NVR's - Associated network risks and mitigations?

Nov 23, 2019
20
6
US
We've been using various Lorex/Dahua NVR's and cameras for many years, haven't had any issues that we know of but the risk is always there. The low cost solutions are perfect for our rental properties, We've seen Amcrest products used in several posts and again these look to be rebranded Dahua products that are also not NDAA compliant and pose certain risks in terms of network security.

Can anyone speak to actual associated risks and offer any mitigations when using these devices or are their better solutions competitively priced as a solution to achieve the same end user functions. Thanks.
 
If this is for your private use, then don't worry about NDAA. That ban only applies to government facilities.

Many, if not most here, have completely isolated their cams from their network and the internet, which allows us to buy the best bang for the buck cameras instead of overpriced NDAA compliant that can still be hacked.....

Well known NDAA compliant companies have been hacked, thus showing that the ban and only using NDAA compliant devices like Verkada doesn't protect you if you give them internet access.

Sadly, too many companies have jumped on the NDAA bandwagon and sell subpar performing cameras and NVRs at a premium price all under the disguise of being secure, which they are not.

It is why we recommend DO NOT LET YOUR CAMERAS OR NVR TOUCH THE INTERNET. You isolate them via VLAN or dual NIC and then access via self-hosted VPN like OpenVPN or WireGuard or Tailscale.
 
Thanks for the reply. These are private networks and agree they don't need to be NDAA compliant it was just a point of reference as we all know there are associated risks. When you say "DO NOT LET YOUR CAMERAS OR NVR TOUCH THE INTERNET" I might be ignorant in asking the question but how will the App work without the NVR using the internet. If there a link or forum post I should be looking for that explains these safeguards when using these branded unit? Thanks again.
 
By default, your cameras are isolated from the internet because the NVR acts as a firewall of sorts. Your LAN may be a subnet of 192.x.x.x while the NVR puts all the cameras on the 10.x.x.x subnet, which means only the NVR can be accessed, not the cameras.

Now there is a distinction in what "access" means - being able to see your camera feeds from the camera is different than actually being able to access the camera.


If one wants to view cameras remotely, they have a few options - port forward, P2P, or hosting a VPN like OpenVPN or using something like Tailscale or Wireguard.

Any system on the internet can be hacked.

Hackers don't care about your camera feed. Hackers use a vulnerable device (NVR or computer or camera or any other IoT) that has ZERO protection on it to get into your LAN and either scrape it for bank info or use your ISP as a bot for DDoS attacks. Your antivirus software and router firewall do not block this crap because you gave an open door directly to your system to bypass these measures.

That is why many of us don't have the Alexa, don't connect smart TVs to our internet, etc.

But many that do have those types of things VLAN them off so they cannot talk to other stuff on the LAN. Doesn't prevent a bot from taking over that specific device to DDoS, but at least it prevents them from scraping your data.

The only way to completely prevent it is to not allow the device to connect to anything and truly be a CCTV system.

But that is unrealistic to most.

Most here will agree that port forwarding directly to your NVR is the least safe. Although the great internet has many articles that state it is OK lol.

Then there is a debate as to if P2P or OpenVPN or something like ZeroTier, Wireguard or TailScale is the next safer option.

Arguments are made both ways.

P2P you are relying on the NVR manufacturer's servers to not be hacked. You have zero control over those. Dahua has recently been shutting down the older P2P servers that were more easily hacked. Many here have confidence in the newest line of P2P security features.

Same with ZeroTier and the like. You are relying on someone else's servers to make that connection. Anytime you are relying on someone else, it can be hacked.

OpenVPN is hosted locally, either native to the router or installed on a computer.

In theory you have the most control over this since it is all in your house.

But it relies on opensource coding that can be hacked as well.

You are relying on your computer and router to be up to date and not allow bad actors in. But that is the same regardless of the solution you are using. At least the computer gets more frequent security updates than an NVR. But Windows is the most common OS that more actors are trying to exploit it than say an NVR.

So you take extra steps like the firewall device @bigredfish has that allows you to monitor everything.

Many of us with BI use Pushover to send notifications that go out to the Pushover email or API servers - in this event all they have access to is your images and not your entire system. You should be able to setup an NVR with the Pushover email option.

Take steps to further minimize access to stuff.

Regardless of which platform you use to access your stuff remotely, have it be isolated from the rest of the system so that the entire system isn't compromised.

Set up procedures that lets you know whenever something connects or logs in to your device. Doesn't necessarily prevent the backdoor exploit, but take any steps possible to eliminate those risks.

Or just say F it and use port forward blindly like most of society. At the end of the day, most don't get hacked. It just sucks if you are one of them that do.




 
  • Like
Reactions: Usafexplorer
By default, your cameras are isolated from the internet because the NVR acts as a firewall of sorts. Your LAN may be a subnet of 192.x.x.x while the NVR puts all the cameras on the 10.x.x.x subnet, which means only the NVR can be accessed, not the cameras.

Now there is a distinction in what "access" means - being able to see your camera feeds from the camera is different than actually being able to access the camera.


If one wants to view cameras remotely, they have a few options - port forward, P2P, or hosting a VPN like OpenVPN or using something like Tailscale or Wireguard.

Any system on the internet can be hacked.

Hackers don't care about your camera feed. Hackers use a vulnerable device (NVR or computer or camera or any other IoT) that has ZERO protection on it to get into your LAN and either scrape it for bank info or use your ISP as a bot for DDoS attacks. Your antivirus software and router firewall do not block this crap because you gave an open door directly to your system to bypass these measures.

That is why many of us don't have the Alexa, don't connect smart TVs to our internet, etc.

But many that do have those types of things VLAN them off so they cannot talk to other stuff on the LAN. Doesn't prevent a bot from taking over that specific device to DDoS, but at least it prevents them from scraping your data.

The only way to completely prevent it is to not allow the device to connect to anything and truly be a CCTV system.

But that is unrealistic to most.

Most here will agree that port forwarding directly to your NVR is the least safe. Although the great internet has many articles that state it is OK lol.

Then there is a debate as to if P2P or OpenVPN or something like ZeroTier, Wireguard or TailScale is the next safer option.

Arguments are made both ways.

P2P you are relying on the NVR manufacturer's servers to not be hacked. You have zero control over those. Dahua has recently been shutting down the older P2P servers that were more easily hacked. Many here have confidence in the newest line of P2P security features.

Same with ZeroTier and the like. You are relying on someone else's servers to make that connection. Anytime you are relying on someone else, it can be hacked.

OpenVPN is hosted locally, either native to the router or installed on a computer.

In theory you have the most control over this since it is all in your house.

But it relies on opensource coding that can be hacked as well.

You are relying on your computer and router to be up to date and not allow bad actors in. But that is the same regardless of the solution you are using. At least the computer gets more frequent security updates than an NVR. But Windows is the most common OS that more actors are trying to exploit it than say an NVR.

So you take extra steps like the firewall device @bigredfish has that allows you to monitor everything.

Many of us with BI use Pushover to send notifications that go out to the Pushover email or API servers - in this event all they have access to is your images and not your entire system. You should be able to setup an NVR with the Pushover email option.

Take steps to further minimize access to stuff.

Regardless of which platform you use to access your stuff remotely, have it be isolated from the rest of the system so that the entire system isn't compromised.

Set up procedures that lets you know whenever something connects or logs in to your device. Doesn't necessarily prevent the backdoor exploit, but take any steps possible to eliminate those risks.

Or just say F it and use port forward blindly like most of society. At the end of the day, most don't get hacked. It just sucks if you are one of them that do.






Great explanation, I really appreciate you taking the time to spell out the details. I know enough to be dangerous, but that's about it. Anyone not within the network security realm is truly limited by "You know what you know" as for myself I just don't know what to search or ask as I just don't know..... Thanks again for the resources, very much appreciated.
 
I go through this aspect not only with IP cameras but phone systems, network printers, bleh bleh bleh.
Once a customer asks me about network security, hacking, anything with routers or firewalls, I advise them (for business insurance and liability purposes) that it's that time in life to hire a professional network person who can keep up day to day, week to week with security vulnerabilities, firewall updates, new this...new that. It is too much to keep upto date with if not in the field.