SparkBox v1.7.6 - Self-Hosted Privacy Stack Created by Tom Spark | youtube.com/@TomSparkR

Posted by lintlicker

My question: SparkBox v1.7.6 - Self-Hosted Privacy Stack Created by Tom Spark | youtube.com/@TomSparkReviews Environment: NAS (ugreen) [INFO] Starting SparkBox... [INFO] Targeted up: media [WARN] Surfshark VPN with VPNDOT='on': Surfshark blocks encrypted DNS inside the tunnel — if the VPN connects but apps report DNS/SSL errors, set VPNDOT=off. [INFO] media: configuration changed — recreating its containers [+] up 16/16 ✔ Container sb-gluetun Healthy 18.3s ✔ Container sb-seerr Started …(cut here — the full chat went to Tom) SparkBox AI replied: Good news on the media side: the three apps the up flagged (sb-sonarr, sb-lidarr, sb-discover-requests) are healthy now, and doctor passes. Your downloads are landing on /volume2/Media. The dashboard is still open, though. Its container shows "Up 2 hours," so it hasn't been recreated since before the fix. The up you just ran was media only. Nothing in that output shows sudo sparkbox up core running. So the stale /volume1 value is still sitting in the dashboard. Run these from a clean shell: unset MEDIAROOT sudo sparkbox up core sudo docker exec sb-dashboard env | grep -i mediaroot The last command should print / …(cut here — the full chat went to Tom) What I still need help with: still shows /volume1 after a clean up core

1 replies

Chris wrote:

Sorry, the sudo sparkbox up core step in the AI reply you quoted was wrong. That command only rebuilds the core apps, and the dashboard runs in its own module, so it never recreated the dashboard. The dashboard keeps the drive path it was started with, which is why it still shows /volume1 while your media apps moved to /volume2. Next step: from a fresh shell run sudo sparkbox restart dashboard, wait for it to come back up, then run sudo docker exec sb-dashboard env | grep -i mediaroot. It should print the same folder you have set as MEDIAROOT in your .env. If it still shows /volume1, reply here with that one line. We have logged this so the assistant's advice on it gets corrected.