Incident History

This incident has been resolved.

This incident has been resolved. We have run all remediations and systems have returned to normal.

From approximately 13:40 to 17:20 UTC on 3 June 2025, MongoDB Atlas was unable to send metrics via Datadog's EU region for some Atlas Projects. Affected users may find gaps during this time in less than half of their metrics if they use Datadog's EU servers. These gaps will not be backfilled.

Cluster health was unaffected, however, alerts may have fired within Datadog – we do not have visibility to this.

Incident Summary

Between May 28th 2025 16:00 UTC and May 28th 2025 19:36 UTC, emails were sent out to users on affected orgs notifying them that they have been moved into the locked status and users from all affected orgs were restricted from performing any actions on their organization.

Root Cause

The root cause of this incident was an outdated internal process that was inadvertently re-enabled during a database migration.

MongoDB Actions

The MongoDB Atlas Billing team has deleted the outdated process and increased alerting on changes in dunning statuses.

Recommended Customer Actions

This issue was fully addressed by the MongoDB Atlas Billing team and does not require customer action.

This incident has been resolved.

Between 18:37 and 21:17 UTC on 16 May 2025, MongoDB Atlas sent out false Host Down alerts to a small portion of our users. Cluster health was unaffected during this time and we sincerely apologize for the confusion.

This incident has been resolved.

This incident has been resolved.

The incident is now resolved. Since May 1st 2025 at 04:52 UTC we falsely reported a limited number of Atlas Clusters as down in the UI and alerted as such.

Alerting latency is back within expected thresholds.