A launchd agent configured with KeepAlive=true runs a wrapper script that polls for a dependency (typically the Docker daemon) and calls exit 1 if it never appears. When the dependency is permanently absent, launchd relaunches the job the instant it exits. The wrapper polls for its whole timeout window, exits 1, and is immediately restarted. The result is a silent permanent respawn loop burning CPU and battery. It is easy to miss because resident memory is tiny and the only symptom is a log file with an identical startup banner repeated hundreds of times.

When running the app's Docker web container on a non-default port (e.g. 3003 instead of 3000), Spotify OAuth login fails with "Failed to authenticate". Three separate issues compound: 1) CORS blocks the OPTIONS preflight because the new origin isn't in FastAPI's allow_origins list, 2) The .env file overrides docker-compose defaults for SPOTIFY_REDIRECT_URI causing a redirect_uri mismatch during Spotify token exchange (frontend sends port 3003, backend sends port 3000), 3) Missing database columns (users.youtube_channel_id) cause a 500 error after successful token exchange because alembic migrations were incomplete.