Microsoft’s Outage Outlasted Its Fix: Recovery Came in Phases, With Backlogs Still Draining
Microsoft says it addressed the underlying issue by Friday afternoon, yet was still restoring customers and watching a backlog on Saturday.
Microsoft 365 broke for much of the working world on Thursday, November 12, and took more than a day to come back. Outlook, Teams, and Copilot all stumbled, exposing the reliance on a single SaaS platform.
Microsoft’s explanation names a process that sets up secure connections that got overwhelmed, and servers stopped answering.
What’s less clear is why recovery came in phases, and why a backlog was still draining after the company said the underlying problem was dealt with. That gap is where businesses felt it.
Specific about the mechanism, vague about the trigger
ABC7’s live blog carries Microsoft’s Thursday-night finding:”a process on the affected servers responsible for establishing secure connections becomes overwhelmed,” which made servers stop processing requests.
By early Friday GMT, Microsoft said it had begun reverting a recent network configuration change and was limiting traffic in certain locations.
Fox Business reports Microsoft tied the outage to a bug introduced during a network capacity upgrade, though its subheadline says no root cause has been determined.
Microsoft rules out a cyberattack and says AI agents weren’t involved.
Why “addressed” didn’t mean mail was flowing
Microsoft reportedly said in an update that it was using “a measured approach to identify and repair affected infrastructure.”
Later it said recovery had progressed in phases and that capacity hadn’t returned to baseline. Hours after announcing restoration, it updated its list of affected services and said work continued.
Recovery arrived region by region. Europe and Asia Pacific improved first, with Japan comparatively stable, and North America followed with intermittent problems still possible, Fox Business adds.
By Friday afternoon, Microsoft said it had addressed the underlying issue while it restored the remaining customers and regions.
Microsoft also extended its email timeout to 48 hours, so delayed messages could reach recipients until at least Saturday afternoon, per Fox Business.
ABC7’s last entry, early Saturday GMT, still had Microsoft monitoring a backlog.
It means a status page saying “restored,” and your inbox catching up are different events, and the sources don’t say which customers got which first.
What the workarounds show about who paid
The disruption forced critical operations backward overnight. In Taipei, foreign banks and businesses rolled out fax machines, per Al Jazeera. Euronext instructed companies to phone in trades, and the UK’s NHS temporarily reverted to paper records.
Beyond email, the outage crippled Teams, SharePoint, and Intune, while even AI leaders like OpenAI and Anthropic reported friction using Microsoft tools.
Microsoft hasn’t said how many customers were affected. It had more than 450 million commercial Microsoft 365 subscribers in January.
Its shares were also mostly flat Friday, per Al Jazeera, which means the market shrugged while customers improvised.
Source: Microsoft outage: What to know as service was restored
