I am using ubuntu 26.04 LTS, I have NordVPN set up with wireguard. according to Tom AI " sparkbox Doctor" I am getting VPN Cycling. I have tried the recommened MTU changes from 1280 to 1200 to 1220.... it does not to seem to solve the issue. I had installed NordVPN with the snap install. then installed the sparkbox. I have since un-installed NordVPN, and have power cycled the computer. every time I changed the MTU, I restarted the media module. any suggestions besides changing VPN providers.
6 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:
Hi Bozz64, Chris here (SparkBox's support assistant, an AI). You don't need a different VPN company, and you can stop changing the packet size: three sizes with no change rules that out. What's most likely going on SparkBox's VPN picks a NordVPN server from a saved list. On a new box that list comes from a copy built into the VPN app, and for NordVPN that copy dates from early 2024. NordVPN has changed many of its servers' keys since then. When the VPN picks one of those servers it connects, but nothing gets through, so it fails its own check, drops the connection and tries again. That is the cycling the checkup saw. Our checkup never looked at how old that list is, so it sent you down the packet-size steps instead. Sorry about that. The checkup now names this first; that change goes out in the next SparkBox update. One step (your box already has this command) In the terminal on the mini PC: sudo sparkbox refresh-vpn-servers It downloads a current server list straight from NordVPN (outside the VPN), then restarts the VPN and the download apps that use it. Jellyfin keeps running. Leave the packet size where it is for now. Then check that it stays connected About 5 minutes after it finishes, paste this: sudo docker logs sb-gluetun --since 5m 2&1 | grep -c "failed to pass the healthcheck" It prints one number. 0 means the VPN stayed connected the whole time. Any other number: run sudo sparkbox doctor and post just the [VPN Tunnel] part here, leaving out any keys or addresses.
Bozz64 wrote:
well, it seems I don't have something right. dahboard shows VPN is ok. but when running the sudo command, I got this reply Refresh the VPN's server list Downloads a current copy of your VPN provider's server list. Fixes the case where the VPN says it's connected but nothing actually goes through it, because the saved list points at servers whose keys your provider has since changed. [ERROR] VPNPROVIDER in .env doesn't look like a provider name: 'NordVPN' Fix it in the dashboard (Settings → VPN) or with: sudo sparkbox set-env VPNPROVIDER surfshark
Chris wrote:
Sorry — I treated the stale server-list explanation as the cause before checking what your command actually returned. This error shows the refresh command stopped at the saved provider name: it expects the lowercase value nordvpn, not NordVPN; it has not refreshed anything yet. The suggested surfshark value would switch providers, so please don't use it. On the Ubuntu terminal, run sudo sparkbox set-env VPNPROVIDER nordvpn, then retry sudo sparkbox refresh-vpn-servers. Please reply with whether the refresh completes or the exact error (leave out any keys or addresses). The dashboard's green VPN status doesn't establish whether traffic is passing, so we'll check that only after this command gets past the provider-name error. I've corrected what I know about this failure so the same unsupported diagnosis isn't repeated for someone else.
Bozz64 wrote:
ok, I used the: sudo sparkbox set-env VPNPROVIDER nordvpn command. NOTE!!! NordVPN didn't work. has to be all small letters. had to : sudo sparkbox restart media. then the :sudo sparkbox refresh-vpn-servers worked correctly the sudo docker logs returned a "0" so i would say it might be fixed now. thank you
Chris wrote:
Thanks for confirming. The provider value is case-sensitive, and your successful refresh followed by zero health-check failures over that five-minute check is a good sign that the cycling has stopped. If it starts cycling again, let me know and we can take the next diagnostic step; no need to change the MTU or switch VPN providers right now.