On 5 December 2025, the internet once again suffered serious disruption caused by problems with Cloudflare services. Users around the world are reporting mass 500 Internal Server Error messages that prevent access to many popular websites and applications. It is another outage for the giant in the security and network optimisation space in a short period of time, raising questions about the stability of critical digital infrastructure.
What happened?
Cloudflare, a company providing CDN (Content Delivery Network) services, protection against DDoS attacks and website optimisation, confirmed a degradation of its services. The problem began around 4:00 a.m. UTC (5:00 a.m. Polish time), when numerous servers started returning a 500 error, meaning an internal server error. Users trying to access sites protected by Cloudflare encountered white screens with error messages, making normal use of services impossible.
According to reports, the outage is linked to backend problems, similar to those in previous incidents. Cloudflare’s status page indicates an active investigation into problems with its Dashboard and API, suggesting the source lies within the company’s internal infrastructure. Interestingly, even the popular outage monitoring service Downdetector.com fell victim to the problem: the Cloudflare status page on Downdetector was not working, further complicating verification of the scale of the event.
Scale of impact
The outage affected millions of users worldwide, paralysing access to key platforms. Among those affected were, among others:
- Trading platforms such as Zerodha and Groww, where traders could not log into accounts or carry out transactions.
- Creative applications such as Canva, where users reported 500 errors while editing projects.
- Cryptocurrency and NFT services, including Jupiter, Meteora, OpenSea and Raydium, CoingGecko, which halted digital asset trading.
- Other services, such as Perplexity, Claude and authorisation systems, which froze due to their dependence on Cloudflare.
Reports indicate that the problem affected at least a dozen or so applications and websites, highlighting how heavily the modern internet depends on the services of a single provider. As one user noted on the X platform: „Awaria Cloudflare zabiera ze sobą połowę internetu” (original: „Cloudflare going down takes half the internet with it”), which captures the scale of this dependence perfectly.
Reactions from users and companies
On the X platform (formerly Twitter), a flood of reports came from frustrated users. One developer wrote: „Cloudflare padł 5 grudnia 2025 😭🤯🤯 Doświadczam błędu 500 Internal Server Error na wielu stronach, w tym Canva i moim własnym projekcie” (original: „Cloudflare down as on 05 Dec 2025 😭🤯🤯 Experiencing 500 Internal Server Error on many website including canva & on my own project”). Another compared the situation to a „technicznego odpowiednika 'kapitan włączył znak zapnij pasy’, ale samolot już wpada w spiralę śmierci” (original: „tech equivalent of 'the captain has turned on the fasten seatbelt sign’ but the plane is already in a death spiral”).
Affected companies such as Groww were quick to respond, informing customers: „We are currently experiencing technical issues due to a global outage at Cloudflare. This is impacting multiple apps and services worldwide” (original: „We are currently experiencing technical issues due to a global outage at Cloudflare. This is impacting multiple apps and services worldwide”). Cloudflare itself is monitoring the situation and promises updates, although for now it has not given the exact cause or a time for repairs.
History of Cloudflare outages
This is not the first time Cloudflare has caused global disruption. Just on 18 November 2025, the company experienced a similar incident caused by a bug in the logic generating bot management features, which knocked out around 20% of global internet traffic. Earlier outages, including those from 2024, often resulted from software bugs or scheduled maintenance that went wrong, as in today’s event, which coincided with planned work at data centres in Chicago and Minneapolis.
Experts stress that in an era of AI and decentralisation, such „choke points” as Cloudflare pose a risk to the entire network. As one commentator wrote: „Szalone przypomnienie, że cały boom AI nadal opiera się na kilku gigantycznych wąskich gardłach” (original: „Wild reminder that the entire AI boom is still built on a few giant choke points”).
Summary
The Cloudflare outage of 5 December 2025 shows how fragile digital infrastructure is. Millions of users are waiting for services to be restored, and businesses are counting their losses. Cloudflare promises a swift fix, but the incident may prompt discussion about diversifying network service providers. Follow the official status pages to stay up to date.
