KEY POINTS:

• Cloudflare confirmed a global network issue at 11:48 UTC on November 18, according to The Mirror US
• Thousands of websites experienced widespread 500 errors and service disruptions, Tom’s Hardware reported
• Services began recovering at 12:21 UTC though higher error rates continued, per Cloudflare’s status page

A major outage affecting Cloudflare’s global network infrastructure disrupted access to numerous websites and online services on November 18, 2025, with the company confirming widespread issues across its systems.

Outage Confirmation and Timeline

Cloudflare acknowledged the global network problems at 11:48 UTC, according to The Mirror US, which reported that the company stated it was “aware of, and investigating an issue which potentially impacts multiple customers.” Tom’s Hardware reported that the company confirmed it was investigating “widespread 500 errors, Cloudflare Dashboard and API also failing.”

The Internet infrastructure firm first acknowledged issues approximately 40 minutes before the main outage report when it reported availability problems with its support portal, according to BleepingComputer’s search results. The Mirror US reported that Cloudflare’s support portal provider was experiencing failures, causing errors when customers tried to view or respond to support tickets.

Scale and Impact

Downdetector received 11,201 reports of issues at 11:37 GMT, according to Tom’s Hardware, which noted the number later decreased to 6,570 reports. BleepingComputer reported in search results that tens of thousands of reports flooded into Downdetector since the outage began.

The Mirror US reported that sites affected included X (formerly Twitter), Facebook, Spotify, Canva, and Grindr according to Down Detector data. Windows Central reported via search results that even Downdetector itself was temporarily taken offline due to its reliance on Cloudflare’s services, though it recovered within minutes.

Tom’s Hardware reported that X (formerly Twitter) experienced intermittent issues as a result of the outage, with the platform’s availability being “patchy for the last hour or so.” Dataconomy noted in search results that users attempting to visit Downdetector were greeted with a blank white screen during the height of the outage.

Technical Details

The Mirror US reported that users received messages stating there was an “internal server error on Cloudflare’s network” with instructions to “please try again in a few minutes.” Dataconomy reported via search results that users encountered infinite loops when browser checks tried to verify if they were human, along with 502/504 “Bad Gateway” errors across thousands of websites.

BleepingComputer reported in search results that Cloudflare’s global network comprises distributed infrastructure in over 330 cities across more than 120 countries, with 449 Tbps global network edge capacity connecting to over 13,000 networks. The same source indicated that in their tests, Cloudflare nodes across Europe were down, including those in Bucharest, Zurich, Warsaw, Oslo, Amsterdam, Berlin, Frankfurt, Vienna, Stockholm, and Hamburg.

Scheduled Maintenance Connection

The Mirror US reported that Cloudflare had announced earlier they would be performing scheduled maintenance in a Santiago datacenter between 12:00 and 15:00 UTC, though the publication noted it was unclear whether this was related to the current issues.

Dataconomy reported via search results that Cloudflare had a heavy schedule of maintenance planned for November 18, 2025, noting that while maintenance is usually seamless with traffic re-routed to other cities, “the volume of simultaneous work combined with the global alert suggests something may have gone wrong during these updates.”

Joburg ETC reported in search results that scheduled maintenance was underway at multiple datacenters including Los Angeles, Miami, Guatemala City, Atlanta, and Chicago, which may have contributed to latency and rerouting issues.

Recovery Efforts

The Mirror US reported that Cloudflare noted services appeared to be recovering at 12:21 UTC, with the company stating: “We are seeing services recover, but customers may continue to observe higher-than-normal error rates as we continue remediation efforts.”

Tom’s Hardware reported in an updated search result that Cloudflare’s System Status page showed the company was still investigating the issue at 12:37 UTC, noting that “services have started to recover, but there will be higher than normal error rates while the sys admins work on fixing the issue.”

BleepingComputer reported via search results at 07:29 EST that Cloudflare was seeing signs of recovery, quoting the company as saying: “We are seeing services recover, but customers may continue to observe higher-than-normal error rates as we continue remediation efforts.”

Current Status

Joburg ETC reported in search results that Cloudflare’s services were “largely up and running again” following the earlier global network issue, with the company confirming it was continuing to investigate the incident. The publication noted that according to the official Cloudflare Status page, the majority of core services including API, CDN, DNS, and Dashboard were marked as Operational.

The same source reported that while the incident “Global Network experiencing issues” remained logged with a status of “Investigating,” the practical impact appeared to have subsided for most users, though Cloudflare’s support portal might still be affected.

Dawn.com reported via search results that NetBlocks, a global internet watchdog, stated the incident was not related to “country-level internet disruptions or filtering.”

Historical Context

The Mirror US noted that the Cloudflare issues echoed major Amazon Web Services outages which disrupted a large portion of the internet weeks ago, impacting apps including Snapchat, Reddit, Fortnite, Roblox, Canva, Duolingo, Ring Doorbell, and McDonald’s. According to The Mirror US, Amazon later announced it had resolved that issue, which occurred at the heart of its cloud computing operations in North Virginia.

BleepingComputer noted in search results that Cloudflare had mitigated another massive outage in June that caused Zero Trust WARP connectivity issues and Access authentication failures across multiple regions.