Skip to the content
AI and Cybersecurity Learning and Practice Center Test automation Open the Learning Center

NewsAnalysis

When the cloud stumbles: an overwhelmed process on Exchange servers took Microsoft 365 down, an engineering failure rather than an attack

Root cause, as far as Microsoft has said: a core process on the affected Exchange servers, the one responsible for establishing secure connections, became overwhelmed, the servers failed their automated health checks and dropped out, and the rest were strained by the load they left behind. Microsoft first blamed elevated load, then said traffic had not grown; it said it found no recent change matching the start, while looking at a few configuration updates as possible contributors, and at first it did not say what overwhelmed the process. It called this an investigation, not a final finding, and the final post-incident report with the root cause had not been published as of November 13. On November 13 it said it was not aware of any malicious activity causing the incident. The fix that worked was a different way of checking server health, which let servers return to service, with automated restarts of those that stopped responding; its first attempts, restarts and a configuration change, did not hold, and it limited which traffic reached recovering servers. At 13:29 UTC on November 13 Microsoft reported encouraging progress, with broader improvement in North America and sustained improvement across much of Europe, but capacity was not back to normal and some intermittent problems could continue; it extended the timeout for delayed email to 48 hours. Later on November 13, MacRumors reports, Microsoft said the issue had been fixed, with broad recovery and stability, after nearly 24 hours, though its status page still said some services might be unavailable. Microsoft's words: "This was a technical issue not related to cyber or AI / agents. The incident was introduced by an issue during an earlier network capacity upgrade", and it promised a full post-incident report once its review is complete. Problems began around 16:00 UTC on November 12 in the Americas and reached any region, Al Jazeera reports; Charles Lamanna of Microsoft said there was no indication of malicious or cyber activity, and on November 13 Microsoft was rolling back a network change without saying whether it was to blame. Government services in Singapore, Japan and Australia were disrupted. Microsoft's status notices name Exchange Online, Teams, SharePoint Online, Microsoft 365 Copilot, Defender XDR, Defender for Office 365, Planner, To Do and Intune. Outage reports for Outlook on Downdetector peaked at about 130,000.

Why it matters

One overwhelmed process took out mail, chat, files, the AI assistant and two security products at once, along with the page that tells administrators what is happening, for most of a day. Check that your team has a second channel, a way to read vendor status that does not depend on the vendor, and agents that fail safely when the service behind them is gone. Expect delayed mail to keep arriving. LazyAdmin notes that service credits under the 99.9 percent uptime commitment are not automatic: they are claimed through a service request in the admin center, or through the partner for licences bought from one.

Sources

Read it with the rest of the news in the Learning Center