Datadog Integrations Outage History

The Datadog integrations installation pipeline experienced an outage that prevented users from installing or updating integrations through the agent command. The issue affected new installations and updates of extra, marketplace, and core integrations, while existing installed integrations continued to function normally without data loss. The problem was identified, fixed, and resolved within 1.2 hours.

Datadog's Microsoft Teams integration experienced a major outage preventing notifications from being delivered to Teams channels for 3 hours. The incident affected the Microsoft Teams component of Datadog Integrations. The issue was resolved after investigation and service was restored.

Datadog's Cloudflare integration experienced a 2.5-hour outage where API metrics were delayed, causing gaps in Cloudflare dashboards and affecting the Azure Monitor component. The issue was identified as originating from Cloudflare's side, and Datadog resolved it by backfilling the missing metrics data.

An issue with Anthropic's API caused the Datadog Anthropic Integration to experience delayed or missing metrics collection, affecting usage metrics including token usage, request volume, and latency. The incident lasted 14.6 hours from March 3rd at 6:30 PM UTC until resolution on March 4th. Datadog coordinated with Anthropic to resolve the underlying API issue.

A partial Oracle Cloud Infrastructure (OCI) outage in the us-ashburn region caused Datadog's OCI Integration metrics to be delayed or missing from monitoring graphs for nearly 50 hours. The incident affected Oracle Cloud Infrastructure components and experienced multiple recurrences during the outage period. The issue was resolved once Oracle fixed their underlying infrastructure problems, restoring normal metric collection and display in Datadog.

A partial Oracle Cloud Infrastructure (OCI) outage in the us-ashburn-1 region caused Datadog's OCI Integration metrics to be delayed or missing for 9.4 hours. The outage affected metric collection from 7pm EST March 2nd until 2am EST March 3rd, resulting in gaps or delays in monitoring graphs displaying OCI metrics from that region. The issue was resolved when OCI restored service, allowing metrics to flow back to Datadog normally.