Infrastructure Resilience Briefing — December 7, 2021 AWS US-EAST-1 Event
An AWS network congestion event in US-EAST-1 disrupted API and console access for hours; operators must document impact and confirm if workloads qualify for EC2 or EBS service credits under AWS’s SLA.
Executive briefing: AWS reported a widespread service event in US-EAST-1 caused by unexpected congestion on internal network devices, degrading APIs, consoles, and dependent services for several hours.1 Customers with EC2 or EBS workloads that experienced availability below SLA thresholds should file credit requests once logs and monitoring confirm the duration and scope of impact.2
Regional impact
- Service reachability: Control-plane calls (including authentication and monitoring APIs) intermittently failed in US-EAST-1, delaying deployments and incident response playbooks.1
- Cross-service blast radius: Dependent services such as Connect, DynamoDB, and EventBridge reported downstream impact because their management traffic rides the same affected backbone.1
- Data-plane protection: Most running EC2 instances continued to serve traffic, but new launches and elastic operations intermittently failed until network capacity was restored.
SLA and credit posture
- Confirm workload availability by comparing RUM, load-balancer, and application metrics to the EC2 SLA targets (99.99% for multi-AZ deployments and 99.5% for single-AZ deployments).2
- Open a support ticket with timestamps and service impact within the reporting window; AWS requires a credit request for each affected account and region.
- Update runbooks to include alternate-region failover validation when control-plane calls stall, reflecting lessons learned from the December 2021 congestion event.