AfriNIC gets 105/8
2010-11-12
Ladies and Gentlemen, history in the making…
Today AfriNIC got the 105/8 block allocated from IANA. This is surprisingly early, I was expecting AfriNIC to request more space in April of next year. This is moving the IANA depletion date much earlier (February of next year instead of April).
ARIN will soon get two blocks from IANA too. I asked John Curran about it at the recent Gogo6live Ipv6 conference and he vaguely admitted that ARIN will request space soon. It wouldn’t surprise me if that would be this week.
This leaves us with 4 x /8 left. Most likely, two for APNIC and two for RIPE. APNIC has around 4 x /8 in its pool which should be sufficient for them until early February of next week. By that time they will only have around 2 x /8 left and they will refill with another 2 x /8. RIPE currently have 2.88 x /8 in their pool. RIPE is currently allocating in a quite slow pace and the 0.88 x /8 will most likely last until late February of next year. They will then have 2 x /8 in the pool and will be able to justify another allocation of 2 x /8. After that there will be no more IPv4 addresses in the central pool.
It is really time to start implementing IPv6…
Loading ...
Remarkable indeed. Could this be the start of a run for the last blocks?
I’m not sure, but there are not that much left to fight over.
I think I’m going to write a post about the fact that Sprint allocated around 4 million v4 addresses yesterday.
/S
[...] [...]
Huston has a different iana exhaustion date, because he only considers iana exhausted when an ADDITIONAL allocation is requested after 2+2+2 are allocated to ARIN, APNIC, and RIPE. This makes about two months difference.
Spectacular to see that the IANA exhaustion time remaining on Huston’s site literally was literally cut in half.
Seeing that the depletion date is coming sooner than exected, i wonder if the number of IPv6-only sites will start to raise before that moment?
As soon as that number will become significant, v4-people will realise that they can only reach a part on internet….
http://bgp.potaroo.net/v6/as2.0/index.html
> Huston has a different iana exhaustion date, because he only considers iana exhausted when an ADDITIONAL allocation is requested after 2+2+2 are allocated to ARIN, APNIC, and RIPE. This makes about two months difference.
Sorry, that’s wrong now. Was I mistaken?
Do the estimates yet take into account 45/8 that was recently announced would be returned to the pool by Interop?
Huston is using a different mathematics (See http://www.ipv4depletion.com/?p=253 for details about the difference) . We have historically been about 6 months apart. The good news is that the difference is very small between the different predictions. There is no doubt about the fact that we are running out in the beginning of next year.
Yes and No. The return from Interop has not yet shown up in the ARIN statistics. So the block is still owned by Interop in my calculations.
The press release from ARIN says:
“ARIN will accept the returned space and not reissue it for a short period, per existing operational procedure. After the hold period, ARIN will follow global policy at that time and return it to the global free pool or distribute the space to those organizations in the ARIN region with documented need, as appropriate”
The return doesn’t really matter for the IANA depletion, because ARIN will allocate an additional 2 blocks anyway before February of next year (probably this week)
Huston has done something strange with his figure 30c graph(ARIN projected address allocations). It looks like he has moved the whole historical part of the graph up by one /8, and eliminated his expectation that ARIN will request a couple /8s in the near future. Maybe he thought the best way to deal with the Interop block return in his calculations was just to pretend it was part of ARINs unallocated pool the whole time, even though it was actually allocated to Interop. Or maybe it’s just a bug in his scripts.
Mindbuilder
This is probably because ARIN have removed the Interop’s allocation of IPv4 addresses and the block is now clear. This is a tad strange, as I thought that Interop were keeping a very small (65k addresses) allocation in the /45 block. Perhaps not everything has filtered through the system yet.
Karen
Hi “Mindbuilder”
Yes it looks like he just moved the ARIN graph up one step. But it really doesn’t matter for the IANA pool because ARIN will still allocate 2 more /8 before the party is over regardless of if they get most of the Interop block back or not. With most of the Interop block back to ARIN they would be able to stay afloat until Jan or Feb next year and allocate at the same time as APNIC is expected to allocate. Without it they would have to allocate now. Again, John Curran from ARIN kind of admitted to me that they are sending in a request (did I misunderstand him?)
I’m not sure how to interpret Huston’s graphs. The green line in figure 31 describing the total pool appears to be linearly going down until the depletion date. This would indicate that there are no additional allocations being made prior the one event where IANA would hand out what is left. I think what he is saying is that with the Interop block going to ARIN the three big RIRs (ARIN, RIPE and APNIC) all will allocate an additional 2 x /8 at more or less the same time in March next year.
About arin:
I think this is just how Huston’s script works.
It just looks at the allocation dates. Now the script sees the 45/8 block as having been assigned to arin all along, and not allocated. So the script figures, hey, arin has has this unallocated block all along. That it had been allocated before can’t be found anymore in the current arin allocation tables. The historic graphs are based on the current, not previous arin allocation tables.
And yes, ARIN, RIPE and APNIC will all allocate in Feb/March.