This is an interesting one. Doing "whois github.com" the nameservers returned from whois.markmonitor.com were at one point:

  Updated Date: 2020-06-22T12:02:10-0700
  [...]
  Name Server: ns-1707.awsdns-21.co.uk
  Name Server: ns-1283.awsdns-32.org
  Name Server: ns-520.awsdns-01.net
  Name Server: dns3.p08.nsone.net
  Name Server: ns-421.awsdns-52.com
  Name Server: dns1.p08.nsone.net
  Name Server: dns4.p08.nsone.net
  Name Server: dns1.p08.nsone.net
  Name Server: dns2.p08.nsone.net
  DNSSEC: unsigned
  URL of the ICANN WHOIS Data Problem Reporting System: http://wdprs.internic.net/
  >>> Last update of WHOIS database: 2020-06-22T12:23:58-0700 <<<

Then as of a bit later:

  Updated Date: 2020-06-22T12:26:23-0700
  [...]
  Name Server: ns-1283.awsdns-32.org
  Name Server: ns-1707.awsdns-21.co.uk
  Name Server: ns-520.awsdns-01.net
  Name Server: ns-421.awsdns-52.com
  DNSSEC: unsigned
  URL of the ICANN WHOIS Data Problem Reporting System: http://wdprs.internic.net/
  >>> Last update of WHOIS database: 2020-06-22T12:34:32-0700 <<<

The problem is:

  $ dig @a.gtld-servers.net ns github.com
  [snip]
  ;; AUTHORITY SECTION:
  github.com.  172800 IN NS ns-520.awsdns-01.net.

Those records have a 48 hour TTL. So while this works around the problem with nsone, if they can't fix those servers it might be a while or need manually flushing caches before github.com fully comes back for everyone.

[edited to add updated date]

Now mentioned at https://www.githubstatus.com/incidents/shdhxntw519h

@dang I saw a number of 'github is down' posts being flagged almost immediately (including this one). Is there a guideline against posting these kinds of messages?

It's hard to see how they are "interesting" in the sense of the guidelines. a post-mortem about what broke might be, but the pure fact that a service is down again? This one here at least has some analysis on what's going on.

I only posted this because I had some immediate insight into the nature of the issue - i.e. one of Github's two DNS providers seems to be having issues. Someone less familiar with DNS could learn a little by running the commands I put in the original post and seeing how the results differ. It might also be interesting to see how the reality of the issue translates into their eventual status page update.

In general though, I absolutely agree that generic 'X is down' messages with no further information are pointless.

I agree with you (though I have to admit I just posted one myself). But then maybe it should be mentioned more explicitly in the guidelines.

I support it, to be honest.

Those threads always devolve into low quality arguments and spam. With occasional marketing effort for other services.

I made one of those posts and it got flagged almost immediately, and I honestly just wanted to know if the issue is on my end or not, GitHub's status page shows that all services are up.

That's also why I came here, and I didn't flag your submission.

But I understand why it happens.

Here we go again. Yet my argument to self-host just becomes more valid. [0][1]

[0] https://news.ycombinator.com/item?id=23102942

[1] https://news.ycombinator.com/item?id=23572532

TBH I only go to GH to access other people's projects.

If I have to use git, I use GitLab...

yep, also noticed that. At first I thought I've messed something up with pihole. But looks like it's something bigger...

https://downforeveryoneorjustme.com/github

whois record updated today. maybe propagation issues. Updated Date: 2020-06-22T12:26:23-0700 Creation Date: 2007-10-09T11:20:50-0700

Same here. 1.1.1.1 seems to resolve it correctly.

can confirm