Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

USB Killer 2.0

usbkill.com

Post reply on HN
Print view View original

21–28 of 28 posts

Re: USB Killer 2.0

#21

Post by joezydeco »

regularfry wrote: post #20
DanBC wrote: post #12

I like patcheudor's comment here: https://news.ycombinator.com/item?id=12467176

Counter-example that isn't too contrived: if you've got security gates controlled by a PC which are designed to fail open, I can see this being a very useful addition to a physical pen-test.

No USB host stack opens itself up just because the transmission circuits have been destroyed.

This isn't the case of a gate that fails open. This is the case of a gate that gets hit by a rocket launcher, which blows the gate open and destroys the datacenter behind it.

Re: USB Killer 2.0

#24

Post by iplaw »

They haven't performed any testing with the USB Kill connected to a USB to Apple lightning adapter or to a USB-to-micro-USB adapter. I think that the device with a micro USB adapter would function identically, as it's just a pin mapping. But a Lightning adapter has a Lightning controller onboard.

Something like this (or even a USB to Lightning charger cable):

https://www.amazon.com/HIOTECH®-Adapter-Lightning-Female-iPh...

Would this destroy the Lightning controller in the adapter/cable, or would it work to destroy the iPhone?

Re: USB Killer 2.0

#26

Post by iplaw »

type0 wrote: post #19
DanBC wrote: post #16

Earlier quoted context omitted.

> Imagine one attacker distracting the victim in a café and the other attacker quickly inserting the Kill stick. Sort of a hardware Denial of Service. You're in a coffee shop. Wouldn't the attackers just "accidentally" spill coffee on your laptop? Some laptops cope well with water from the top (over the keyboard) but not in the air vents.

> Wouldn't the attackers just "accidentally" spill coffee on your laptop? The difference is deniability, you can always see that someone killed your laptop with coffee or smashed it with a hammer, with this you wouldn't know until you can examine the circuits.

And even then, power surges do happen. Circuitry get fried from time to time. There is still a level of plausible deniability.

Re: USB Killer 2.0

#27

Post by regularfry »

joezydeco wrote: post #21
regularfry wrote: post #20

Earlier quoted context omitted.

Counter-example that isn't too contrived: if you've got security gates controlled by a PC which are designed to fail open, I can see this being a very useful addition to a physical pen-test.

No USB host stack opens itself up just because the transmission circuits have been destroyed. This isn't the case of a gate that fails open. This is the case of a gate that gets hit by a rocket launcher, which blows the gate open and destroys the datacenter behind it.

> No USB host stack opens itself up just because the transmission circuits have been destroyed.

I'm not talking about the host, I'm talking about the gates. I've seen a laptop bricked because it got the wrong voltage on the USB pins (I've still got it; motherboard SMT fuse blew and I've not got round to replacing it). From a fire safety perspective, it makes perfect sense for gates to fail open (or undriven, so you can push them open manually) when their controller dies. If you can get to the USB port, that's a perfectly feasible route in.

Re: USB Killer 2.0

#28

Post by compsciphd »

regularfry wrote: post #27
joezydeco wrote: post #21

Earlier quoted context omitted.

No USB host stack opens itself up just because the transmission circuits have been destroyed. This isn't the case of a gate that fails open. This is the case of a gate that gets hit by a rocket launcher, which blows the gate open and destroys the datacenter behind it.

> No USB host stack opens itself up just because the transmission circuits have been destroyed. I'm not talking about the host, I'm talking about the gates. I've seen a laptop bricked because it got the wrong voltage on the USB pins (I've still got it; motherboard SMT fuse blew and I've not got round to replacing it). From a fire safety perspective, it makes perfect sense for gates to fail open (or undriven, so you c…

I think the author's point is that if you have exposed USB ports then there is nothing one can do. i.e. no matter how hardened they are, you can just keep on increasing the voltage (relatively easily vs hardening) and you will kill the PC. so one can just as easily pentest by removing the power from the PC without any fatally destructive behavior.
Post reply on HN