As the title says, I want to know the most paranoid security measures you’ve implemented in your homelab. I can think of SDN solutions with firewalls covering every interface, ACLs, locked-down/hardened OSes etc but not much beyond that. I’m wondering how deep this paranoia can go (and maybe even go down my own route too!).
Thanks!
I’m an enterprise guy, so that’s the explanation for non home use things.
I bet this can still get exploited, just would take effort hopefully none does for a home network.
I’m still one shitty windows zero day click away from getting my workstation or browser tokens owned though, I can feel it.
Sophos is so cool! It locks down my work Windows computer so hard, and hook on every file open, that it makes the build time 4 times slower on projects. It keeps the computer so safe that I just don’t use it, I learned about namespaced vpn connections in Linux and just use my main computer with namespaces to work lol
Also laughing because that’s how some companies get owned, IP stolen, etc.
There has to be balance, if your life using their system sucks so hard you can’t do your job or meet production marks, you get creative.
My industry has to prioritize security over productivity. It’s almost impossible to get work done.
I understand that. But among the peers working there are some using windows without any further protection. Why do I need to be the one getting IO-bombed by a software that scans the same files that were gathered from an internal git server anyway, when there are people whose protection is literally “pls don’t tresspass”
I trust my system way better: data at rest is encrypted with LUKS instead of bitlocker’s sucky encription; openvpn conf was upgraded by me because it admins use 128 bits keys for some reason. Etc.
Your working environment sounds gross :)
IT is hard. Finding good IT people is harder in my opinion. Working for a company that is not super squared away with good security and great usability sucks. At least you found some work arounds and are trying to do it well.
It really is
From what I get the higher ups wanted to implement those measures to comply with some certification or whatever
Problem is there are workstations from before this decision that are completely open and will probably never be upgraded; and you get new ones that are completely closed to the point workers would rather use their own hardware
If it’s possible to bypass this locked shitshow and just connect through another machine, then it’s really just a half assed measured don’t you think?
I mean it’s not straightforward doing it and probably the guys who would be the easiest victims and entry points can’t bypass the VPN connection to another machine like I could. But then, those who can do it, can also set up stricter firewall rules and control from their own machines, rather than using windows
Ha yeah.
Id say the same for trellix.
You should try doing things with installs or updating apps when the edr product blocks write access to all temp locations. You have to do an exclusion for every installer, signing cert, or turn it off to install programs.
As somebody taking like 0% of all that measures and not having any problem, luck was involved for sure, unless they have a good reason to attack you in particular… I feel like you will be fine…
Ha, probably. It’s fun to learn stuff though.
Working in this field, almost every company has been beached, IP stolen, etc.
Sometimes your home IP gets hit in an automated scan for a vulnerability and then auto exploited by automation. I’m hoping not to get random chance added to a botnet.
You seem to have a great setup. Since this comment touches on slightly advanced topics, I’ll ask this here:
Your measures seem to be focussed more on security than privacy - which is great! It’s my fault for not specifying it in the post, but I’d definitely like to know if you have done anything specific to keep your network private as well as secure.
Thanks for your wonderful comment - saved!
I guess the firmware is as good as possible. All network devices are just computers and can be exploited. I use a Cisco router as my actual gateway. Sophos is inline after that.
Privacy. 🤔
Not much. I have certain traffic go through a VPN to the Internet, but that’s split tunneled.
I use incognito? That doesn’t really do anything, ha.
I’m slowly killing web browser tracking and cookie stuff that group policy allows.
I didn’t know MS exchange could be used as a WAF. Will need to read more about that.
Can I host Intune completely on-prem?
What do you mean by “My Sophos is self-contained”?
Does your Cisco router get updates? My problem with these companies is that they build backdoors in their firmware for agencies to use. Are you monitoring the network usage of your Cisco gateway?
Using AD/RADIUS on-prem is an intriguing idea. I didn’t consider it because if my AD server goes down I’m essentially locked out of my services. I need to think more on this. Thanks.
I have the older Sophos utm, which doesn’t use the Sophos cloud central manager.
I think their new firewall utm can work disconnected, but I don’t know.
Sophos has a home use license that’s free for non business use.
I love companies that do community edition or free home use.
Sophos, Veeam has nfr, Elastiflow has community edition, which is a netflow.
This is the first time I’ve come across Elastiflow, thanks for mentioning it. Seems like an intriguing service to add.
I was considering using Suricata/installing Security Onion to do IDS from the certificate from a private CA. Sophos firewall seems pretty good too.
Sorry for confusion. I use Sophos utm as a WAF for exchange. Basically reverse proxy that is specifically programmed for exchange attacks. It allows OWA to keep working.
I put the exchange admin URL behind authentication, so you try to go to /ecp, it Sophos intercepts and make you authenticate to Sophos utm first, which is passing to ad with radius.
MS got rid of intune on prem. It’s only Azure service now. I think.
My router is my biggest vuln. Oddly the most important. It’s an enterprise ISR. It’s updated as far as possible. My paranoia ends with the US gov/NSA. I don’t care if they want back door oddly. I don’t want China using me for attack relay however.
Loads of monitoring. You do a span/mirror port to your IDS like security Onion. Let it analyze all your traffic. Apparently there are some state sponsored exploits that allow them to owe a router at kernel level and hide their activities from you and monitoring, but that’s a level I can’t deal with.
As far as lock out, you create a break glass on everything. Emergency account with non rememberable ridiculous password, saved in a safe place.
This is such a great and a simple idea. Thanks.
I think I followed your setup at a high level, but because I don’t have hands-on experience with AD I didn’t quite catch the scope of it. Thanks for letting me know, I’ll get some reading done when I get the time!