The issue has been resolved, and page load times should be back to normal.

The root cause was a denial-of-service attack on one of our customers, taking advantage of inadequate rate limiting on a particular endpoint. We've fixed the rate limiting to prevent this from happening again, and we apologize for the inconvenience.

We have tentatively identified the problem and are implementing measures to mitigate it. Sites are coming back online.

We're investigating reports of timeouts when attempting to access sites hosted in our US datacenter.