This incident is resolved. Task runs are executing normally.

Systems are back to normal

Maintenance completed. Backend infrastructure is healthy

The increase of API errors was caused by an Azure network service degradation that is now resolved.