<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Sandboxing</title><link>https://blog.ast.moe/en/tags/sandboxing/</link><description>Recent posts: Gorimuchu</description><generator>Hugo</generator><item><title>How to Clean Up a Sandbox When Its Supervisor Dies</title><link>https://blog.ast.moe/en/blog/2026-11-15/</link><pubDate>Sun, 15 Nov 2026 22:21:26 +0900</pubDate><guid>https://blog.ast.moe/en/blog/2026-11-15/</guid><description>&lt;p&gt;The current focus in izanagi-tempest, a system for evaluating AI agents in isolation, is resource cleanup after a supervisor terminates abnormally. Stopping the workload is not sufficient: if the parent and supervisor delete the same resources concurrently, or if cleanup failure erases the original fault from the record, both isolation and evaluation remain problematic.&lt;/p&gt;
&lt;p&gt;While working through these problems, I looked for how other systems handle them. Papers on failure detection, crash-only design, stale owners, and fault injection into recovery code addressed similar difficulties. containerd also has a cleanup path for when a supervisory process dies. These sources helped me organize the conditions our Linux probe needs to check.&lt;/p&gt;</description></item></channel></rss>