Several Microsoft cloud customers experienced service disruptions on Wednesday after what the company described as an “inadvertent configuration change” hit its Azure platform. The issue disrupted Azure Front Door — Microsoft’s content delivery network that helps businesses boost application speed and performance.

The problem started around 1600 GMT and caused outages across multiple regions. Microsoft updated clients at 2230 GMT, confirming that it had rolled back to its “last known good” configuration. However, the company warned that “some users may experience intermittent failures” as systems recover.

Widespread Impact on Major Websites

The outage affected several popular consumer websites. According to DownDetector, the disruption impacted Xbox, Alaska Airlines, and Costco, among others.

System configuration updates are a common part of cloud operations. Companies routinely make these changes to improve performance, fix bugs, or add new features. However, even a minor mistake can trigger widespread problems due to the interconnected nature of global cloud networks.

Microsoft explained, “We are currently recovering nodes and re-routing traffic through healthy nodes across our fleet. This recovery effort involves reloading configurations and rebalancing traffic across a large number of nodes to restore full operational scale.”

Similar Amazon Outage Last Week

Just a week earlier, Amazon Web Services (AWS) suffered a large-scale outage that took down several major online services. Platforms such as Prime Video, Disney+, Perplexity AI, Fortnite, Airbnb, Snapchat, and Duolingo were hit.

The outage even affected mobile services and messaging apps like Signal and WhatsApp in parts of Europe, according to DownDetector. Users also faced difficulties accessing Amazon’s e-commerce website.

Several banks, including Lloyd’s, also reported problems and identified AWS as the source of the issue.

Amazon later announced that its systems had returned to “pre-event levels”, though it cautioned that clearing the data backlog caused by the outage would take several hours.

A Reminder of Global Dependence on Cloud Services

The back-to-back outages at Microsoft and Amazon underline how much the digital world depends on a few cloud providers. Even a small glitch in configuration or infrastructure can cause ripple effects across multiple sectors — from banking and travel to gaming and streaming.