Information Security Stack Exchange is a question and answer site for information security professionals. It only takes a minute to sign up.
Sign up to join this community
If I open up a dedicated server on my home network to accept incoming connections (HTTP, SSH, etc) and that machine is compromised by an outside attacker, what are the risks to other devices on the network?
Is it enough to keep important personal files off the server itself, or is the entire LAN potentially open to the outside world?
Without a defense in depth security strategy, once the perimeter defense of your network has been penetrated through your compromised server there are a few threats to the rest of your network.
The vulnerability exploited by the attacker could be available on other machines on your network, allowing an attacker to move to another machine with ease.
A poor access control or authentication scheme to other machines on the network could allow an attacker access to those machines or unauthorized use of service. i,e. Intended for ease of use, A web application server that will gracefully accept any kind of traffic from the compromised machine.
Malware could be present on the compromised server, and may even propagate via a network vulnerability itself. i,e. a worm.
To answer your second question, given that your server has been compromised; The potential is there. Any information that you would like to keep the Confidentiality and integrity would best be placed on another machine with limited access. I would recommend some kind of defense in depth strategy. Or a firewall with well configured policy and traffic monitoring. Snort is a free network based Intrusion Detection System if you want to take it up a notch.
Risk is the calculated cost of a threat agent carrying out an attack on asset. (I offer this definition for completeness because some people use risk and threat interchangeably.) Risk = (probability threat will occur) * (Value of the asset).
A threat is the potential for a vulnerability to be exploited by a threat agent/attacker. (Note, acts of god are considered threats, and their risk should be accounted for!)
In general a compromised device on the LAN is a pretty serious risk to the rest of the devices on the network. An attacker can launch probes from the compromised machine to look for weaknesses in other machines - for example, do you have remote desktop enabled anywhere? File & Printer sharing? A machine that is not fully patched and up to date, which could allow an attacker to remotely infect it with malware? A device with a web-based control panel (printer, network camera, smart thermostat, router, etc.) that still has the default login and password (or none at all)? This is by no means an exhaustive list of potential weaknesses, but any of these things could allow an attacker to spread to other machines and/or cause more serious damage.
You could go around making sure every device is properly hardened, but even if there are no weaknesses in other devices on the LAN, an attacker could still, say, launch an ARP spoofing attack to redirect all network traffic to the compromised machine, allowing him to monitor all of your Internet and browsing traffic.
If the compromised computer is on a windows domain and the attacker gets system access he can dump local password hashes.
So if you have the same administrator accounts on all machines youre entire network is compromised.
Also the attacker can lay low and wait for a domain user to log on and steal that password so that probably also gives access to log into a lot of machines on a windows domain.
With one foot in the door full network compromise is only a matter of time (think vulnerability scanning inside the network etc...).
Always plan for the worst case (isolation, segmentation, monitoring, and more...)
