Our engineering team was alerted to increased error rates and began to investigate at 9:31 AM ET. Investigation led to some problematic systems configuration with our database. The configuration was rectified which led to partial incident resolution for many npm users and customers, at 9:54 am ET. Follow-up SRE work led to a full restoration of service at 11:06 am ET.
We apologize for the inconvenience and are working on post-incident steps to prevent future such issues.
Posted Sep 26, 2019 - 15:46 UTC
This incident affected: Registry Reads, Package publishing, and npm Enterprise.