MongoDB Outage History

MongoDB experienced increased error rates when users attempted to create and manage Search and Vector Search indexes, affecting database search functionality. The incident lasted 2.2 hours before a fix was implemented and deployed. The service has been restored and the incident was resolved after monitoring confirmed the fix was successful.

MongoDB experienced connection and querying issues with Atlas Data Federation and Online Archive instances across GCP regions due to networking problems. The incident lasted 1.3 hours and prevented users from accessing these services in affected GCP regions. The networking issue was identified and mitigated, with connections restored to all GCP regions.

MongoDB Atlas for Government users experienced gaps in metrics being sent to their DataDog Gov integrations starting at approximately 17:00 UTC. The issue affected monitoring data delivery for government users with DataDog integrations. The incident lasted 1.3 hours and was classified as a minor service disruption.

MongoDB Cloud experienced authentication failures affecting search queries on some clusters for 6.4 hours. The issue specifically impacted clusters with colocated search nodes, while clusters with dedicated search nodes remained unaffected. The root cause was identified and mitigated, with the incident fully resolved after monitoring confirmed no residual impact.

MongoDB Atlas for Government experienced a service disruption where data plane node logs (including mongod and audit logs) were unavailable for clusters in GCP and AWS gov-cloud regions from May 14th 4am UTC for approximately 6 days. The incident was resolved after 1.8 hours of active troubleshooting once the root cause was identified, with log download functionality fully restored.

MongoDB Cloud experienced impaired cluster operations for 2.6 hours due to a Let's Encrypt certificate authority outage. The service was restored after Let's Encrypt returned to healthy status, with MongoDB continuing to monitor until cluster operations fully normalized.

AWS Availability Zone use1-az4 in the US-EAST-1 region experienced an impairment that affected MongoDB Atlas clusters, causing some cluster nodes to have degraded health and delays in operations like instance scaling and configuration changes. The Atlas system implemented healing actions to restore affected nodes and provisioned new nodes in healthy availability zones while AWS worked to restore full availability. The incident was resolved after 36.1 hours of monitoring and remediation efforts.

MongoDB AWS clusters experienced degraded DNS changes that prevented cluster topology modifications from completing successfully or caused significant delays. The issue was traced to corrupted metadata that disrupted DNS operations and downstream cluster updates. The problem was resolved after 7 hours by patching the faulty metadata, restoring normal DNS functionality and cluster update processes.

MongoDB Atlas experienced degraded performance affecting a subset of backup export jobs for 23.2 hours. The issue caused failures in backup export functionality while cluster health and data remained unaffected. The problem was resolved after implementing a fix and monitoring the results.

MongoDB experienced a 48-hour partial outage in Azure West Europe where customers encountered nodes reporting as "down" and delayed cluster modifications. The incident affected MongoDB Cloud services in the region, with MongoDB handling primary failover across the three available availability zones. The issue was resolved after instances began recovering, and all Atlas clusters in the region returned to healthy and fully operational status.