Reboot server

During a planned maintenance operation on node configuration, 4 compute nodes rebooted and became temporarily unavailable due to a latent configuration fault that surfaced on restart. Compute capacity was reduced for approximately 1 hour.

Impact: Degraded worker capacity on the cluster; some workloads were rescheduled for the duration of the incident. The cluster API and control plane remained continuously availabl - no control-plane service interruption.

Resolution: The configuration was corrected and reapplied to the affected nodes, all of which returned to an operational state. The cluster is fully restored.

Follow-up: The fix is being rolled out progressively across the rest of the fleet to prevent any recurrence.

We’ve identified a formatting issue with a new configuration. Our team is actively working on it and we’ll share an update shortly. Thanks for your patience

A manual error on our side caused a service interruption. Our servers are currently restarting and the service should be fully restored within the next few minutes.

We apologize for the inconvenience. We’ll post an update as soon as everything is back to normal.