Scaleway Outage History

Scaleway's Qwen-3-235B-A22-Instruct generative AI model in the fr-par region experienced increased response times for 8.6 hours. The performance degradation affected users of this specific AI model with slower than normal API responses. The incident was resolved after investigation and remediation efforts.

Scaleway experienced a DNS resolution failure affecting MongoDB SRV record lookups on their Private Network in the FR-PAR region, preventing customers from connecting to their MongoDB instances through Private Network endpoints. The issue impacted the DBaaS service and lasted 40 minutes before being resolved by engineering teams.

Scaleway performed mandatory maintenance operations on Serverless Containers in the fr-par region on April 1st and April 8-9, 2026 to prepare for their new v1 API. During the maintenance, namespaces temporarily entered "upgrading" status blocking write operations for up to 15 minutes, container instances were created causing potential scaling anomalies, long-running requests could fail, and cron jobs might trigger multiple times. The operations were completed successfully, maintaining container reachability throughout the process while preparing the infrastructure for enhanced API features.

Scaleway performed a mandatory maintenance operation on Serverless Containers in the fr-par region to prepare for their new v1 API upgrade. During the operation, users experienced temporary namespace upgrades (blocking write operations for up to 15 minutes), multiple container instances running simultaneously, potential failures of long-running requests, and cron jobs triggering multiple times. The maintenance was completed successfully, with containers remaining reachable throughout the process except for application-level errors.

Scaleway's Cockpit service in the nl-ams region experienced errors when querying Logs Datasources due to querier pods failing to restart properly after a Kubernetes upgrade. The incident affected users' ability to view and analyze logs in the nl-ams-1 component for approximately 14 days. A fix was implemented and monitored before the incident was fully resolved.

A malfunctioning network switch in Scaleway's fr-par-1 data center caused a complete loss of internet connectivity for Dedibox dedicated servers located in a specific rack (DC3 Room 4-6 Rack A08). The outage lasted 5.3 hours while technicians worked to replace the faulty switch hardware. All services were restored to normal operation once the switch replacement was completed.

Scaleway's Account API in the fr-par-1 region experienced a technical issue that prevented users from receiving login emails. The incident lasted approximately 158.5 hours (about 6.6 days) and had major impact on user authentication. The issue was resolved after implementing a fix and monitoring period.

Scaleway's Managed Opensearch service in the fr-par-1 region experienced failures in create-deployment workflows, preventing users from creating new deployments. The incident affected the Excellence component and lasted approximately 210 hours before being resolved on March 26th, 2026.

Scaleway's Cockpit service in the fr-par-1 region experienced high error rates due to numerous workloads failing to become ready. The incident lasted approximately 10 days before being resolved. The service disruption was classified as minor impact.

Scaleway's Object Storage service experienced issues for 16.8 hours where clients using AWS CLI could not download objects when specifying a starting range greater than 0, though the standard GetObject command remained functional. The error was identified and a fix was deployed, with the incident fully resolved after a monitoring period.