Microsoft says 365 outage fix is 'not taking as expected'

Microsoft told customers on Friday that its efforts to repair a global Microsoft 365 outage were 'not taking as expected,' as Outlook, Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender XDR remained patchy more than a day after the incident began. The failure started around 18:00 UTC on November 12, per the BBC's live blog, with Downdetector logging at least 43,000 reports of Outlook problems across 72 countries.

The company said the cause was 'a fault introduced during an earlier upgrade to its network capacity' that surfaced after a more recent network change, and that 'there is no indication this is the result of malicious or cyber activity.' Microsoft also said the incident was not related to AI agents. An initial fix using an alternative service health check method brought a subset of affected servers back online, but failed under load once traffic returned to them.

Microsoft 365 Copilot was out alongside Exchange and Teams for the duration. By late Friday the company said it was seeing 'encouraging recovery progress,' and at 17:50 UTC on 13 November, about a day after the outage began, it reported broad recovery and said the underlying issue was fixed.