-1

This is in reference to advice on various security blogs about the Mirai IoT DDoS attacks. The average security blog will say something like, "reboot your IoT device and then change the default password and bingo bango, you're secure!"

Except what? My (smart) blu-ray player doesn't have a "password". Neither does my Amazon Fire Stick, my Samsung DVR box, or about 9 other random IP devices in my home. What do they mean when they say to change the password of your IoT devices?

9
  • 4
    I'm voting to close this question as off-topic because this question is next to impossible to answer unless one actually reads what the reference article says. Additionally, Amazon Fire Sticks, are not really considered to be an IoT device. Mirai also doesn't affect every IoT device.
    – Ramhound
    Jan 31, 2018 at 22:33
  • Well then how about webcams? I don't have one, but as a theoretical question, I've never heard of a webcam that has a "password". Some kind of software it interacts with might have a user account with a password, but the device itself doesn't. And those were a big part of the DDoS attacks. Jan 31, 2018 at 22:33
  • Lots of webcams have the ability to view remotely viewed. Do you have any of those devices on your network? What they mean is, if your password has the ability to remotely controlled, then you should change the username and password associated with the function and reboot the device.
    – Ramhound
    Jan 31, 2018 at 22:38
  • 1
    @DanielB Haha what? The Mirai IoT advice is NOT about home networks? Most of the compromised devices are on home networks, where the device uses UPnP and therefore circumvents the NAT by design. Jan 31, 2018 at 22:40
  • 1
    @WakeDemons3 You're making a lot of assumptions and showing some ignorance. It's in a good pursuit, but please begin by calming down and researching how these devices are set up and how they can be communicated with. Any of these devices that CAN communicate over the internet are going to do so using one of a small handful of protocols. The controls inside them that provide this function are mostly likely one of a small handful of possible models. These models and protocols have their own methods of securing, but in general, they all have passwords, and most of these can be changed. Jan 31, 2018 at 23:21

1 Answer 1

3

There is no way to authoritatively answer this question here. There are too many types of devices and too much information required to secure all of them. However, understanding a few basic principles will allow you to begin the process of securing your own systems.

  • Each of these devices has a way they can be connected to. In some cases this will be a miniature web server and you can get into them via a web browser. In other cases it'll be telnet or ssh, and you'd use a telnet or ssh client to connect.
  • Each of these devices will have an IP address on your network. This IP address is what you'll point your browser or telnet/ssh client at to connect.
  • The default username and password for each device can probably be found deep in the technical documentation. If not there, they're probably on the internet somewhere.
  • The best security is security-in-depth. That means you shouldn't only focus on your IoT devices; you should make sure the firmware and security of ALL devices on your network are up to date and strong. This means your router, your modem, your mobile devices, your computers. A generally strong network can help protect a weak member. The reverse of this is also true: one strong network can be compromised by a weak member.
  • The easiest way to secure these devices will be to make sure their firmware is updated fully. If the manufacturer does not note that they have resolved Mirai vulnerabilities in a firmware update, you may need to change the passwords yourself manually.
  • Manually changing passwords on devices could "brick" them, making them unusable. It's better to contact the manufacturer and see if they have recommended steps to secure the devices before you just try changing passwords.
2
  • So if I Nmap my home subnet and get the IP addresses of various devices, I can just browser to that IP address and an interface will become accessible? Feb 1, 2018 at 0:37
  • 2
    That's a good start. If the device has a webserver, the browser will work. If it requires some other protocol to connect nmap would likely be able to identify this. Feb 1, 2018 at 0:38

Not the answer you're looking for? Browse other questions tagged .