RIP routing.

[nizamshaikh2504]

Here i have created a 3 network of IP Addresses 198.162.16.0/20 , 32/20 , 48/20. 

configured both the router but while routing the network using ...#network 198.162.16.0 and ....#network 198.162.32.0 in the router-rip mode , network is not getting added .

but whenever i m taking CIDR as /24 the network is adding.

why so?

help me ASAP!

[Joseph W. Doherty]

Original RIP is a classful routing protocol and 192.168.0.0 is a class C (/24).

If you want to work with non-classful subnets, you'll want to run RIPv2.

(At least I believe above - generally I don't use RIP and I recall RIPv2 works a bit oddly when using non-classful networks.  If Rick see this post, he is likely to know for sure.)

[Kevin Rivest]

Joseph is correct for the most part, but just a little bit to add.

RIPv1 is a completely classful routing protocol. You cannot advertise any networks other than at classful boundaries because the subnet mask is not included in the routing update.

RIPv2 has support for VLSM. It does include the subnet mask in routing updates. The configuration at the command line still reflects the original classful nature of RIP though and assumes the classful mask. However if you configure version 2 and no auto-summary it will be able to advertise subnets that fall within the range of the network command.

RIPv2 however does not have support for CIDR (supernetting) through summarization. You can however advertise the networks with a network command and auto summarization turned off.

Hopefully this answers your question.

[paul driver]

Hello

Just like to add, ripv1 will only advertise prefixes that match the subnet mask of its interface, so in your case what you are seeing is correct because as stated by others rip is classful, the down side to that is your stub networks cannot be advertised correctly.

If the interconnects between the two advertising rts in this topology had ip addresses that were within the classful subnet, then your stub networks would be advertised ONLY within their classful ip range however with this topology above 192.162.32.0/20 and 192.162.48.0/20  dont fall into a classfull network so they won’t be advertised at all.

Ripv2 by default is still classful with auto-summarisation being enabled by default and even if you disable that you can still can only apply classful network statements to the routing stanza, It will only recognize variable length subnetmask like in your topology above if the interconnects between the two rts have each an ip address within a classful subnet, you have enable rip version2 and disabled summarisation

Redistribution is also a effected by the default nature of this protocol, So if you try and redistribute between say a cidr’d prefix from a classless routing protocol into ripv1, then rip wont except it, it will only except classful prefixes that match its classful interfaces

Using Ripv1.
1) only send - receive ripv 1 updates
2) Can disable auto -summarization but it has no effect
3) the interconnects between the two advertising rts need to each an ip address within a classful subnet.
4) your stub networks will only be advertised as classful even if they have a longer subnetmask

Using Ripv2

1) Can be configured to send-receive Ripv1 or Ripv2 updates
2) Can disable auto -summarization
3) the interconnects between the two advertising rts still need each ip address within a classful subnet.
4) your stub networks will be advertised with their correct subnet mask

res
Paul

[nizamshaikh2504]

could you please make video and explain it to me with both the version using same ip addresses.

that would be very helpful for me.

PLEASE

[paul driver]

Hello

Apologies, I dont make videos however there is plenty of materiel on CCO explaining this routing protocol and no doubt videos will be available somewhere

res
Paul

[nizamshaikh2504]

need links