A significant global internet disruption occurred on Wednesday, affecting a wide range of websites and online services, including major platforms such as Heathrow Airport, NatWest bank, and the popular game Minecraft. The outage, linked to Microsoft’s cloud services, caused widespread inconvenience for users across the UK, the US, and other regions, highlighting the vulnerability of modern internet infrastructure dependent on a handful of large cloud providers.
The problems began to surface in the afternoon, with Microsoft’s Azure cloud computing platform—an essential backbone for many internet services—reporting a “degradation of some services” around 16:00 GMT. The root cause was identified as “DNS issues,” a technical term relating to the system that translates website names into IP addresses. This was notably the same underlying problem that led to a massive Amazon Web Services (AWS) outage just the previous week. However, Amazon confirmed that its AWS platform was operating normally during this incident.
Downdetector, a popular outage tracking website, recorded thousands of reports from users experiencing difficulties accessing numerous websites worldwide. In the UK, alongside Heathrow Airport’s website, supermarket chain Asda and mobile operator O2 were among those affected. Meanwhile, in the United States, customers reported problems with the websites of Starbucks and Kroger, two major retail chains. Some of the most notable disruptions involved Microsoft’s own services, including Microsoft 365, where users encountered delays and error messages, particularly with Outlook email. At one point, Microsoft’s service status page itself was inaccessible, forcing the company to post updates on the social media platform X (formerly Twitter).
By 21:00 GMT, Microsoft had rolled back a recent update that was believed to have caused the issues, allowing many affected websites to return to normal operation. However, not all services recovered immediately; for example, the Marks & Spencer (M&S) website remained offline late into the evening. Microsoft acknowledged that the outage stemmed from an “inadvertent configuration change,” meaning that a modification to the system was made unintentionally and triggered widespread disruptions.
The impact extended beyond commercial websites to public institutions. In Scotland, the Scottish Parliament was forced to suspend its business due to technical problems with its online voting system. This disruption led to the postponement of a crucial debate on land reform legislation, which could have allowed the government to intervene in private land sales and mandate the breakup of large estates. A senior parliamentary source indicated to the BBC that these technical issues were likely connected to the Microsoft outage.
While the outage was disruptive, some essential services remained operational. For example, although NatWest’s website was temporarily down, its mobile banking app, web chat, and telephone customer support continued to function. The UK consumer rights organization Which? emphasized the importance of businesses keeping customers informed and supported throughout such incidents. Lisa Webb, a consumer law expert from Which?, advised customers to retain records of any failed or delayed payments and to proactively contact companies to avoid penalties, such as late fees, due to the service interruptions.
This latest outage once again exposed the risks inherent in the heavy reliance on a small number of cloud service providers like Microsoft, Amazon, and Google. Microsoft Azure alone is estimated to hold about 20% of the global cloud market, meaning that technical issues within its infrastructure can have far-reaching consequences. Dr. Saqib Kakvi, a researcher at Royal Holloway University, pointed out that economic pressures have driven the consolidation of web hosting services into just a few major players. While this centralization offers cost benefits, it also means that a single failure can disrupt hundreds or even thousands of applications and systems simultaneously.
Experts agree that the complexity of modern cloud services contributes to these vulnerabilities. Gregory Falco, an engineering professor at Cornell University, described platforms like Azure and AWS not as monolithic systems but as vast assemblages of thousands of interconnected components. Some of these components are managed directly by the cloud providers, while others are maintained by third-party vendors. Falco cited an example from last year, when a software update from cybersecurity firm CrowdStrike affected over eight million computers running Microsoft systems, illustrating how external factors can also influence the stability of cloud services.
The recent spate of outages, including this Microsoft incident and the prior AWS disruption, has sparked broader conversations about the fragility of today’s internet infrastructure. The dependence on a few giants for critical services means that technical errors or configuration mistakes can cascade into widespread service interruptions, affecting millions of users worldwide. As cloud technologies continue to underpin everything from online shopping and banking