# AWS CNI lessons from a Production Outage

DevFeed: [AWS CNI lessons from a Production Outage](<https://devfeed.tech/articles/aws-cni-lessons-from-a-production-outage-5016.md>)

Original publisher: [Read original article](<https://neon.com/blog/aws-cni-lessons-from-a-production-outage>)

Author: Em Sharnoff

Published: 2025-06-06T20:57:44Z

Content type: article

Language: en

Sources: [Blog -- Neon Docs](<https://devfeed.tech/sources/blog-neon-docs.md>)

Topics: [incident](<https://devfeed.tech/topics/incident.md>)

Tags: [aws](<https://devfeed.tech/tags/aws.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [outage](<https://devfeed.tech/tags/outage.md>), [post-mortem](<https://devfeed.tech/tags/post-mortem.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [production](<https://devfeed.tech/tags/production.md>), [serverless](<https://devfeed.tech/tags/serverless.md>), [vpc](<https://devfeed.tech/tags/vpc.md>)

## AI overview

A post-mortem of Neon's AWS us-east-1 outage examines IP-allocation failures in the AWS VPC CNI plugin. It describes how failed idle-compute termination and subsequent CNI configuration changes contributed to subnet IP exhaustion and recovery problems.

## Source excerpt

_This post is the last in a series discussing the Neon outages on 2025 05 16 and 2025 05 19 in our AWS us east 1 region. In this post, we cover the IP allocation failures that persisted through the majority of the disruption. For further details, read our_ top level Post Mortem h...