IPv4 should be all gone, except that it isn’t…

Those of you that have been watching the various “IPv4 exhaustion sites” (including Geoff Huston’s Potaroo site and Stephan Lagerholm’s IPv4Depletion site, as well as the official graph page at APNIC) will know that IANA should have run out of IPv4 addresses last weekend, under normal circumstances.  However this hasn’t happened, and it appears that no-one has “pushed the button”, that is, the day when a RIR asks for the last 2 /8s available for allocation, and thereby triggering a distribution of the remaining 5 /8s one each to each of the Regional Internet Registries (RIRs).

According to comments made from people who should know, the reason this hasn’t happened is that someone, no idea whether IANA, APNIC, or IANA plus all the RIRs, wants to turn this into a massive PR stunt.  So it appears there is an exhaustion day, but it’s a secret.  And that irritates me.  It’s not as if IANA didn’t know this was coming, they could have had the press releases written months ago.

APNIC’s pool is getting lower by the day (at the time of writing this was 1.67 /8s, or 28,017,950 addresses, and the usual threshold for asking for more addresses is about 2.0 /8s.  There are a number of dates it could be, for example, there’s January 31st, Feb 14th, or even the ICANN meeting in March when it is rumoured that Bill Clinton will be coming to speak (but personally I don’t think they can last that long).  It could be today (except that today is a weekend).

I really just wish everyone would stop stalling for time and push the button to start Stage 2.  Whatever day or time it happens, the press are going to be all over it, there’s going to be mass panic in some quarters of the industry, and it wouldn’t surprise me if one or more ISPs or service providers disappear in the next year or two because they just weren’t ready in time and got caught out.  So let’s just push the button now, please IANA? (Well, okay, when you get up!)

28,017,950.72

One thought on “IPv4 should be all gone, except that it isn’t…

  1. There’s a good reason to put it off to a significant date (e.g. 31st January); you want to get as many of the talking heads who’ll say “this is a major milestone towards the end of the Internet as we know it – if you’re not IPv6 Ready, you’re about to run out of Internet” free and ready to talk to the press.

    If you don’t, you run the risk that the only “experts” available to the press are the ones who say “it’s OK. No need to panic. CGNAT will save us all”. And that’s not what we want; we want the people who understand all this stuff to be available to explain why we need IPv6, not just forty-two layers of NAT.

Comments are closed.