At $DAYJOB, we produce several embedded devices, with access to a management network that often is not connected to the Internet. We also have a management console, running on a stadnard Red Hat or CentOS Linux, which has access to both the management network and the Internet.
I am trying to get the embdded devices to authenticate themselves over the Internet towards a machine that we control over HTTPS, and am evaluating tinyproxy as a possible solution, but I am at a bit of a loss how to configure it.
In this setup, I want
- to only allow the
CONNECT
method; and - to only allow the requests to terminate at a specific host (DNS resolvable).
Is it possible to use tinyproxy for this scenario using the standard configuration options? Hacking the sources is of course always an option, but using the standard RPMs is a lot easier.