Subnetting Query

[ITexpert]

Hello Guys,

My company allot me 10.110.x.x subnet for internal Subnetting. 

I am planing to reserve it like this :

10.110.1.x  to 10.110.20.x  - Location 1

10.110.21.x  to 10.110.40.x  - Location 2

10.110.41.x to 10.110.60.x  - Location 3

Follow this trend if company grows ...

I am confused which subnet mask i will use :   255.255.0.0  or 255.255.255.0  , please advice with reason, I am confused between subnetting skills.

My manager ask me to DO this  and i am not able to figure out the reason for this.

Statement :

Our phone PBX is located Location 1  on subnet 10.110.10.xxx, is needs to move to less intrusive subnet such as 10.110.253.xxx or recommendation.

What should be the purpose of moving from this subnet to 10.110.253.c ?

Thanks

[ITexpert]

Guys, please Suggest ?

[Richard Burts]

You posted the question 35 minutes ago and already you are posting again asking us for responses? You should develop some patience.

It is difficult to know which subnet mask you should use. Your post is a bit vague but suggests that your company has given you 10.110.0.0/16. Is that correct? There are a few things to consider which may help you to determine how you will approach your subnetting plan. First consider how many locations will you need to support which will tell you how many subnet groups you need to support and that may shed some light on what is the maximum size of subnet that you will use.

You should consider whether locations are all about the same size or are some larger and some are smaller? This will lead you to consider whether all subnetting should be same size or some subnets larger and some subnets smaller.

You can consider what you think the largest location will be and how many devices are in that location. Then consider what size subnet will accommodate that many devices.

You will find that subnetting works best when it is done based on powers of 2 (2 or 4 or 8 or 16 or 32 etc). Setting up 1 to 20, then 21 to 40 looks nice and makes sense when we think in decimal numbers. But IP addressing is based on binary numbers and not decimal numbers. So locations with 1 to 15, and 16 to 31 would work better.

Doing effective subnetting is frequently trying to solve questions when you have several constraints that you must consider, such as how many groups will I need, how large is the largest group, do all groups need to be the same size, how much room do we need to provide for growth.

HTH

Rick

[ITexpert]

Hello Richard,

Thanks for always being an helpful Person.  everything make sense to me. But still some doubts.

Yeah its correct i got 10.110.x.x/16  by management.  I am confused that even when i will use the subnets for particular vlans for ex. Vlan 10 --> 10.110.10.0 So i will use mask  16 or 24. How to analyze this thing,   whats the industry standard ?

Other example :

What to do in case if we need small subnet like for example cameras which is less than 50 and will never increase.For example, vlan 50 - cameras 10.110.50.0 , this subnet can provide up to 253 addresses (10.110.50.2 to 10.110.50.254). If i limit the DHCP scope i will waste the addresses.

How to preserve these addresses ?

The other thing, My manager planning to move one subnet from 10.110.10.x to 10.110.253 .x and this move does not make any sense in my mind.

Thanks

[Richard Burts · Accepted solution]

Yes one of the things to consider when planning subnetting is the question of whether all subnets should be the same size or to use different sizes. It is easiest when you make all subnets the same size. But that is not efficient. As you mention your subnet for cameras has a small number of devices and if subnets are the same size then you will be wasting addresses in the camera subnet. So many people adopt the approach of varying sizes of subnets.

I might suggest that you use an approach something like this as you plan how to do your subnetting. You have been assigned 10.110.0.0/16 (255.255.0.0) and the question is how to break it up to support subnets at multiple locations. Start by determining how many locations you need to support which tells you how to break up your address allocation. Perhaps you are going to have 32 locations. So you need to break up your /16 into 32 different groups. Using a mask of 255.255.248.0 will provide 32 groups. So location 1 gets 10.110.0.0 through 10.110.7.0

location 2 gets 10.110.8..0 through 10.110.15.0

location 3 gets 10.110.16.0 through 10.110.23.0

and so forth for the 32 locations.

Then within each location you need to break it up into subnets to support the various functions within that location. Lets start by determining how big is the largest function within that location. Perhaps there is an office (or a building or whatever) that might have 200 hosts in its network. So to support 200 hosts you would need a subnet with mask of 255.255.255.0. So lets think about the first location and perhaps it needs 3 subnets of the larger size. So you would allocate

10.110.0.0 255.255.255.0 to the first office/building

10.110.1.0 255.255.255.0 to the second office/building

10.110.2.0 255.255.255.0 to the third office/building

Then we can allocate a subnet for cameras. If there are 50 cameras you might want to allocate a subnet with 64 addresses (255.255.255.192). So you could allocate 10.110.3.0 255.55.255.192 for cameras.

Then you might use another part of 10.110.3.0 for another function, perhaps a subnet for loopback interfaces to use for device management. So 10.110.3.128 255.255.255.128 for loopback assignments. You would continue on in this way determining which function, with need for how many addresses, and generate the subnet that fills that need.

I am not sure why your manager wants to move one subnet. But if he does it should not be that big of a deal. Recognize that if you do move that subnet then it impacts planning for whichever location would use that last group of subnets.

HTH

Rick

[ITexpert]

Hello Richard, Thanks for Solid explanation !

I am working IN networking from 2-3 years now but never face loopback interface.

Can you explain, when in production environment we need loopback interfaces and whats their cons and pros ?

@Richard Burts

Thanks

[Richard Burts · Accepted solution]

You are welcome. I am glad that my explanations are helpful. These forums are excellent places to ask questions and to learn about networking. I hope to see you continue to be active in the forums.

A loopback interface is a virtual interface that can be configured. Since it is a virtual interface it is not tied to any physical interface and is not dependent on any physical interface. Loopback interfaces are frequently used for things where we want the interface to always be available (things like system management or perhaps for things like neighbor addresses in BGP). Let us think of an example where you have a router that has two FastEthernet interfaces. You want to be able to telnet or SSH to the router to be able to manage the router. Let us assume that you have set up system management to user FastEther0/0. As long as the physical interface is up/up you have access to the device. But what happens if something happens and FastEther0/0 is down. Now you have lost management access even though it might be possible to access the device using FastEther0/1. So we configure interface loopback0 on the device and configure an IP address and mask for the interface. Now our system management is accessing the device using the address of loopback 0. If FastEther0/0 is up we will use it to get to the device. But if that interface is down then we can access the loopback using FastEther0/1.

HTH

Rick

[ITexpert]

Thanks... Great !