I was wondering how tools like teamviewer work. I’m not using the words server and client because I am not quite sure who is who. So I will talk about local and remote.
I have a tool installed on a remote machine behind firewalls and NAT, so only port 80 is okay for outbound connections. Now the local machine wants to connect to the remote machine. The only way I can image connection to work, is if the remote polls the local machine to check if it wants to connect, and then the remote establishes a connection to the local machine.
Teamviewer uses the same mechanism only with one entity in the middle, so the remote doesn’t have to know the local in advance? So the remote always polls the entity in the internet?
Is that the way this kind of connection works? It seems quite a wast, always having to poll the local or some dealer-webserver. Especially if the connection has to work quickly, the polls must happen every second?
Am I missing something?
If someone is to scared to answer, a simple yes or no does the trick 🙂
The hole punshing approach, so it’s a yes.