VLSM Issue

sentimetal

Hello! I am having an issue with VLSM. I'll start with a scenario:

I'm given an address block of 192.168.1.0/24. On the Fa0/0 side of Router 1 I have a LAN that needs at least enough host address for 20 hosts. On the other side of the router, Se0/0/0 needs just 2 addresses for the WAN link. I cannot use the zero subnet.

On the Fa0/0 side of R2, I'm to use the third subnet. Again, I need at least 20 host addresses. So I use a /27 mask for optimal space.

So I figure:

192.168.1.32/27 is my first subnet for R1's LAN
192.168.1.64/30 is my second subnet, used for the WAN links.
192.168.1.68/27 for R2's LAN.

When I assign the ip addresses to R2's serial interface that connects R1 and R2, I get a message saying the addresses overlap, but how am I supposed to use the 3rd subnet for a WAN link while getting the most efficient address space? If I were to start at 192.168.1.0/30 wouldn't I get:
.0, .4, .8, .12, .16, .20, .24, .28, .32, .36, .40, .44, .48, 52, .56, .60, 64, .68 so wouldn't it be a vaild subnet either way?

SubnetZero

Personally I take a single /24 and use it for all of my point-to-points, no need to mix and match from the same /24. Is there some reason you're doing it this way?

Either way if you want to make this work you can use the /30 after the two /27's to make it contiguous

192.168.1.32/27 for R1's LAN
192.168.1.64/27 for R2's LAN
192.168.1.96/30 for the WAN links

So on R1

interface FastEthernet0/0
ip address 192.168.1.33 255.255.255.224
no shut
!
interface Serial0/0
ip address 192.168.1.97 255.255.255.252
no shut

on R2

interface FastEthernet0/0
ip address 192.168.1.65 255.255.255.224
no shut
!
interface Serial0/0
ip address 192.168.1.98 255.255.255.252
no shut


Alternatively you could have blocked out your /30's all the way to .95

sentimetal

SubnetZero wrote: »

Personally I take a single /24 and use it for all of my point-to-points, no need to mix and match from the same /24. Is there some reason you're doing it this way?

Either way if you want to make this work you can use the /30 after the two /27's to make it contiguous

192.168.1.32/27 for R1's LAN
192.168.1.64/27 for R2's LAN
192.168.1.96/30 for the WAN links

So on R1

interface FastEthernet0/0
ip address 192.168.1.33 255.255.255.224
no shut
!
interface Serial0/0
ip address 192.168.1.97 255.255.255.252
no shut

on R2

interface FastEthernet0/0
ip address 192.168.1.65 255.255.255.224
no shut
!
interface Serial0/0
ip address 192.168.1.98 255.255.255.252
no shut


Alternatively you could have blocked out your /30's all the way to .95

The lab instructions said not to use the zero subnet and that subnet 1 was for R1 LAN, subnet 2 was for the WAN link, and subnet 3 was for the the R2 LAN. Would I actually have to keep a /30 mask all the way up to .96 to change to a /27 mask? I can see how .68/27 doesn't make sense, but yeah.

SubnetZero

Lab it up and see what happens! I can tell you that I just did it and it works

lanrexng2

Thanks for that breakdown Sir Zero

SubnetZero

sentimetal wrote: »

The lab instructions said not to use the zero subnet and that subnet 1 was for R1 LAN, subnet 2 was for the WAN link, and subnet 3 was for the the R2 LAN. Would I actually have to keep a /30 mask all the way up to .96 to change to a /27 mask? I can see how .68/27 doesn't make sense, but yeah.

It's all binary. I'll point you to a site that explains this well when I get home tonight.

That being said there is kind of method when subneting with VLSM. Basically (like my first method) you should order your subnets by capacity in decreasing numbers.

for our example I put the two /27's first and the /30 as the last one. That’s the order in which you should assign the addresses ranges.

Doing a /27, a /30, and then a /27 is possible but kind of odd so you should always be going from bigger (at the beginging of the range) to smallest towards the end of the range whenever possible

HTH

sentimetal

SubnetZero wrote: »

It's all binary. I'll point you to a site that explains this well when I get home tonight.

That being said there is kind of method when subneting with VLSM. Basically (like my first method) you should order your subnets by capacity in decreasing numbers.

for our example I put the two /27's first and the /30 as the last one. That’s the order in which you should assign the addresses ranges.

Doing a /27, a /30, and then a /27 is possible but kind of odd so you should always be going from bigger (at the beginging of the range) to smallest towards the end of the range whenever possible

HTH

I hate resorting to binary for my subnetting. : ( I suppose it's good for the overall understanding. It DID work, I think for some reason I gave .64 a /27 subnetmask and didn't catch it while checking last time.

alxx

Why ?
Don't be scared of binary.

Learn it and use it.

SubnetZero

sentimetal wrote: »

I hate resorting to binary for my subnetting. : ( I suppose it's good for the overall understanding. It DID work, I think for some reason I gave .64 a /27 subnetmask and didn't catch it while checking last time.

But that's all this stuff is man, 1's and 0's!

So you said you got this to work with your OG configuration? How? That shouldn't work and doesn't work on any of my routers

R2(config)#interface FastEthernet0/0
R2(config-if)# ip address 192.168.1.69 255.255.255.224


R2(config-if)#interface Serial0/0
R2(config-if)# ip address 192.168.1.65 255.255.255.252
% 192.168.1.64 overlaps with FastEthernet0/0

However if I configure it with .97 there's no overlap

R2(config)#interface FastEthernet0/0
R2(config-if)# ip address 192.168.1.69 255.255.255.224


R2(config-if)#int s0/0
R2(config-if)#ip address 192.168.1.97 255.255.255.252

In fact every address up to 97 will fail because the router expects us to be using subnets in increments of 32 (0,32,64,96, etc) however by throwing a /30 in between the two /27's you're violating this rule.

If this is the PacketTracer lab I'm thinking of I don't think you need the /30 between your WAN interfaces (even though that's best practice)

Also, since they want the second subnet to be used for your WAN interfaces you should probably just do three /27's. In this way you are going in increments of 32 and you wont get the overlap error.