Cloudflare Outage November 18, 2025
early morning on Tuesday, November 18, 2025, Cloudflare—a key provider of web infrastructure, security, and performance services handling about 20% of global internet traffic—experienced a widespread internal service degradation. This has led to intermittent connectivity issues, "Error 522" (connection timed out) and "internal server error" messages for users worldwide. The outage peaked around 6:40 AM ET, affecting millions of sites and apps, but Cloudflare has implemented a fix, with most services now recovering or back to normal levels. Some residual errors may persist as propagation completes across their global network.
Current Status
- Resolution Progress: Cloudflare's status page confirms the incident is resolved, with error rates returning to pre-incident baselines for most affected services like Cloudflare Access and WARP. They are still monitoring application services and the support portal, which saw brief disruptions. A planned maintenance in their Santiago datacenter may have contributed, but the root cause (possibly an unexpected traffic spike) is under review.
- User Reports: Downdetector shows a sharp spike in complaints, with 62% citing server connection issues, 27% website problems, and 11% hosting errors. Peaks hit around 2:43 PM UTC, but reports are declining rapidly post-fix. Affected regions include North America (e.g., Texas, Seattle), Europe (e.g., France, UK), and Asia.
- Global Impact: Over 100 million websites rely on Cloudflare, so the ripple effects were immediate and broad, halting access to everything from social media to gaming and e-commerce. Confirmed Affected Services
Based on Cloudflare's updates, user reports, and outage trackers, here's a rundown of high-profile impacts (many now recovering):
| X (formerly Twitter) | Login and feed loading failures; "something went wrong" errors. | Recovering; partial access on some desktops. |
| OpenAI (ChatGPT) | API and web access disruptions; intermittent 500 errors. | Back online for most users. |
| Spotify | Streaming and app connectivity issues. | Resolved. |
| Discord | Server connections and voice chat drops. | Normal. |
| Amazon Web Services (AWS) | Select services (e.g., some APIs) affected via Cloudflare integration. | Minimal ongoing impact. |
| Canva | Design tool loading errors. | Resolved. |
| Tinder | Match loading and messaging failures. | Back to normal. |
| League of Legends | Game server logins and matchmaking offline. | Online. |
| Letterboxd | Site-wide access blocked with error pages. | Resolved. |
| Downdetector | Brief self-outage, skewing real-time tracking. | Functional again. |
| Others | Vinted (shopping), Grindr (dating), bet365 (betting). | Recovering. |
What Caused This?
Cloudflare hasn't pinpointed the exact trigger yet, but early indicators point to an internal network issue, possibly exacerbated by a traffic surge or the Santiago maintenance. Unlike the recent AWS outage, this was shorter-lived (under 3 hours at peak) but highlighted the fragility of centralized web infrastructure— a single point like Cloudflare can freeze vast swaths of the internet.
Tips for Users
- Check Status: Monitor [Cloudflare's status page](https://www.cloudflarestatus.com/) for real-time updates or subscribe to notifications via their dashboard.
- Workarounds: Try mobile data/VPN if Wi-Fi is affected, or incognito mode to bypass cached errors. For X users, the app may work when the web doesn't.
- Broader Lesson: This echoes recent disruptions (e.g., AWS last month), underscoring why decentralization (like blockchain) stayed resilient—posts on X noted crypto services chugging along unaffected.
If you're still hitting snags, drop details on your region/service, and I'll dig deeper. The web's bouncing back fast—stay tuned!