Why must you go from largest subnet to smallest for VLSM?

Pupil

I tried settings up an example VLSM'd network in Packet Tracer. I began with the smallest subnets (/30 for WAN links) and ended with the largest (/27 blocks for example users), but I kept getting hit with error overlapping networks. However, when I went from largest to smallest, it worked without a hitch. Do you always have to go from largest to smallest?

RouteMyPacket

Pupil wrote: »

I tried settings up an example VLSM'd network in Packet Tracer. I began with the smallest subnets (/30 for WAN links) and ended with the largest (/27 blocks for example users), but I kept getting hit with error overlapping networks. However, when I went from largest to smallest, it worked without a hitch. Do you always have to go from largest to smallest?

Based on your findings, what do you think?

Jon_Cisco

don't just look at the subnet.
Writing out the address range when learning this will help.
What seems tricky in your head sometimes looks obvious on paper.

Let us know what you come up with.

davenull

You don't have to go from the largest to the smallest, but it's easier and a good practice to do so. Whenever I think of VLSM subnetting I think of cutting a pizza or a pie - you want to get the largest slices out of the way first for the most hungry person.

Pupil

Jon_Cisco wrote: »

don't just look at the subnet.
Writing out the address range when learning this will help.
What seems tricky in your head sometimes looks obvious on paper.

Let us know what you come up with.

I did write it out on paper twice and verified that there were no overlapping subnets. But Cisco's IOS still wouldn't accept me starting from smallest sized subnets to the largest.

RouteMyPacket

So how did you write it out and "verifyZ" there was no overlap? Any reason why you can't post your process? So why doesn't the IOS take it? It told you why, it overlaps.

Let's try to cut up a basic 192.168.1.0/24

192.168.1.0/30 - Ok, we will have a range of 192.168.1.0 - 192.168.1.3

Now we want to create a /27 for say HR

192.168.1.4/27

Hmmm, why can't you add this? What network does 192.168.1.4/27 belong too? A /27 gives us how many addresses? 32 right? Hmm, so that network would be 192.168.1.0 - 192.168.1.31 hence why it will not accept your form of subnetting. We subnet largest to smallest to avoid overlapping networks.

That's it, nothing special about it...it is what it is.

Typically you use a completely different subnet for L3 links so there is no worry about overlapping, you simply use it when you need it.

L3 - 172.31.1.0/24

Site A to Site B - 172.31.1.0/30
Site C to Site D - 172.31.1.4/30
Site E to Site A - 172.31.1.8/30

Pupil

That makes a lot of sense. I understand now; you can't simply use any range you want. They are fixed. Hence, /25 is always .0-.127 and .128 -.255. So, we must go from largest subnet to smallest otherwise there will be overlap. Thanks for clarifying that for me.

Scorp6

Pupil wrote: »

That makes a lot of sense. I understand now; you can't simply use any range you want. They are fixed. Hence, /25 is always .0-.127 and .128 -.255. So, we must go from largest subnet to smallest otherwise there will be overlap. Thanks for clarifying that for me.

Precisely. I did the EXACT same thing during my CCNA course. Instructor asked me to write the ranges on the board. As soon as as he asked me what the Network addresses were for both subnets I just dropped the marker and erased it. "Big slices of the ip pie first, I got it", I said.

Jon_Cisco

Good work Pupil.

Have you used the subnetting.net site yet?
Once you learn the basics it's a great site to practice on.

Jon

her.yang

Now, you don't necessarily HAVE to but it's good practice. It makes route summarization much, much easier down the road. Going back to your example, it would actually work but what you'd have to do is start your /27 subnet where it would logically begin had you started subnetting using /27s in the first place.

For example, let's subnet out 192.168.1.0/24 into slices of /27.
192.168.1.0/27
192.168.1.32/27
192.168.1.64/27
192.168.1.96/27
...

If you wanted to start subnetting using a /30 mask, you could use up the entire 192.168.1.0/27 subnet for /30 networks (it would give you 8 networks) and use the next network, 192.168.1.32/27, for users. You could then use the next /27 subnet again for /30 networks if you wanted to; no problem. In fact, here's an example from a multilayer switch:

SW1#sh ip route
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS level-2
ia - IS-IS inter area, * - candidate default, U - per-user static route
o - ODR, P - periodic downloaded static route


Gateway of last resort is not set


192.168.0.0/24 is variably subnetted, 5 subnets, 2 masks
C 192.168.0.32/27 is directly connected, Vlan2
C 192.168.0.8/30 is directly connected, Vlan4
C 192.168.0.12/30 is directly connected, Vlan5
C 192.168.0.0/30 is directly connected, Vlan1
C 192.168.0.4/30 is directly connected, Vlan3

I guess the short answer is: your subnets have to fall on the bit boundaries. Hope this helps.

Pupil

Yeah that definitely helps. So in the end it's doable but much simpler to just go from largest to smallest. Thanks everyone.