No IP Classless

Mikdilly · August 2007

Could someone explain the following statement:

The 'no ip classless' command prevents a router from using the default route when a packet's destination classful network has component subnets in the routing table.

I think i understand the default route in the beginning of the sentence as being the gateway of last resort that would be configured using ip default-network but i don't understand what a component subnet would be.

dtlokee · August 2007

The assumption with classful routing is that all of a classful address belongs to one organization. The idea of a component subnet would be any subnet of a classful network. When you look at the routing table it lists the subnets based on the classful networks like this:

Network 192.168.1.0/27 is subnetted 2 subnets
192.168.1.32 ....
192.168.1.64 ....

The 192.168.1.32 and 192.168.1.64 subnets would be the "component subnets" of the 192.168.1.0 network.

So the point is that if you tried to route a packet to 192.168.1.151 it would be dropped even if there was a default route in the routing table. The reason fo this is that the router knows of subnets that belong to the major network (192.168.1.0 in this case), but does not have an entry for the specific subnet.

Mikdilly · August 2007

dtlokee wrote:

The assumption with classful routing is that all of a classful address belongs to one organization. The idea of a component subnet would be any subnet of a classful network. When you look at the routing table it lists the subnets based on the classful networks like this:

Network 192.168.1.0/27 is subnetted 2 subnets
192.168.1.32 ....
192.168.1.64 ....

The 192.168.1.32 and 192.168.1.64 subnets would be the "component subnets" of the 192.168.1.0 network.

So the point is that if you tried to route a packet to 192.168.1.151 it would be dropped even if there was a default route in the routing table. The reason fo this is that the router knows of subnets that belong to the major network (192.168.1.0 in this case), but does not have an entry for the specific subnet.

Thanks for the explanation, so it's the no ip classless command that tells the router to drop the packet if the packet is destined for a subnet that is not in the routing table? And the opposite would be true in that it would send the packet to the default route if you were using 'ip classless'?

Netstudent · August 2007

If you wanted those questions to be answered then, EXACTLY!

Mikdilly · August 2007

Netstudent wrote:

If you wanted those questions to be answered then, EXACTLY!

Thanks, I got the original statement from ICND practice exam cd, i know you said you used this for your preparation, was wondering if you remembered this ques., i believe one if the answers they gave as being correct was wrong. It had to do with 3 routers, two of which they give the routing tables for in an exhibit, the third router has an unadvertised route to some subnet that i think is not in the routing tables of either two. The other 2 routers have default routes configured that show in the exhibit. One of the answers they claim as being correct says that with no ip classless configured one of the routers with a default route configured will forward the packet to network configured as the default network. Going by what dtlokee explained the packet should be dropped, no? I may be missing something in the question, but their answer didn't seem correct. I can get the exact wording of the ques, if needed.

Netstudent · August 2007

hmmm..

The only question i remember on that CD that involved the IP classless rules was a question where it asked something like. What would happen to a packet that was destined for x.x.x.x.

Then the asnwer choices are a series of kinda complicated looking routing table entries. I can't remember if the routing table entries were just a default route or a whole routing table.Then you have to know the rules.

Each answer choice said something like If the if the ip classless was issued then it would do this and if the no ip classless command was issued it would do this.

I remember it being a pretty involved and complicated question. It took me some time to think through it but I got it right on the first try. I'm pretty sure I am thinking of the one you are talking about. I don;t remember it being wrong though. Did you read through the explanation carefully?

The router looks at the destination address in the packet and then compares it to the routing table to see where to shoot it out. IF the ip classless is issued it will send the packet out the default route if it does not have that destination subnet in the routing table no mater what. When the no ip classless is issued, it will look at the classfull portion of the destination address. If the classfull portion is in the routing table it will then look at the next octet or what I like to call the subnet identifying octet. If the routing table does not have an entry for that subnet in the subnet identifying octet, the packet is dropped. IF the routing table does not even have an entry for the classfull portion of the packet's destination then it will go ahead and use the default route.

Mikdilly · August 2007

Yeah, it's a killer, would really suck to get it on the exam. But in every answer choice that involved having either ip classless or no ip classless each one said it would forward the packet to the default network, none of them said that the packet would be dropped. Wouldn't it be accurate to say that if something starts out saying no ip classless is used then it follows that the packet is always going to be dropped? There's no scenario that exists where configuring no ip classless results in forwarding the packet to the default network, is there? Maybe i've missed something in the ques. or in the use of no ip classless, i'm going to see if i can find it tonight and go back over it.

Netstudent · August 2007

There's no scenario that exists where configuring no ip classless results in forwarding the packet to the default network, is there?

Yes sir there is. If and only if the classfull portion IS NOT in the routing table.

So like Lokee said,
you have 2 routes to 192.168.1.32 and 192.168.1.64

IF a packet comes in destined for 192.168.1.97 then this poor packet is going to the bit bucket. IF this packet is destined for 200.200.200.1, then this packet will be routed out the default route. ONLY because the classfull portion 200.200.200 is not in the routing table.

Mikdilly · August 2007

Netstudent wrote:

There's no scenario that exists where configuring no ip classless results in forwarding the packet to the default network, is there?

Yes sir there is. If and only if the classfull portion IS NOT in the routing table.

So like Lokee said,
you have 2 routes to 192.168.1.32 and 192.168.1.64

IF a packet comes in destined for 192.168.1.97 then this poor packet is going to the bit bucket. IF this packet is destined for 200.200.200.1, then this packet will be routed out the default route. ONLY because the classfull portion 200.200.200 is not in the routing table.

That must explain it then, i'll consult your answers as i'm looking at the ques. tonight and hope that it clicks in, thanks again.

No IP Classless

Comments