<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Production Pitfalls on SRE Engineering Practice</title><link>https://www.sre.wang/en/tags/production-pitfalls/</link><description>Recent content in Production Pitfalls on SRE Engineering Practice</description><generator>Hugo -- gohugo.io</generator><language>en</language><lastBuildDate>Sat, 08 Aug 2026 10:38:00 +0800</lastBuildDate><atom:link href="https://www.sre.wang/en/tags/production-pitfalls/index.xml" rel="self" type="application/rss+xml"/><item><title>Managed Clusters Aren't Set-and-Forget: 5 Production Pitfalls After Migrating from Self-Hosted K8s to ACK</title><link>https://www.sre.wang/en/posts/cloud-ack-managed-cluster-pitfalls/</link><pubDate>Sat, 08 Aug 2026 10:38:00 +0800</pubDate><guid>https://www.sre.wang/en/posts/cloud-ack-managed-cluster-pitfalls/</guid><description>After migrating from self-hosted K8s to Alibaba Cloud ACK managed cluster, the control plane is indeed managed for you, but these 5 pitfalls will wake SREs at 3 AM: network plugin selection, CSI cross-zone storage scheduling, Cluster Autoscaler scale-down delay, image pull rate limiting, and control plane observability gaps.</description></item></channel></rss>