Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

IP Addresses Through 2025

potaroo.net

Post reply on HN
Print view View original

91–100 of 158 posts

Re: IP Addresses Through 2025

#91

Post by cyberax »

RiverCrochet wrote: post #75
dist-epoch wrote: post #44

Earlier quoted context omitted.

But has the nice side-effect of working as a firewall, before traffic gets to you.

- Did you disable UPnP on your router? If not, any device behind the router can simply ask the router to open a port, typically without authentication, bypassing this "firewall" completely. - TURN and STUN trivially bypass this side-effect, and a side effect of that is a third party has to often be involved, which can be collecting data later leaked or used against you. - The monstrosity of NAT is that it's the core…

I've yet to see UPnP work...

Re: IP Addresses Through 2025

#92

Post by rmoriz »

There are plenty of vectors left to squeeze the existing IPv4 space especially all the Legacy assignments held by deceased companies and individuals. There is no procedure to reclaim them. Even when you invest time and money to find the relatives, the RIR may decline a transfer so nobody invests here as long as plenty of former hosting, colocation and regional access providers leave the market after their customers moved to the US hyperscalers or out or business.

I think around 2000 every new LIR at RIPE got a /19 allocation. Smaller companies are now almost 30 years old and the founders divest their assets step by step unless someone buys everything.

Re: IP Addresses Through 2025

#93

Post by huslage »

dlcarrier wrote: post #41
Fiveplus wrote: post #17

The collapse in IPv4 transfer prices is what caught my eye here, dropping from a ~$55 peak in 2021 to a mean of $22 in early 2026 (figure 12). This validates my hypothesis that the run-up in 2020–2022 was an artificial scarcity bubble driven largely by hyperscalers. AWS was right up there stockpiling before they shifted their pricing model. Once AWS introduced the hourly charge for public IPv4 addresses (effectively…

As someone with a background in electronics who doesn't manage any internet-connected equipment but has multiple embedded devices connected to a WAN, I'm glad that IPv4 still seems to have a bit of life left in it. When IPv6 was developed, over 30 years ago, connecting everything to the internet seemed like a great idea. I know that IPv6 can be made secure, but I don't have the background or research time to learn ho…

NAT is not a security measure at all. It just obscures what's behind a firewall, but that is leaky and not reliable from a security perspective. It might make you feel better, but that is not security.

Re: IP Addresses Through 2025

#94

Post by radlad »

cyberax wrote: post #89
radlad wrote: post #46

Earlier quoted context omitted.

> the NAT-by-default of IPv4 IPv4 is not NAT-by-default. The reality of the world we live in today is that most home networks have a NAT, because you need multiple devices behind a single IP. That said, I agree: it's quite unknowable how many services I've turned on on local machines with the expectation that a router firewall sat between me and potential clients. But that doesn't go away with IPv6 - the NAT does, th…

> NAT is not the firewall. NAT _is_ a firewall. And a much safer one than IPv6 firewalls, because NAT will fail safe if misconfigured.

NAT is not a firewall: all it does is rewrite packets, it does not drop them.

Re: IP Addresses Through 2025

#95

Post by patmorgan23 »

speedgoose wrote: post #60
1970-01-01 wrote: post #38

I'm interested in any new successful startups going full IPV6 from the beginning. Once we cross that bridge, where your internal IPV4 knowledge is equivalent to token ring knowledge, there's nothing else to watch.

It would be a strange and unnecessary risk to take for a startup in my opinion.

It's really not a risk, as long as you dual stack your edge.

Re: IP Addresses Through 2025

#96

Post by cyberax »

snowwrestler wrote: post #24

I pay close attention to IPv4 addresses for outgoing emails. At work we use several email services and pay for a dedicated IP(v4) at each. And when we provision a new service, we expect our new IP address to be “clean,” by which I mean it is ideally not found on any email reputation list. For websites and services I don’t care. Some hosting platforms publish via CNAME, and some via A and AAAA records. Most seem to us…

> dig AAAA github.com > dig AAAA amazon.com

Hm...

It's more likely that the widespread deployment of CGNAT and 464XLAT in mobile networks made the IPv4 scarcity a non-issue. The some CGNAT solutions can multiplex more than 20000 devices onto a single IPv4 address.

I'm a very early adopter of IPv6, and I _still_ have operational issues with it.

Re: IP Addresses Through 2025

#97

Post by psim1 »

In 2021 I speculated on IP and acquired a /23 block by ARIN wait list. I figured on running some services from the IP space for a while and after the 5 years mandated wait time would cash in when surely it would fetch $100k from some party desperate for IPv4.

At this point the services I am running are far more lucrative than the IP space itself is turning out to be.

Re: IP Addresses Through 2025

#98

Post by 1vuio0pswjnm7 »

What happens when a so-called "tech" company that cannot be trusted wants to punch holes in the user's firewall without prior consent from the user

Purely hypothetical, of course

For example, WhatsApp tries to connect to at least two servers on UDP port 3478 without asking the user if this is what they want to do or explaining the purposes of these connections

Example server addresses are

57.144.221.54

31.13.70.48

3478 is the port used for "Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)", or "STUN" for short

https://www.ietf.org/rfc/rfc3489.txt

Perhaps IPv6 would obviate the need for STUN

Re: IP Addresses Through 2025

#99

Post by tormeh »

1970-01-01 wrote: post #88
oasisbob wrote: post #84

Earlier quoted context omitted.

Exclusively IPv6 without any transitional mechanisms would be difficult to succeed with. However, there are network upstarts like Jio (India) which made huge v6 investments from day one which use 464xlat for subscribers to access v4-only resources.

>Exclusively IPv6 without any transitional mechanisms would be difficult to succeed with. That's my point; why is it still difficult? What exactly are the pain points for a fully commercialized native IPV6-only business, and why do we think it will be easier to maintain the status quo?

Many wired networks are IPv4-only, so you've excluded a bunch of consumers. It'd be like not supporting the Edge browser.

Re: IP Addresses Through 2025

#100

Post by jonathanlydall »

radlad wrote: post #94
cyberax wrote: post #89

Earlier quoted context omitted.

> NAT is not the firewall. NAT _is_ a firewall. And a much safer one than IPv6 firewalls, because NAT will fail safe if misconfigured.

NAT is not a firewall: all it does is rewrite packets, it does not drop them.

The article actually remarks on this kind of argument.

While you are technically correct about NAT not being a firewall, it is in practice a widely used front-line defense which even if not “perfect”, it has indisputably proven to be quite effective against a lot of malicious activity.

Against highly determined malicious actors you will of course want a proper firewall, but for 99% of people, NAT is enough to keep from being bothered by run of the mill malicious actors.

Kind of like physical home security, a lot of it is very easy to bypass, but it’s good enough for the common threats.

Post reply on HN