- Sources: Cloudflare R2 incident, Cloudflare email delivery incident, Cloudflare network performance incident
- Summary: Cloudflare recorded six incidents dated 2026-07-24. The most severe, rated major, ran 18:49 to 19:19 UTC with elevated error rates for R2 buckets hosted in the APAC region. An email delivery incident from 18:33 to 19:28 UTC delayed the mail Cloudflare itself sends, which it lists as including password reset and MFA token messages. A network performance incident in western North America ran 11:24 to 11:59 UTC and was attributed to partner network instability over shared interconnections, resolved by an infrastructure partner rather than by Cloudflare. Three further entries cover elevated errors from congestion reaching APAC between 13:43 and 14:13 UTC, GraphQL rate limit errors in the analytics dashboards from 14:07 to 16:44 UTC, and Content Security Policy rules not rendering in the dashboard from 21:34 to 21:59 UTC while the policies themselves kept working. No entry names a root cause beyond the partner interconnection problem.
- Why it matters: The email incident is the one worth planning around, because a provider's own mail path failing takes password reset and MFA delivery with it while every other health signal stays green.
- Follow-up: Watch for a Cloudflare writeup on the R2 APAC errors and whether the western North America interconnection instability recurs.
send feedback on this story