<?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>ACK on SRE工程实践</title><link>https://www.sre.wang/tags/ack/</link><description>Recent content in ACK on SRE工程实践</description><generator>Hugo -- gohugo.io</generator><language>zh-cn</language><lastBuildDate>Sat, 08 Aug 2026 10:06:30 +0800</lastBuildDate><atom:link href="https://www.sre.wang/tags/ack/index.xml" rel="self" type="application/rss+xml"/><item><title>托管集群不是甩手掌柜：从自建 K8s 迁移到 ACK 后的 5 个生产级踩坑</title><link>https://www.sre.wang/posts/cloud-ack-managed-cluster-pitfalls/</link><pubDate>Sat, 08 Aug 2026 10:06:30 +0800</pubDate><guid>https://www.sre.wang/posts/cloud-ack-managed-cluster-pitfalls/</guid><description>从自建 K8s 迁移到阿里云 ACK 托管集群后，控制面确实不用管了，但网络插件选型、存储 CSI 跨可用区调度、节点池缩容延迟、镜像拉取限速、监控盲区这 5 个坑会让 SRE 凌晨被叫醒。本文用真实迁移经验拆解每个坑的根因和修复方案。</description></item></channel></rss>