Elevated error rates on metrics ingestion and API calls

Incident Report for Datadog EU

Resolved

We suffered two incidents with our data ingestion pipelines on Wednesday Dec 18th:
* Elevated error rate for the Agent metrics pipeline from 13:24 - 13:28 UTC. There are no gaps in graphs as the Datadog Agent has retries mechanisms, but false positives notifications might have been sent depending on monitor configurations
* Elevated error rate for the API and APM Statistics pipelines from 14:55 - 15:02 UTC, leading to potentials gaps in graphs if API calls were not retried, and false positive notifications might have been sent depending on monitor configurations

Posted Dec 18, 2019 - 12:34 EST

This incident affected: APM and Metrics and Infra Monitoring.