Nebius incident
Managed Service for SkyPilot: cluster creation and update requests failing in eu-north1
Nebius experienced a major incident on September 14, 2026 affecting Managed Service for SkyPilot, lasting 1d. The incident has been resolved; the full update timeline is below.
Affected components
Update timeline
- investigating Sep 15, 2026, 09:56 AM UTC
We are investigating an issue affecting Managed Service for SkyPilot in eu-north1. Requests to create or update SkyPilot clusters, as well as start and stop requests, are failing with HTTP 503 errors. Instances that are already running are not affected and continue to operate normally.
- monitoring Sep 15, 2026, 12:19 PM UTC
The error rate has dropped to zero. Requests to create, update, start, and stop SkyPilot clusters in eu-north1 are completing successfully again. We are continuing to investigate the root cause and are monitoring the service closely. Instances that were already running have not been affected at any point. We will provide a further update as the investigation progresses.
- resolved Sep 15, 2026, 03:33 PM UTC
This incident has been resolved.