Hi there I'm having some difficulty setting up my vpn for my media server. When I fill in my vpn details (using PIA for context) my vpn container prints the following error message: ERROR updater settings: VPN provider name is not valid: value is not one of the possible choices: privateinternetaccess must be one of airvpn, cyberghost, expressvpn, fastestvpn, giganews, hidemyass, ipvanish, ivpn, mullvad, nordvpn, perfect privacy, privado, private internet access, privatevpn, protonvpn, purevpn, slickvpn, surfshark, torguard, vpnsecure, vpn unlimited, vyprvpn or windscribe I tried using the "custom" option as my Vpn Provider instead of "Private Internet Access", thinking that's what it was referring to, but still got the same error. I also tried using wireguard instead of openvpn to see if it would make a difference, but it didn't. Running sudo sparkbox doctor also gave the same error message. If it helps, I'm running debian bookworm, and I'm also running a pi-hole. Thanks in advance!
8 replies
Chris wrote:
Got it — this is an automatic acknowledgment from Chris, your AI support representative. Your post is in the support queue, and we’ll get to it as soon as we can. Legend members receive priority replies. If you’d like faster support while helping fund the project’s goals and continued development, consider becoming a Legend backer. <!-- chris-support-queue-ack:v1 --
Chris wrote:
That error is telling you the exact problem: you entered privateinternetaccess (one word), but the system needs private internet access (with spaces) to match the list.\n\nThe valid provider names are lowercase with spaces where needed: private internet access, not privateinternetaccess or Private Internet Access.\n\nTo fix it:\n\n1. Go to your SparkBox dashboard → Settings → Network → VPN\n2. For VPN Provider, select private internet access (exactly as shown, lowercase with spaces)\n3. Enter your PIA credentials\n4. Save and wait for the VPN container to come up\n\nThen run sudo sparkbox doctor again — the VPN section should clear.\n\nIf you still see an error after selecting private internet access from the dropdown (not typing it manually), reply with the exact error and we can dig deeper.
Plex wrote:
This is the only option I have for PIA is shown below. I then refilled my openvpn details and still got the same error. Another thing I tried was editing the provider name in the .env file to this: VPNPROVIDER=private internet access (Note: it was written as privateinternetaccess by default) Then I ran sudo sparkbox restart media, and got a new error: ERROR VPN settings: provider settings: server selection: for VPN service provider private internet access: the country specified is not valid: one or more values is set but there is no possible value available For this I tried leaving VPN Server Country blank, then writing my country of choice, restarting both times, but it still gives the error mentioned above. I should've also mentioned before that I'm running version 1.6.681.
Chris wrote:
Thanks for showing me that screenshot — I see the problem now. Your dashboard dropdown is showing only privateinternetaccess (one word), but you're correctly using private internet access (with spaces) in the .env, which is what gluetun needs. That dashboard value issue is something we need to track down, but good news: your manual .env edit is the right fix. The "country is not valid" error you're now hitting is separate and happens when gluetun can't load PIA's server list. Here's what to try: 1. Verify the container has internet access: Run sudo sparkbox logs media | tail -50 and look for any mention of gluetun failing to fetch updates or server lists. If you see timeout/DNS errors, the VPN container itself can't reach the internet yet. 2. Force a full restart with the new .env: Run sudo sparkbox restart media — this ensures gluetun boots with the corrected private internet access name and tries to load PIA's country list fresh. 3. After restart, run doctor again: sudo sparkbox doctor should now pass the VPN section if gluetun was able to reach its config. If the error persists after the restart, paste the output from sudo sparkbox logs media | grep -i "country\|server\|provider" | tail -20 and we can dig into what gluetun is reporting.
Plex wrote:
I tried running sudo sparkbox logs media | tail -50 and got the following error: Error response from daemon: No such container: media I then ran sudo sparkbox status to see if media was healthy, and it was. I tried restarting it, then re-running the command, but to no avail. I'll attach the logs from running sudo sparkbox restart media in case it might help (Note: I don't know what the first few warnings are about. I've corrected the vpn provider name even though it says otherwise): [WARN] VPNENDPOINTIP is set but VPNPROVIDER='privateinternetaccess', not 'custom'. [WARN] Custom endpoints are only used when VPNPROVIDER=custom — gluetun will otherwise [WARN] ignore the endpoint and connect to random privateinternetaccess servers. [WARN] Set VPNPROVIDER=custom in .env (or the dashboard) to use your endpoint. [INFO] Restarting module: media (recreating to pick up .env changes) WARN[0000] Found orphan containers (sb-pihole, sb-dashboard, sb-npm, sb-portainer, sb-homepage) for this project. If you removed or renamed this service in your compose file, you can run this command with the --remove-orphans flag to clean it up. [+] up 11/11 ✔ Container sb-bazarr Started 59.1s ✔ Container sb-deunhealth Started 53.6s ✘ Container sb-gluetun Error dependency gluetun failed to start 65.5s ✔ Container sb-jellyfin-media Started 52.1s ✔ Container sb-seerr Started 59.1s ✔ Container sb-byparr Recreated 11.1s ✔ Container sb-prowlarr Recreated 11.6s ✔ Container sb-sabnzbd Recreated 11.6s ✔ Container sb-sonarr Recreated 10.2s ✔ Container sb-qbittorrent Recreated 11.6s ✔ Container sb-radarr Recreated 11.6s dependency failed to start: container sb-gluetun is unhealthy [INFO] Reconnecting sb-qbittorrent to the recreated VPN network namespace... [INFO] Reconnecting sb-prowlarr to the recreated VPN network namespace... [INFO] Reconnecting sb-sonarr to the recreated VPN network namespace... [INFO] Reconnecting sb-radarr to the recreated VPN network namespace... [INFO] Reconnecting sb-sabnzbd to the recreated VPN network namespace... [INFO] Reconnecting sb-byparr to the recreated VPN network namespace... [OK] Re-bounced 6 media app(s) onto the live VPN netns. [WARN] Some apps were created but never started (a dependency — usually the VPN tunnel — wasn't healthy in time). Starting them now: [ERROR] These apps are still NOT running: qbittorrent sonarr radarr prowlarr [ERROR] They ride on the VPN tunnel and can't start until it's healthy. [ERROR] Check what's wrong: sudo sparkbox doctor [ERROR] Then bring them up: sudo sparkbox restart media [ERROR] Restart did NOT finish cleanly — some apps may not be running. [ERROR] Check: sudo sparkbox status [ERROR] Diagnose: sudo sparkbox doctor
Chris wrote:
Got it — the fact that docker logs can't find the container tells us exactly what's happening: gluetun failed to start, so the entire media namespace never came up, which is why the other containers can't find it. I see the problem in your restart output. You have two separate VPN settings conflicting: VPNENDPOINTIP is set but VPNPROVIDER='privateinternetaccess' Here's what's happening: - Your .env still says VPNPROVIDER=privateinternetaccess (one word) - But you also set VPNENDPOINTIP, which only works when VPNPROVIDER=custom - When gluetun sees privateinternetaccess + VPNENDPOINTIP together, it rejects the config To fix this, pick ONE path: Option A: Use PIA normally (drop the custom endpoint) 1. Remove any custom endpoint: go to SparkBox Settings → Network → VPN and clear VPN Server 2. Edit .env and change VPNPROVIDER to exactly: private internet access (with spaces) 3. Remove VPNENDPOINTIP from .env if it's there 4. Run: sudo sparkbox restart media 5. After it's up, run: sudo sparkbox doctor Option B: Use a custom endpoint 1. Change .env to VPNPROVIDER=custom (not privateinternetaccess) 2. Keep your VPNENDPOINTIP and add VPNENDPOINTPORT if needed (check your PIA docs) 3. Run: sudo sparkbox restart media Try Option A first — it's the standard PIA setup. If gluetun comes up healthy after restart, the media apps will follow. Let me know what the doctor output shows.
Plex wrote:
The problem has been fixed! I first tried option A, but after restarting it still gave me this error: ERROR VPN settings: provider settings: server selection: for VPN service provider private internet access: the country specified is not valid: one or more values is set but there is no possible value available Then I tried option B, and now gluetun and the vpn is working properly. Interestingly though, I still can't run sudo sparkbox logs media | tail -50. It still gives this error, even after restarting: Error response from daemon: No such container: media I don't know if this is a bug or not. If you want more details to iron out any of these issues, I'd be happy to help, or else thank you very much for your help!
Chris wrote:
Great to hear option B worked! Custom mode with your endpoint is a valid setup, and it sounds like your gluetun is healthy now. On the sudo sparkbox logs media issue — that's an interesting one. The command is looking for a container literally named "media", but in SparkBox, the media services run as individual named containers (sb-gluetun, sb-jellyfin-media, sb-sonarr, etc.), not under a single "media" container. If you want to see the full media stack logs, try: sudo docker-compose -f /opt/sparkbox/docker-compose.yml logs gluetun | tail -50 Or for Jellyfin specifically: sudo docker-compose -f /opt/sparkbox/docker-compose.yml logs jellyfin-media | tail -50 Replacing gluetun or jellyfin-media with whichever service you want to check. That should resolve the "No such container" error. The sudo sparkbox logs wrapper may have a documentation gap here — we'll look into making that clearer. Your VPN is up and running now though, so you're all set! Let us know if anything else comes up.