Solved: ip classless command

firewall1981 · ‎01-18-2009

Dear Team,

If we disable ip classless command in routers then the router will discard the packet for unknown subnets which fall winthin a kniwn major network but it forward packets via default route which belongs to different networks.

Now i want to know why the router discard packet which only fall within the known major network why not for unknown networks.

//br

Giuseppe Larosa · ‎01-18-2009

Hello Naven,

the

no ip classless was the default behaviur before IOS release 12.0.

The classful behaviour of a router is related to the classful old routing protocols like RIPv1 and IGRP.

These routing protocols don't carry the subnet mask in their updates this doesn't mean that subnetting is impossible but that fixed length subnet mask has to be used.

the missing information about the subnet mask is inferred from the subnet mask of another subnet in the same major network.

Because at major network boundaries these protocols perform auto-summarization no detail about far major network is known.

And also a major network cannot be partitioned without causing connectivity problems.

So the roots of this scenario are:

we are connected to this major network X, this major network is a whole, we know all subnets in use in this major network

if a packet arrives that doesn't match any known subnet the packet is discarded.

Here known major networks means major networks to which the router is connected: it has an interface at least whose ip address belongs to the network.

it is impossible to know what are the valid subnets in a non connected major network because just the major network can be known.

Actually packets not belonging to connected major networks follow the default route otherwise the default route would be useless.

Hope to help

Giuseppe

View solution in original post

Giuseppe Larosa · ‎01-18-2009

Hello Naven,

the

no ip classless was the default behaviur before IOS release 12.0.

The classful behaviour of a router is related to the classful old routing protocols like RIPv1 and IGRP.

These routing protocols don't carry the subnet mask in their updates this doesn't mean that subnetting is impossible but that fixed length subnet mask has to be used.

the missing information about the subnet mask is inferred from the subnet mask of another subnet in the same major network.

Because at major network boundaries these protocols perform auto-summarization no detail about far major network is known.

And also a major network cannot be partitioned without causing connectivity problems.

So the roots of this scenario are:

we are connected to this major network X, this major network is a whole, we know all subnets in use in this major network

if a packet arrives that doesn't match any known subnet the packet is discarded.

Here known major networks means major networks to which the router is connected: it has an interface at least whose ip address belongs to the network.

it is impossible to know what are the valid subnets in a non connected major network because just the major network can be known.

Actually packets not belonging to connected major networks follow the default route otherwise the default route would be useless.

Hope to help

Giuseppe

Richard Burts · ‎01-18-2009

Naveen

The answer from Giuseppe contains a lot of information. Much of it is good information but not really related to the answer to your question. But there is a key statement in his response which is the key to answering your question about why packets to an unknown subnet of a network for which we know some subnets are discarded. The key concept is this:"this major network is a whole, we know all subnets in use in this major network". A key assumption of classful routing is that if we know any subnet of a network then we will know every subnet that is used in that network.

So the routing logic in classful routing is fairly simple: if we are forwarding a packet to a network about which we know some subnets then we must know the particular subnet to which this packet will go. But if the packet is going to a network about which we know no subnets then it is appropriate to use the default route.

Perhaps another way of looking at the difference between classful routing logic and classless routing logic is that classful believes why should we send a packet to the default route when we believe that the subnet to which it should go does not exist? While classless routing believes that the destination subnet may exist even though we do not have a route to it in our routing table so it makes sense to use the default route.

HTH

Rick

HTH

Rick

firewall1981 · ‎01-20-2009

Your Message was helpful to me.

Thanks a lot