Multiple Regions - Signing With Microsoft is momentarily impacted

Update timeline

1. investigating May 01, 2026, 05:59 PM UTC

   Signing With Microsoft is momentarily impacted

2. identified May 01, 2026, 06:10 PM UTC

   The issue has been identified, and the team is actively working on mitigation. Further updates will be shared as progress continues.

3. monitoring May 01, 2026, 06:35 PM UTC

   The issue has been mitigated, and services are now operating normally. We will continue to monitor the situation.”

4. resolved May 01, 2026, 06:37 PM UTC

   The issue has been mitigated, and services are now operating normally. We will continue to monitor the situation.”

5. postmortem Jul 15, 2026, 01:43 AM UTC

   ## Customer Impact Between May 1, 2026 at 5:20 pm UTC and approximately 6:12 pm UTC, customers attempting to sign in to Automation Cloud services using Microsoft \(Sign in with Microsoft\) were unable to complete authentication. Other sign-in methods and existing logged-in sessions were unaffected. The impact was global for customers using Microsoft-based authentication. ## Root Cause The cause was the expiration of a Microsoft Entra ID \(formerly Azure Active Directory\) client secret used to federate Microsoft sign-in to Automation Cloud. When the credential expired, our authentication provider could no longer complete federation with Microsoft, and all new Microsoft sign-in attempts failed. The credential had been due for routine rotation. Internal notifications intended to warn UiPath in advance of expiration did not reach the responsible team in time, and the rotation did not occur before expiration. ## Detection and Response Automated monitoring detected the issue at 5:28 pm UTC, and engineers were paged immediately. The team identified the cause within minutes using existing runbooks, generated a new credential, and applied it to our authentication configuration. The change took approximately 10–15 minutes to propagate across regions of our authentication provider. The last observed sign-in failure occurred at 6:12 pm UTC, and the incident was resolved on our public status page at 6:37 pm UTC. ## Follow-up To reduce the likelihood of recurrence, we are reviewing the credential expiry notification process to ensure the team responsible for rotation reliably receives advance warnings, and auditing ownership coverage across our authentication credentials so notifications reach the right people. We are also evaluating additional preventative alerting on credentials approaching expiration, along with improvements to our incident response tooling for faster coordination on high-severity alerts. We sincerely apologize for the disruption.