0

Some brief technical notes on Venom

Like you, I was displeased by the lack of details on the "Venom" vulnerability, so I thought I'd write up what little I found.

The patch to the source code is here. Since the note references CVE-2015-3456, we know it's venom:
http://git.qemu.org/?p=qemu.git;a=commit;h=e907746266721f305d67bc0718795fedee2e824c

Looking up those terms, I find writeups, such as this one from RedHat:
https://securityblog.redhat.com/2015/05/13/venom-dont-get-bitten/

It comes down to a typical heap/stack buffer overflow (depending), where the attacker can write large amounts of data past the end of a buffer. Since this is the kernel, there are no protections like NX or ASLR. To exploit this, you'd likely need some knowledge of the host operating system.

The details look straightforward, which means a PoC should arrive by tomorrow.

This is a hypervisor privilege escalation bug. To exploit this, you'd sign up with one of the zillions of VPS providers and get a Linux instance. You'd then, likely, replace the floppy driver in the Linux kernel with a custom driver that exploits this bug. You have root access to your own kernel, of course, which you are going to escalate to root access of the hypervisor.

People suggest adding an exploit to toolkits like Continue reading

0

How to fix the CFAA

Someone on Twitter asked for a blogpost on how I'd fix the CFAA, the anti-hacking law. So here is my proposal.

The major problem with the law is that the term "authorized" isn't clearly defined. You non-technical people think the meaning is obvious, because you can pick up a dictionary and simply point to a definition. However, in the computer world, things are a bit more complicated.

It's like a sign on a store window saying "No shirt, no shoes, no service" -- but you walk in anyway with neither. You know your presence is unwanted, but are you actually trespassing? Is your presence not "authorized"? Or, should we demand a higher standard, such as when the store owner asks you to leave (and you refuse) that you now are trespassing/unauthorized?

What happens on the Internet is that websites routinely make public data they actually don't want people to access. Is accessing such data unauthorized? We saw that a couple days ago, where Twitter accidentally published their quarterly results an hour early on their website. An automated script discovered this and republished Twitters results to a wider audience, ironically using Twitter to do so. This caused $5-billion to drop off Continue reading

0

Ultron didn’t save the world

The movie Avengers: Age of Ultron has a message for us in cybersec: In our desire to save the world, we are likely to destroy it.

0

Review: Avengers, Age of Ultron

Today was the opening of the movie "Avengers: Age of Ultron". The best way to describe it is this. On the first date, you went and saw "The Avengers". You felt the rush of something new, and you were quite satisfied. This movie, "Age of Ultron", is the second date. You already know what to expect, but that doesn't matter, because you progress past the holding-hands stage. You didn't go all the way, but you know that's coming on the third date, with "Avengers: Infinity Wars".

0

Some notes on why crypto backdoors are unreasonable

Today, a congressional committee held hearings about 'crypto backdoors' that would allow the FBI to decrypt text messages, phone calls, and data on phones. The thing to note about this topic is that it's not anywhere close to reasonable public policy. The technical and international problems are unsolvable with anything close to the proposed policy. Even if the policy were reasonable, it's unreasonable that law enforcement should be lobbying for it.

Crypto is end-to-end

The debate hinges on a huge fallacy, that it's about regulating industry, forcing companies like Apple to include backdoors. This makes it seem like it's a small law. The truth is that crypto is end-to-end. Apple sells a generic computer we hold in our hand. As a user, I can install any software I want on it -- including software that completely defeats any backdoor that Apple would install. Examples of such software would be Signal and Silent Circle.

It seems reasonable that you could extend the law so that it covers any software provider. But that doesn't work, because software is often open-source, meaning that anybody can build their own app from it. Starting from scratch, it would take me about six-months to write my Continue reading

0

The hollow rhetoric of nation-state threats

0

Solidarity

The government's zealous War on Hackers threatens us, the good hackers who stop the bad ones. They can't tell the good witches from the bad witches. When members of our community get threatened by the system, we should probably do more to stand in solidarity with them. I mention this because many of you will be flying to SFO this coming week for the RSA Conference, which gives us an opportunity to show solidarity.

Today, a security researcher tweeted a joke while on a plane. When he landed, the FBI grabbed him and confiscated all his stuff. The tweets are here:

Find myself on a 737/800, lets see Box-IFE-ICE-SATCOM, ? Shall we start playing with EICAS messages? "PASS OXYGEN ON" Anyone ? :)
— Chris Roberts (@Sidragon1) April 15, 2015

Chris Roberts' area of research is embedded control systems like those on planes. It's not simply that the FBI grabbed him because of a random person on a plane, but specifically because he's a security researcher. He's on the FBI's radar (so to speak) for things like this Fox News interview.

I suggest we all start joke tweeting along these lines,  from the airplanes, like:

DFW->SFO. Playing with airplane wifi. I Continue reading

0

Masscanning for MS15-034

So Microsoft has an important web-server bug, so naturally I'd like to scan the Internet for it. I'm running the scan now, but I'm not sure it's going to give any useful results.

The bug comes from adding the following header to a web request like the following

Range: bytes=0-18446744073709551615

As you can see, it's just a standard (64-bit) integer overflow, where 18446744073709551615 equals -1.

That specific header is harmless, it appears that other variations are the ones that may cause a problem. However, it serves as a useful check to see if the server is patched. If the server is unpatched, it'll return the following error:

HTTP/1.1 416 Requested Range Not Satisfiable

From the PoC's say, a response that looks like the following means that it is patched:

The request has an invalid header name

However, when I run the scan across the Internet, I'm getting the following sorts of responses from servers claiming to be IIS:

HTTP/1.1 200 OK
HTTP/1.1 206 Partial Content
HTTP/1.1 301 Moved Permanently
HTTP/1.1 302 Object moved
HTTP/1.1 302 Found
HTTP/1.1 302 Redirect
HTTP/1.1 401 Unauthorized
HTTP/1.1 403 Forbidden
HTTP/1.1 404 Object Not Found
Continue reading

0

Should billionaires have to disclose their political contributions?

The Intercept has a story about a billionaire sneaking into a Jeb Bush fundraiser. It points out:

"Bush’s campaign operation has taken steps to conceal the names of certain big-money donors. ... Bush’s Right to Rise also formed a 501(c)(4) issue advocacy wing, which, like a Super PAC, can raise and spend unlimited amounts of money — but unlike a Super PAC, never has to reveal donor names."

This leads me to ask two questions:

1. Should billionaires be allowed to spend unlimited amounts of money promoting their politics?
2. If they can spend unlimited amounts, should they be forced to disclose them?

If you know me, you know that I'm asking a trick question. I'm not referring to venture capitalist Ron Conway, the billionaire mentioned the story. I'm instead referring to Pierre Omidyar the billionaire founder of eBay who funds The Intercept, which blatantly promotes his political views such as those on NSA surveillance. Can Omidyar spend endless amounts on The Intercept? Should he be forced to disclose how much?

This question is at the heart of the Supreme Court decision in Citezen's United. It comes down to this: the Supreme Court was unable to craft rules that could tell Continue reading

0

Scalability of the Great Cannon

Here is a great paper on China's Great Cannon, which was used to DDoS GitHub. One question is how scalable such a system can be, or how much resources it would take for China to intercept connections and replace content.

0

Stop making the NSA the bogeyman of privacy

Snowden is my hero, but here's the thing: the NSA is the least of our worries. Firstly, their attention is foreign, not domestic. Secondly, they are relatively uncorrupt. Our attention should be focused on the corrupt domestic law-enforcement agencies, like the ATF, DEA, and FBI.

0

No, 75% are not vulnerable to Heartbleed

A little-known company "Venafi" is suddenly in the news implying 75% of major systems are still vulnerable to Heartbleed. This deserves a rating of "liar liar pants on fire".

The issue isn't patches but certificates. Systems are patched, but while they were still vulnerable to Heartbleed, hackers may have stole the certificates. Therefore, the certificates need to be replaced. Not everyone has replaced their certificates, and those that have may have done so incorrectly (using the same keys, not revoking previous).

Thus, what the report is saying is that 75% haven't properly updated their certificates correctly. Naturally, they sell a solution for that problem.

However, even this claim isn't accurate. Only a small percentage of systems were vulnerable to Heartbleed in the first place, and it's hard to say which certificates actually needed to be replaced.

That's why you have the weasely marketing language above. It's not saying 3 out of 4 of all systems, but only those that were vulnerable to begin with (a minority). They aren't saying they are still vulnerable to Heartbleed itself, but only that they are vulnerable to breach -- due to the certificates having been stolen.

The entire report is so full of this Continue reading

0

The .onion address

A draft RFC for Tor's .onion address is finally being written. This is a proper thing. Like the old days of the Internet, people just did things, then documented them later. Tor's .onion addresses have been in use for quite a while (I just setup another one yesterday). It's time that we documented this for the rest of the community, to forestall problems like somebody registering .onion as a DNS TLD.

One quibble I have with the document is section 2.1, which says:

1. Users: human users are expected to recognize .onion names as having different security properties, and also being only available through software that is aware of onion addresses.

0

Pin-pointing China’s attack against GitHub

For the past week, the website "GitHub" has been under attack by China. In this post, I pin-point where the attack is coming from by doing an http-traceroute.

GitHub is a key infrastructure website for the Internet, being the largest host of open-source projects, most famously Linux. (I host my code there). It's also a popular blogging platform.

Among the zillions of projects are https://github.com/greatfire and https://github.com/cn-nytimes. These are mirrors (copies) of the websites http://greatfire.com and http://cn.nytimes.com. GreatFire provides tools for circumventing China's Internet censorship, the NYTimes contains news stories China wants censored.

China blocks the offending websites, but it cannot easily block the GitHub mirrors. It's choices are either to block or allow everything on GitHub. Since GitHub is key infrastructure for open-source, blocking GitHub is not really a viable option.

Therefore, China chose another option, to flood those specific GitHub URLs with traffic in order to pressure GitHub into removing those pages. This is a stupid policy decision, of course, since Americans are quite touchy on the subject and are unlikely to comply with such pressure. It's likely GitHub itself can resolve the issue, as there are a zillion ways to respond. If Continue reading

0

War on Hackers: a Clear and Present Danger

President Obama has upped his war on hackers by declaring a "state of emergency". This triggers several laws that grant him expanded powers, such as seizing the assets of those suspected of hacking, or taking control of the Internet.

One one hand, this seems reasonable. Hackers from China and Russia are indeed a threat, causing billions in economic damage every year, by stealing money and intellectual property. This declaration specifically targets these issues. Presumably, in the next few weeks, we'll see announcements from the Treasure Department seizing assets from Chinese companies known to have stolen intellectual property via hacking.

But on the other hand, it's problematic. Declarations of emergency tend to be permanent. We already operate under 30 declarations of emergencies dating back to the Korean war. Once government grabs new powers, it tends not to give them back. Also, this really isn't an "emergency", the hacking it addresses goes back a decade. It's obvious corruption of the "emergency" provisions in the law for the President to bypass congress and rule by decree.

Moreover, while tailored specifically to the threats of foreign hackers, it ultimately affects everyone everywhere. It allows the government to bypass due process and seize Continue reading

0

Message to Errata employees

Dear employees,

Starting next week, Errata Security will be following RSA Conference's lead and institute a "Morality Dress Code" in order to deal with the problem of loose women on the premises.

Attire of an overly revealing or suggestive nature is not permitted. Examples of such attire may include but are not restricted to:

• Tops displaying excessive cleavage;
• Tank tops, halter tops, camisole tops or tube tops;
• Miniskirts or minidresses;
• Shorts;
• Lycra (or other Second-Skin) bodysuits;
• Objectionable or offensive costumes.

"Shalim" by Zivya - Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons - http://commons.wikimedia.org/wiki/File:Shalim.JPG#/media/File:Shalim.JPG

0

x86 is a high-level language

Just so you know, x86 machine-code is now a "high-level" language. What instructions say, and what they do, are very different things.

I mention this because of those commenting on this post on OpenSSL's "constant-time" calculations, designed to avoid revealing secrets due to variations in compute time. The major comment is that it's hard to do this perfectly in C. My response is that it's hard to do this even in x86 machine code.

Consider registers, for example. Everyone knows that the 32-bit x86 was limited to 8 registers, while 64-bit expanded that to 16 registers. This isn't actually true. The latest Intel processors have 168 registers. The name of the register in x86 code is really just a variable name, similar to how variables work in high-level languages.

So many registers are needed because the processor has 300 instructions "in flight" at any point in time in various stages of execution. It rearranges these instructions, executing them out-of-order. Everyone knows that processors can execute things slightly out-of-order, but that's understated. Today's processors are massively out-of-order.

Consider the traditional branch pair of a CMP (compare) followed by a JMPcc (conditional jump). While this is defined as two separate instructions as Continue reading

0

What ever it is, CISA isn’t cybersecurity

In the next couple months, Congress will likely pass CISA, the Cybersecurity Information Sharing Act. This is a bad police-state thing. It will do little to prevent attacks, but do a lot to increase mass surveillance.

They did not consult us security experts when drafting this bill. If they had, we would have told them the idea doesn’t really work. Companies like IBM and Dell SecureWorks already have massive “cybersecurity information sharing” systems where they hoover up large quantities of threat information from their customers. This rarely allows them to prevent attacks as the CISA bill promises.

In other words, we’ve tried the CISA experiment, and we know it doesn’t really work.


While CISA won’t prevent attacks, it will cause mass surveillance. Most of the information produced by countermeasures is in fact false-positives, triggering on innocent anomalies rather than malicious hackers. Your normal day-to-day activities on the Internet occasionally trigger these false-positives. When this information gets forwarded to law enforcement, it puts everyone in legal jeopardy. It may trigger an investigation, or it may just become evidence about you, for example, showing which porn sites you surf. It’s mass surveillance through random sampling.

That such mass surveillance is the goal Continue reading

0

GitHub won because it’s social-media

0

No, the CIA isn’t stealing Apple’s secrets

The Intercept news site by Glenn Greenwald is activism rather than journalism. Their stories don't reference experts knowledgeable about subjects, but only activists who are concerned about the subjects. This was demonstrated yet against in their piece claiming "The CIA Campaign to Steal Apple's Secrets". Yes, the Snowden documents are real, but pretty much everything else is made up.

Here's the deal. Terrorist leaders use iPhones. They are a status symbol, and status symbols are important to leaders. Moreover, since Apple's security is actually pretty good, terrorists use the phones for good reason (most Android devices suck at security, even the Blackphone).

When CIA drones bomb a terrorist compound, iPhones will be found among the bodies. Or, when there is a terrorist suspect coming out of a dance club in Karachi, a CIA agent may punch them in the face and run away with their phone. However, it happens, the CIA gets phones and wants to decrypt them.

Back in 2011 when this conference happened, the process of decrypting retrieved iPhones was time consuming (months), destructive, and didn't always work. The context of the presentation wasn't that they wanted to secretly spy on everyone's phones. The context was Continue reading