Microsoft 365 outage stretches into second day as Exchange Online recovery continues

Microsoft is still working to fully restore Microsoft 365 after an Exchange Online problem stretched into its second day, leaving some business users without reliable access to Outlook email and disrupting a much wider set of services, from Teams to SharePoint.

According to TechCrunch, the incident began with an issue in Exchange Online that caused email delays, delivery failures and authentication problems for a portion of Outlook users. Microsoft made significant progress on Monday, but as of Tuesday, September 1, the company had still stopped short of declaring the outage fully resolved.

What actually broke

Microsoft's status page for business and enterprise customers attributed the disruption to an issue with a "core authentication configuration used by multiple Microsoft 365 services," TechCrunch reports. That shared dependency explains the unusually broad blast radius: alongside Exchange Online and Outlook, the outage touched SharePoint, Copilot, Teams, Purview, Defender XDR, the Microsoft 365 Admin Center and Universal Print.

Microsoft later shared on X that its investigation had traced the trouble to a misconfiguration that kept authentication components from "deploying as expected to a portion of infrastructure." In practice, that meant the fix was not a simple matter of restarting a service — Microsoft had to get a broken deployment rolled out correctly across the affected systems.

The company first reviewed recent changes that could have introduced the problem, then tested and implemented a series of mitigations. As those rolled out, Microsoft told Outlook users to expect a "gradual recovery" of mail flow rather than an instant return to normal.

A staged recovery

The restoration played out in stages. By 10 p.m. ET on Monday, Microsoft reported that mail flow was improving, although search functionality in Outlook remained degraded. Shortly after 3 a.m. ET on Tuesday, the company said search was showing improvement as well.

Microsoft's most recent update on X, cited by TechCrunch, said mitigation actions were "continuing to progress within the remaining affected infrastructure," that telemetry indications remained positive and that service availability was improving. The company added that it was entering a period of extended monitoring to confirm a full resolution.

The wording matters. Many users are likely seeing normal behavior again by now, but Microsoft has not said everything is back to baseline — a distinction IT administrators will weigh when deciding whether their own escalations can stand down.

Why it matters

Email remains core infrastructure for most organizations, and a multi-day Exchange Online disruption is more than an inconvenience: delayed or failed messages stall deals, support queues and internal approvals. The incident also illustrates the structural risk of consolidated cloud platforms. Because a single authentication layer is shared across Microsoft 365, one misconfiguration cascaded into an outage spanning email, collaboration, security tooling and even the admin console administrators would normally use to investigate.

It is also a reminder of how recovery from large cloud incidents tends to unfold. Mitigations roll out gradually across affected infrastructure, so users experience uneven restoration — mail flowing while search still lags — and vendors communicate in hedged, incremental status updates rather than a clean all-clear.

For IT teams, the practical lessons are familiar but reinforced: keep fallback channels for critical communication, monitor vendor status feeds directly rather than relying on user reports, and treat "improving" as meaningfully different from "resolved."