly connected, Serial0

D 10.0.0.0/8 is a summary, 00:03:18, Null0

Lab_A#ping 172.16.20.1

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 172.16.20.1, timeout is 2 seconds:

.....

Success rate is 0 percent (0/5)

Lab_A#

Here the reaction is as descripted in the course book and if i use the no auto-summary command than of course ping works again....see output below:

Lab_A#sh ip ru

%SYS-5-CONFIG_I: Configured from console by consoloute

Codes: C - connected, S - static, I - IGRP, R - RIP, M - mobile, B - BGP

D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area

N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2

E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP

i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, * - candidate default

U - per-user static route, o - ODR

Gateway of last resort is not set

172.16.0.0/16 is variably subnetted, 2 subnets, 2 masks

C 172.16.10.0/24 is directly connected, Ethernet0

D 172.16.0.0/16 [90/2195456] via 10.1.3.2, 00:00:15, Serial0

10.0.0.0/24 is subnetted, 1 subnets

C 10.1.3.0 is directly connected, Serial0

Lab_A#ping 172.16.20.1

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 172.16.20.1, timeout is 2 seconds:

!!!!!

Success rate is 100 percent (5/5), round-trip min/avg/max = 4/4/4 ms

Lab_A#

So i would say that the boson NetSim is not handeling the situation as it should, correct me if i am wrong.

Thanks any way for your help.

Yves