DigitalOcean Outage History

The outage began on May 6, 2025 1:00 PM UTC with Droplet Metrics in the BLR1 region reporting intermittent unavailability, impacting users who rely on metrics data for Droplets in that region. London’s LON1 network maintenance started at 8:00 PM UTC and contributed to the disruption. In a later update at May 7, 2025 11:10 PM UTC, DigitalOcean described intermittent gaps in BLR1 metrics and noted ongoing investigations and testing across potential root causes. The BLR1 metrics issue was resolved by May 8, 2025 1:05 PM UTC, with metrics returning to normal and services restored.

DigitalOcean reported a regional issue in the NYC1 zone affecting Droplet creation, with users seeing 504 Gateway Timeout errors when attempting to create droplets in NYC1. The outage began at 2:20 PM UTC and was resolved by 3:50 PM UTC, lasting about 1 hour and 30 minutes. The impact was limited to the NYC1 region and affected the Droplets / NYC1 component.

DigitalOcean conducted Core Infrastructure Maintenance starting at 1:00 PM UTC. During this window, Spaces in the FRA1 region experienced degraded availability with higher error rates when accessed. The FRA1 Spaces component moved to warn at 1:10 PM UTC as engineers investigated the issue. The outage lasted approximately 8 hours, ending when the maintenance completed and the FRA1 Spaces component returned to up at 9:00 PM UTC.

DigitalOcean Managed Kubernetes (DOKS) in the SFO2 region experienced a networking issue that affected new worker nodes. Users attempting to create or autoscale workers could encounter connectivity problems within the clusters they join. The incident began on Apr 29, 2025 10:30 PM UTC and was resolved by Apr 30, 2025 12:15 AM UTC, lasting about 1 hour and 45 minutes. The impact was isolated to the SFO2 region, with no global service disruption reported. The event prompted attention from DigitalOcean engineering as they investigated the root cause and worked to restore normal networking for new nodes.

DigitalOcean reported a regional incident in the FRA1 data center where Droplet event processing experienced delays, potentially slowing or stalling actions such as creating, resizing, powering on/off, or deleting Droplets. The issue began on 1:45 PM UTC and persisted for about 2 hours and 50 minutes, until 4:35 PM UTC. The impact was centered on the Event Processing / FRA1 component, with operations in that region taking longer than usual to complete.

DigitalOcean’s status page documented the investigation and subsequent recovery. The FRA1 event processing status moved from up to warn at Apr 24, 2025 1:45 PM UTC and returned to up at Apr 24, 2025 4:35 PM UTC, spanning two official updates during the incident.

DigitalOcean experienced delays in delivering transactional emails to customers due to an upstream provider issue, impacting important notifications such as sign-in verification codes, sign-ups, and maintenance notices. The Cloud Control Panel was listed in a degraded state during the incident. The outage began around 8:55 PM UTC and lasted approximately 7 hours and 50 minutes, with full restoration by 4:45 AM UTC.

DigitalOcean reported a brief issue in the Spaces creation flow through the Cloud Control Panel, where the Data Center options were not listing correctly for new Spaces. The problem affected the Spaces / Global component and could prevent users from selecting a Data Center during Space creation. The issue began at Apr 22, 2025 6:05 PM UTC and was resolved by Apr 22, 2025 6:35 PM UTC, lasting about 30 minutes. DigitalOcean posted two status updates during the incident as engineers investigated and then confirmed resolution. The service returned to normal after the issue was addressed.

DigitalOcean reported a degraded operation affecting MongoDB 8 database creation in the AMS3 region, causing errors for users attempting to create new Mongo 8 databases in AMS3. The issue was acknowledged on DigitalOcean's status page with an Investigating note, listing the Managed Databases / AMS3 component. The outage began on 4:55 PM UTC and lasted into the next day as engineers investigated and worked toward a fix.

The AMS3 component returned to a healthy state and the status was updated to up at 1:05 AM UTC. The incident lasted roughly 8 hours and 10 minutes and appears to have been limited to the AMS3 region, with no broader global impact reported.

DigitalOcean reported a regional networking issue in the LON1 region that could cause connectivity loss to resources hosted there. The outage began at 12:17 PM UTC and was resolved by 2:00 PM UTC, lasting about 1 hour and 43 minutes. The status page listed the incident as Investigating at 12:20 PM UTC and included two updates before restoration. The impact was limited to the LON1 region and affected Networking / LON1 components.

DigitalOcean's GenAI Platform experienced degraded performance starting at 9:10 PM UTC. Users attempting to create agents using certain language models, specifically OpenAI and Llama 3.1 8B, encountered errors and were shown an agreement checkbox that would fail when accepted, preventing agent creation.