OVHcloud Outage History

OVHcloud experienced a MongoDB service incident from March 16, 2026 14:00 UTC to 20:05 UTC affecting multiple global regions including GRA, DE, UK, RBX, MUM, WAW, SGP, SBG, BHS, EU-WEST-PAR, and EU-SOUTH-MIL. Customers were unable to create new MongoDB services during this period, though existing database services continued to operate normally. The incident was caused by an unexpected underlying infrastructure malfunction and was resolved after teams deployed a fix.

OVHcloud's GRA11 region experienced a software issue that prevented customers from creating new Public Cloud instances and caused network instabilities that blocked network configuration modifications on existing instances. The incident lasted 4.5 hours from 15:00 to 19:30 UTC on March 12, 2026, affecting the Compute - Instance service. The issue has been resolved and normal functionality restored.

OVHcloud's Managed Kubernetes Service experienced slow APIServer access from Worker Nodes affecting all MKS Clusters on GRA11, which impacted applications that frequently interact with the APIServer. The incident was caused by a software issue and occurred from 09:11 to 10:21 UTC on March 12, 2026. The service was resolved and returned to normal operation with continued monitoring for stability.

OVHcloud's MongoDB Discovery service in the GRA region experienced a major incident from March 10-13, 2026, caused by an underlying infrastructure malfunction and software issues. The incident prevented new service creation and management operations, while existing database services continued running but with degraded self-healing capabilities. The service was fully restored after 95.6 hours with ongoing monitoring to ensure stability.

A software issue in OVHcloud's GRA11 region prevented customers from creating new Public Cloud instances and modifying network configurations on existing instances for 2 hours and 35 minutes on March 11, 2026. The incident affected the Compute - Instance service from 17:00 UTC to 19:35 UTC and has been resolved.

OVHcloud's MongoDB service experienced a major incident across multiple regions (GRA, DE, SBG, UK, BHS, WAW) lasting 7 hours from March 10, 2026 16:40 UTC to 23:40 UTC. Management operations like adding users and updating flavors were temporarily unavailable, while existing database services remained operational and continued serving customer workloads. The incident was caused by an unexpected underlying infrastructure malfunction and was fully resolved after teams identified and fixed the root cause.

OVHcloud's Managed Private Registry service in the GRA region experienced a major incident from 09:47 to 17:33 UTC on March 10, 2026, causing some registries to become temporarily unavailable. Customers were unable to access their registries in the affected region due to a related infrastructure incident. The service was fully restored after approximately 7 hours and 46 minutes once teams identified and resolved the root cause.

OVHcloud's CEPH storage service in the GRA2 region experienced a major incident caused by faulty network equipment, making some storage volumes unavailable and causing performance degradation with reduced data redundancy for customers. The incident lasted approximately 10 hours from 07:26 to 17:33 UTC on March 10, 2026, and was resolved after teams replaced the faulty network equipment.

OVHcloud experienced a major storage incident in the GRA2 region from 07:26 to 10:11 UTC on March 10, 2026, caused by faulty network equipment. Some CEPH volumes became unavailable, causing customers to experience performance degradation and deteriorated data redundancy. The issue was resolved after teams replaced the faulty network equipment, with service fully restored by 10:05 UTC.

OVHcloud experienced a major API Network incident in the GRA11 region affecting their Public Cloud service from 13:30 to 17:10 UTC on March 9, 2026. Customers encountered slow API responses and intermittent HTTP 500 errors when performing operations through the API. The incident was resolved after 3 hours and 40 minutes, with the root cause identified as an unexpected underlying infrastructure malfunction.