Microsoft Global Outage: What We Know About Fixes and Lingering Disruptions

Services Remain Disrupted

The outage began around midday Thursday and affected Microsoft email, cloud and AI services. Microsoft documented the affected services and said recovery efforts were still underway Friday morning.

The Microsoft 365 Status account said the company had extended email timeout to 48 hours to help preserve mail delayed during the disruption. Microsoft also warned that intermittent issues could continue as traffic returned.

Public Services and Businesses Hit

Businesses, hospitals, police departments and schools worked around the outage. In Japan, government IT infrastructure was disrupted, deputy chief cabinet secretary Masanao Ozaki said. NBC News reported that some doctors could not send patients updates about medication changes or follow-up care.

The number of affected users is unclear. Microsoft estimated in January that paid Microsoft 365 commercial subscriptions exceeded 450 million.

Why Recovery Took Time

Microsoft said a portion of its infrastructure was experiencing high load across multiple regions, but said increased traffic was not the cause. The company attributed the problem to servers failing automated health checks and becoming unavailable.

Microsoft tried to prevent more server outages and validate alternate health checks. Four hours later, it said an overwhelmed core process was causing capacity loss, and described isolating infrastructure, restarting components and gradually reintroducing traffic to test stability.

Some servers became unresponsive again after receiving traffic, creating another obstacle. Microsoft said its earlier actions had not brought the expected relief and that it was pursuing multiple ways to reduce the impact.

Microsoft Promises Further Review

Microsoft said it had no evidence of malicious activity connected to the incident. It promised a full post-incident report covering the root cause and additional defensive measures after mitigation is complete. Ciaran Martin, the former head of the UK’s National Cyber Security Centre, told NBC News the disruption appeared to rank among the biggest outages ever, and suggested it could be worse than comparable incidents involving Amazon Web Services and CrowdStrike.