Navidrome Daily: Cleaning Up After 0.63.0
A cluster of fixes addresses fallout from the 0.63.0 release—a 32-bit startup crash and a broken database migration—while new CI guards aim to stop both classes of bugs from recurring.
Duration: PT2M23S
Episode overview
This episode is a short developer briefing from Navidrome Daily.
It explains recent repository work in plain language.
- Show: Navidrome Daily
- Published: 2026-07-10T10:00:45Z
- Audio duration: PT2M23S
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 July 10th, 2026, and this is Navidrome Daily.
Today's story is really about the release process catching up with itself after 0.63.0 shipped with two serious defects—and the team closing the gaps that let them through.
Start with the 32-bit crash. PR 5739 from Deluan fixes a segfault and illegal instruction error hitting armv7, v6, v5, and 386 standalone binaries. The real root cause took real digging: once the binary's code section crossed 16 megabytes in 0.63.0, GNU's linker started emitting corrupt relocations for libatomic's…
Second, PR 5748 fixes a database problem: some 0.63.0 databases have a migration marked as applied in the migration log, but its actual schema change never took effect, leaving bpm and bit-depth columns non-nullable when the scanner now expects to write nulls. That breaks every scan. The fix is a new migration that…
Both bugs share a root cause: something recorded as done that wasn't actually true — a build tag that didn't propagate, a migration that didn't apply. That theme shows up again in PR 5750, which adds a CI check to catch out-of-order database migrations before they merge, closing exactly the kind of gap that let…
Elsewhere, PR…