Sandboxing FTW

After I reminded people that I offer a complimentary attachment checking service through my office I got a submission from one of our warehouse operators in Texas. It was an oddly-named attachment called UmjSJCk.zip. I saved it to my Mac and opened Terminal. Then unpacked the zip file and it unpacked to Quotation.exe. I giggled a bit when I ran the file command on it and saw that it was a Windows executable. Exactly what I expected. So I put it in a folder called sandbox and started my copy of Windows XP that I have in VirtualBox. The OS has it’s hard drive set to immutable, so any changes or write activities that the OS does is not sent to the VHD image, but rather to a “snapshot” VHD image on the side. Each time I start the OS, it’s as if I am starting it for the first time, because when an immutable VM finds something (anything) in the snapshot folder, it dumps it first then creates a new snapshot image for writes. I make sure the sandbox can’t see anything beyond my Mac by assigning it’s LAN connection as a Host-Only Adapter. That means that the VM can only see VirtualBox’es fake network host and nothing else.

So start this sandbox Windows XP, mount the sandbox folder as a drive to the sandbox – set as Read Only also, by the way, no baby-backwash here… and then double-clicked on Quotation.exe. It loaded a process and started to grope the network connection. Of course it did. So, with the bug trying it’s best to reach out and fetch it’s payload I clicked on the little red close control and told VirtualBox to power off the virtual machine.

Poof. All gone. Changes and everything. Then I dumped the sandbox contents.

I think whats more concerning here is that my scan using ClamAV on my Mac in regards to this data showed no infected data. Well, it certainly was trying to be nasty.

Then I start to wonder about the inherent usefulness of VirtualBox when it comes to airgapped computing when it comes to privacy and really being paranoid about encryption. But then I realize that when I turn off my Airport on my MBP, that it’s just as good as anything I could screw around with in VirtualBox. An infection in my MBP? Heh… piff.

Geek Excursions: BitMessage

Along with my curiosity surrounding Bitcoin, there is a similar technology that has been released for public use called BitMessage. This system is a really neat way to securely communicate in a secure method that involves absolutely no trust whatsoever. It’s a completely decentralized email infrastructure and has captured a lot of my spare attention. BitMessage works a lot like how Bitcoin does, you can create email addresses on the fly, they are a long sequence of random characters that your system can display because you have both a public key and a private key. In a lot of ways BitMessage deals with the biggest problem surrounding PGP/GPG, which is key management. Nobody really wants to manage keys or use the system because it’s extra work. Plus even with PGP/GPG, your identity is written on your keys for everyone to see.

Getting started with BitMessage is a snap. First you need to download the BitMessage client, and you can get that at bitmessage.org. There’s a Windows and Mac client available, you can start it and be instantly attached to the BitMessage network, ready to create new “BitMessage Addresses” and throw them away just as easily. So, for example, you could reach me by sending me a BitMessage to this address: BM-2cWAk99gBxdAQAKYQGC5Gbskon21GdT29X. When you send a message using BitMessage, its to this address and from an address that your client makes, so the conversation occurs securely and since every node has a copy of the data it’s impossible to tell who is getting what information. I think an even more secure method would be to cross BitMessage with a PGP/GPG key. The only problem with a key like that is that classically PGP/GPG keys require that you include your email address as a subkey so that you can be identified by a human-readable email address when looking for your public key or when someone else is looking for it, to verify a signature for example. The PGP/GPG system doesn’t require an email address, you can of course create a public and private keypair using PGP/GPG and make the email address up from whole cloth, and instead just let people know the key ID that you want them to use. So technically if Alice wanted to secretly communicate with me, we could give each other our public keys to start and then use BitMessage as the messaging mule. I don’t see how any eavesdropper could make sense out of any of that data flow. It’s unclear what the contents are, the PGP/GPG encryption keeps the contents of the message secure, and BitMessage itself seriously obfuscates if not outright eliminates being able to tell where the messages are ultimately going to or coming from.

I have to admit that BitMessage is very user friendly and very handy to have. My only issue with it is that I don’t know anyone who uses it, but perhaps this blog post will change that. If you are interested in this bleeding-edge crypto/privacy software, I encourage you to chat me up on BitMessage for serious matters or for fun.

OS Tryouts 1: PC-BSD

PC-BSD

System Setup

The PC-BSD initial setup was pleasant enough, there was only brief exposure to the horror of the console as cryptic text scrolled past. I can imagine consumers panicking when they see these sorts of screens, pages of text they can’t comprehend without a solid understanding that much of it really is meaningless unless the system doesn’t work, and then it rockets from being worthless to priceless. Generally when I think of designing operating systems for consumers, you want to suppress this behind some pretty pictures or a progress bar, which is a clearer representation that everything is proceeding according to plan. Even when everything is working properly in systems like these you can spy error reports in the startup console text screens. The developers either don’t care or expect the errors and they are “worthless” issues because the system starts up normally. To consumers, if they are reading along and have a little bit of training about what they are looking at, they could be unsettled by a line that looks like an error even if it’s a throwaway warning.

After the initial setup, the standard installation questions are rather straightforward. Language and locale settings, however it is good to note that these days the really good systems automatically fetch much of this material from the indigenous Internet address. I would argue that if the IP is in the United States then it’s likely English, and if you know the IP, then you know the location, so time zones are easily set as well. The hostname selection is always different from system to system I’ve found. Some systems are computer-before-person and some are person-before-computer. Since you can set this to whatever you like, it’s not really a quibble.

PC-BSD does a very good job at clearly separating the difference between root access and user access. You create the password for the root account, and then it automatically leads you to create a user account afterwards, with the option for encryption presented immediately, which is a nice touch.

First Login

I was presented with a login dialog box, I selected my window manager to be Cinnamon as it was an installer option when I set up this system. The system attempted to start X Windows and then the desktop manager crashed. I tried to restart it twice and then when that wasn’t working I clicked Cancel and the system started into X Windows without a desktop manager. There are no clear ways on the display to proceed forward unless I wish to use “AppCafe”, “PC-BSD Control Panel”, or the “PC-BSD Handbook”. I tried to use the magic keyboard combination of Control-Alt-Backspace to exit out of X Windows to no avail, the key combination does not work. I then inserted Control-Alt-Delete which reset the system and led me directly back to the login window. This time I selected the default window manager, of KDE and logged in. The system did at this point proceed properly.

I tried to start a basic application, in this case I wandered through the applications and selected “Marble” in the education category. The app failed silently. After that I went to system update and started the update search. The wait for progress was rather long at about five minutes, but I did see there were “Base System Updates” available, what they are is not stated, but I elected to install them anyways. The progress bar does not really fill up in the way that a consumer would expect, but rather as a quarter-inch blue rippled box that bounces slowly left and right.

Generally when the system is installed and updated it seems to be competent. The fact that you can’t really stray from the KDE interface is a little bit of a concern, but generally not a huge problem. I would say that PC-BSD really isn’t ready for prime time consumer use yet. Then again, no Linux OS is, at least yet.

WIL WHEATON dot TUMBLR, So any journalist passing through London’s Heathrow has now been warned: do not take any documents with you. Britain is now a police state when it comes to journalists, just like Russia is.

WIL WHEATON dot TUMBLR, So any journalist passing through London’s Heathrow has now been warned: do not take any documents with you. Britain is now a police state when it comes to journalists, just like Russia is..

This post by Wil Wheaton is a really great reminder that when you are traveling, and I wouldn’t necessarily just put this as international to Britain but even when visiting the next town or crossing state lines even. Rights are being trampled everywhere you go, wether it be from a out-of-control cop, a bloodthirsty Sheriffs deputy or even a sticky-fingered TSA agent there is no lack of potential thugs, enemies, and thieves in your midst.

There are ways to secure your data and keep it handy as well. Store everything in an encrypted disk image or TrueCrypt archive on a cloud service like Dropbox or Google Drive and duplicate the same things in your memory sticks. If the thugs take your devices then you can rest assured that all you lost was the material itself, but no content.

I’m surprised that journalists and people who know journalists don’t all use GPG to secure their communications. I would think that if you were a whistleblower or had contact with a whistleblower that these little checkboxes would be foremost on your mind and already checked off.

You can’t trust any government, any cop, or any Vampire to keep their word. This goes for everyone as well, including your carrier and service providers. What should Verizon know? Shit. How about Dropbox? The same. Trust nobody and you’ll be safer than someone who trusted someone else. Trust is earned and right now, very very few people have it.

Encrypt Everything

Lavabit and Silent Circle have given up when it comes to providing encrypted email communications. Mega plans on providing something to cover the gap and in general the only real way to deal with privacy-in-email is end-to-end encryption. There was talk that at some point email might give way to writing letters and using the US Postal Service but there as well you’ve got Postmasters writing commands taped to mail about how everything has to be photocopied and stored – so even the US Postal Service is full of spies, the only thing the US Postal Service can be trusted to carry is junk mail.

What is the answer? Pretty Good Privacy. PGP, or rather, the non-Symantec version of it which is the GNU one, the GPG. If you really want to keep what you write private when you send it to someone else, the only way to do that is for everyone to have GPG installed on their email system so you can write email using their public key, which converts your email to cyphertext, secure from even the NSA’s prying eyes, and requires your recipient to unlock the message using their secret key, which they have.

I’ve been playing with PGP and GPG now for a very long time and I decided I would at least make a route available if anyone wanted to contact me with privacy intact – my public keys are on my blog, they are also on all the keyservers including the one hosted and run by MIT and the GPG Keyserver as well. To send me a private message via email all you need to do is get GPG, set it up, create your secret and public key, get my public key, use it to write me an email and only I’ll be able to read it. The NSA will just flag the encrypted contents for later analysis and thanks to AES–256, they’ll be hard pressed to get to the plaintext in your message.

That’s the way around all of this. GPG for everything. GPG public keys for email, for chat, for VPN, for files, and HTTP-in-GPG. Everything pumped through GPG. Since the government won’t stop spying on us, it’s our duty as citizens to secure our own effects against illegal search and siezure, and technology exists to do so.

Encrypt everything.

Encrypted Time Machine Drive Botch in Mac OSX 10.8.2 Mountain Lion

We had a Firewire 800 drive botch when it came to whole-volume encryption in Mac OSX 10.8.2 Mountain Lion. We lost the password and couldn’t recover it. The drive refused to erase, all the options were grayed out. I refuse to believe that a software change can render hardware junk, so there had to be a way, and I found it. Here’s the procedure:

  • Attach botched drive to computer, since the password won’t work, cancel the unlock dialog box
  • Open Terminal
  • Enter command: diskutil CoreStorage list
  • You will get a long list, you are looking for the UUID of the “Logical Volume Group” at the very top of the list, for the drive that is affected.
  • Enter command: diskutil CoreStorage delete [UUID]
  • The system will eject the volumes, destroy the grouping, erase the disk, then initialize the disk, mount it and finish.
  • Done!

It's silly, and you should stop doing it.

Email confidentiality footers annoy me. I see them frequently on many emails that I get and I think of them as meaningless text that really should be ignored. That an email is somehow a private exchange of information is laughable. Email is sent in plaintext using an open protocol and on the wire it’s all unencrypted.

What really brings this to the forefront is when I see these meaningless bits of mental flotsam and jetsam clogging up my email box because someone set a vacation autoresponse and their membership on a email list is causing them to constantly reply with a “I’ll be out from…” email with this stupid block of text at the bottom asserting that the email is the property of blah blah blah.

Writing email has the same security protections as writing a postcard and tying it to a bird and letting it fly off. Your assertion that your communications are somehow proprietary or classified is utterly hilarious.

If people really wanted to make this not so utterly irrelevant, they should use public-key encryption or at least something like ROT–13 encryption so that the text isn’t readily apparent and takes some work to decode. Sending plaintext with this silly block at the bottom just musses up the display and doesn’t mean anything to anybody. So stop it.

Dropbox Lied to Users About Data Security, Complaint to FTC Alleges | Threat Level | Wired.com

Dropbox Lied to Users About Data Security, Complaint to FTC Alleges | Threat Level | Wired.com.

Read the above article, it’s quite good and covers the problems that many geeks have with Dropbox. I have to admit that I’m quite fond of finding ways to “Have my cake and eat it too” and in the spirit of that saying it’s important to highlight a core issue that needs to be covered: If you don’t manage your own security, you don’t have any.

Every service is vulnerable to a search and seizure order as long as it’s hardware exists within the United States. Any company that claims that they protect your data even from this basic assumption is lying to you. You can help them by helping yourself. The people who run Dropbox certainly have aims to secure your data, otherwise nobody but a scant few would be willing to store their data in the cloud. This situation is only half-way to what is really required to make a service like Dropbox a real charmer. It comes down to security and I’ve written about it at length before. The end user has to meet Dropbox for the other half of the way. Dropbox encrypts their data using AES-256 and they have a master key that they use along with yours so that they can maintain a backdoor in case of a search and seizure order to fulfill. Protect yourself by using any number of applications, ranging from TrueCrypt, iCrypt, openssh, to encrypted DMG files. If you create one of these encrypted files to store your private information then send it to Dropbox, even if they have to divulge the file to the authorities all they can provide them is another AES-256 encrypted file that they don’t have a key to. When the authorities try to pry open the file, all they’ll see is noise, because they don’t have your key.

It’s really quite easy when you think of it, Dropbox is at most 50% secure. You can provide another 50% making your use of Dropbox 100% secure. It all comes down to going that little extra inch with any of the tools covered above. I can’t help but really love encrypted DMG files as they are the most convenient to use with Macs. You just double-click on the DMG file, enter in your password, and the volume is mounted as if it were a drive on your computer. All the files are plain and easy to use. Ejecting the drive after you are done using it closes it and the data lives 100% secure in the cloud.

Getting bent because Dropbox only gives you 50% security is rather dumb. Anyone at all has to assume that it maxes out at 50% irrespective of what Dropbox claims. If you are smart and secure your own effects, then you’ve nothing to worry about and can get over this silly thing without a single thought. Makes sense to me.