ipv4

Internet Protocol version 4

IPv4 is a pervasive but dying protocol. In fact, it is planned to end of life on February 1, 2021. However, at least for now, it remains the standard for network address assignment.

Subnet Mask

used by the device to determine what subnet it belongs to. This determines what devices the device can send data to. The mask tells you how many bits in the IP address must agree (counting from the left) for the address to be accessible. For example, if a device has an IP address of 10.10.10.10 and it’s subnet mask is 255.255.255.0, then the first three octets must agree. So, any address in 10.10.10.0 and 10.10.10.255 are reachable, but 10.10.11.0 is not.

By convention, a subnet mask must have a binary form with all 1s on the left and all 0s on the right. This is because we always mask the most significant bits first. For example, 255.224.0.0 has the binary form 11111111.11100000.00000000.00000000 and is a valid submask.

255.36.0.0 has the binary form 11111111.00100100.00000000.00000000 is not valid because the 1s are not left aligned.

As a general rule, each octet of the subnet mask must take a value in 255, 254, 252, 248, 240, 224, 192, 128, and 0.

Examples of Accessible Addresses

* In reality, every address is not reachable because of the reserved private ranges (see below). ** This is a bit tricky to see. But if you work it out in binary form the result becomes clear.

Connection to CIDR

It is useful to look at the binary representation of the subnet mask: 255.255.240.0 = 11111111.11111111.11110000. In this representation it is easy to see that the first 20 bits of the address are masked. Therefore, in CIDR notation this is a /20 mask.

Default Gateway

The IP address that the router uses to talk to devices inside the intranet. If the device needs to communicate with a device outside of the subnet, it will need to send the data first to the default gateway, and the router will forward this data.

Classful Address

In the beginning of networking These classes are too rigid, and they aren’t enforced any more, but it is useful to know them. Most networks still default to these values.

Classes D and E

Classes D and E are not available for public use. Class D addresses are reserved for multicasting and are not assigned to hosts. Class E address are reserved for research and experimental purposes.

Classful Addresses in Binary Form

It is useful to think of this in binary form too:

Classless Inter-Domain Routing

TODO: VLSM

Reserved Private Ranges

A list of IP addresses reserved for private networks. These cannot be used for any public facing devices, i.e. any devices exposed directly to the internet.

Loopback Range

A list of addresses reserved for a device to talk to itself. Usually only one loopback interface is needed on a device, but more loopback interfaces can be created if necessary. Any address between 127.0.0.0 and 127.255.255.255 can be used as a loopback address.

ZeroConf Range

Quick Maths