Microsoft says Teams, Outlook and Copilot fully restored after 24-hour global outage

What happened

Microsoft's outage began in the afternoon, Greek time, and lasted more than 24 hours. The problem started in the United States and reached 74 countries within about an hour, according to the NewsIT count, while other outlets put the figure at more than 70. Downdetector, the outage tracker, logged reports from users in more than 70 countries, with the number of reports running into the hundreds of thousands. More than 20 Microsoft products and services were affected, including Outlook, Teams and the Copilot AI assistant. Microsoft has said the disruption came from a technical fault and not from a cyberattack.

Cause and repair work

According to a message Microsoft sent to affected customers, the company had upgraded its network gateways some months earlier to increase capacity, and the error had caused no known problems in the months before. A network change then triggered the fault, which set off a chain reaction of failures and extra load on one system. More than 1,000 engineers and other staff joined video calls under the leadership of Scott Guthrie, who heads the company's cloud services, and Charles Lamanna, who oversees Microsoft 365 and other work applications. Engineers first tried to bring servers back, but the sudden return of millions of users knocked them offline again. To break that cycle, they throttled data flow by region, rolled back earlier software versions and reconnected users slowly, checking stability at each stage. Microsoft has said it will publish a full report once its required checks are complete.

Consumer and business recovery

Microsoft said consumer services were fully restored at 16:10 GMT on the day of the announcement. Its business suite, Microsoft 365, was still recovering, and the company is monitoring the volume of emails that piled up during the outage before it confirms that enterprise services are fully back. Microsoft said it does not yet consider the problem fully resolved, while describing progress as encouraging. The company said it did not know when full service would return.

Impact on governments, health and transport

In Britain the outage reached the National Health Service, police and schools. Police there could not take reports over the internet and urged people to call their phone lines instead. A court in Manchester postponed a trial, with the judge citing the risks of relying on cutting-edge technology. In Japan, communications between government ministries were disrupted, and in Australia systems at security services were affected, including those of the country's intelligence agency. The Philippine Supreme Court, rail services, hospitals and courts in North America were also hit. Germany's government estimated the economic cost would be heavy because millions of working hours were lost, even though state institutions saw no widespread problems. In Greece, rail passengers could not reach their electronic tickets because their email was inaccessible.

Scale and context

The company had earlier said there were no signs of a cyberattack and attributed the problem to server malfunctions. Microsoft had announced in January more than 450 million Microsoft 365 subscriptions, and NEWS 24/7 put the number of people worldwide using the affected products at more than 430 million. in.gr compared the scale of the outage with the CrowdStrike incident of 19 July 2024. The outage has intensified concern over dependence on cloud-hosted services.