Does Edge DX Remote Control Leave a Trace afer Ending Remote Session?

A user asked about remote control functionality in Edge DX and whether it leaves traces that the security department can pick up. @Member gave an explanation that the client used in the agent is a VNC client, but is secured and connected to other parts of the system via a web-socket. This means that the VNC client cannot be externally connected, thus not leaving traces.

Read the entire ‘Does Edge DX Remote Control Leave Traces?’ thread below:

Hi community folks, i’m searching for more information related to the remote control functionality in Edge DX. It’s based on VNC but does it leave traces, tokens or other stuff that the security dept. don’t like behind when you end the remote session?

as far as I know it doesn’t but @member knows all about it 🙂

Thanks Wouter, if there is also some documentation or statement that would be helping.

Hey there…

So the client (the part in the agent) is a VNC client but it’s secured and customized. Basically it does not listen for external connections, let me elaborate…

When you initiate a remote control connection from the UI, two things happen:

1. The UI opens a secure (HTTPS) web-socket to the tenant (cloud).
2. The tenant sends a message to the device (agent) to initiate a remote control connection.

The way in which the agent gets this message (#2) is the same way it does every other message, either via it’s operations web-socket (which every agent opens upon starting), or via REST polling if web-sockets are not available.

Once the agent gets the command, it too initializes a web-socket to the tenant.

So at this point, both the UI and the agent have opened web-sockets to the tenant. Note these are outbound sockets, so nothing connecting externally to the agent / device.

Once the agent connects it’s web-socket, it spins up an instance of the VNC client, and binds it to the web-socket. So the VNC client cannot be connected to from an external source. Basically it’s only communications are locally, VNC -> EDGE AGENT -> -> Tenant

Once the VNC client has started, the tenant connects the socket from the UI and the socket from the agent together, basically forming a secure tunnel through the tenant.

On the UI side, we use a HTML5 based VNC client (a standard library and lot of our own code) which is designed to work through the web-socket.

So basically, while we use a VNC client on the endpoint, it’s not used like a standard VNC client, and nothing connects to it directly.

Hopefully this makes sense?

Thanks @member that makes a lot of sense!

Continue reading and comment on the thread ‘Does Edge DX Remote Control Leave a Trace afer Ending Remote Session?’.  Not a member? Join Here!