Earlier quoted context omitted.
What happens when DNS is down
What happens when DNS is down (IPv4 edition)?
You type in the IP that's extremely easy to remember as it's just 4 numbers seperated by periods
IPv6 is the only way forward
301–310 of 350 posts
Re: IPv6 is the only way forward
#302Earlier quoted context omitted.
> except Android That alone is significant. Furthermore, DHCPv6 holds you back from various desirable things like privacy addresses and (arguably even more importantly) IPv6 Mostly.
> Furthermore, DHCPv6 holds you back from various desirable things like privacy addresses and (arguably even more importantly) IPv6 Mostly. Why would DHCPv6 hold back privacy addresses? Can't DHCPv6 servers generate random host address bits and assign them in DHCP Offer packets? Couldn't clients generate random addresses and put them in Request packets? See perhaps OPTION_IA_TA (Temporary Address): * https://datatrac…
> See perhaps OPTION_IA_TA (Temporary Address):
I was unaware of this, so thanks. Sounds like it addresses (pun intended) my concern.
> How does DHCPv6 hold back IPv6-mostly? First, most clients will send out a DHCPv4 request in case IPv4 is the only option, in which case IPv6-mostly can be signalled
It's not the signalling that's the problem--it's the configuration of the CLAT which requires SLAAC, afaiu. This is in fact the subject of the latest IPv6 Buzz podcast episode: https://packetpushers.net/podcasts/ipv6-buzz/ipb197-slaac-an...
Re: IPv6 is the only way forward
#303IPv4 has been "in crisis" for the entire 20 years I've worked in tech and we seem to be managing alright. Not to say things can't be better or we shouldn't try to improve. But I'll be surprised if v4 isn't still the default for most use cases in another 20 years.
Part of the issue is this affects different countries differently, based on residential IP allocations to household ratios. I am currently on CGNAT in Australia split 256 ways, and any site that doesn't support IPv6 can be borderline unusable. I can't imagine what it's like in countries with worse ratios, like India.
It's been in crisis for decades, but it's also getting increasingly worse every year.
Re: IPv6 is the only way forward
#304Earlier quoted context omitted.
I dismiss ISPs that don't support IPv6.
I should dismiss my ISP that's worked for something like 20 years, works now, and will in all likelyhood still be working in 20 years (baring M&A nonsense or the apocalypse)? Sorry, IPv6 is absolutely not the hill I'm going to die on.
If it doesn't support IPv6 it doesn't work.
Re: IPv6 is the only way forward
#305The same annoying and trite IPv6 haters come out of the woodworks with every article posted.
Re: IPv6 is the only way forward
#306Someone should’ve thought about the UX of IPv6 before declaring it to be “the way”. It’s like having to learn Klingon just to setup your printer. IPvNext could sort that out… maybe it’s time to consider moving on.
SLAAC + mDNS makes IPv6 basically invisible.
If by invisible you mean "things randomly won't work and you'll smash your head in the wall trying to figure out why".
Re: IPv6 is the only way forward
#307Re: IPv6 is the only way forward
#308Earlier quoted context omitted.
> Unless all of these devices are running a dedicated full time server that must be reachable inbound by everyone this is not required. I think this is a lack of imagination. The fact that (CG-)NAT is in the way could be precluding the development of software that could take advantage of incoming/P2P connections. It's a form of (negative/inverse) survivorship bias: kind of like zoning for only single-family homes and…
I think this is a lack of imagination. No imagination required. P2P works fine if at least 20% to 30% have ports open inbound. 70%+ need not have open inbound ports. Where this could theoretically be a problem is if a specific sub-set of CG-NAT users were the only people seeding and downloading something. This non existent problem can be worked around using a VPN mesh. Tinc is an open source VPN that operates in user…
> No imagination required. P2P works fine if at least 20% to 30% have ports open inbound. 70%+ need not have open inbound ports. Where this could theoretically be a problem is if a specific sub-set of CG-NAT users were the only people seeding and downloading something.
"Seeding"? "Downloading"? I think applications besides BitTorrent could be invented and become popular. Even now, existing things like SIP and WebRTC would probably be much less onerous.
> This non existent problem can be worked around using a VPN mesh. Tinc is an open source VPN that operates in user-space and while not as fast as Wireguard it can do things Wireguard could never dream of such as user space mesh routing, always discovering the shortest path.
So you're introducing another layer of software because the underlying network does not have the functionality available (just like STUN/TURN/ICE had to be invented to deal with NAT).
Here's another idea: have IPv6, and if folks want to have end-to-end encrypted communications, start up an IKEv2 process (that opens a hole for its port via UPNP/PCP), and we have IPsec (which is built into most OSes anyway) encrypted communications opportunistically enabled.
Re: IPv6 is the only way forward
#309Earlier quoted context omitted.
> So you actually agree with me, that making all addresses public was stupid to begin with. If an address is not public how can you start an connection from it, or end a connection at it? A web server needs a public address if you want to have people reach it. And you, at some point, also have to have a public address if you want to connect to pubic services: either on your end-host, at your CPE/router's WAN interfac…
So having all devices on public addresses was stupid to begin with on IPv4 and it was arrogantly stupid on IPv6.
> So having all devices on public addresses was stupid to begin with on IPv4 and it was arrogantly stupid on IPv6.
"Yeah? Well, you know, that's just like uh, your opinion, man." — The Dude
Publicly addressable ≠ publicly reachable.
When I was with my last ISP which had IPv6, my printer had a public address, but the only people who could reach it were those on my home network.
Re: IPv6 is the only way forward
#310Earlier quoted context omitted.
> Furthermore, DHCPv6 holds you back from various desirable things like privacy addresses and (arguably even more importantly) IPv6 Mostly. Why would DHCPv6 hold back privacy addresses? Can't DHCPv6 servers generate random host address bits and assign them in DHCP Offer packets? Couldn't clients generate random addresses and put them in Request packets? See perhaps OPTION_IA_TA (Temporary Address): * https://datatrac…
> See perhaps OPTION_IA_TA (Temporary Address): I was unaware of this, so thanks. Sounds like it addresses (pun intended) my concern. > How does DHCPv6 hold back IPv6-mostly? First, most clients will send out a DHCPv4 request in case IPv4 is the only option, in which case IPv6-mostly can be signalled It's not the signalling that's the problem--it's the configuration of the CLAT which requires SLAAC, afaiu. This is in…
> It's not the signalling that's the problem--it's the configuration of the CLAT which requires SLAAC, afaiu.
This operational difficulty has been recognized and alternatives are being put forward:
* https://datatracker.ietf.org/doc/html/draft-ietf-v6ops-clato...