I appreciate IPv6 link local. mDNS + ipv6 ll addresses on my home network means I can connect to local devices, by name (mydevice.local) regardless of whether or not the router or DHCP/RA+DNS is working.
40% of Google users now connect via IPv6
141–150 of 353 posts
Re: 40% of Google users now connect via IPv6
#142Are there any advantages to IPv6 other than more addresses ?
In theory, IPv6 could be slighty faster over IPv4. https://superuser.com/questions/1055584/how-does-ipv6-affect... - Smaller headers - Simpler routing
v6 actually has longer headers on average (40B instead of 20B) but is generally more efficient to process because it's a fixed header length.
ON the other hand, theoretically IPv4 header length could vary depending on the presence of IP options, so you'd need to compute the offset to continue processing the packet.
(I say theoretically because conventional wisdom is that IP options are unreliable in the face of middleboxes, so they're mostly unused. But compliant IPv4 processors have to calculate the header size anyways...)
Re: 40% of Google users now connect via IPv6
#143Earlier quoted context omitted.
A local ISP told me that they want to get into IPv6 as soon as possible. NATs are getting costly and moving just YouTube traffic to IPv6 would actually help a lot. The problem is with end user devices that do not use a stable DUID and when the client hits the reset button, it changes. We are probably going to work around this by responding from the closest hop and taking the MAC into account.
Why are NATs getting costly? Don't quite understand.
because maintaining state for GCNAT tables is far more complex then just forwarding packets. routers doing NAT are thus more expensive then those just doing simple forwarding.
Also, in some countries ISP's need to map the use of a specific ip address to a specific subscriber for law enforcment purposes. GCNAT is no exception to this and creates a large amount of overhead because the public IPV4 prefix space is shared between multiple customers.
Re: 40% of Google users now connect via IPv6
#144Anyone go out of their way to disable IPV6 on all their systems, even at the hardware level, like on routers etc? What's a good reason to avoid IPV6?
I did it before because I assumed it was causing issues and not configuring things on my network properly.
I was wrong and the issues were elsewhere, but it remained disabled on the router for a long time.
I’m generally a person who resists change and I can’t tangibly see the benefits of ipv6; until I realised that “port-forwarding” is an exclusively NAT problem and it’s much easier with ipv6 to just natively open a port on the firewall if I want.
Re: 40% of Google users now connect via IPv6
#145Anyone go out of their way to disable IPV6 on all their systems, even at the hardware level, like on routers etc? What's a good reason to avoid IPV6?
Yes at my router for privacy. No nat means no source device obfuscation
Re: 40% of Google users now connect via IPv6
#146Earlier quoted context omitted.
> So I never added an AAAA record. IPv4 servers are reachable by IPv6-only clients. ipv6test.google.com famously has only an A (IPv4) record for a reason. > God knows what subtle breakages would happen when IPv6 requests come in. Unless your product deals with the nitty gritty of the networking stack itself, pretty sure everything continues to hum along just fine, if the basics are covered.
pretty sure everything continues to hum along just fine Haha. Wishful thinking. Have you ever worked with a live system that grew for 10 years? There could be tons of IP specific things in there. To enforce rate limits. For security. To do statistics. To do A/B tests...
Yes, there could be. But in the end, those are just parameters of the specific networks/services that make up the internet.
Considering that routing is not deterministic, you could get vastly different characteristics from point A to point B across the internet at specific times anyways.
Re: 40% of Google users now connect via IPv6
#147That was easy.
Re: 40% of Google users now connect via IPv6
#148Earlier quoted context omitted.
And adds 20ms+ latency, another complex point of failure, and potential problems depending on the reputation of the IP address block of wherever you're hosting the endpoint. I've used VPNs to smooth over various Starlink problems since I got the service and it helps but it's not a great solution.
My he.net ipv6 tunnel adds about 2ms of latency on average, which I can totally live with.
At least when I was doing this, it also meant that I blocked Netflix and a few other services for my entire network.
Re: 40% of Google users now connect via IPv6
#149Well… only about 10 years after headlines were fearmongering about the world breaking due to IP addresses not being available.
The “breakage” is happening. It’s just in the form of crappy solutions like CGNAT and DS-lite to work around the shortage rather than the internet just not working at all.
also, ipv4 address space is becoming intrenched by large players because address pricing is becoming very expensive.
Want to start a new service and require global connectivity, good luck winning IPv4 auctions because AWS is buying up all address space, even if they are not using it.
Re: 40% of Google users now connect via IPv6
#150“In December 1998, IPv6 became a Draft Standard for the IETF,[2] which subsequently ratified it as an Internet Standard on 14 July 2017.” 40% adoption after 25 years? Really highlights how terrible IPv6 is in terms of backwards compatibility. What we needed was an internet protocol with the benefits of IPv6 that runs as an extension to the IPv4 stack. The current approach to duplicate everything into IPv6 is wasteful…
Would you care to elaborate what "an internet protocol with the benefits of IPv6 that runs as an extension to the IPv4 stack" actually looks like at the technical level that would give the "backward compatibility" that IPv6 you claim is lacking ?