Microsoft is working to restore several of its major services, including Outlook, Microsoft Teams and Microsoft 365, after a widespread outage disrupted access for thousands of users across multiple regions. The company acknowledged that its recovery efforts were not progressing as expected, with affected servers continuing to become unresponsive even after being restarted.
The disruption also affected Microsoft 365 connectors used by artificial intelligence platforms OpenAI and Anthropic, raising concerns about the wider impact of the service interruption.
Microsoft said it had paused deployments, redirected additional traffic away from the affected infrastructure and begun restarting impacted servers. However, the company indicated that the recovery process was proving more complicated than anticipated.
“We've monitored the status of the changes applied to improve serviceability and confirmed they're not taking as expected,” Microsoft wrote on X in an update on Friday.
The company added that it was continuing to assess the effectiveness of its interventions and identify further measures to stabilise services across affected regions.
ALSO READ: Microsoft Rolls Out Voluntary Exit Scheme For Employees In India, US, UK And Canada
Microsoft Outage: What Went Wrong And How The Company Is Responding
Microsoft's initial response involved stopping all ongoing deployments and diverting additional traffic away from the infrastructure experiencing problems. Its engineering teams also began restarting affected servers in an attempt to restore normal operations.
However, the recovery process encountered another complication. Microsoft revealed that some servers that had been brought back online became unresponsive again once they started receiving traffic.
“The servers we returned to service can become unresponsive again once they receive traffic; we're analysing diagnostic data to determine what's triggering the behaviour,” the company said in its update.
Microsoft also indicated that recent configuration changes could be contributing to the disruption. The company began evaluating whether reversing some of these updates could accelerate the recovery process.
“We're actively restarting and validating the affected servers to return them to service. In parallel, we're assessing whether reverting a small number of recent configuration updates we identified could help accelerate recovery,” the company wrote on X.
In another update, Microsoft said it had identified an alternative method of checking server health. This approach had helped a portion of the affected infrastructure return to service successfully.
“We've validated an alternative service health check method, which allowed a subset of affected servers to successfully return to service. We're now applying this change more broadly. We're also preparing automated recovery systems to continually detect and restart unresponsive servers, to maintain service stability while this change progresses,” Microsoft said.
The company also sought to reassure users that its engineering teams were pursuing multiple recovery options.
"We're aggressively pursuing all pathways to reduce impact," Microsoft said in an earlier statement.
Outlook, Teams And Microsoft 365 Outage: Thousands Report Problems
The outage affected several widely used Microsoft products, with Outlook recording the highest number of user complaints.
According to Downdetector, which aggregates outage reports from users, nearly 130,000 people reported problems accessing Outlook shortly after 1 pm Eastern Time, or approximately 11:30 pm IST.
Users reported difficulties loading their inboxes, sending emails and receiving messages. Although the number of complaints subsequently declined, reports of disrupted access continued.
Microsoft 365 also recorded more than 38,000 reports on the outage tracking platform, while approximately 4,500 users reported problems accessing Microsoft Teams.
The scale of the disruption highlighted how problems affecting Microsoft's underlying infrastructure could impact multiple services simultaneously.
Microsoft had earlier announced that it had resolved a separate Teams disruption affecting users in North America. However, subsequent updates indicated that the wider connectivity problems could also affect Microsoft 365 users in other geographical regions.
The company has not provided a definitive explanation of the underlying technical failure, although its references to unresponsive servers and recent configuration updates offer some indication of the challenges facing its recovery teams.
ALSO READ: Microsoft Boss Satya Nadella Warns Firms Are 'Paying Twice' For AI
OpenAI And Anthropic Also Report Microsoft 365 Connector Issues
The disruption extended beyond Microsoft's own products, affecting integrations used by external artificial intelligence platforms.
OpenAI said it was investigating an issue involving Microsoft connectors, warning that users could encounter problems when establishing new connections and that certain tools might not function as expected.
Anthropic also confirmed that it had identified a problem involving an upstream provider that was affecting its Microsoft 365 connector.
These disruptions underline the extent to which third-party platforms depend on Microsoft's infrastructure and connectivity services to deliver their own features.
Microsoft's engineering teams continued working on server recovery, alternative health checks and automated systems to identify and restart unresponsive infrastructure. However, the company's admission that its initial corrective measures were not delivering the expected results suggested that restoring stability across all affected services could take additional time.