On our lab network, we are going to have a router, switch, three clients and a server. The clients will be running Windows 7, the Server will be running Windows Server 2008 OS on a small LAN with a Class C private IPv4 address range.
The subnet mask will be 255.255.255.0, meaning the first three blocks of numbers (192.168.0) will be the network (subnet) and the last block will identify the device (client, server, router etc). This will give us a maximum of 254 devices, which is the maximum number allowed in each block. Remember you can’t allocate 0 or 255 to a device – these two are reserved.
As a simple example, we have installed a simple web server and its bound to port 80, as we can see from the settings on the screen. This means that a client using a web browser such as internet explorer can connect to the server 192.168.0.10 on port 80 to access the test website.
Software Installed is Windows 7 with Internet Explorer web browser.
On one of our clients on the network, we can open up internet explorer and type in the server’s IP address.
The web browser will by default connect via port 80 since this is the port http uses to connect.
The web server software receives the request, generates the webpage and sends the HTML code back to the browser. The browser then reads the HTML code and renders the page accordingly on the screen to the client.
Now typing and remembering IP addresses is a pain. It would be far nicer to be able to type in a name such as elluminetpress.com. This is known as a domain name. We can introduce another server onto the network called a DNS server (domain name resolution server). This server contains a list of IP addresses and the corresponding domain name. So in our DNS records we could have
So now all I have to do is enter elluminetpress.com into the browser and it will query the DNS server with the domain name, and the DNS server will return the IP address. The network will then use the IP address to route the request to the server.
Now say we had a few hundred devices and machines on our network. If we did this manually we would have to go around each machine and manually input and IP address. We can introduce another server onto our network called a DHCP (dynamic host configuration protocol).
When a new device is plugged into the network and is configured to use DHCP, it will broadcast a DHCP request to the whole subnet (255.255.255.255). The DHCP server will pick an IP address from its pool of addresses in a range, called a lease, and send the packet back to the MAC or hardware address where the request came from. The DHCP client running on the device will then allocate the IP address and any DNS or router settings to the client’s ethernet or wifi card network settings.
As you can see below, the DHCP server has allocated an IP address 192.168.0.2 to the device.has added the IP address of the gateway which is the router 192.168.0.1 the IP address of the machine running the DNS server in this case 192.168.0.10
The router connects different networks to each other through the internet in order for data to flow between networks. If we were to increase the size of our network to many hundreds or even thousands of machines, we would divide the network up into smaller networks called subnets. You will need a router to act as a gateway to each subnet. Here’s a simplified example of our school network.
The router looks at each packet coming in and out of the network and if the network part of the IP address matches the subnet of its network, the router passes the packet to its clients if it doesn’t match, the router passes it on.
A router usually has two IP addresses, an external address and an internal address. The external address is usually supplied by an ISP.