MongoDB Outage History

A billing bug in MongoDB Atlas caused some customers to receive higher-than-expected backup charges on their April 24, 2026 invoices. The issue was purely billing-related and did not affect backup configurations, cluster health, or customer data. MongoDB resolved the bug within 3.6 hours and automatically refunded all overcharges, with correct usage amounts reflected by April 28th.

MongoDB Atlas for Government experienced a major outage where the Business Intelligence Connector failed completely, preventing users from connecting to and querying their BI Connectors for nearly 93 hours. The issue was identified on April 17th and resolved on April 21st after implementing a fix. While BI Connector functionality was unavailable, underlying cluster health remained unaffected and data stayed accessible through regular MongoDB clients.

MongoDB Cloud experienced delays in cluster creation and adding new nodes specifically in the Azure East US 2 region due to Azure capacity constraints. Users encountered provisioning delays for new clusters and expansion of existing clusters in this region. The issue was resolved after 1.4 hours when Azure capacity constraints were lifted, restoring normal provisioning capabilities.

MongoDB Atlas experienced intermittent HTTP 503 errors affecting the Atlas Admin API and automation tooling like Terraform for nearly 40 hours. The errors caused control plane operations to fail temporarily, though requests typically succeeded on retry and database connectivity remained unaffected. The engineering team identified the root cause, applied a fix, and confirmed resolution after monitoring showed no further error spikes.

MongoDB Atlas Data Federation experienced a major issue preventing it from reading data from customer Azure blob storage containers. The problem affected all Atlas Data Federation queries attempting to access Azure blob storage data sources. The issue was identified, fixed, and resolved after 1.2 hours of service disruption.

MongoDB Atlas experienced a 49-minute outage where the Clusters page failed to load, displaying a blank page for affected users. The issue only impacted the MongoDB Cloud interface and did not affect actual cluster health or functionality. The incident was resolved after implementing a fix and monitoring the results.

MongoDB Atlas for Government experienced SSO login failures due to an SSO routing issue that prevented some users from accessing the UI via saved SSO URLs or federated identity provider links, with affected users encountering 400 errors or incomplete authentication flows. The incident lasted approximately 13 days starting March 18, 2026, and was resolved after implementing a fix to restore full SSO login functionality. Users were able to work around the issue by logging in directly through the standard email prompt at the MongoDB Government account portal.

MongoDB experienced high failure rates when provisioning clusters in Azure Virginia-East2 (eastus2) due to an underlying Azure service issue affecting virtual machine operations in that region. The incident impacted MongoDB Cloud services, causing delays and failures in cluster creation for users attempting to deploy in the eastus2 region. The issue lasted 10.1 hours while MongoDB engaged with Azure support to resolve the underlying virtual machine capacity constraints and deployment failures.

MongoDB Atlas cluster snapshots were failing for 3.2 hours, preventing backup operations across the MongoDB Cloud platform. The engineering team identified the root cause and implemented a fix. The incident was resolved after monitoring confirmed the snapshots were functioning normally again.

MongoDB Atlas experienced a major incident where email notifications were failing to reach some users. The issue affected MongoDB Cloud services and prevented users from receiving important email communications from the platform. The incident was resolved after 39 minutes, with email notifications restored to normal functionality.