Dashboard Storage tile stuck red despite mount being fully healthy
Posted by noliver2761
Setup: UGREEN NAS, MEDIAROOT=/volume1/data, media module (Radarr/Sonarr/qBittorrent/gluetun). Timeline: Ran an update today, Storage tile in the media pipeline widget went red with "folder where finished movies/shows should go can't be reached." Root-caused a separate real issue along the way — qBittorrent's WebUI password had regenerated, Radarr/Sonarr were stuck retrying the old one and got IP-banned (403 "too many failed authentication attempts"). Fixed by restarting sb-gluetun to clear the ban and updating the qBit password in both arr apps' download client settings. VPN and Download App tiles went green after that. Storage tile stayed red the whole time, before and after the qBit fix, and after sparkbox restart media. Verified healthy, manually: grep MEDIAROOT /opt/sparkbox/.env → /volume1/data, correct. ls -la /volume1/data → folders present and populated (downloads, media/movies, media/tv), correct ownership. Radarr root folder /data/media/movies, Sonarr root folder /data/media/tv — both correctly configured, matches disk layout. docker exec sb-radarr touch /data/media/movies/test.txt → succeeded, cleaned up. docker exec sb-sonarr touch /data/media/tv/test.txt → succeeded, cleaned up. df -h /volume1/data → 6.7T free, 9% used, mount is live. Jellyfin plays existing library fine from the same MEDIAROOT the whole time. Conclusion: every actual write/read path works. The Storage check in the dashboard's media pipeline widget appears to be reading stale state, or checking the wrong thing, and isn't clearing even after sparkbox restart media recreated the containers. Possibly related to the gluetun/media restart cycle timing, but the check should've re-evaluated cleanly after that.
1 replies
tomspark wrote:
Excellent debugging—you've ruled out the mount itself. This looks like the dashboard's storage-check is reading stale state or hitting a cache issue. Two quick checks: clear your browser cache (Ctrl+Shift+Del) and reload, and if it persists, check the dashboard logs: docker logs sb-dashboard | grep -i storage might show what the check is actually testing. If those don't clear it, this is filed as a bug and being tracked—you've got the workaround in the meantime (everything genuinely works, just the widget hasn't caught up).