I do feel for the engineers working on whatever is going on over there btw. I worked on a "Self-DDoS" incident about a decade ago. It sucked. I wrote about it in my book "Hands-on Digital Forensics and Incident Response".
@SecureOwl Really, it would be quite impressive if they managed to *not* cause an issue like this. Or something else. This sort of rate limit isn't a small change. Doing that without breaking things would require a lot of planning and testing, and take a while before it could actually be rolled out.
While I imagine they were basically only given the time to implement the obvious solution, deploy to production, and hope for the best.
@tsupasat funnily enough ExtraHop was one of our main troubleshooting tools during this incident ;-)
Also funnily enough I posted this today because a company got bought out by someone who has no clue how to run it ;-) ;-)