I often get asked questions along the lines of “How bad is IPv4 address exhaustion, honestly?”. I take the responsibility of answering questions like these seriously; I think it’s important that a broad cross section of people understand the issue.
First, I explain there are three options to obtain IPv4 addresses, via:
- Your local friendly Regional Internet Registry (RIR), who can possibly allocate a block of addresses to you, if you qualify under that RIR’s rules (policies).
- Your friendly upstream ISP, who can possibly assign a temporary address of block of addresses, if they have them available.
- The open market, which is available in some parts of the world.
I then provide the following explanation as to what IPv4 shortage means for each of these options.
There are five RIRs whose role is to allocate IP addresses to those who need them.
All of the RIRs have entered a stage of ‘IPv4 exhaustion’, which means that their supply of IPv4 addresses is now limited.
That said, each one has a remaining ‘rationed’ address pool, which can be allocated in small quantities and under special conditions in each case. The availability and conditions are specific to each RIR, but in most cases, a new member of an RIR can receive a small block, and only one block, of public IPv4 address space.
Read how APNIC currently allocates its remaining IPv4 addresses to current and new Members.
The main impact of ‘IPv4 exhaustion‘ is that a network operator can no longer receive large IPv4 blocks from a RIR that are sufficient to address any significant Internet infrastructure. Therefore, they need to rely on one of the other mechanisms for obtaining public address space.
As a customer of an ISP, you may receive one or more addresses that are assigned to you automatically on connection, or semi-permanently, while you remain a customer of that ISP. Because ISPs themselves have a shortage of available addresses, it is generally harder to obtain an assignment of public IPv4 in this manner. It depends on the ISP and their own policies and pricing (and obviously on the amount of addresses that they have available within their own allocated pools).
Since the advent of IPv4 exhaustion in each of the RIR regions, most have adopted rules (policies) allowing IPv4 address blocks to be transferred between consenting parties, under specific conditions. In some cases, and under specific conditions, transfers may be obtained from another RIR region (for instance a transfer may occur between the ARIN region and the APNIC region).
Read how APNIC Members can transfer IPv4 addresses
The primary condition for a transfer is that the parties (source and recipient) both consent to the transfer, however, the RIR(s) involved take no part in the transaction, which may be on commercial terms. There are ‘brokers’ operating in the IPv4 market who can locate sellers and/or buyers, and provide advice on pricing and availability of IPv4 transfers.
It’s time for IPv6
The ultimate solution to IPv4 exhaustion is, of course, the complete transition of the Internet to IPv6, however, this will take time and until then there will be (by definition) networks and sites that only support IPv4. This requires other networks and sites, even if they support IPv6, to maintain IPv4 connectivity, which in turn requires some number of IPv4 addresses. For that reason, IPv4 exhaustion certainly is an issue to be understood and dealt with, especially by those who are building new networks and services.
The views expressed by the authors of this blog are their own and do not necessarily reflect the views of APNIC. Please note a Code of Conduct applies to this blog.
It seems that everyone is focused on the IPv4 address pool exhaustion, but not investigating how are they being uses. In particular, how many are sitting idle. Here is a draft proposal to IETF called EzIP (phonetic for Easy IPv4):
Basically, the EzIP approach will not only resolve IPv4 address shortage issues, but also largely mitigate the root cause to cyber security vulnerabilities, plus open up new possibilities for the Internet, all within the confines of the IPv4 domain. These should relieve the urgency to deploy the IPv6 for an appreciable length of time.
Thoughts and comments will be very much appreciated.
Abe (2018-07-12 09:59)
Being an IPv4 broker is see a lot of resources unused and unfortunately the common answer is.; ”We have no human resources to change our (legacy !) /16 for a /24” (that would indeed suffice).
There is a different and simpler approach to this subject. Please have a look at the following paper.
To save your time, you may want to start the reading from Appendix C.
Abe (2019-06-14 08:40)
I think that the issue is less IPv4 address exhaustion and more the bad habits that are permeating the internet in efforts to extend IPv4s useful life (NAT and poor assumptions about how to use the larger address space).
Rather than philosophical opinions, we have come up with a somewhat unorthodox (depending on your perspective) scheme to deal with these issues from the ground up. It is based on traditional telecom disciplines and conventions. Please have a look at the following paper.
There is nothing new here, and this solution does not require any development effort. All it took was logical analysis and derivation based on the original IPv4 protocol and a long-reserved yet not used address pool.
The main body of the document and the first two appendices are rather tedious, by necessity. You may want to start from Appendix C, where we outline the current status of demonstrating the realizablity of this proposal.
Abe (2019-06-14 09:04)
You may like to have a look at the feasibility demonstration report below about our proposal:
It should provide some material for furthering the dialog.
Abe (2020-08-29 15:40 EDT)