Security Hacks

1532 Articles

LEDs Light The Way To This Backdoor

September 27, 2019 by 41 Comments

A curious trend for some years in the world of PC hardware has been that of attaching LEDs to all the constituent parts of a computer. The idea is that somehow a gaming rig that looks badass will somehow be just a little bit faster. As [Graham  Sutherland] discovered when he wanted to extinguish the LEDs on his new Gigabyte graphics card, these LEDs can present an unexpected security hazard.

The key to their insecurity comes in the Gigabyte driver. This is a piece of software that you would normally expect to be an abstraction layer with an interface visible to your user level privilege, and a safe decoupling between that and the considerably more security sensitive hardware layer from which the LED bus can be found. Instead of this, the Gigabyte driver is more of a wrapper that simply exposes the LED bus directly to the user level. It’s intended that user-level code can easily bit-bang WS2812 LEDs without hinderance, but its effect is to provide a gaping hole in the security layers intended to keep malicious code away from the hardware. The cherry on the cake is provided by the discovery of a PIC microcontroller on the bus which can be flashed with new code, providing an attacker with persistent storage unbeknownst to the operating system or CPU.

The entire Twitter thread is very much worth reading whether you are a PC infosec savant or a dilettante, because not only should we all know something about the mechanisms of PC backdoors we should also be aware that sometimes a component as innocuous as an LED can be a source of a security issue.

Thanks [Slurm] for the tip.

Gigabyte motherboard picture: Gani01 [Public domain].

Copying High Security Keys With OpenSCAD And Light

September 24, 2019 by 24 Comments

The ability to duplicate keys with a 3D printer is certainly nothing new, but so far we’ve only seen the technique used against relatively low hanging fruit. It’s one thing to print a key that will open a $15 Kwikset deadbolt from the hardware store or a TSA-approved “lock” that’s little more than a toy, but a high-security key is another story. The geometry of these keys is far more complex, making them too challenging to duplicate on a consumer-level printer. Or at least, you’d think so.

Inspired by previous printed keys, [Tiernan] wanted to see if the techniques could be refined for use against high security Abloy Protec locks, which are noted for their resistance to traditional physical attacks such as picking. The resulting STLs are, unsurprisingly, beyond the capabilities of your average desktop FDM printer. But with a sub-$300 USD Anycubic Photon DLP printer, it’s now possible to circumvent these highly regarded locks non-destructively.

Of course, these keys are far too intricate to duplicate from a single picture, so you’ll need to have the physical key in hand and decode it manually. [Tiernan] wisely leaves that step of the process out, so anyone looking to use this project will need to have a good working knowledge of the Abloy Protec system. Hopefully this keeps bad actors from doing anything too nefarious with this research.

Once you have the decoded values for the key you want to duplicate, you just need to provide them to the OpenSCAD library [Tiernan] has developed and print the resulting STL on your sufficiently high-resolution printer. Generally speaking, the parts produced by resin-based printing have a high tensile strength but are very brittle, so perhaps not the kind of thing you want to stick in your expensive Abloy lock. That said, there are some “Tough Resin” formulations available now which produce parts that are at least as strong as those made with thermoplastics. So while the printed keys might not be strong enough for daily use, they’ll certainly work in a pinch.

FIDO2: The Dream Of Password-Free Authentication On The WWW

September 23, 2019 by 29 Comments

Of all the things which are annoying about the modern World Wide Web, the need to create and remember countless passwords is on the top of most people’s lists. From dozens of passwords for everything from social media sites to shopping, company, and productivity-related platforms like Github, a large part of our day is spent dealing with passwords.

While one can totally use a password manager to streamline the process, this does not absolve you from having to maintain this list and ensure you never lose access to it, while simultaneously making sure credentials for the password manager are never compromised. The promise of password-less methods of authentication is that of a world where one’s identity is proven without hassle, and cannot ever be stolen, because it relies on biometrics and hardware tokens instead of an easily copied password.

The FIDO2 project promises Web Authentication that means never entering a password into a website again. But like everything, it comes with some strings attached. In this article, we’ll take a look at how FIDO2 plans to work and how that contrasts with the state of security in general.

Continue reading “FIDO2: The Dream Of Password-Free Authentication On The WWW” →

This Week In Security: Zeroconf Strikes Again, Lastpass Leaks Your Last Password, And All Your Data Is Belong To Us

September 20, 2019 by 18 Comments

VoIP cameras, DVRs, and other devices running the Web Services Dynamic Discovery (WSDD) protocol are being used in a new type of DDoS attack. This isn’t the first time a zeroconf service has been hijacked as part of a DDoS, as UPnP has also been abused in similar ways.

Feel like alphabet soup yet? A Denial of Service attack is one where the target is simply made unavailable, rather than actually compromised. The classic example of this is the SYN flood, where an attacker would open hundreds of connections to a web server at once, exhausting the server’s resources and interrupting legitimate use of that server. As mitigations for these attacks were developed (SYN Cookies, for example), DoS attacks were replaced by Distributed Denial of Service (DDOS) attacks. Rather than attack a weakness on the target machine, like available RAM or CPU cycles, a DDoS generally targets available network bandwidth by hitting the target website from many, many locations at once. No clever software tricks can help when your Internet connection is fully saturated with junk traffic. Continue reading “This Week In Security: Zeroconf Strikes Again, Lastpass Leaks Your Last Password, And All Your Data Is Belong To Us” →

This Week In Security: Mass IPhone Compromise, More VPN Vulns, Telegram Leaking Data, And The Hack Of @Jack

September 6, 2019 by 15 Comments

In a very mobile-centric installment, we’re starting with the story of a long-running iPhone exploitation campaign. It’s being reported that this campaign was being run by the Chinese government. Attack attribution is decidedly non-trivial, so let’s be cautious and say that these attacks were probably Chinese operations.

In any case, Google’s Project Zero was the first to notice and disclose the malicious sites and attacks. There were five separate vulnerability chains, targeting iOS versions 10 through 12, with at least one previously unknown 0-day vulnerability in use. The Project Zero write-up is particularly detailed, and really documents the exploits.

The payload as investigated by Project Zero doesn’t permanently install any malware on the device, so if you suspect you could have been compromised, a reboot is sufficient to clear you device.

This attack is novel in how sophisticated it is, while simultaneously being almost entirely non-targeted. The malicious code would run on the device of any iOS user who visited the hosting site. The 0-day vulnerability used in this attack would have a potential value of over a million dollars, and these high value attacks have historically been more targeted against similarly high-value targets. While the websites used in the attack have not been disclosed, the sites themselves were apparently targeted at certain ethnic and religious groups inside China.

Once a device was infected, the payload would upload photos, messages, contacts, and even live GPS information to the command & control infrastructure. It also seems that Android and Windows devices were similarly targeted in the same attack.

Telegram Leaking Phone Numbers

“By default, your number is only visible to people who you’ve added to your address book as contacts.” Telegram, best known for encrypted messages, also allows for anonymous communication. Protesters in Hong Kong are using that feature to organize anonymously, through Telegram’s public group messaging. However, a data leak was recently discovered, exposing the phone numbers of members of these public groups. As you can imagine, protesters very much want to avoid being personally identified. The leak is based on a feature — Telegram wants to automatically connect you to other Telegram users whom you already know.

By default, your number is only visible to people who you’ve added to your address book as contacts.

Telegram is based on telephone numbers. When a new user creates an account, they are prompted to upload their contact list. If one of the uploaded contacts has a number already in the Telegram system, those accounts are automatically connected, causing the telephone numbers to become visible to each other. See the problem? An attacker can load a device with several thousand phone numbers, connect it to the Telegram system, and enter one of the target groups. If there is a collision between the pre-loaded contacts and the members of the group, the number is outed. With sufficient resources, this attack could even be automated, allowing for a very large information gathering campaign.

In this case, it seems such a campaign was carried out, targeting the Hong Kong protesters. One can’t help but think of the first story we covered, and wonder if the contact data from compromised devices was used to partially seed the search pool for this effort.

The Hack of @Jack

You may have seen that Twitter’s CEO, Jack [@Jack] Dorsey’s Twitter account was hacked, and a series of unsavory tweets were sent from that account. This seems to be a continuing campaign by [chucklingSquad], who have also targeted other high profile accounts. How did they manage to bypass two factor authentication and a strong password? Cloudhopper. Acquired by Twitter in 2010, Cloudhopper is the service that automatically posts a user’s SMS messages to Twitter.

Rather than a username and password, or security token, the user is secured only by their cell phone number. Enter the port-out and SIM-swap scams. These are two similar techniques that can be used to steal a phone number. The port-out scam takes advantage of the legal requirement for portable phone numbers. In the port-out scam, the attacker claims to be switching to a new carrier. A SIM-swap scam is convincing a carrier he or she is switching to a new phone and new SIM card. It’s not clear which technique was used, but I suspect a port-out scam, as Dorsey hadn’t gotten his cell number back after several days, while a SIM swap scam can be resolved much more quickly.

Google’s Bug Bounty Expanded

In more positive news, Google has announced the expansion of their bounty programs. In effect, Google is now funding bug bounties for the most popular apps on the Play store, in addition to Google’s own code. This seems like a ripe opportunity for aspiring researchers, so go pick an app with over 100 million downloads, and dive in.

An odd coincidence, that 100 million number is approximately how many downloads CamScanner had when it was pulled from the Play store for malicious behavior. This seems to have been caused by a third party advertisement library.

Updates

Last week we talked about Devcore and their VPN Appliance research work. Since then, they have released part 3 of their report. Pulse Secure doesn’t have nearly as easily exploited vulnerabilities, but the Devcore team did find a pre-authentication vulnerability that allowed reading arbitraty data off the device filesystem. As a victory lap, they compromised one of Twitter’s vulnerable devices, reported it to Twitter’s bug bounty program, and took home the highest tier reward for their trouble.

High Voltage Protects Low Denominations

September 6, 2019 by 17 Comments

How do you keep people out of your change jar? If you didn’t say with a 3D printed iris mechanism and high-voltage spark gap, then clearly you aren’t [Vije Miller]. Which is probably for the best, as we’re not sure we actually want to live in a world where there are two of these things.

Regular Hackaday readers will know that [Vije] has a way of using electromechanical trickery to inject a bit of excitement, and occasionally a little danger, into even the most mundane aspects of life. His latest project is an automated change jar that uses a pinpad to authenticate users, while everyone else gets the business end of a spark gap if the PIR sensor detects them getting to close.

You can see a demonstration of the jar in the video after the break, where he shows the jar’s ability to stop…himself, from getting access to it. Hey, nobody said it was meant to keep out real intruders. Though we do think a similar gadget could be a fun way to keep the kids out of the cookie jar before dinner, though we’d strongly suggest deleting the high-voltage component from the project before deploying it with a gullet full of Keebler’s best.

[Vije] was able to adapt a printable iris design he found on Thingiverse to fit over the mouth of the jar, and uses servos in the base to rotate the whole assembly around and open it up. The internal Arduino Nano handles reading from the pinpad, controlling the stepper, and of course firing up the spark generator for 1000 milliseconds each time the PIR sensor detects somebody trying to be cute. Just the sound of the arc should be enough to get somebody to reconsider the value of literal pocket change.

Some of the design elements used in this change jar’s high voltage components were influenced by the lessons learned when [Vije] was building his plasma-powered toilet air freshener. There’s a sentence we bet you never expected to read today.

Continue reading “High Voltage Protects Low Denominations” →

ESP8266 And ESP32 WiFi Hacked!

September 5, 2019 by 48 Comments

[Matheus Garbelini] just came out with three (3!) different WiFi attacks on the popular ESP32/8266 family of chips. He notified Espressif first (thanks!) and they’ve patched around most of the vulnerabilities already, but if you’re running software on any of these chips that’s in a critical environment, you’d better push up new firmware pretty quick.

The first flaw is the simplest, and only effects ESP8266s. While connecting to an access point, the access point sends the ESP8266 an “AKM suite count” field that contains the number of authentication methods that are available for the connection. Because the ESP doesn’t do bounds-checking on this value, a malicious fake access point can send a large number here, probably overflowing a buffer, but definitely crashing the ESP. If you can send an ESP8266 a bogus beacon frame or probe response, you can crash it.

What’s most fun about the beacon frame crasher is that it can be implemented on an ESP8266 as well. Crash-ception! This takes advantage of the ESP’s packet injection mode, which we’ve covered before.

The second and third vulnerabilities exploit bugs in the way the ESP libraries handle the extensible authentication protocol (EAP) which is mostly used in enterprise and higher-security environments. One hack makes the ESP32 or ESP8266 on the EAP-enabled network crash, but the other hack allows for a complete hijacking of the encrypted session.

These EAP hacks are more troubling, and not just because session hijacking is more dangerous than a crash-DOS scenario. The ESP32 codebase has already been patched against them, but the older ESP8266 SDK has not yet. So as of now, if you’re running an ESP8266 on EAP, you’re vulnerable. We have no idea how many ESP8266 devices are out there in EAP networks,  but we’d really like to see Espressif patch up this hole anyway.

[Matheus] points out the irony that if you’re using WPA2, you’re actually safer than if you’re unpatched and using the nominally more secure EAP. He also wrote us that if you’re stuck with a bunch of ESP8266s in an EAP environment, you should at least encrypt and sign your data to prevent eavesdropping and/or replay attacks.

Again, because [Matheus] informed Espressif first, most of the bugs are already fixed. It’s even percolated downstream into the Arduino-for-ESP, where it’s just been worked into the latest release a few hours ago. Time for an update. But those crusty old NodeMCU builds that we’ve got running everything in our house?  Time for a full recompile.

We’ve always wondered when we’d see the first ESP8266 attacks in the wild, and that day has finally come. Thanks, [Matheus]!