1

I've got a terminal server set up in a data center that also serves as a console server. This console server only accepts regular telnet requests. To this server are a bunch of routers and switches connected with console cables. For example, to connect to a router I telnet to the server's public IP-address and a port. I need this action to be taken place from the webbrowser: telnet://ip-address which will open an external terminal application on my PC.

It is not unknown that telnet is an incredible nonsecure communication protocol. My question therefore is: how do I telnet to my console server via a webbrowser in a secure way? If I force my webserver to do this via HTTPS, will this be secure enough?

  • I'm pretty sure that "secure telnet" is an oxymoron. – rook Nov 26 '14 at 16:47
  • @Rook Telnet's only as insecure as the network itself is; if you have a secure network (say, a crossover cable between two machines sitting next to each other in a locked room), it's perfectly fine. – cpast Nov 26 '14 at 19:48
1

If the webserver is running on the same box as the telnet system (or directly connected on a physically secure connection), then you could safely relay information to the webserver and have it relay it, via telnet, to the system that you want to talk to. The important thing to realized, however, is that your communication is only protected between your system and the webserver. The web server's connection is still unencrypted and open to being snooped on unless that network is under your control and otherwise protected.

If this provides sufficient security for you, an alternate (and possibly easier) approach would be to setup a simple VPN to the network and allow telnet traffic over the VPN. VPN is more natively designed for this kind of forwarding and can be configured to encrypt traffic. It would allow you to use a telnet client of your choice on your PC instead of having to use some funky web based interface.

  • Of course, that was something I missed in my story! I talked to my boss today and he said that we have a VPN connection from the webserver to the network with the console server. So the only thing that rests me to do is to find something to telnet from my webserver to any host on that network. – Beeelze Oct 27 '14 at 14:16
  • In that case, the security related answer is either chain the VPN connections or use an encrypted web server connection. Designing a website that could be used as a Telnet client is off topic though as it isn't security related. It sounds like the security related portions of the question are answered now. – AJ Henderson Oct 27 '14 at 14:19
0

The browser is just executing the telnet program on your local system. This will establish its own connection to the terminal server, independent from the browser connection. So it will be insecure, even if the link itself was served via https.

Check your console server if it support SSH instead, because SSH (available on Windows with putty and others) will have an encrypted connection.

  • My console server doesn't support SSH, otherwise I would've used it already :) Are there by any chance solutions possible to make this telnet connection from my webserver. If I do this via HTTPS this way then it should be secure right. Because it is the webserver itself that will establish the connection to the console server. – Beeelze Oct 27 '14 at 12:12
  • There are products like jsTerm you need to install at the side of the web server and which then can provide a terminal interface inside the browser. Search for 'telnet emulator inside browser'. – Steffen Ullrich Oct 27 '14 at 12:21
  • Thx for your answers! I will try out some of those products and make my website serve requests from HTTPS from now on :) – Beeelze Oct 27 '14 at 12:42
0

If you use HTTPS, then you will benefit from the protection of SSL between the browser and the Web server, and nowhere else. HTTPS is not Telnet; this is another protocol; meaning that a Telnet client will not use HTTPS or SSL, even if you find a way to trigger it from an HTTPS-served Web page.

If I understand your case correctly, then you have:

  • Several devices that expect Telnet connections; these devices are in a given network N.
  • A Web server, such that communications between that server and N are somehow secured (same local network, VPN...).
  • A human user with a Web browser, located outside of N.

Your problem is to make sure that the data exchanged with the devices ultimately makes its way to the human user, but without fearing compromise or alterations in the transit between the Web server and the browser. This points at the following conceptual solution: the Web server runs a Telnet client, and forwards all input and output to the browser as some kind of Web page update.

Anyterm is supposed to do just that.

  • Good answer. My question should have been something more like: "How can I use my webserver as a telnet client". I was more looking into the security part of what I need to do. – Beeelze Oct 28 '14 at 11:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.