RIPv1 is pretty outdated and as you know the it checks the outgoing and incoming masks to compute against the routes it already knows.

This is the explanation on the why it doesnt entertain classless..

http://www.cisco.com/en/US/tech/tk364/technologies_tech_note09186a0080093f1e.shtml

Please add two more line under your "router rip"

ver 2 <---- use version 2

no auto <--- not to automatically summarize

The response from Tharak was a good explanation of how to work around VLSM issues when using RIP. But that is not what the original poster was asking, if I read the question correctly the original poster was asking how he could experiment and see the effect in RIP v1 of trying to use VLSM and to see how it did not work.

To do that I believe that you need at least 2 routers. What was described in the original post is a good start. Let us assume that we have R1 with this configuration:

interface fasteth0/0

description interface with /29 mask

ip address 192.168.10.1 255.255.255.248

interface fasteth0/1

description interface with /28 mask

ip address 192.168.10.97 255.255.255.240

router rip

network 192.168.10.0

and let us assume that there is a second router R2 with this configuration

interface fasteth0/0

description interface with /29 mask

ip address 192.168.10.2 255.255.255.248

router rip

network 192.168.10.0

This gives a router running RIP over the /29 connection. So on R2 do show ip route. You should find that the router is not learning the /28 subnet. This is the demonstration of the problem with VLSM when using RIP v1.

HTH

Rick