Resolved
All services have returned to normal capacity. We have confirmed stability throughout our monitoring period.
Monitoring
The issue has been mitigated and our services have returned to normal capacity. We are continuing to monitor the platform closely to ensure full stability and will provide a final update shortly.
Identified
During a maintenance operation on one of our cluster nodes, the orchestrator entered a retry loop trying to reassign tasks that referenced a previous node identity. This blocked new task placement across the cluster for approximately 30 minutes, briefly affecting the availability of some services.