Rails Daily: Making Rails Safe for Ractors
A cluster of pull requests this week targets Ractor safety across Rails, fixing unshareable procs, deadlocks, and frozen reflection caches so apps can boot and query under strict concurrency modes. Separately, PostgreSQL gets two reliability fixes around advisory locks and referential integrity triggers.
Duration: PT2M18S
Episode overview
This episode is a short developer briefing from Rails Daily.
It explains recent repository work in plain language.
- Show: Rails Daily
- Published: 2026-09-04T06:02:02Z
- Audio duration: PT2M18S
Transcript excerpt
This excerpt keeps the crawler page concise. Listen to the episode or use the RSS feed for the full update.
Good morning, it's September fourth, twenty twenty-six, and this is Rails Daily.
The biggest signal in today's activity: Rails is hardening itself for Ractors, and it's touching more corners of the framework than you'd expect.
Start with boot time. PR fifty-eight thousand six hundred fifty-five, from Edouard-chin, fixes a callback registered during initialization that captured an outer variable and broke apps when Ractor safety mode is set to raise. PR fifty-eight thousand six hundred fifty-nine builds on that same pattern, extracting…
That same theme shows up deeper in Active Record. Kyle Fazzari's fix, PR fifty-eight thousand six hundred fifty-one, removes a Ractor-safe storage call that was actually causing a deadlock in schema loading — a subtle case where a supposedly safe primitive triggered reentrant calls into itself. And Edouard-chin's PR…
Second theme: PostgreSQL correctness. PR fifty-eight thousand six hundred fifty-eight addresses advisory locks silently surviving a dropped connection since Rails 7 — a correctness regression that could let two processes hold the same lock at once. And PR fifty-eight thousand six hundred thirty-five, now merged,…
Smaller but worth noting:…