A major outage at internet infrastructure company Cloudflare is disrupting large parts of the web today, Tuesday 18 November 2025. Popular services including X (formerly Twitter), ChatGPT, OpenAI’s other tools, Letterboxd, Canva, and several online games are either slow to load, partially available, or completely unreachable for many users around the world. [1]
According to Cloudflare’s own status page, the company is currently experiencing a “Major Outage” affecting its network and core services, with partial outages reported across multiple regions. [2]
This is a developing story; details below reflect what is known as of late 18 November 2025.
What is happening with Cloudflare today?
Cloudflare has confirmed that its global network is experiencing problems that are impacting multiple customers and services at once.
At 11:48 UTC, the company posted an incident titled “Cloudflare Global Network experiencing issues,” saying it was aware of a problem affecting multiple customers and that users might see widespread 500-series errors. The same update noted that Cloudflare’s dashboard and API were also failing. [3]
A follow‑up status update at 12:03 UTC stated that engineers were continuing to investigate the issue, with no root cause yet publicly confirmed. [4]
On the main status page, Cloudflare lists: [5]
- Cloudflare Sites and Services: Major Outage
- Network: Major Outage
- Multiple geographic regions (including parts of Africa and Asia) flagged with partial outages or maintenancestatus for various data centers.
Put simply: Cloudflare is not completely offline, but a significant portion of its network is degraded, and because so many services depend on it, the impact feels like parts of the internet have simply stopped working.
When did today’s Cloudflare outage start?
Timelines vary slightly by region and by which service users noticed first, but current reporting points to a late morning UTC start:
- Around 6:00 AM ET (approximately 11:00 UTC), Cloudflare’s support portal provider began experiencing issues, which coincided with first signs of degradation on Cloudflare-powered services. [6]
- Shortly afterwards, outage monitoring site Downdetector saw a sharp spike in reports for X, with thousands of users in both the US and UK reporting that the platform was down just after 6:08 AM ET. [7]
- At 11:48 UTC, Cloudflare officially declared a global network incident, noting widespread 500 errors and failures in its dashboard and API. [8]
- Media outlets and tech reporters across Europe, Asia and North America began confirming broad access problemsto Cloudflare-backed sites soon after, with several describing it as a “major web outage” or “parts of the internet down.” [9]
As of this writing, the incident on Cloudflare’s status page remains in the investigating phase, with no “resolved” marker yet published. [10]
Which websites and apps are affected?
Because Cloudflare sits in front of millions of sites as a content delivery network (CDN), DDoS protection layer, and performance optimizer, the exact list of impacted services is long and still evolving. However, multiple outlets and monitoring data highlight several high‑profile names:
- X (formerly Twitter) – one of the first widely noticed failures, with tens of thousands of outage reports on Downdetector and both the web and app versions showing errors or failing to load posts. [11]
- ChatGPT and other OpenAI services – users and reporters have seen connection errors and timeouts; some regions cannot access the service at all. [12]
- Letterboxd – the popular film logging and review site has shown Cloudflare-branded error pages or failed to load entirely for many visitors. [13]
- Downdetector – ironically, the service many people use to check whether sites are down has itself been affected, due to its own reliance on Cloudflare. [14]
- Canva and various productivity tools – reports indicate users struggling to log in or load assets in the browser. [15]
- Online games and gaming platforms – services such as League of Legends, Valorant and other multiplayer titles have experienced connectivity issues, lobby failures, or login problems in some regions. [16]
Outage dashboards and media coverage also show elevated problem reports for Spotify, AWS, Google services and other major platforms, though in some of these cases it is not yet clear whether Cloudflare is directly involved or whether users are simply encountering secondary network problems during the wider incident. [17]
In many cases, users are seeing:
- Cloudflare-branded error pages (such as 500, 502, or 504 errors)
- Endless security or “bot check” loops that never complete
- Blank pages or generic “Something went wrong” messages
What Cloudflare has said so far
Cloudflare has shared only limited information publicly, mostly through short status updates.
Across its status page and mirrored notifications, the company says it is: [18]
- Aware of an issue impacting multiple customers
- Seeing widespread 500 errors
- Experiencing failures in the Cloudflare Dashboard and API itself
- Working to understand the full impact and mitigate the problem
Mirage News and other outlets have reproduced Cloudflare’s wording, which confirms the global nature of the incident but stops short of naming a cause. [19]
As of late 18 November 2025:
- There is no official confirmation that this is the result of a cyberattack, DDoS campaign, or misconfiguration.
- No detailed technical post‑mortem has been published yet; Cloudflare typically releases those days after major incidents, not while they are still unfolding. [20]
Is the outage linked to Cloudflare’s scheduled maintenance?
Cloudflare’s status page shows a heavy schedule of maintenance across multiple data centers on 18 November 2025, including sites in Atlanta, Chicago, Miami, Buenos Aires, Quito, Santiago de los Caballeros and others. [21]
Some coverage notes that this maintenance was already underway when the global incident was declared, and speculates that there may be a connection—for example, a configuration change or routing issue that scaled beyond the intended scope. [22]
However:
- Cloudflare has not confirmed that scheduled maintenance is responsible for the outage.
- Public statements so far refer only to an issue impacting the “Global Network” and multiple customers, without tying it directly to any particular maintenance window. [23]
Until Cloudflare publishes a root-cause analysis, any explanation beyond this remains informed speculation.
How big is Cloudflare’s role in the internet – and why does this outage feel so huge?
Today’s disruption is drawing attention not only because big-name platforms are affected, but because it exposes how centralized parts of the modern web have become.
Cloudflare operates a vast network that: [24]
- Covers 330+ cities in over 120 countries
- Provides content delivery, DDoS protection, web application firewalling, and performance optimization
- Connects to more than 13,000 networks, including major ISPs, cloud providers and enterprises
- Advertises hundreds of terabits per second of edge capacity
Because many websites use Cloudflare for multiple layers of their stack—DNS, CDN, security and sometimes even zero‑trust access—a single systemic problem can:
- Prevent browsers from reaching the site at all
- Break login flows or security checks
- Cause an entire region’s traffic to time out, even if the origin servers are healthy
That is why users today are experiencing failures that feel like a complete site outage, even though the underlying application or cloud provider might not be having problems of its own.
How today’s Cloudflare outage is affecting users and businesses
For everyday users
Typical symptoms today include:
- Timelines and feeds that refuse to load on social networks like X
- ChatGPT sessions that never start or stall before generating replies
- Shopping sites, ticket portals and small business websites showing Cloudflare errors
- “Check if you’re human” pages looping indefinitely without letting users through [25]
Because Cloudflare sits between you and the website, there is almost nothing a normal user can do to “fix” this locally—restarting Wi‑Fi or clearing cookies will not resolve a Cloudflare network problem.
For companies and site operators
Organizations that rely on Cloudflare may be seeing:
- Increased error rates and failed requests from certain regions
- Inability to log into the Cloudflare dashboard or manage DNS and firewall rules
- Difficulty even filing support tickets if Cloudflare’s own support portal is degraded [26]
In some cases, businesses with multi-CDN strategies or backup DNS providers may be able to fail over traffic away from Cloudflare. But many smaller sites and SaaS platforms use Cloudflare as their primary (or only) edge provider and must wait for the company’s fix.
What can you do if Cloudflare is down for you?
While you cannot solve the underlying issue yourself, there are practical steps to take while the outage continues:
- Confirm it’s not just your connection
- Check a neutral source (such as independent news outlets or your ISP’s status page) to confirm the incident.
- Because Downdetector itself relies on Cloudflare and has been unstable today, treat its graphs with caution. [27]
- Avoid risky troubleshooting myths
- Repeatedly clearing cookies, altering DNS settings at random, or disabling security tools will not fix a Cloudflare‑side incident.
- Be especially cautious about any browser extensions or “fix tools” that claim to bypass Cloudflare errors.
- Look for official communication from the services you use
- Many platforms post real‑time updates on X, status pages, or support centers—ironically, those updates may also be affected if they sit behind Cloudflare, but often they are hosted separately.
- Delay sensitive transactions if pages are half‑loading
- If a checkout, payment, or login page is loading inconsistently, it’s safer to wait until the provider confirms the issue is resolved rather than risk duplicate payments or incomplete orders.
- For IT teams: review redundancy plans
- Large outages like this are a reminder to audit dependencies on single vendors (Cloudflare, AWS, other CDNs) and consider backup arrangements for critical workloads.
How does this compare to previous Cloudflare and cloud outages?
Major internet outages have become uncomfortably frequent in recent months:
- A significant AWS outage last month took down many popular services and is still being dissected by analysts and cloud providers. [28]
- Cloudflare itself has published several post‑mortems in 2025 describing incidents that impacted services like object storage (R2) and Zero Trust connectivity, typically resolving them within hours and then offering detailed technical breakdowns days later. [29]
Today’s event appears similar in scale of impact—once again illustrating how a relatively small number of infrastructure providers sit at the center of enormous volumes of internet traffic.
Many observers on industry forums have also pointed out that this is one of several “internet backbone” incidents in a short span of weeks, fueling renewed debate about resilience, over‑reliance on a few providers, and whether regulators should push for more diversification. [30]
When will Cloudflare be back up?
Right now, that is the question everyone is asking—but no clear public timeline has been given.
What we do know:
- Cloudflare engineers say they are actively investigating and mitigating the issue. [31]
- Some outage trackers and media outlets are seeing early signs that error rates are starting to improve in certain regions, although services are not yet fully restored everywhere. [32]
- Historically, Cloudflare has resolved major outages within hours and then released a technical post‑incident report—so a detailed explanation is likely to follow once the immediate disruption is under control. [33]
Until Cloudflare marks the incident as “Resolved” on its status page, users should expect intermittent problems when accessing Cloudflare‑backed sites, especially during peak traffic times.
Key takeaways
- Cloudflare is experiencing a major, confirmed global network outage today (18 November 2025), leading to widespread errors and timeouts across many websites and services. [34]
- High‑profile platforms including X, ChatGPT, OpenAI services, Letterboxd, Canva, Downdetector and several online games are among those affected. [35]
- Cloudflare has acknowledged widespread 500 errors and dashboard/API failures, but has not yet disclosed a root cause or full impact analysis. [36]
- Extensive scheduled maintenance was underway at several data centers today, but any direct link to the outage remains unconfirmed. [37]
- Users cannot fix the problem locally; the best course of action is to monitor official status updates, avoid risky troubleshooting, and delay critical transactions until services stabilize.
This article will remain accurate as a snapshot of the situation on 18 November 2025; readers should consult Cloudflare’s status page and official service announcements for the very latest updates.
References
1. www.techradar.com, 2. www.cloudflarestatus.com, 3. www.bleepingcomputer.com, 4. www.miragenews.com, 5. www.cloudflarestatus.com, 6. www.windowscentral.com, 7. www.techradar.com, 8. www.bleepingcomputer.com, 9. www.techradar.com, 10. www.miragenews.com, 11. www.techradar.com, 12. www.windowscentral.com, 13. www.techradar.com, 14. www.techradar.com, 15. www.windowscentral.com, 16. www.windowscentral.com, 17. www.bleepingcomputer.com, 18. www.miragenews.com, 19. www.miragenews.com, 20. blog.cloudflare.com, 21. www.cloudflarestatus.com, 22. dataconomy.com, 23. www.miragenews.com, 24. www.bleepingcomputer.com, 25. dataconomy.com, 26. www.bleepingcomputer.com, 27. dataconomy.com, 28. www.financialexpress.com, 29. blog.cloudflare.com, 30. www.reddit.com, 31. www.miragenews.com, 32. www.techradar.com, 33. blog.cloudflare.com, 34. www.cloudflarestatus.com, 35. www.techradar.com, 36. www.tomshardware.com, 37. www.cloudflarestatus.com
