Reset operations no longer leave your box half-broken
Posted by tomspark
SparkBox 1.5.8 fixes a critical bug in Soft Reset, Reset to Defaults, and Factory Reset where the reset script would die partway through because it was running inside the dashboard container — the same container that docker compose down kills as its first step. The wipe would stall at 'Stopping all SparkBox services...' and leave your stack corrupted. I've moved all three reset flows to a short-lived privileged helper container (using the same pattern self-update already uses) that mounts the docker socket and SBROOT directly, so the work completes even after the dashboard shuts down. If you've been holding off on a factory reset because a previous attempt went sideways, you're good to go now. - Fixed: Soft Reset now completes reliably - Fixed: Reset to Defaults no longer stalls mid-operation - Fixed: Factory Reset carries through to completion Head to the Updates tab in your dashboard to pull 1.5.8. If you hit any issues during reset, post in d/sparkbox and I'll dig in. --- Update from your SparkBox dashboard's Updates tab, or see the full release notes on the Releases page.