2026-10-18 13:50 UTCIncidents Detected

Brief Cloudflare US East Slowdown

No widespread internet outage is in progress this Sunday morning (checked around 13:50 UTC). The one notable event so far was short and is resolved: a Cloudflare latency incident in the eastern United States.

**Cloudflare: latency and timeouts in Virginia and Ohio (resolved)**

From 08:00 to 08:50 UTC, Cloudflare customers served through its Ashburn, Virginia (IAD) and Columbus, Ohio (CMH) locations saw higher latency and timeouts. Cloudflare says an external provider unexpectedly withdrew prefixes on peering links in those cities. Service fully recovered once the prefix advertisements were restored. Ashburn is one of the busiest interconnection hubs in the world, so some users in the US East region may have noticed slow or failing sites for a while. The incident was posted after it ended. [Cloudflare incident](cloudflarestatus.com)

Cloudflare's status page also lists a number of individual data centers as degraded or re-routed. That is routine for Cloudflare's network, where traffic moves to nearby locations during maintenance. Its long-running WARP geolocation issue (open since August 27) is still marked as identified.

**Minor, ongoing**

**Twilio:** SMS sent from alphanumeric sender IDs is failing to reach subscribers on several networks in Turkey (identified since 23:14 UTC on October 17). A few long-running regional messaging and voice issues are also still open (Thailand voice post-dial delay, GCI US MMS delays, Australia MMS). [Twilio status](status.twilio.com)

**Azure DevOps:** an advisory for service maintenance is in effect, with no reported customer impact.

**Recently resolved (past 72 hours)**

**Cloudflare, October 15–16:** a busy stretch. Network performance issues in Eastern North America (October 15, 20:48 to 21:38 UTC) and London (October 15) were both rated major. There were also elevated HTTP 502 errors, DNS update delays and Durable Objects errors (October 15), plus HTTP 5xx errors in Ashburn, Cloudflare Access availability problems, and an outage of the FedRAMP dashboard and Access (October 16, 17:56 to 19:09 UTC). All are resolved.

**OpenAI, October 15:** "widespread feature degradation" across ChatGPT from 16:29 to 18:56 UTC, resolved. [OpenAI status](status.openai.com)

**Twilio, October 17:** 5300 errors on the Singapore client edge (10:07 to 15:11 UTC), resolved.

**Zoom, October 16:** a brief degradation of Zoom Chat compose and AI Companion SMS summaries, resolved within an hour.

**All clear**

AWS, Google Cloud, Azure, GitHub, Akamai, Anthropic (Claude), Discord, Atlassian, Slack, Zoom, Vercel, Heroku, DigitalOcean, Stripe, Cursor, Groq, Replicate, ElevenLabs, Cohere and Stability AI all report normal operation. Searches of crowd-sourced outage reports turned up no major consumer platform outage today. Meta's status page could not be read automatically, and Fastly's status page blocks automated checks, so their status is based on the absence of public reports.

The internet is up.

← Home View all history

Brief Cloudflare US East Slowdown

Last 30 Days