MongoDB Atlas reported this major-impact incident on its official status page on Aug 5, 2026, 05:17 UTC. It was resolved after 6h 57m.
Right now MongoDB Atlas is operational. Live MongoDB Atlas status →
We believe the symptoms of this have largely subsided, looking at both customer cases and our internal metrics. We will share any more information as we get it, but for now resolving this incident as Atlas is fully operational.
We are investigating an issue with Google Cloud Platform (GCP) capacity that is affecting some MongoDB Atlas customers across multiple GCP regions and geographies. We expect the scope of this issue to continue to expand as our investigation progresses. It currently includes, but is not limited to, multiple GCP regions across North America, Europe, and Asia Pacific/Australia, as well as M50 and S70L instance types. Customers may experience errors or delays when creating or modifying Atlas clusters, particularly deployments with search nodes. Some customers may also see host-down alerts while this issue is ongoing. MongoDB is working with GCP to understand the scope of the issue and restore capacity. Customers encountering provisioning issues may retry later or consider using a different GCP region, cloud provider, or instance size. If you are experiencing a time-sensitive production impact, please contact MongoDB Support through the Support Portal or the in-product support chat. We
Our 5-minute checks didn't record a change in MongoDB Atlas's overall status around this incident. Smaller or regional incidents often leave a provider's overall status green.
Overlapping incidents aren't necessarily related.
MongoDB Atlas reported 15 incidents in the last 90 days, 7 of them major or critical. A typical incident lasted 2h 31m. See MongoDB Atlas's uptime and incident history