Hi,
Is there a way to securely connect to my device running Torizon like a vnc server that you recommend ? I need to be able to connect remotely with more than ssh.
Thanks!
Hi @jaimeibk,
So by default we really only have SSH as a mechanism to securely access a Torizon device. For further options you would need to build in additional tooling into the Torizon image yourself.
Theoretically you might be able to utilize some container that hosts a vnc server or something similar on the device to add the capability to your Torizon system. Though I don’t believe anyone here has tried this approach before so I can’t comment how well it would work out.
Can I ask why you need an alternative to SSH? I can bring it up with the team to see if it’s a feature we could either add or maybe at least document in a how-to guide.
Best Regards,
Jeremias
Hi, @jeremias.tx
Our support team sometimes need to access the system remotely to help our customers. With ssh, they would only get access to the shell, and maybe they would need to access the UI, or something else system-wise.
Best!
Jaime
I understand now. Given that your UI would most likely run in a container then the approach of having a vnc server or something similar running in the container or maybe a separate debug container would make sense.
In fact some quick research shows blogs (https://qxf2.com/blog/view-docker-container-display-using-vnc-viewer/) that do just that.
I’ll share this use case with the team and see if we could at least maybe document it as a how-to.
Best Regards,
Jeremias
Hi, @jeremias.tx
–I’ll share this use case with the team and see if we could at least maybe document it as a how-to.
This would be really helpful. Please update me as this unfolds.
Jaime
Hi, @jeremias.tx
I have been giving this some thought. The vnc server should then run inside the UI container? How would the debug container scenario work, though?
Hi @jaimeibk,
Just thinking conceptually here, it would be ideal to not have such debug/support interfaces in your end container. The idea and speaking generally here would be to either have a separate container that shares volumes or information with the container you want to debug/interact with.
Or you’d have two identical containers one with just the end application and another with the application plus debugging tools. Then in your case a customer might be able to spin up the debug container, which has tools that would allow your team to debug/access it remotely.
Again I’m just speaking conceptually and generally here.
Best Regards,
Jeremias