Microsoft Explains Root Cause of Massive Outlook and Exchange Outage, Reveals Three Failures Behind Global Meltdown


microsoft x account hacked
Image credit: Microsoft

Microsoft has published a detailed explanation of what caused its massive November 12 outage. The company says a hidden software defect, increased network traffic, and reduced server capacity combined to trigger a global failure. At the worst point, more than 80% of its networking gateway servers became unresponsive.

The incident affected several Microsoft 365 services, including Exchange Online, Teams, OneDrive, SharePoint, and Office apps. Microsoft has now explained why the disruption spread so quickly and why restoring services took much longer than expected.

Three issues combined to trigger Microsoft 365 global outage

In a detailed post on X, Charles Lamanna, EVP Copilot, Agents, and Platforms at Microsoft, noted about three main factors behind the incident. The first was a latent software defect inside an internal networking gateway process. Under certain traffic conditions, lock contention caused servers to become overwhelmed and stop responding.

The defect came from a fix prepared specifically for Microsoft 365 networking gateway servers. Microsoft says it did not affect broader Windows deployments. The change first started rolling out in July 2026 and reached global deployment over several weeks. No problems appeared during the following three months.

Interestingly, Microsoft’s normal load testing failed to catch the defect. Its test labs used different traffic patterns and software configurations. Testing also focused more heavily on older hardware with fewer processor cores. The problem became worse on newer machines with more cores.

The second factor was a sudden increase in network gateway traffic. Around 15 hours before the outage, Teams failed over from Microsoft’s alternative gateway to its primary gateway. The primary system also supports Exchange Online, increasing its workload and changing traffic patterns.

The third factor was reduced server capacity. Nearly 14% of networking gateway capacity was already unavailable because of an unrelated issue from the previous day. Microsoft says this reduction alone did not affect service because sufficient capacity remained.

How the outage spread across Microsoft’s network

These three conditions combined during peak US traffic hours. The increased workload and reduced capacity exposed the hidden software defect, causing requests to pile up until servers stopped responding.

Microsoft’s traffic management systems then redirected users to healthy endpoints. Normally, this protects services from individual server failures. This time, however, redirected traffic overloaded the remaining servers and triggered a cascading failure.

Within less than two hours, the defect had spread across Microsoft’s entire gateway fleet. At its lowest point, more than 80% of servers were unresponsive.

Microsoft also discovered that its usual isolation measures failed to contain the disruption. The gateway normally uses thousands of capacity units organized into separate regions. A separate issue prevented the usual barriers between these partitions from stopping the cascading failure.

This allowed the outage to spread globally instead of remaining contained within individual regions.

Why recovery took longer than expected

Services such as Teams, OneDrive, SharePoint, and Office apps could fail over from the faulty gateway infrastructure. That helped restore several services earlier, while Teams chats and meetings continued working for most of the outage.

Exchange Online could not fail over in the same way, which explains why email access remained affected for longer. Recovery also created another problem. Servers returning online were immediately overwhelmed by waiting traffic and became unresponsive again. Microsoft’s automated recovery processes were not designed for such widespread server failures.

The company eventually identified the defect and prepared an update to restore the previously working code. Engineers took groups of servers offline, applied the fix, restarted them, and gradually returned them to service.

Microsoft says its final Post-Incident Review will provide more details later this week. The findings also explain why the outage lasted so long despite Microsoft’s network redundancy and recovery systems.

More about the topics: microsoft, Microsoft Exchange, Outlook

Readers help support Windows Report. We may get a commission if you buy through our links. Tooltip Icon

Read our disclosure page to find out how can you help Windows Report sustain the editorial team. Read more

User forum

0 messages