RE: Simple and Often Over Looked RIP

From: Carlos Campos Torres \(ccampost\) (ccampost@cisco.com)
Date: Wed May 24 2006 - 00:26:58 ART


Hi Godswill,

I also wondered the same things while doing some labs. I think it's just
impossible to avoid and sometimes I realized other protocols would get
priority over RIP and would display the routes according to what asked
in the exam due to AD.
I wouldn't care a lot about doing distribute-lists or filtering this
somehow unless explicitly told.

Fortunately, we will only have this problems with RIP in the exam and I
think it will not be the dominant routing protocol in most of the cases.

I think this still falls into healthy paranoid =)

Regards,

Carlos Campos
(919) 392-6285
 

-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Godswill Oletu
Sent: Sunday, May 21, 2006 10:24 AM
To: ccielab@groupstudy.com
Subject: Simple and Often Over Looked RIP

Hi,

Hope I am not getting too paranoid.

I find RIP, to be simple but very very deadly. It is not rich in
features and flavors like the other IGPs and can be sometime Stubborn to
troubleshoot or resolve problems in RIP than in EIGRP, OSPF or BGP.

Most of the commercial labs, that I am been using, employ the class B
address space all over the pod, let take a simple topology like this,
that is using the 172.16.0.0/16 address space?

R1(172.16.10.1)----------------------------(172.16.10.2)-R5-(172.16.20.2
)----------------------(172.16.20.1)R2
 
|
 
(172.16.30.2)
                                                                       |
                                                                       |
                                                                       |
 
(172.16.30.4)
                                                                       |
                                                                     R4

I do not know if the line diagram above will make it accross in good
shape. However, R5 have 3 Fasthethernet interfaces (Fa0/0; Fa0/1 &
Fa0/2) )with IP addresses 172.16.10.2; 172.16.20.2 and 172.16.30.2; all
/24.

Task:
Enable RIPv2 on R5 and advertise Fastethernet0/0 into the RIP domain.

Question:
Since RIP will default to the classfull network when adding interfaces
into the RIP domain, even with RIPv2 and 'no auto-summary' enabled,
should one filter out Fa0/1 and Fa0/2 addresses from the advertisement
into RIP domain?

I am in the habit of filtering them and advertising just what the task
say should be advertised into RIP, but some of the commercial labs I
have seen, do not filter out the other advertisements. Wouldn't the
proctor be cranky to hop on R4 and see not only Fa0/0 address advertised
by RIP but also Fa0/1 address, which he/her did not ask for?

If the keyword 'ONLY' is somewhere in the task, it will lead one to
filter the advertisement and be more restrictive; but with the absence
of that keyword, what is the best practice?

Is it?

!
router rip
version 2
no auto-summary
network 172.16.0.0
!

And move on OR

router rip
version 2
no auto-summary
network 172.16.0.0
distribute-list 1 out
!
access-list 1 deny 172.16.20.0 0.0.0.255 access-list 1 deny 172.16.30.0
0.0.0.255 access-list 1 permit any !



This archive was generated by hypermail 2.1.4 : Thu Jun 01 2006 - 06:33:22 ART