Scaleway Outage History

Scaleway's BookMyName email redirection service experienced a complete outage, preventing all email redirects from functioning properly. The incident lasted 52 minutes while the team investigated and resolved the underlying issue. The redirection service has been restored and is now operational again.

Public switches failed in Scaleway's DC2 Room 203-B Rack J4, causing servers in that rack to lose network access and become unreachable while RPN network remained functional. The issue affected Dedibox services in DC2 for 1.5 hours until the faulty switch was replaced, restoring full service.

Scaleway's webhosting service experienced database connection issues starting March 7th, with customers encountering "Too many connections" errors that caused websites to malfunction or become unavailable. The incident affected the Hosting component and was resolved after 35.3 hours on March 9th at 12:57 PM.

Scaleway's Serverless Containers service in the nl-ams region experienced issues where containers were stuck in restart loops due to an unannounced change that converted user-configured health checks from readiness probes to liveness probes. This caused containers with failing health checks to continuously restart before they could properly initialize. The issue was resolved by rolling back the probe configuration change, returning health checks to their previous readiness-only function and stopping the aggressive container restarts.

Scaleway's Cockpit service experienced elevated error rates when querying logs from their datasource in the fr-par region, likely due to an internal migration process. The issue affected the fr-par-1 and fr-par-2 components for 21.7 hours. The incident was resolved after investigation and remediation efforts.

Since March 3rd, 2026, daily lifecycle jobs failed to execute for some Object Storage buckets in Scaleway's Paris region. The root cause has been identified and a fix is ready for deployment, after which jobs will run immediately to clear the accumulated backlog.

Scaleway's Elastic Metal EM-RV servers in the fr-par-2 region became unavailable for new orders due to servers being stuck in a cleaning state, lasting approximately 14 days. Existing instances continued to operate normally and were not impacted. The issue was resolved on March 18th, restoring the ability to order new EM-RV servers.

A fraction of Scaleway's Cockpit log query services in the nl-ams-1 region went down, causing errors when users attempted to query logs. The affected services were restarted and the incident was resolved within 4 minutes.

Scaleway's Managed OpenSearch service experienced an issue where the dedicated offer was not displayed in the console for nodes in the fr-par region. This console display problem affected users' ability to view the dedicated offering option for their OpenSearch deployments. The incident was resolved after 23.5 hours of investigation and remediation.

Following scheduled maintenance, Scaleway's Serverless Containers in the nl-ams region experienced intermittent errors including TLS handshake failures, 502 errors, connection timeouts, and increased latencies due to load balancer connection saturation that caused backends to fluctuate between "up" and "down" states. The issue occurred in two phases totaling approximately 15 hours over two days, during which requests made during "down" phases were rejected. The problem was resolved by rebalancing backends and clearing idle connections to free up capacity, ensuring backends remained consistently available to receive traffic.