DigitalOcean Outage History

DigitalOcean experienced a brief outage affecting access to the DigitalOcean Support Portal (cloudsupport.digitalocean.com). Users attempting to reach the Support Center could not load the portal during the incident. The issue began at Jun 14, 2025 2:40 PM UTC and was resolved by Jun 14, 2025 3:35 PM UTC, lasting about 55 minutes. DigitalOcean posted two status updates: an initial note that engineering was investigating the issue, followed by a resolution update and a workaround directing users to submit support requests via the general contact form. The impact was limited to the Support Center component, with no broader DigitalOcean services indicated as affected.

DigitalOcean reported a brief degradation in the DBaaS logs retrieval API, affecting all non-MongoDB clusters. Users attempting to fetch logs via the API may have encountered 500 errors, and logs did not appear in the Recent Logs tab. The incident began at 4:30 PM UTC and lasted about 2 hours, with service restored by 6:30 PM UTC. The issue impacted the Managed Databases / Global component and appeared to be a global issue for log visibility across DBaaS clusters.

DigitalOcean conducted scheduled maintenance in the NYC3 region that impacted several core services. During the maintenance window, users may have experienced intermittent connectivity or degraded performance for API access, Networking/NYC3, the Cloud Control Panel, and the Cloud Firewall. The maintenance was regional to NYC3 and affected four components.

The maintenance window began at 10:05 PM UTC and completed by 12:00 AM UTC. After the window, all four components returned to normal operation: API, Networking/NYC3, Cloud Control Panel, and Cloud Firewall were restored.

DigitalOcean suffered a multi-region outage across global regions, impacting Droplets and related services such as Load Balancers and Kubernetes. Users experienced elevated failures when creating Droplets and using Droplet-based products, and App Platform along with Managed Databases control plane operations (listing, creating, scaling resources) were affected, disrupting deployment and management workflows across the platform. The issue spanned across all regions, with services intermittently degraded during the incident.

DigitalOcean reported a regional outage in the BLR1 region impacting Spaces object access with elevated error rates and Container Registry operations. App Platform deployments were also delayed globally during the incident. The outage began at 7:45 AM UTC and was resolved by 11:00 AM UTC, lasting roughly 3 hours and 15 minutes. It affected three components: Spaces / BLR1, App Platform / Global, and Container Registry / BLR1, resulting in degraded object access and slower or failing deployments during the event.

DigitalOcean's engineering team investigated the issue and posted updates. A fix was implemented around 9:10 AM UTC with improving error rates, and by 11:00 AM UTC services were recovering. The incident included four status updates from DigitalOcean during the period, with stability returning for Spaces in BLR1, App Platform deployments globally, and Container Registry in BLR1.

DigitalOcean reported a brief degradation in App Platform Deployments, causing delays in creating new apps and, in particular, delays when linking domains during the deployment flow. Existing apps and domains already configured were not affected. The issue appeared to be global across App Platform deployments. DigitalOcean posted status updates during the incident as they investigated and then confirmed resolution.

The incident began at May 21, 2025 8:15 PM UTC and was resolved by May 21, 2025 9:25 PM UTC, lasting about 1 hour 10 minutes. DigitalOcean issued two status updates during the event as the team investigated and then restored normal deployment performance. The disruption affected new app deployment workflows and domain linking, but did not impact existing apps or configured domains.

The GenAI Platform on DigitalOcean experienced degraded performance after GenAI Platform agents using Llama 3.3 70B failed to respond, affecting users attempting GenAI tasks. This issue was limited to the GenAI Platform component and could cause delays or incomplete results for GenAI workloads.

The incident began at May 15, 2025 3:00 PM UTC and was resolved by May 15, 2025 5:05 PM UTC, lasting about 2 hours and 5 minutes. DigitalOcean issued 2 status updates during the incident to inform users and track progress.

DigitalOcean reported a brief degraded state for the GenAI Platform as agents stopped responding. The incident began at 5:45 AM UTC and was resolved by 6:20 AM UTC, lasting roughly 35 minutes. DigitalOcean posted two status updates during the event while the team investigated and confirmed restoration. The impact appeared limited to the GenAI Platform component, with no related user reports and no broader service disruption.

The Droplet search feature in DigitalOcean's Cloud Control Panel experienced a partial outage, with users potentially unable to search for and locate Droplets. The incident began at May 14, 2025 12:35 PM UTC and was resolved by May 14, 2025 3:45 PM UTC, lasting approximately 3 hours and 10 minutes. The Cloud Control Panel component transitioned from normal to degraded during the incident and returned to normal after the issue was addressed. Two official updates were posted during the event: the first at May 14, 2025 12:35 PM UTC indicating an investigation, and a follow-up at May 14, 2025 3:45 PM UTC confirming resolution.

DigitalOcean's GenAI Platform, including the DeepSeek model, briefly degraded after an issue that could display the message: 'It looks like the agent ran out of tokens while reasoning. Please try again or increase the max tokens.' The incident began at May 13, 2025 9:55 PM UTC when engineers started investigating and the GenAI Platform was listed as degraded, and was resolved by May 13, 2025 11:25 PM UTC as service returned to normal operation.

The DigitalOcean status page documented two updates during the event, confirming ongoing investigation and eventual restoration. The impact was limited to the GenAI Platform (DeepSeek model), with no broader DigitalOcean services affected.