<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Mysql on Luther Monson</title>
    <link>https://luth.io/tags/mysql/</link>
    <description>Recent content in Mysql on Luther Monson</description>
    <generator>Hugo</generator>
    <language>en</language>
    <lastBuildDate>Thu, 27 Aug 2026 20:30:00 -0700</lastBuildDate>
    <atom:link href="https://luth.io/tags/mysql/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Security and Performance Hardening of a Rust Database Proxy</title>
      <link>https://luth.io/blog/2026/08/hardening-rust-database-proxy/</link>
      <pubDate>Thu, 27 Aug 2026 20:30:00 -0700</pubDate>
      <guid>https://luth.io/blog/2026/08/hardening-rust-database-proxy/</guid>
      <description>&lt;p&gt;&lt;a href=&#34;https://github.com/ephpm/litewire&#34;&gt;litewire&lt;/a&gt; is the piece of ePHPm that lets PHP apps talk MySQL to a SQLite file. I&#xA;wrote about &lt;a href=&#34;https://luth.io/blog/2026/07/sqlite-production-database-protocol-translation/&#34;&gt;how it works&lt;/a&gt; and about &lt;a href=&#34;https://luth.io/blog/2026/08/faking-mysql-quirks-wordpress/&#34;&gt;the MySQL quirks&#xA;WordPress depends on&lt;/a&gt;. This one is the less visible half of August:&lt;/p&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;an audit pass on litewire&lt;/li&gt;&#xA;&lt;li&gt;a multi-tenant mode that turned a database proxy into a security boundary&lt;/li&gt;&#xA;&lt;li&gt;a performance regression I shipped and then had to go find&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;p&gt;It&amp;rsquo;s written in Rust. Memory safety is supposed to be the part I don&amp;rsquo;t have to think about, and it mostly is. The&#xA;use-after-free in this story was in a dependency. The worst bug wasn&amp;rsquo;t a memory bug at all.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Faking MySQL&#39;s Quirks Well Enough to Fool WordPress</title>
      <link>https://luth.io/blog/2026/08/faking-mysql-quirks-wordpress/</link>
      <pubDate>Sun, 16 Aug 2026 15:50:00 -0700</pubDate>
      <guid>https://luth.io/blog/2026/08/faking-mysql-quirks-wordpress/</guid>
      <description>&lt;p&gt;Last month I wrote about &lt;a href=&#34;https://luth.io/blog/2026/07/sqlite-production-database-protocol-translation/&#34;&gt;running SQLite behind a MySQL socket&lt;/a&gt;&#xA;with &lt;a href=&#34;https://github.com/ephpm/litewire&#34;&gt;litewire&lt;/a&gt;. I ended on &amp;ldquo;passing an install suite is not the same thing as&#xA;behaving like MySQL&amp;rdquo;. That turned out to be accurate. On August 10th a few things happened at once:&lt;/p&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;I put a real WordPress 7.0.3 site through a benchmark on ePHPm.&lt;/li&gt;&#xA;&lt;li&gt;I A/B tested the new in-process database bridge against the wire path.&lt;/li&gt;&#xA;&lt;li&gt;I started building &lt;a href=&#34;https://github.com/ephpm/db-wordpress&#34;&gt;ephpm/db-wordpress&lt;/a&gt;, a &lt;code&gt;wpdb&lt;/code&gt; drop-in that talks to the&#xA;database in-process instead of over a socket.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;p&gt;Between them they produced six litewire issues in about three hours. None of them broke the install. All of them made&#xA;WordPress quietly do the wrong thing.&lt;/p&gt;</description>
    </item>
    <item>
      <title>Database Connection Pooling with Smart Reset</title>
      <link>https://luth.io/blog/2026/08/db-connection-pooling-smart-reset/</link>
      <pubDate>Sun, 09 Aug 2026 16:25:00 -0700</pubDate>
      <guid>https://luth.io/blog/2026/08/db-connection-pooling-smart-reset/</guid>
      <description>&lt;p&gt;PHP&amp;rsquo;s share-nothing model is one of its best properties. It has one expensive side effect: a request can&amp;rsquo;t keep a&#xA;database connection for the next request. php-fpm gives you &lt;code&gt;PDO::ATTR_PERSISTENT&lt;/code&gt; and &lt;code&gt;mysqli&lt;/code&gt;&amp;rsquo;s &lt;code&gt;p:&lt;/code&gt; hosts. But&#xA;those are per-worker-process. PDO&amp;rsquo;s keep whatever session state the last request left behind. mysqli at least resets&#xA;the session with &lt;code&gt;COM_CHANGE_USER&lt;/code&gt; on reuse. And almost nobody turns them on. So in practice every PHP request does a&#xA;TCP connect, a MySQL handshake and an auth exchange. Then it maybe sends some &lt;code&gt;SET&lt;/code&gt;s, runs its queries and sends&#xA;&lt;code&gt;COM_QUIT&lt;/code&gt;. Multiply that by your request rate and MySQL spends a real share of its time authenticating clients it&#xA;just said goodbye to.&lt;/p&gt;</description>
    </item>
    <item>
      <title>SQLite as a Production Database via Protocol Translation</title>
      <link>https://luth.io/blog/2026/07/sqlite-production-database-protocol-translation/</link>
      <pubDate>Sat, 25 Jul 2026 20:40:00 -0700</pubDate>
      <guid>https://luth.io/blog/2026/07/sqlite-production-database-protocol-translation/</guid>
      <description>&lt;p&gt;Most of the PHP sites I&amp;rsquo;d put on &lt;a href=&#34;https://github.com/ephpm/ephpm&#34;&gt;ePHPm&lt;/a&gt; don&amp;rsquo;t need a MySQL server. A brochure&#xA;WordPress, an internal Laravel tool, a preview environment that lives for an afternoon. What they do need is something&#xA;that &lt;em&gt;looks&lt;/em&gt; like MySQL, because &lt;code&gt;mysqli&lt;/code&gt; and &lt;code&gt;pdo_mysql&lt;/code&gt; are the only database drivers their code has ever been&#xA;pointed at. I mentioned this in the &lt;a href=&#34;https://luth.io/blog/2026/06/ephpm-pragmatism/&#34;&gt;ePHPm pragmatism post&lt;/a&gt; as a one-paragraph&#xA;&amp;ldquo;because&amp;rdquo;. This is the longer version, now that WordPress 7.0 and Laravel 13 both install and run end to end on it.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
