<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>Velantis</title>
  <link>https://lisdodde.cfd/</link>
  <atom:link href="https://lisdodde.cfd/feed.xml" rel="self" type="application/rss+xml"/>
  <description>Notes on Kubernetes, Postgres, platform engineering, and on-call.</description>
  <language>en-us</language>
  <lastBuildDate>Tue, 06 May 2026 12:00:00 +0000</lastBuildDate>
  <image>
    <url>https://lisdodde.cfd/assets/og-preview.png</url>
    <title>Velantis</title>
    <link>https://lisdodde.cfd/</link>
  </image>

  <item>
    <title>Kubernetes resource limits are not what I thought they were</title>
    <link>https://lisdodde.cfd/blog/k8s-resource-limits.html</link>
    <guid>https://lisdodde.cfd/blog/k8s-resource-limits.html</guid>
    <pubDate>Mon, 14 Apr 2026 09:00:00 +0000</pubDate>
    <description>CPU throttling in Kubernetes is subtler than the docs let on. Here's what actually happens when you set limits.cpu and why your p99 latency might be lying to you.</description>
  </item>

  <item>
    <title>Six months with autovacuum: what I got wrong initially</title>
    <link>https://lisdodde.cfd/blog/postgres-autovacuum.html</link>
    <guid>https://lisdodde.cfd/blog/postgres-autovacuum.html</guid>
    <pubDate>Tue, 03 Mar 2026 09:00:00 +0000</pubDate>
    <description>Postgres autovacuum tuning isn't hard, but it takes longer to validate than you expect. Here's what I learned running a 400GB database through several rounds of adjustments.</description>
  </item>

  <item>
    <title>On-call at 3am: a personal retrospective</title>
    <link>https://lisdodde.cfd/blog/oncall-notes.html</link>
    <guid>https://lisdodde.cfd/blog/oncall-notes.html</guid>
    <pubDate>Wed, 21 Jan 2026 09:00:00 +0000</pubDate>
    <description>After four years on rotating on-call, some things I've learned about alert design, runbooks, and keeping incidents from ruining your sleep.</description>
  </item>
</channel>
</rss>
