Live data from Hacker News

GitHub seems to be enabling IPv6 for GitHub Pages

github.com

41–50 of 57 posts

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#41
I recently moved to a new city and signed up with a medium-sized regional ISP. I was disappointed when I set up my router and saw I didn’t have an IPv6 address. I searched the web for “${MY_ISP} IPv6” and found their FAQ page with the following entry (paraphrased):

> Q: Does $ISP support IPv6? > A: Yes! $ISP supports IPv6. For instructions on configuring your router….

I spent an hour trying and failing to get it working. Frustrated that I’d just invested so much time and not wanting it to be a total waste, I called customer service and, as you can imagine, I had an extremely difficult time trying to convince anyone that I didn’t need to turn my router off and on again.

Over the span of a few weeks, I tried calling twice and using their live chat once to ask about IPv6. At this point it was a matter of principle, if their website said they supported it I was going to hold them to it. Each time I communicated with them, I got a different answer as to why IPv6 wasn’t working: one person told me they do support it but not in my area, another person told me they don’t support it at all, and the third person told me that they did “offer” it but did not “support” it. Each time I read them the FAQ page entry which explicitly said they do support it with no conditions. By their responses, it was obvious that they thought I was a weirdo, but I just really wanted them to give me the IPv6 that they’d promised.

After all of this failed, I tried one last thing: I submitted a complaint with the FCC about misleading advertising. Before you call me a Karen, I wasn’t trying to get anyone in trouble or to get special treatment or whatever. I just wanted them to provide the services they promised. I specifically mentioned that I had nothing but positive experiences with the staff that I’d interacted with and that my issue was just a technical/marketing one. Well, whaddya know, I got a lovely email within 6 hours thanking me for bringing the issue to their attention, saying that they had identified a configuration issue in my local area, and that they had scheduled maintenance for the next day to fix it. And within 24 hours I (and presumably hundreds of my neighbors) had working IPv6!

Was it worth the time and effort? For IPv6, no. But for the satisfaction of getting a corporation to follow through with their word? Also, no. But for the ability to write this comment? Well, no. And I suspect this is how just about everyone who has ever spent more than an hour banging their head against a wall trying to get IPv6 working has felt in the end. Like, wow, I can now look at literally the same exact cat pictures as I did before but now I have access to 79 octillion times as many addresses, yay?

That’s my best guess as to why we’re still in the dark ages of 2^32 internet addresses.

Side note, their FAQ page has been changed to say “IPv6 is supported in most areas”. I’m at least proud to know I’m the reason for a few more weasel words on a corporate marketing website. Success.

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#42

I recently moved to a new city and signed up with a medium-sized regional ISP. I was disappointed when I set up my router and saw I didn’t have an IPv6 address. I searched the web for “${MY_ISP} IPv6” and found their FAQ page with the following entry (paraphrased): > Q: Does $ISP support IPv6? > A: Yes! $ISP supports IPv6. For instructions on configuring your router…. I spent an hour trying and failing to get it work…

The world owes you a debt of gratitude. I've often thought about siccing the FTC on companies, but you pulled it off. How long did it take between your report and the company responding?

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#43

Apologies for the silly question, but what’s so difficult in adopting v6? Cloud providers spend billions on new hardware every year, I’m honestly surprised at reading that some of them don’t fully support it yet? Is there some specific technical issue preventing them from mass adoption?

It's a chicken and egg problem - websites, etc don't want to enable IPv6 because clients have broken IPv6 implementations and that causes delays and issues, and clients have broken IPv6 implementations because the internet providers don't care to fix it because nobody uses it.

(Many computers THINK they have IPv6 connectivity when they do NOT and so if they see an AAAA record and try to connect to it, it can cause a delay while that fails. Most browsers are better at this now (they will try to connect to both and show whichever answers faster), but not all network things are browsers.)

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#44

I recently moved to a new city and signed up with a medium-sized regional ISP. I was disappointed when I set up my router and saw I didn’t have an IPv6 address. I searched the web for “${MY_ISP} IPv6” and found their FAQ page with the following entry (paraphrased): > Q: Does $ISP support IPv6? > A: Yes! $ISP supports IPv6. For instructions on configuring your router…. I spent an hour trying and failing to get it work…

The world owes you a debt of gratitude. I've often thought about siccing the FTC on companies, but you pulled it off. How long did it take between your report and the company responding?

It was really fast like I said, under 6 hours until I got a response. I submitted it to the fcc in the morning and had the email by the afternoon, and had IPv6 by the next day at 10am. Goes to show what you can accomplish if you go up the chain, I could have probably called their tech support every day for the next 5 years and gotten nowhere.

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#45
post #5

Too bad I often have to disable IPv6 because my ISP suddenly decides to stop routing it from time to time. This is Comcast, so arguing with them doesn't get me anywhere, all I get is "Thanks for expressing your concern." "Rest assured that I will solve all your problems." from the support dude in some random call/chat center.

I use both Comcast and Monkeybrains. I've had to disable IPv6 only once in the last 18 months: for my Monkeybrains connection.

Just to clarify, this was just once, for about a day.

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#46

Apologies for the silly question, but what’s so difficult in adopting v6? Cloud providers spend billions on new hardware every year, I’m honestly surprised at reading that some of them don’t fully support it yet? Is there some specific technical issue preventing them from mass adoption?

In the past, one of the factors was SEO. A few years ago (in 2018, to be precise), the company I worked for tried enabling IPv6 on some of our websites, and observed some drop in Google ranking, that recovered when the IPv6 experiment was undone.

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#47
post #10

I'm surprised they didn't have it until now. Every cloud provider's load balancer and CDN has had IPv6 for a while. That they don't have it for the site or SSH is less surprising, but GitHub Pages is static content.

Maybe AWS, Azure, and GCP, but not places like Digital Ocean. IPv6 is still very niche, and it's suffering from that effect where it being very niche keeps it very niche. Nobody wants to be an early adopter (and momentum seems to be dying), and the result is that there's no adoption. I think we've been trying to make IPv6 happen since 1998, and ... it's not happening.

Linode fully supports IPv6 and it works quite nicely.

I used to have a Hurricane Electric IPv6 tunnel but I've abandoned mine as my small rural cable provider turned IPv6 on quietly (all you have to do is have your router ask for a /48).

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#48

Apologies for the silly question, but what’s so difficult in adopting v6? Cloud providers spend billions on new hardware every year, I’m honestly surprised at reading that some of them don’t fully support it yet? Is there some specific technical issue preventing them from mass adoption?

In the past, one of the factors was SEO. A few years ago (in 2018, to be precise), the company I worked for tried enabling IPv6 on some of our websites, and observed some drop in Google ranking, that recovered when the IPv6 experiment was undone.

Sorry but this is just too anecdotal for me. There are so many factors that go into Google rankings and Google itself makes it pretty easy to discover any crawling or indexing issues on a website that it seems the root cause of the issue should have been discovered (if it was due to a crawl/index) and I doubt it was a properly working ipv6 setup (something broke somewhere on your company's end is what I'm getting at). Cloudflare uses ipv6 and I doubt all their customers websites were being penalized by Google

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#49

Earlier quoted context omitted.

In the past, one of the factors was SEO. A few years ago (in 2018, to be precise), the company I worked for tried enabling IPv6 on some of our websites, and observed some drop in Google ranking, that recovered when the IPv6 experiment was undone.

Sorry but this is just too anecdotal for me. There are so many factors that go into Google rankings and Google itself makes it pretty easy to discover any crawling or indexing issues on a website that it seems the root cause of the issue should have been discovered (if it was due to a crawl/index) and I doubt it was a properly working ipv6 setup (something broke somewhere on your company's end is what I'm getting at)…

Well, it was just a stock Hetzner dedicated server. All that was added/removed is a DNS AAAA record pointing to the server's IPv6 address. We used IPv6 before in order to access LXC containers hosted on that server. If THAT counts as a broken setup (which I doubt, see below), then, sorry, I don't see how we could do better.

I have personally checked the result from my then-home connection (unfortunately, through he.net tunnel) and it worked, and it was properly logged by Apache. The CEO had native IPv6 on the phone, and it also worked for him, and, according to Apache logs, IPv6 was also used. I have also checked that the site was still accessible over IPv4. So I indeed don't know what went wrong.

On the other hand, I did not have access to Google webmaster console and had to trust the CEO that the SEO problem exists. It might have been a coincidence. And again, the SEO problem was in 2018, and is likely irrelevant today even if it really existed back then, that's why I said "in the past".

Re: GitHub seems to be enabling IPv6 for GitHub Pages

#50
post #30

Earlier quoted context omitted.

Also, A6 is probably a better name, but it has a worse (unworkable?) format, so AAAA is a reasonable second choice.

It's not that hard to say anyway. You read it as "quad a".

Fun fact: I mulled over whether to write “an AAAA record” or “a AAAA record”. I ended up deciding to go with “an AAAA record” even though I would read it “a quadruple A record” which is physically easier to say than “an A A A A record”.
Post reply on HN