[ballenf]

A possible explanation is that the traffic from active use of 1.1.1.1 caused some backend service to get overloaded with traffic due to a faulty assumption that the address would never be used by customers. Anyone keep traceroutes while before the patch to see if there were errant stops or delays?

They had the choice of "fix the whole backend" or "block 1.x on the user end".

Guess we know which one was easier. If all this wild speculation is true, maybe they're working on a fix to the root cause and will roll back the patch when complete.

This would make the situation both due to incompetence and intentional.

[Phylter]

1.1.1.1 is well known (based on the announcement from cloudflare anyway) to have tons of random traffic. That's part of the reason it wasn't implemented by others as a valid address for anything. Could the fact that they're simply allowing traffic at that address cause additional stress on AT&T's network?

I ask because I don't know. I figure any traffic headed that direction would go anyway it just wouldn't get routed very far with no valid destination.

[Piskvorrr]

Yeah. And there's also a lot of traffic going in Facebook's direction, for example. Hey, let's blackhole that too - and alleviate the stress on our network that comes from people using it. (In non-sarcastic tone: that doesn't make any sense.)

[Phylter]

Based on what I understand, the amount of traffic headed to 1.1.1.1 is much more significant. I agree with you though, that wouldn’t be justification to block it. It looks like they’re also blocking 1.0.0.1 and the relevant ipv6 addresses which shouldn’t have the same traffic issue.

[tracker1]

I doubt it's all that significant, it's a really small portion of traffic compared to a web page, javascript, css or images... and with caching even less of an impact.

[Phylter]

The problem isn’t DNS traffic. The problem is that for years people have been using 1.1.1.1 in the configuration of software and devices when they didn’t have an up address to configure. The result is that when 1.1.1.1 becomes routable all that additional traffic flows there and AT&T along with other provides carries that traffic. I was wrong that AT&T was blocking it for honorable reasons but this is a still a significant amount of traffic.