UiPath incident
Singapore - Document Understanding - Partial Outage
UiPath experienced a major incident on August 25, 2026 affecting Document Understanding, lasting 1h 14m. The incident has been resolved; the full update timeline is below.
Affected components
Update timeline
- investigating Aug 25, 2026, 03:44 AM UTC
We are investigating an issue affecting customers using the Extended OCR functionality in the Document Understanding service in the Singapore region. Our teams are working to identify the root cause and will provide further updates as more information becomes available.
- monitoring Aug 25, 2026, 04:04 AM UTC
The issue has been mitigated, and we are closely monitoring the service to ensure it continues to operate as expected. We will provide further updates as more information becomes available.
- resolved Aug 25, 2026, 04:59 AM UTC
The issue has been resolved, and the service is operating as expected. No further impact has been observed.
- postmortem Aug 27, 2026, 03:42 AM UTC
## Customer impact On August 25, 2026, Extended OCR requests in the Document Understanding service failed with 500 status code in the Singapore region for approximately 33 minutes, between 03:08 and 03:41 UTC. The cause was a Microsoft regional service disruption in Singapore. All other Document Understanding functionality was unaffected, and no other region was impacted. ## Root cause The failure was attributed to a regional Microsoft service disruption in Singapore affecting resources used by the Extended OCR capability. No changes had been made on our side, and Microsoft subsequently updated their own status page to reflect a Singapore regional issue. A formal root cause analysis has been requested from Microsoft. ## Detection An automated alert for the Document Understanding service fired on August 25, 2026 at 3:13 am UTC. The alert was acknowledged promptly, and the incident was declared customer-impacting within minutes. ## Response The on-call engineer scoped the impact to the Singapore region. Responders ruled out a recent service update as the cause, as the same update had been deployed to other regions with no comparable impact, which pointed to a regional dependency failure outside our infrastructure. Because the failure originated in an upstream Microsoft service, no mitigating action was available or required on the UiPath side. Requests began succeeding again at 03:41 UTC as the Microsoft dependency recovered. Responders held the incident open to verify sustained recovery: ten minutes of clean traffic were confirmed at 03:51 UTC, the incident moved to Monitoring at 04:04 UTC, and it was resolved at 04:59 UTC after continued stability with no further failures. ## Follow up 1. Requested a root cause analysis from Microsoft for the Singapore regional disruption affecting Extended OCR dependencies, including how recurrence can be prevented.