DigitalOcean Outage History

The DigitalOcean SFO3 region underwent scheduled core infrastructure maintenance beginning at 9:15 AM UTC affecting Droplets, Managed Databases, API, Cloud Control Panel, Kubernetes, and Load Balancers. Users in the SFO3 region may have experienced intermittent access or degraded performance during the maintenance window. A small number of user reports were received around 8:45 AM UTC prior to the official maintenance. The window lasted until 3:00 PM UTC when the remaining components were restored. The API and Cloud Control Panel were restored earlier, by 9:25 AM UTC.

During the incident, DigitalOcean issued 3 official updates to communicate progress. All remaining components were back online by 3:00 PM UTC. The geographic scope was limited to the SFO3 region.

DigitalOcean’s Managed Databases control plane experienced degradation starting at 4:34 PM UTC, affecting the ability to list backups via API and UI across multiple engines (PostgreSQL, MySQL, OpenSearch, Redis, Kafka). The issue did not appear to affect the data plane, so active databases remained operational. The incident lasted about 4 hours and 51 minutes, with services returning to normal by 9:25 PM UTC.

DigitalOcean’s status page documented the incident with multiple updates, beginning with an investigation and then reporting degradation in the Managed Databases control plane. The provider issued five official updates between 6:05 PM UTC and 9:25 PM UTC, noting ongoing investigation and degraded performance before restoration. The issue had global scope across Managed Databases, while the data plane remained unaffected.

DigitalOcean began multi-region infrastructure maintenance in the LON1 region, with Droplets, Managed Databases, Kubernetes, Load Balancers, API, and Cloud Control Panel listed as affected. This led to partial outages and degraded performance for these services. Guardrails latency on the GenAI Platform was identified around 8:07 PM UTC, with updates at 8:55 PM UTC and 9:10 PM UTC noting longer response times for agents with guardrails. Over the course of the incident, several components transitioned between maintenance and up as engineers implemented fixes, and the situation evolved across regions.

DigitalOcean conducted scheduled Core Infrastructure Maintenance in the SFO2 region. The maintenance window began at 9:00 AM UTC and ended at 3:00 PM UTC. During this window, key components were temporarily affected in SFO2, including Kubernetes / SFO2, Load Balancers / SFO2, API, Cloud Control Panel, Managed Databases / SFO2, and Droplets / SFO2. This regional maintenance may have caused service disruption or degraded access for users with resources in the SFO2 region.

The DigitalOcean SYD1 region underwent a scheduled maintenance window beginning at 4:00 PM UTC. The work affected six components in SYD1: Droplets / SYD1, Load Balancers / SYD1, Cloud Control Panel, Managed Databases / SYD1, Kubernetes / SYD1, and the API. The DigitalOcean status page indicated the maintenance was in progress and that updates would be provided as necessary.

The maintenance concluded at 9:45 PM UTC, with all listed components returning to normal. The incident spanned roughly 5 hours and 45 minutes, and the provider issued two official status updates during the window.

DigitalOcean reported a multi-service disruption affecting Spaces, API, App Platform, Load Balancers, GenAI Platform, and Managed Databases, causing degraded performance and intermittent failures for users. The incident was linked to an upstream provider issue and expanded from an initial impact on a subset of services to additional components. The outage began at 12:30 PM UTC and most services started recovering by 7:35 PM UTC, with signs of improvement appearing earlier during the incident.

The provider published 3 status updates during the event, detailing the scope and progression. Updates noted that provisioning or management actions for new clusters could be impacted, while existing clusters were not affected. The disruption appeared global across listed components, and services returned to normal operation following the final update at 7:35 PM UTC.

DigitalOcean experienced a partial outage affecting the creation of resources across Spaces, App Platform, DOCR, and related Event Processing. Users attempting to create new Spaces, deploy apps on App Platform, or add items to Container Registry reported errors. The incident began at Nov 16, 2025 8:35 AM UTC when the status page transitioned to a degraded state for the affected components, and it lasted for about 1 hour and 30 minutes until Nov 16, 2025 10:05 AM UTC.

DigitalOcean issued 2 status updates during the incident as engineers investigated and then confirmed restoration. By Nov 16, 2025 10:05 AM UTC, all affected components had returned to normal, marking the end of the incident. The issue affected 4 components globally, indicating a broad but partial degradation rather than a full service outage.

DigitalOcean's BLR1 region experienced a regional connectivity outage with intermittent network access. Between 5:28 PM UTC and 5:37 PM UTC, engineers observed a major connectivity loss and rerouted traffic via an alternate upstream provider to mitigate the impact. On Nov 8, 2025 6:10 PM UTC, the DigitalOcean status page posted a warning for the BLR1 networking component, noting that the issue stems from a broader Internet issue and that connectivity was improving. The region began to recover and connectivity returned to normal by Nov 9, 2025 7:15 AM UTC. The outage lasted about 13 hours and 5 minutes, and DigitalOcean issued two status updates during the incident detailing mitigation steps and recovery.

The Gradient AI Platform agent creation process on DigitalOcean encountered issues for customers using a VPC, with some deployments getting stuck on 'Waiting for Deployment.' The incident began around 6:50 PM UTC and the status was elevated to degraded at 8:30 PM UTC as engineers investigated the cause. The issue lasted until 12:00 AM UTC, when the GenAI Platform returned to normal.

DigitalOcean experienced a global issue with Garbage Collection in the Container Registry, causing the GC process to fail or become stuck. This led to degraded performance and possible errors for users performing cleanup tasks across DigitalOcean's Container Registry globally. The incident began at 11:30 AM UTC and was resolved by 7:10 PM UTC, lasting about 7 hours 40 minutes.