MongoDB confirmed that its Atlas API was unavailable for six hours in AWS’s me‑central‑1 and me‑south‑1 zones, disrupting cluster management and raising questions about cloud‑native database reliability.

MongoDB Atlas API suffered a six‑hour outage on October 13, 2026, affecting the me‑central‑1 and me‑south‑1 AWS regions and temporarily disabling cluster management operations for customers.

Scope of the interruption

The outage began at approximately 02:15 UTC and persisted until around 08:20 UTC, during which the Atlas API returned error responses for all calls related to cluster creation, scaling, and monitoring.

MongoDB’s status page confirmed that the issue was isolated to the two Middle East AWS zones, leaving other regions such as us‑east‑1 and eu‑west‑1 unaffected.

Root cause analysis

According to the post‑mortem, a misconfiguration in the load‑balancer health‑check settings triggered a cascading failure in the API gateway layer. The faulty configuration prevented health checks from succeeding, causing the gateway to mark the service as unavailable.

MongoDB engineers rolled back the change and restored normal traffic flow after confirming stable health‑check responses across the affected zones.

Impact on customers

During the downtime, customers could not perform automated scaling, deploy new clusters, or retrieve real‑time metrics via the API. Manual interventions through the MongoDB Atlas console remained possible, but many automation pipelines halted.

  • Inability to trigger autoscaling policies
  • Failed deployment scripts for CI/CD pipelines
  • Delayed monitoring alerts that rely on API endpoints

Preventive measures

MongoDB announced several steps to mitigate future incidents, including stricter change‑management reviews for load‑balancer configurations and expanded health‑check redundancy across multiple AWS zones.

The company also plans to introduce a fallback API endpoint that can serve read‑only requests while primary services recover.

For readers interested in broader cloud‑infrastructure reliability topics, see the Infrastructure coverage on uxc.news.

Source: Probecast coverage of MongoDB Atlas API outage

Track coverage on Momentum · More in Infrastructure.