Error: Request error: tunneling socket could not be established, cause=socket hang up


one of my devices Intel NUC has following issue when I try to reboot/restart the device.

Request error: tunneling socket could not be established, cause=socket hang up.

Additionally when trying to Shell into main:
Spawning shell…
Application container must be running for a terminal to be started.
SSH session disconnected

I can only reboot via reboot now (in HostOS shell).

I rebooted and get the same problem. Does anyone know what is going on? Balena daemon corrupted? Thanks


Can you grant support access to this device so that we can take a look?

Hey, sent permission via PM last week. Any news on this?

Hi, some of my other sensors have the same problem now.

On one of the devices I updated balenaOS, since I did not see any other option. It says Update successful, rebooting. Now it is not coming online anymore.

Hello, my first suggestion would be to look into device health checks to see if there is an obvious problem. You can run them from the DIAGNOSTICS tab on the device dashboard. Please let us know how that goes.

Thanks for the info @rquant. Can you please update balenaOS to the latest version?
Relevant docs are here:

We see that check_balenaOS failed because your device is running a version that we pulled back as it had issues. So it’s best for your device(s) to run on the latest reliable balenaOS version.

Let us know if you encounter any issues with the balenaOS update or experience any further issues once the update goes through.

Hello, I did that yesterday on a device which had the same problems.

It says as Status: Update successful, rebooting (yellow text)

Last Online

Online (for 4 hours)

I still cannot reach the app container console or use Public URL to reach the device. Nor does the status change. I can enter the BalenaOS terminal and type reboot now. It will reboot and show the same status again and app container not running.

Let’s take a further look at the device. Could you please grant us support access and send us the device UUID via private message again?

Hi there @rquant,

Thanks for providing those UUIDs and enabling support access. In order to recover the devices I may need to remove & re-pull the images, is that something you are okay with at the moment?

I was able to fully recover the device running balenaOS 2.39, which is now in the process of updating to the latest release. We had not seen this failure case before, but have begun tracking it internally to fix going forward.

The other device dropped offline as I was investigating the issue (there is a different fundamental problem). If you see it come back online, feel free to ping us again and we can continue debugging the issue.

Thank you for reporting these issues!

Hello the device is online again. The reason why its offline sometimes is because its after opening hours of the store.

Hi there,

I was able to get that second device cleaned up and it is now also updating.

Please let us know if you experience further issues!