Customer impact

Between August 28, 2026 at 3:23 pm UTC and August 28, 2026 at 10:57 pm UTC, a subset of customers experienced errors when opening queue item details panels in Orchestrator. The affected path returned a 404 page for LINKED queues when the user did NOT have access to the original folder in which the items were created.

The impact affected only Orchestrator UI interactions and it spanned across all regions running the affected software version. Orchestrator application programming interface access was unaffected, and exporting queue data to CSV was available as a workaround. The total duration was approximately 7 hours.

Root cause

The incident was caused by a regression in the Orchestrator user interface for linked queues accessed from other folders. The regression did not correctly handle the linked-queue details flow when the requesting user lacked access to the original folder, which caused the queue item details panel to route to a 404 page instead of displaying the expected information.

Detection

The issue was detected through an automated incident alert for the Orchestrator front end on August 28, 2026 at 3:23 pm UTC.

Response

At 3:31 pm UTC on August 28, 2026, our engineering team described the issue as customers being unable to access queue item details panels due to a regression. At 3:41 pm UTC, a public status update confirmed that the root cause had been identified, that a fix was being deployed, and that application programming interface access was unaffected.

At 5:27 pm UTC, the scope was narrowed to linked queues where the user did not have access to the original folder. At 5:29 pm UTC, the team determined that the initial fix did not fully address the affected scenario, and a corrected fix was developed. At 5:30 pm UTC, the affected scenario was reproduced and the corrected fix was validated locally. At 5:39 pm UTC, a status page update documented CSV export as a workaround.

Deployment of the corrected fix continued across affected regions. At 10:50 pm UTC, the fix was confirmed deployed everywhere. At 10:57 pm UTC, the incident was marked resolved and the status page was updated to confirm the service was operating as expected.

Follow up

Formal follow-up actions are being tracked through the post-incident review process initiated at resolution.

Action items

• Expanding our automated test cases to include this scenario, and other similar scenarios related to linked objects or cross-folder scenarios.
• Ensuring we can safely turn off any changes via a flag for a faster response time.