Static

Microsoft configuration change leaves SharePoint pages drawing a blank

First reported by The Register ·

The signal ●○○○ Compiled by AI from The Register, the single source so far
Why you might care

Some SharePoint pages might not load for you on September 16, 2026, and the fix is automatic.

What happened

Microsoft experienced a widespread outage of SharePoint Online pages on September 16, 2026, from 1604 to 1730 GMT. Users were unable to load SharePoint sites and pages, encountering a "Sorry, something went wrong: Thread was being aborted" error. The issue was attributed to a faulty configuration change impacting how servers deliver code for page rendering. Microsoft reverted the change, confirming resolution through service telemetry. The company stated it is investigating its change validation and deployment processes to prevent future incidents. This incident follows a pattern of Microsoft cloud services being disrupted by configuration changes, including past issues with Azure, Microsoft 365, and Exchange Online.

What it means

This incident highlights a recurring vulnerability in Microsoft's cloud infrastructure, where seemingly minor configuration changes can have significant, widespread impacts on critical services like SharePoint. The company's admission that it is "further assessing the process by which we validate and deploy configuration changes" suggests a systemic issue in quality control that has led to multiple disruptions across its cloud offerings. The recurring nature of these failures, despite promises to enhance detection and mitigation, indicates that preventing faulty changes from reaching production remains a challenge for Microsoft.

The immediate rollback and subsequent investigation underscore the reactive nature of Microsoft's response to such incidents, rather than a proactive prevention strategy. Users and organizations relying on SharePoint for day-to-day operations experienced downtime, impacting productivity and potentially business continuity. The commitment to a post-incident report within a few business days is standard, but the underlying problem of deployment quality control needs sustained attention to build user confidence in the reliability of Microsoft's cloud services.

AI-written summary. May contain errors.