Service worker cache now fully clears on server ID mismatch
Posted by tomspark
v1.5.129 fixes a gap in v1.5.128's cache invalidation. If you're running SparkBox over HTTPS (common with reverse proxies like NPM), the service worker could keep serving stale resources even after localStorage was cleared on a server ID mismatch. We now unregister all service workers and delete all CacheStorage entries before reload, so you get a truly clean slate. This is a strict superset of the previous patch — HTTP installs see no change (caches require secure contexts anyway), while HTTPS setups gain full cache invalidation. The test suite now verifies CacheStorage clearing on secure contexts, and all three test scenarios pass on our UGREEN. Head to your Updates tab to pull v1.5.129. If you hit any issues, post here in d/sparkbox. --- Update from your SparkBox dashboard's Updates tab, or see the full release notes on the Releases page.