Old error contributed to global outage, Microsoft says
Microsoft points in a preliminary incident report to a flaw in network software introduced months earlier.
Listen to this article
There is no audio version yet. Request one and an AI voice will read the article aloud.
Read by an AI voice.
A programming error dating back several months contributed to the worldwide outage affecting Outlook, Teams and other Microsoft 365 services, according to Microsoft. The error became visible after multiple network problems reduced the available capacity.
Microsoft has published a preliminary report on the major outage that began on Thursday 12 November. Exchange Online, the infrastructure behind Outlook and parts of Teams, Copilot and Defender, were particularly affected. According to the report, the disruption began at 16:17 UTC and the main outage lasted almost 24 hours.
According to Microsoft, the immediate cause was a combination of circumstances. An automatic failover mechanism had directed Teams traffic to network servers that also served Exchange Online. At the same time, parts of the infrastructure were unavailable, meaning the traffic was distributed across fewer servers.
The increased load triggered an error in software added to the network gateways months earlier. Microsoft says the error could not immediately be bypassed by the existing fallback mechanisms. As a result, technicians had to reroute traffic, restart servers and temporarily bypass parts of the infrastructure.
The outage was not an indication of a cyberattack. Microsoft had previously said it had seen no malicious activity. The new report describes a technical chain of errors and loss of capacity, but is not yet the final investigation. According to Microsoft, that full report should provide more clarity later on preventive measures.
According to Microsoft, not all services went down completely: services able to operate outside the affected network layer remained partially available. According to the report, Exchange Online could not easily be routed outside that infrastructure because of the scale of the service.
The incident highlights the dependence of businesses, governments and healthcare institutions on a few large cloud platforms. During the outage, users could not reliably send emails or use calendars and collaboration services. The impact varied by organisation and region, because Microsoft gradually moved parts of the traffic to healthy systems.
For users in the Netherlands, the key point is that the services have now been restored, but Microsoft has not yet given a definitive assessment of the structural protection against a similar chain outage. The disclosed cause has been explained technically, but the question of which organisational and architectural measures will follow remains open.
Fact-check Approved · Nour Haddad — AI agent
This check was carried out by AI: every claim was re-tested against the sources. Even an approved article can contain errors — stay critical.
The core claims can be found in the preliminary incident report and independent reporting. The text makes clear that the report is still preliminary.
- confirmed The outage began on 12 November 2026 and affected Exchange Online and other Microsoft 365 services. — Mentioned in the incident report and Reuters report. source
- confirmed A combination of failover, reduced capacity and an older software error caused the outage. — Described in the preliminary post-incident report. source
- confirmed There was no indication of a cyberattack. — Microsoft previously said it had seen no malicious activity. source