Encrypting absolutely everything, even within the LAN

11,022

Solution 1

I use IPsec here for everything. The reasoning is that most attacks are made by insiders anyway - the bad side/good side thinking is flawed. (If anyone makes off with the servers they can have fun trying to break the full-disk encryption, so no problem there, either.)

It's also fun to use telnet, NIS, NFS and FTP without any worries - feels like the good old days! :-)

Solution 2

IPSec is the standard for this. It comes in different forms and there is a lot of vocabulary to it.

I recommend this guide to IPSec to get you started.

Solution 3

On the windows side what you are looking for is Domain and Server isolation. You can play with it using this labcast

My default windows implementation plan for a new install has this included. In windows it's not hard to set up and offers lots of extra security (and there's really no "extra" administration you might need a few more security groups).

NAP doesn't require IPsec (or even the PKI - that's only if you want to run native mode). SCCM is an additional product- nothing extra is required for domain and server isolation. NAP is primarily designed for send "health" information about a client and if that client doesn't pass your "health" check then it is deferred to only communicate with a remediation server(s). Health is defined as patch requirements AV settings, security settings etc. You certainly can use SCCM to set up IPSEC, but it's not a requirement.

With Domain and server isolation I don't have the ability to check those "health"- nor is it the point. When it's set up I am encrypting the traffic and ensuring that and additionally guaranteeing that the servers and clients are only allowed to communicate with the servers and clients required for that business function (eg HR workstations are the only workstations allowed to communicate with the HR server).

Solution 4

Do you have a threat model where unfettered access to your LAN infrastructure is expected? If so, yes, deploying IPSEC to all end-points is probably what you want.

However, in most threat models there's enough perimeter security that you can essentially ignore the LAN infrastructure as a cheap access method.

The picture changes if you have WiFi installed, you'd need something between the WiFi network(s) and the wired network(s) to ensure that you don't have any information leakage via that route.

Share:
11,022

Related videos on Youtube

user3745402
Author by

user3745402

Real name: Christian Lercher Started programming when I was 7 years old (Basic on C64, then Assembler, later C/C++, and now mostly Java). First Windows experience: ca 1993 (Win 3.11) First Unix experience: 1998 (Solaris, then SuSE Linux, now mostly Debian/Ubuntu and MacOS). I'm just as much interested in technical things, as in social topics, and economics. It really depends.

Updated on September 17, 2022

Comments

  • user3745402
    user3745402 over 1 year

    Has anybody tried that approach already? I'm really considering it: Instead of relying on network based IDS etc., every packet must use encryption which was initiated by a certificate issued by my own CA.

    • Every client gets a unique client certificate
    • Every server gets a unique server certificate
    • Every service additionally requires to login.

    Both SSL and SSH would be ok. Access to the internet would be done via an SSL tunnel to the gateway.

    Is it feasible? Does it create practical problems? How could it be done and enforced? What do you think?

    More details

    My goal is to simplify the LAN's security concept - I'm not yet sure, if that's a crazy idea! But I feel, that securing a HTTPS or SSH server from internet threats (if using mutual authentication) is sometimes easier than monitoring everything that can happen in the wild world of a LAN.

    On a non-encrypted LAN, I feel it's really hard to be a good step ahead of a potential attacker, because of threats like:

    • Low level attacks like ARP spoofing, Port stealing, ...
    • WLAN access (e.g. every developer will be allowed to access the SVN server from the (W)LAN - I don't think it will be through a VPN...)

    => For simplicity, isn't it easier to make the assumption, that there is always an attacker in the LAN?

    => Could I end up simplifying a (small company's) LAN security concept by treating it like a WAN? Or would I rather complicate it?

    IPSec and alternatives

    IPSec sounds very promising, but I'd be interested in alternatives to IPSec, too - Using SSL/SSH individually per service and creating an Stunnel to the Gateway? Using Kerberos maybe? ... What are the advantages of IPSec or the others?

    If you can help me with getting a better grasp on IPSec, please see my follow-up question specifically on IPSec.

    • Grizly
      Grizly about 14 years
      There is a standard scheme for this: IPSEC.. is that what you mean? A lot of VPN's use it over TLS, but LAN's implement it easily.. windows computers can be enabled and enforced to use it with GPO's.. makes troubleshooting a bit harder, but its not that difficult..
    • user3745402
      user3745402 about 14 years
      @Grizly: I'm not 100% sure! Maybe it can be done with IPSEC (?) It's important for me to authenticate both client and server via a certificate, and additionally make sure, that the client is logged in to the service (in cases where laptops are stolen etc.)
  • Kyle Brandt
    Kyle Brandt about 14 years
    I only use it for WAN tunnels over the Internet myself, so you will have to wait for someone who has experience deploying it on entire LAN. It would provide extra security, but would come with a complexity cost for my environment. I generally rely on ssh and ssl to provide the network level security where needed. I would guess you won't find IPSec deployed on LANs in small/medium companies too often unless they deal in a buisness that requires very high security, I could be wrong though. I would guess this because updates, troubleshooting, backup etc probably take up the admin's time.
  • Philip
    Philip about 14 years
    We had this discussion at my work last year. It boiled down to 'physical security trumps all'. If you've got access to the packets (LAN), you've got access to stealing the whole server (in our case); so no point in putting a padlock on a 3' fence.
  • Spiff
    Spiff about 14 years
    IPSec is probably what he wants, but for the sake of accuracy I feel the need to point out that it doesn't encrypt "absolutely everything" in the context of a LAN. It only encrypts IP-based protocols. It's not uncommon to see non-IP protocols on LANs even today. ARP, STP, CDP, 802.1X, PPPoE, AoE, and legacy protocol stacks in some dark corners of the enterprise (Novell IPX, AppleTalk, NetBEUI, DECnet, etc.)
  • user3745402
    user3745402 about 14 years
    @Vatine: You guessed correctly: WiFi access :-)
  • user3745402
    user3745402 about 14 years
    +1 Sounds good. I'm mostly on Linux, but I like to hear all solutions!
  • user3745402
    user3745402 about 14 years
    Thanks - is this different from SCCM and NAP (from the answer by GAThrawn)? I don't use Windows very much... But it almost seems, as if the thing I want to do is something that's more common in the Windows world?
  • raja
    raja about 14 years
    See my edits above
  • Vatine
    Vatine about 14 years
    Well, with wifi, there's still things you can do to keep it within a secured perimeter, but I guess covering the building in conductive foil can be more expensive than rolling out IPSEC.
  • user3745402
    user3745402 about 14 years
    Your PDF presentation you linked to over there (fukt.bsnet.se/lectures/2007-02-15_IPsec) also answered a big part of this question here. Thanks!