Amazon Web Services (AWS) faced a significant disruption impacting its on-demand cloud computing service. CloudFront users experienced widespread disruptions with reports of encountering 5xx errors across the platform.
The problem commenced at 12:45 am PDT today. AWS Support acknowledged the issue in a service update at 1:57 am PDT, stating they were investigating increased 5xx errors for CloudFront customers using VPC Origins connectivity.
Following complaints from users about the lack of updates, AWS Support reassured them that their team was actively working on resolving the issue and updating the Health Dashboard with new information.
Downdetector recorded over 350 user reports concerning AWS service issues. In an update at 3:18 am PDT, AWS indicated they believed they had identified the root cause of the problem related to a packet processing subsystem routing requests from CloudFront’s edge locations to resources within customer VPCs.
AWS recommended customers temporarily change their origin type to address the errors. The service, provided by online retailer Amazon, plays a critical role in supporting online applications and internet infrastructure for companies like Netflix, BMW Group, Booking.com, Epic Games, and Canva.
AWS confirmed the global impact of the issues and mentioned testing a mitigation strategy to resolve them, which would be deployed in phases. Last October, AWS experienced a significant outage affecting services like Snapchat, Fortnite, Roblox, McDonald’s, and Canva, with over 6.5 million global reports received by Down Detector.
Amazon later announced the resolution of the issue that occurred at the core of its cloud computing operations in North Virginia. AWS, a top cloud computing service provider competing with Google and Microsoft, offers computing power, data storage, and digital services to numerous companies and institutions.