OVHcloud Outage History

OVHcloud experienced a major incident affecting their Public Cloud Compute instances in the GRA1, GRA3, GRA5, and GRA9 regions due to an unexpected infrastructure failure, making some instances unreachable and preventing customers from accessing their instances. The incident lasted 8.3 hours, starting at 06:31 UTC on May 4, 2026, and was fully resolved by 15:13 UTC the same day.

OVHcloud experienced a major infrastructure failure that made compute instances and Redis databases unreachable in the LIM1 region for 31.9 hours from May 2-3, 2026. The incident was caused by an unexpected underlying infrastructure malfunction that prevented customers from accessing their instances and databases in the affected region. Service was gradually restored with most instances becoming operational within the first day, and the remaining instances were fully restored by May 3rd.

OVHcloud experienced a major incident from 07:19 to 08:20 UTC on May 1, 2026, where some compute instances in the ERI1 region became unreachable due to a network equipment failure. Customers were temporarily unable to access their affected instances during the 1.1-hour outage. The incident was resolved after teams identified the network equipment issue and restored service.

OVHcloud's Managed Kubernetes Service experienced a security vulnerability (CVE-2026-31431) that could allow container users to gain elevated access on underlying worker nodes, affecting clusters across multiple global regions. The vulnerability was limited to individual clusters and posed higher risk for customers running shared or multi-tenant workloads within the same cluster. OVHcloud resolved the incident by releasing updated versions after 8 days, while providing temporary mitigation instructions during the remediation period.

OVHcloud's Data & Analytics platform in the FR region experienced a software issue that prevented customers from building APIs and applications for 6.4 hours on April 30, 2026. The incident began at 08:18 UTC and was resolved at 19:14 UTC, with the root cause identified as a software problem affecting the build functionality. Service was fully restored after teams mobilized to address the identified issue.

OVHcloud's Object Storage S3 service in the GRA region experienced a major incident caused by a software issue from 10:25 to 11:30 UTC on April 28, 2026. Customers encountered 500/503 errors when attempting to download or upload objects to the storage service. The incident was resolved after 65 minutes once teams identified and fixed the underlying software problem.

OVHcloud experienced a major incident in the RBX10 region where some compute instances became unreachable due to an unexpected infrastructure failure, preventing customers from accessing their instances. The incident lasted 2.5 hours from 14:40 to 16:42 UTC on April 24, 2026, and was fully resolved after teams identified and fixed the infrastructure issue.

OVHcloud's storage services in the BHS region experienced a major outage for 48 minutes on April 22, 2026, from 13:22 to 14:20 UTC. Customers encountered 503 errors and difficulties accessing their storage services due to an unexpected underlying infrastructure malfunction. The incident was resolved after teams identified and fixed the root cause.

OVHcloud experienced a major incident affecting Cloud Compute instances in the BHS1, BHS3, and BHS5 regions, making some instances unreachable due to an unexpected underlying infrastructure malfunction. The incident lasted 2.1 hours from April 21, 2026 11:45 UTC to 13:59 UTC, preventing customers from accessing their instances in the affected regions. The issue was identified as an infrastructure malfunction and was fully resolved by OVHcloud's technical teams.

OVHcloud experienced a major incident affecting their Managed Private Registry service in the DE region from 07:51 to 09:40 UTC on April 21, 2026. Some registries became unreachable, preventing customers from performing management operations on their container registries. The incident was resolved after 1.4 hours and was caused by a software issue.