Why does my computer have so many IPv6 addresses?

How many IPv6 addresses are you supposed to have?

It is a valid question to ask any network expert who considers the subject in terms of numbers.

The standard answer, the one that most people have adopted, is to use /29's, meaning that you have a maximum of 29 IPv6 addresses assigned to you by the provider.

There are several reasons why this might be right for some organizations and yet still be wrong for others. For one thing, there are different types of networks which have radically different requirements for the number of IPv6 addresses that they use. As an example, mobile operators need very large numbers of IPv6 addresses since their networks are constantly changing as customers move around them. This has lead to the term 'floating-point' address, meaning a dynamic number of IPv6 addresses that will be allocated to them. A second type of network which uses dynamic addressing is the Internet of Things (IoT). In this case, the number of devices using the network is much greater than the total amount of data capacity on the network itself, so there is no need for a fixed number of addresses.

A second reason is that the standards used to allocate addresses vary from organization to organization. The RFC 6724 document that describes the most recent version of the Internet Protocol (IPv6) has stated that organisations should use /32's and /48's, but this hasn't been universally adopted as the standard, with some providers using other addressing types.

Yet another reason to use a smaller number of IPv6 addresses is that it means that the network will have less IP congestion problems. When there are more than about 30,000 hosts on the network, that will become noticeable, leading to the first of the dreaded blackholing issues.

As a very broad guide, let's take a look at what a company will need based on an average data/host count. The number of hosts on the network: Average data rate per host: 1TB of data per month = 8,000 GB per month. That adds up to about 16,000 GB per year = 176 TB per year. If we have a total IPv6 address pool of 1,000, we have a maximum of 10 IPv6 addresses per host on the network.

Is it normal to have two IPv6 addresses?

We have two IPv6 addresses for our company.

I'm a bit confused as to why we have two of them, since I thought the 6th bit meant it was an IP version 4 address, not a 6. I'm sure the answer is really simple but I can't seem to find it.

It is normal. There are two ways of numbering IPv6 addresses, and the order of numbers indicates which one you are using: The most common way is to use "6" as the first digit, meaning that the prefix length is 64 bits. So if your subnet is "10.0/24", then you would have addresses in the format "10.1-10.254" with a prefix length of "64".

The second way is to use "0" as the first digit, meaning that the prefix length is 32 bits.0/32", then you would have addresses in the format "10.0-10.

I think you are confusing how IP version 6 works with IP version 4. IPv6 has a 128 bit address size. There is a way to think of these 128 bits in the form of two 32 bit numbers. These 32 bit numbers are the A and B subnets. The most significant bit is the first (the A subnet) and the least significant bit is the last (the B subnet).

So for example, the address 10.0 would have the first 32 bits as 0.0 and the last 32 bits as 0.255.

For IPv4, you have the "Network" number and the "Host" number. The network portion is a 32 bit number and the host portion is a 32 bit number.

For example, the address 1.4 would have the network 1.0 and the host 1.

For IPv6, the first 32 bits of the address are the host portion and the last 32 bits of the address are the network portion. For example, the address 10.