Scaleway Outage History

A public network switch in Scaleway's PAR2 datacenter (Room9, rack C22) failed, causing all servers in the affected rack to become unreachable and requiring them to be powered down. The datacenter field team resolved the switch issue and restored all servers to online status. The incident lasted 24 hours and was fully resolved with continued monitoring to ensure stability.

Scaleway's Cockpit monitoring service experienced a disruption where product metrics and logs stopped being ingested starting at 15:50. The Platform Frontend Engine (PFE) component was unable to receive this telemetry data, affecting monitoring capabilities across all regions. The issue was resolved after 50 minutes of downtime.

Scaleway's Cockpit service experienced an issue where custom metrics were not being ingested properly, starting at 15:50. The problem affected the platform's ability to receive and process custom metrics from users. The issue was resolved after 11 minutes, with all custom metrics ingestion returning to normal operation.

Scaleway's Cost Manager experienced a crash when users attempted to switch date filters, beginning around 2PM Paris time. The bug prevented normal navigation within the cost management interface. The incident was resolved after 19.4 hours of investigation and remediation efforts.

Since February 10, some Scaleway customers with serverless containers or functions in sandbox v1 configuration were unable to view CPU and memory metrics in the Cockpit monitoring interface. The issue was identified and resolved within 10 minutes on February 23.