What happened
Yesterday morning there was a significant global outage of Cloudflare's services, which affected many major websites and applications such as ChatGPT, X (formerly Twitter), Canva and others.
According to Cloudflare, the main cause was a sudden spike in unusual traffic to one of its services, which triggered internal errors in the network's traffic management system.
In particular, the company acknowledged that an automatically generated configuration file (used to manage traffic related to threats) "grew beyond its expected size and caused the software system that handles several Cloudflare services to crash".
This meant that part of the web traffic routed through Cloudflare was not served correctly, causing errors and outages: that is why our website was also offline or unreachable for some users.
Impact on our company and our official website
Because our official website relies on Cloudflare's services for protection, content delivery and performance optimisation, the outage meant that:
- some users could not reach the site or saw "Internal Server Error" messages or a blank page;
- the site's loading speed and availability were affected for a while;
- we activated internal procedures to monitor the situation, inform our clients and users about the issue and make sure there were no further problems on our own servers.
What we did and what we will do
- We checked our internal system logs and confirmed that the problem did not lie with our hosting or CMS, but came from the external Cloudflare services in between.
- We informed our users and clients through our social channels and newsletter, apologising for the inconvenience and explaining the external nature of the problem.
- Together with our IT team, we are assessing contingency plans (for example a fallback service and active monitoring) to reduce the impact of any future outages of external infrastructure.
What Cloudflare is for and how it works
What Cloudflare is
Cloudflare is a web infrastructure platform providing services such as a Content Delivery Network (CDN), Web Application Firewall (WAF), DDoS protection, DNS, site speed optimisation, content caching, traffic routing and much more.
In practice, many companies and websites rely on Cloudflare to improve performance, increase resilience to cyber attacks and deliver content as efficiently as possible.
How it works, in broad terms
- When a user visits your website, part of the traffic is routed through Cloudflare's network of "edge" servers, geographically close to the user. This makes it possible to serve content faster and reduce latency.
- Cloudflare intercepts potentially dangerous requests (for example DDoS attacks, bots, malicious requests) with its WAF and other security systems, protecting the site from vulnerabilities and overloads.
- It caches static content (images, CSS/JS files, etc.), reducing the load on the origin server and improving scalability.
- It also manages DNS records and routing, and can act as a reverse proxy: the public IP address of your site can be Cloudflare's, while the real server is "protected" behind this infrastructure.
Why it matters to us
In our business, the decision to rely on Cloudflare was driven by:
- the need for high availability of our official website, as a communication tool and company showcase;
- the wish to offer fast performance to visitors wherever they are;
- the need for protection against cyber attacks and overloads, given the sensitivity of our content and our company's reputation.
However, as yesterday's events show, even the most robust infrastructure is not immune to failures: it is essential to be prepared.
Conclusion
Yesterday's outage was not caused by a targeted external attack on our site, but by a malfunction of Cloudflare's network due to an unexpected traffic spike and the overload of an internal system. Although we were not directly responsible, we recognise the impact it had on our users' experience and we are committed to improving our site's resilience. Thank you all for your patience; we remain at your service.
