A massive Microsoft 365 outage that started Thursday at around 18:00 UTC has stretched beyond 20 hours, leaving organizations unable to access email, attend Teams meetings, use cloud PCs, or consult Copilot, according to a report published by The Register on November 13, 2026. The disruption has been so pervasive that in Australia, it's being called the "Microsoft long weekend" — an outage severe enough to render Friday essentially a write-off for companies dependent on Redmond's cloud services. The software giant's repeated attempts to fix the problem have failed to restore service, and the company still has no estimated time for full recovery.
The outage affects Microsoft 365 services that interact with Exchange Online and email services, with roughly 15 hosted services experiencing disruptions, while services that don't require Exchange Online remain operational. As of 12:00 PM UTC on Friday, users were still unable to connect to multiple Microsoft 365 services. By 1300 UTC, the company reported continued recovery progress across affected regions, with broader improvements observed in North America and sustained improvements across much of Europe, though some users in Asia Pacific continued to experience intermittent access issues or delayed email delivery. Microsoft's latest update at 2144 UTC indicated that infrastructure was stabilizing, with the company draining queued messages and rebalancing traffic, though service access could recover before mail flow fully normalizes.
Microsoft initially attributed the outage to a "problem that's causing servers to fail automated health checks and become unavailable," and the company believed it had found a fix but quickly acknowledged that "Actions taken thus far have not provided the anticipated relief." The company's status feed reveals a litany of failed remediation attempts: scaling up virtual machines encountered issues with machines becoming stuck in a bad state, preventing changes from syncing as expected; traffic flow optimizations and capacity improvements failed to result in sustained recovery; and changes applied to improve serviceability weren't taking as expected. Microsoft has also shifted its language regarding the possibility of a cyberattack, moving from stating at 04:11 UTC that it could find "no indication this is the result of malicious activity" to saying at 05:46 UTC that "As of now, we are not aware of any malicious activity that is causing this incident" — wording that hasn't completely ruled out a cyberattack as the cause.
According to the report, Microsoft is taking a measured approach to identify and repair affected infrastructure while continuing to pursue multiple workstreams to pinpoint the underlying cause of the traffic influx and advance the recoverability of backend servers. The company's challenges stem from what it describes as sustained traffic management imbalances and servers showing failures that require specific actions to improve stability. The software giant is tracking results to verify that servers are showing the anticipated recovery and performance intended, though no timeline for complete restoration has been provided. The scale and duration of this outage underscore the fragility of centralized cloud dependencies, particularly when automated failover mechanisms themselves become part of the failure cascade. Organizations relying solely on a single vendor's cloud ecosystem may need to reconsider whether the convenience of integration justifies the risk of synchronized collapse across critical business functions.

