Restarted servers kept falling over

The trouble started in Exchange Online, the email service at the centre of Microsoft 365, when users in the Americas could not open their mailboxes. Within about two hours, Microsoft had widened the incident to every service that relies on Exchange, including Teams, SharePoint, Microsoft 365 Copilot and its Defender XDR security tools. Consumer apps such as Outlook.com, OneDrive and OneNote were also hit. Microsoft first blamed elevated load, then corrected itself: traffic was normal, and the strain came from servers that had dropped out, leaving a core process on the rest overwhelmed.

Engineers paused all deployments, moved traffic and restarted servers, yet machines brought back online failed again once traffic returned. By Friday morning, Microsoft said its fix was “not taking as expected”. It has since added virtual machines, rebuilt some servers on an earlier configuration and restricted certain types of traffic in some locations, which explains why recovery is patchy. Microsoft executive Charles Lamanna said there was "no indication this is the result of malicious or cyber activity".

From inboxes to police stations

Because so much daily work runs through Exchange, the effects reached well beyond email. Government offices in Japan switched to text messages, and economy minister Ryosei Akazawa said the disruption had a “major impact”. In Australia, the spy agency and federal police were affected. Several UK police forces asked the public to report crimes by phone instead of online. In India, airlines reported problems with their email support. OpenAI and Anthropic also faced issues with their Microsoft 365 connectors, which allow their AI assistants to access users’ emails and files.

Outages are adding up for Microsoft

Microsoft 365 experienced an outage lasting nearly ten hours in January. Its Azure cloud platform also suffered two outages within about 40 hours, between 29 September and 1 October. Notably, Microsoft promises 99.9% monthly uptime for Exchange Online, Teams and SharePoint under its service level agreement.

Microsoft has promised to publish a report explaining the cause of the latest outage once services are restored. But the incident raises a more immediate concern for IT leaders: when email and Teams stop working together, employees lose their main channels for coordinating a response. Businesses should have an alternative communication channel ready before the next outage.