<?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>蓝绿部署 on SRE工程实践</title><link>https://www.sre.wang/tags/%E8%93%9D%E7%BB%BF%E9%83%A8%E7%BD%B2/</link><description>Recent content in 蓝绿部署 on SRE工程实践</description><image><title>SRE工程实践</title><url>https://www.sre.wang/share-card-square.jpg</url><link>https://www.sre.wang/share-card-square.jpg</link></image><generator>Hugo -- gohugo.io</generator><language>zh-cn</language><lastBuildDate>Sat, 17 Oct 2026 00:05:38 +0800</lastBuildDate><atom:link href="https://www.sre.wang/tags/%E8%93%9D%E7%BB%BF%E9%83%A8%E7%BD%B2/index.xml" rel="self" type="application/rss+xml"/><item><title>切流量切到一半数据库连接池炸了：蓝绿与金丝雀发布的 6 个生产级决策与流量治理实录</title><link>https://www.sre.wang/posts/cicd-blue-green-canary-deployment/</link><pubDate>Sat, 17 Oct 2026 00:05:38 +0800</pubDate><guid>https://www.sre.wang/posts/cicd-blue-green-canary-deployment/</guid><description>蓝绿部署和金丝雀发布不是切个流量那么简单。本文从 K8s 原生 Deployment 的局限讲起，拆解 Argo Rollouts + Istio 的渐进式交付架构，覆盖数据库迁移兼容、流量镜像验证、自动回滚门禁等 6 个生产级决策，附完整可复制的 YAML 配置和性能对比数据。</description></item></channel></rss>