<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ecr on ferkakta.dev</title><link>https://ferkakta.dev/tags/ecr/</link><description>Recent content in Ecr on ferkakta.dev</description><generator>Hugo</generator><language>en-US</language><copyright>Copyright fizz.</copyright><lastBuildDate>Tue, 17 Feb 2026 09:00:00 -0600</lastBuildDate><atom:link href="https://ferkakta.dev/tags/ecr/index.xml" rel="self" type="application/rss+xml"/><item><title>Your terraform apply is silently rolling back your container images</title><link>https://ferkakta.dev/state-aware-ecr-image-resolution-github-actions/</link><pubDate>Tue, 17 Feb 2026 09:00:00 -0600</pubDate><guid>https://ferkakta.dev/state-aware-ecr-image-resolution-github-actions/</guid><description>&lt;p&gt;Every &amp;ldquo;deploy to EKS with GitHub Actions&amp;rdquo; tutorial solves the same problem: build an image, push to ECR, deploy it. The tutorial ends at &amp;ldquo;your pod is running.&amp;rdquo; Nobody talks about day two.&lt;/p&gt;
&lt;h2 id="the-silent-rollback"&gt;The silent rollback&lt;/h2&gt;
&lt;p&gt;Day two: you have a running EKS cluster with three services per tenant. You need to change an IAM policy. You open a PR, touch one line of Terraform, run &lt;code&gt;terraform apply&lt;/code&gt;.&lt;/p&gt;
&lt;p&gt;Your IAM policy updates. Your container images also update — to whatever was hardcoded in &lt;code&gt;variables.tf&lt;/code&gt; as the default. That default was correct three months ago. Your services just rolled back to a three-month-old image and nobody noticed because the deployment succeeded.&lt;/p&gt;</description></item></channel></rss>