Which partially explains how they managed to have a 10 Mbit port at an internet exchange. Although that seems weird even for 2010? I assume this would be a 100 Mbps or Gbit port artificially throttled to 10 Mbps?
That said, 50 Mbps of traffic sounds like a non-issue for someone like Cloudflare, so I'm not surprised they took the prefix.
Edit: A post by Cloudflare (linked in another subthread here) mentioned 10 Gbps of traffic at the time Cloudflare started using it. Which, as it would be spread over multiple locations, is probably a relatively minor annoyance.
TL;DR: 10 Gbps of bad traffic at the time they took it over, but the main issue seems to have been that many networks were unable to reach the IP due to misconfigured equipment on the route.
If the misdirected traffic was 50 Mbps, I would expect Cloudflare to be able to simply ignore it. (Edit: it was 10 Gbps in 2018 when Cloudflare took over, which I expect to be an annoyance but not a problem if spread over multiple internet exchanges).
Given the number of people now relying on Clouflare 1.1.1.1 to "get Internet" (ie using 1.1.1.1 as recursive name server), I can't imagine APNIC deciding to stop Clouflare using this range.
It seems "too late" to revert this decision. Otherwise people will experience "Internet stopped working", blaming their ISP.
APNIC may decide to keep a working DNS server on 1.1.1.1, but ethically, routing traffic to someone else than Cloudflare is not great.
If just temporary assigned to Cloudflare, APNIC shouldn't care if it sees a better use for the range. Supporting unintended uses only encourages various types of abuse. And changing DNS settings is easy enough.
That said, if a lot of people rely on 1.1.1.1 as DNS, it's worth considering whether reassignment qualifies as 'better' use of this resource. Not to mention the hassle caused by making changes to popular [anything].
Fixed IP addresses change and are deprecated all the time. It's of zero concern to APNIC that customers of Cloudflare or various ISPs can no longer access the internet because they relied on a temporary IP assignment, after the service was gracefully terminated and deprecated, with ample lead time.
That being said, the use by Cloudflare is an excellent way to reclaim this part of the IP space, I don't see why they would terminate this collaboration.
> Only recently, in 2008, 1/8 was moved from "the IANA reserved" to the "IANA unallocated" pool of addresses. In January 2010 it was finally allocated to APNIC in order to be distributed to Local Internet Registries in the Asia-Pacific region.
How do creators/maintainers of these non-renewable resources such as IPv4 addresses, computer ports, wireless spectrum, etc. keep making the same mistake of allocating such a high percentage of the space to specific players, leaving the scraps for the most numerous users? Is it just greed, or not good forward thinking?
We engineers have accepted the available private address space <https://en.wikipedia.org/wiki/Private_network> but just think about how intuitive it is to an outsider? We make millions of people memorize 192.168.X.X to setup their home routers instead of the common sense letting the millions use the 1/8 privately? Just my opinion, but perhaps holding us back from IPv6 deployment is UI people don't see how an ordinary user would be able to remember a full "fd00::" address for day-to-day home networking or typing into their iPhone.
I don't have as much knowledge in computer ports background story, but last I checked <https://en.wikipedia.org/wiki/List_of_TCP_and_UDP_port_numbe...> it's full of defunct, barely-used single players hogging up the coveted lower numbers, again leaving the millions to deal with possibly stepping on others' toes by using them "unofficially" and causing "pollution".
Last example is thinking about how many billions of wifi and cellular devices are forced into the tiniest slivers of the spectrum <https://upload.wikimedia.org/wikipedia/commons/c/c7/United_S...>. The engineers working so hard to cram even more signals into the 2.4GHz and 5GHz slivers need someone who is going to look up and realize it's time to expand the walls of the box they are trapped in. The only firm allocations should be for physics constraints, such as weather radar, but also means Wifi 6 should be a non-start due to not penetrating concrete, one of the most common materials of home construction.
Given these examples, the argument I'm trying to make is I find it insulting to consider the fact that people used the 1/8 block "pollution". It's the natural human guess at usable example numbers and the like, and much in the same way that repeated "human-error" in accidents is being re-argued as "engineering-error", I find the misallocation of the 1/8 block to be directly the fault of the RFC creators for not giving it to the people.
Long term, IPv4 allocation won't matter and it's too late to fix that problem
Port numbers are kinda obsolete on the internet since everyone uses port 443 to bypass middleboxes that block all the other ports in the name of "security". Hardcoding a number for each protocol instead of including IP+port in DNS responses was a stupid mistake
All consumer devices being forced to share ISM bands is BS, but let's be real even if more spectrum was opened up back in the 2000s, a lot of devices would still use the same range of frequencies because they wanna use the same cheap ICs and antennas
Good wifi performance requires an AP in every room. The lack of 6ghz penetration is great for high-end setups because it eliminates the hidden node problem
"SRV" records include the port in DNS responses. They've been available for almost 30 years. Unfortunately, outside of a few protocols (SIP, XMPP, etc.) they never caught on.
Your argument is that someone should have tried to predict what random numbers people who didn’t read the spec would choose, and reserve those? That seems…quixotic. Would you reserve other numbers in this scheme, like all
possible birthdays, just in case?
I would argue that if anyone had considered the possibility of people randomly choosing 1/8 addresses, the error was in reserving 1/8 in the first place. This only happened because there were no consequences to making up addresses.
There is no licensing for internet providers and that's a good thing. Most ISPs in the US already provide IPv6. 45% of users access Google over IPv6 and this has been increasing linearly for the last decade: https://www.google.com/intl/en/ipv6/statistics.html
The final barrier is crappy consumer routers with terrible default settings. Not much we can do but wait for these devices to die because the average user won't buy a new one until that happens
> Most ISPs in the US already provide IPv6. 45% of users...
Is that distorted by cellular providers?
I know my wired ISP does not provide IPv6, even when I explicitly asked for it. They do CGNAT for most customers and assign a static IPv4 to anyone who complains (and then try to start charging for it a year later).
The ISP at my previous home was also IPv4 only (although, thankfully, no CGNAT).
Probably a bit, though I have noticed an upward trend in v6 support with ISPs.
When I was using Spectrum, was pleased to find they had (barely-passable) support, SLAAC wasn't a thing IIRC. Google Fiber has done well and so did AT&T's offering
Ok, fair enough. If Spectrum and AT&T both support v6 now (edit: Comcast/Xfinity too!), then that probably does cover a good portion of the wired ISP market.
Last I checked, most of that IPv6 traffic was mobile, and most residential ISPs don't offer IPv6, or if they do in areas, they don't always provide IPv6-ready equipment. We're in a similar situation up here in Canada - adoption is slowly trickling in through mobile networks, but many large ISPs still aren't offering it.
TL;DW: if you need to use example IPs in your documentation, use the ones listed in RFC3849 and RFC5737!