Maestro Performance Degradation in Europe Region

Update timeline

1. monitoring Jun 24, 2026, 09:43 PM UTC

   We observed increased latency affecting Maestro workflow executions in the Europe region between June 12, 2026, 20:00 UTC and June 24, 2026, 15:00 UTC. Impact: Some Maestro workflows experienced slower-than-normal execution times during this period. No other services were impacted. The issue was isolated to Maestro workflow execution in the Europe region. Our team has taken corrective actions to restore normal performance and continues to monitor the service to ensure stability. We apologize for any inconvenience this may have caused.

2. resolved Jun 25, 2026, 12:06 AM UTC

   We observed increased latency affecting Maestro workflow executions in the Europe region between June 12, 2026, 20:00 UTC and June 24, 2026, 15:00 UTC. Impact: Some Maestro workflows experienced slower-than-normal execution times during this period. No other services were impacted. The issue was isolated to Maestro workflow execution in the Europe region. Our team has taken corrective actions to restore normal performance and continues to monitor the service to ensure stability. We apologize for any inconvenience this may have caused.

3. postmortem Jul 22, 2026, 10:10 PM UTC

   ## _Customer Impact_ Between June 12, 2026 at 20:00 UTC and June 24, 2026 at 15:00 UTC, some customers running Maestro workflows in the Europe region experienced slower-than-normal execution times. Individual workflow steps that normally complete within seconds occasionally took significantly longer to start, which slowed overall workflow execution. All affected workflows completed successfully. No workflows failed because of this issue, and no data was lost. ## _Root Cause_ A recent platform update introduced a performance bottleneck in an internal workload-management mechanism used during Maestro workflow execution. **What went wrong:** During periods of high traffic, the mechanism created resource contention on some of the servers that execute workflow steps. This delayed some workflow steps before they started, although the steps ran normally once execution began. **What should have happened:** Internal workload-management operations should not delay workflow execution, including during periods of high traffic. ## _Detection_ The issue was not detected by internal monitoring or alerting. It was reported by a customer on June 22, 2026, and escalated through our incident-management process on June 24, 2026. The investigation confirmed increased Maestro workflow execution latency in the Europe region. ## _Response_ Once the incident was declared, the response team identified the recent platform change responsible for the delays and applied a mitigation across the affected environments. Workflow execution latency returned to normal within the hour, and monitoring confirmed that performance remained stable. ## _Follow-Up_ To reduce the likelihood and impact of similar incidents, we are: * Strengthening performance safeguards for changes that affect workflow execution. * Enhancing monitoring and alerting for unusual increases in workflow execution latency. * Improving rollout and validation controls for performance-sensitive platform changes.