advertising a non-existent network?

mguymguy Member Posts: 167 ■■■□□□□□□□
I have a general question. two, actually. been bugging me.

#1 How come I can't have RIP advertise a non-existent network?

I'm trying to practice NATing and I want R2 to think R1 has the network I just performed NAT on.

(R1)ip nat inside source list 1 1.1.1.1 overload

(R1)router rip
network 1.1.1.1

Result: R2 will not see I that R1 has network and will just drop the packet

#2 This bugs me
When you perform a network statement how come you don't have to include the mask?

Say I have a network: 192.168.1.8 255.255.255.255.248

So..
(R1) router rip
(R1) version 2
(R1) network 192.168.1.0

but tadah! it knows the mask.. automagically.. ? icon_confused.gif:

Comments

  • stlsmoorestlsmoore Member Posts: 515 ■■■□□□□□□□
    I think for #2 RIPv2 assumes the subnet mask is 255.255.255.248 because there's an interface on that router a part of that particular subnet already. Hopefully, someone smarter than me can chime in though lol.
    My Cisco Blog Adventure: http://shawnmoorecisco.blogspot.com/

    Don't Forget to Add me on LinkedIn!
    https://www.linkedin.com/in/shawnrmoore
  • NetworkVeteranNetworkVeteran Member Posts: 2,338 ■■■■■■■■□□
    #1 How come I can't have RIP advertise a non-existent network?
    You can advertise any network you like, whether it exists in your autonomous system or not!

    R1(config)#ip route 2.0.0.0 255.0.0.0 null0
    R1(config)#router rip
    R1(config-router)#network 1.0.0.0
    R1(config-router)#redistribute static
    R1(config-router)#end
    R1#sh
    *Mar 1 09:15:41.477: %SYS-5-CONFIG_I: Configured from console by console
    R1#show ip rip database
    1.0.0.0/8 auto-summary
    1.1.1.0/24 directly connected, FastEthernet0/0
    2.0.0.0/8 auto-summary
    2.0.0.0/8 redistributed
    [1] via 0.0.0.0,

    #2 This bugs me
    When you perform a network statement how come you don't have to include the mask?... (R1) network 192.168.1.0
    192.168.1.0 is a class C address so the network mask is 255.255.255.0. As for 192.168.1.8/30, it doesn't know the subnet mask "auto-magically", it checks the RIB (show ip route connected).
    Hopefully, someone smarter than me can chime in though lol.
    You're out of luck! It's just me, and I've already had a couple beers tonight. :p
  • networker050184networker050184 Mod Posts: 11,962 Mod
    I think your confusion comes from what the network command actually does. The network command is there to tell the router what interface to include in the routing process. It doesn't actually tell the router what network to advertise, that is taken from the actual configuration on the interface.
    An expert is a man who has made all the mistakes which can be made.
  • YFZbluYFZblu Member Posts: 1,462 ■■■■■■■■□□
    @networker050184: Great post, beat me to it
  • mguymguy Member Posts: 167 ■■■□□□□□□□
    So the network command doesn't make the router say "Hi other routers! I know about this network!!(1.1.1.0)" but rather it is merely activating the interfaces that fall into the network.

    Thus,

    (R1) network 1.1.1.0

    without an actual interface configured with that IP address, doesnt activate the sending of advertisements on a valid interface address, and thus null. So basically, you can't advertise a non-existent network because you will need an interface in that non-existend network in the first place?! At least until ICDN2.. dunno about redistribution.. yet..
  • ZartanasaurusZartanasaurus Member Posts: 2,008 ■■■■■■■■■□

    Right. The network command is including interfaces into the RIP process and advertising the network associated with that interface. If none of your interfaces match the network statement, then it doesn't include anything in the process. The network statement is pretty vanilla in RIP vs EIGRP/OSPF where you can use a mask to be more specific with which interfaces you want to include (OSPF even takes it further by allowing you to activate OSPF directly under the interface).

    Also remember that RIPv2 includes the subnet mask in its update messages, whereas RIPv1 doesn't. So it's RIPv1 that makes assumptions about what the mask is based on the receiving interface. RIPv2 neighbors are told explicitly.

    Now if you ever get to BGP, forget everything we just told you about how the network command works. icon_lol.gif
    Currently reading:
    IPSec VPN Design 44%
    Mastering VMWare vSphere 5​ 42.8%
  • NetworkVeteranNetworkVeteran Member Posts: 2,338 ■■■■■■■■□□
    dunno about redistribution.. yet..
    You DO know that redistribution can cause RIP to redistribute a network that doesn't exist in the local autonomous system, because I already showed you that above.
    So basically, you can't advertise a non-existent network because you will need an interface in that non-existend network in the first place?!

    Even ignoring the redistribute command, there are multiple ways to advertise networks that don't really physically exist in your autonomous system--

    R1(config)#router rip
    R1(config-router)#network 1.0.0.0
    R1(config-router)#default-information originate
    R1(config-router)#end
    R1#sh ip r
    *Mar 2 07:38:00.448: %SYS-5-CONFIG_I: Configured from console by console
    R1#sh ip rip database
    0.0.0.0/0 auto-summary
    0.0.0.0/0 redistributed
    [1] via 0.0.0.0,
    1.0.0.0/8 auto-summary
    1.1.1.0/24 directly connected, FastEthernet0/0

    This is the entirety of what RIP actually sends on the Fa0/0 interface--

    Routing Information Protocol
    Command: Response (2)
    Version: RIPv1 (1)
    IP Address: 0.0.0.0, Metric: 1
    Address Family: IP (2)
    IP Address: 0.0.0.0 (0.0.0.0)
    Metric: 1
Sign In or Register to comment.