Archive for the ‘speculation’ Category

Paranoia, anybody: redux, part II, reloaded, the sequel, extended, directors cut.

Friday, February 19th, 2010

It’s been roughly 24 hours since I posted  about paranoia and foursquare. I was correct in my foresight expecting people to respond somewhat forcibly, or strongly – but I got my responses from ENTIRELY the wrong crowd I was trying to speak to: my infosec friends.

I wanted to acknowledge valid points that were brought up in conversations carried on after the fact and transmogrify the undertone from my last post into an overtone in this one. My suspicion is that my previous snarkiness may have obfuscated the clarity of the point I was trying to make.

  • Yes, absolutely, I agree that over-sharing your location creates a vulnerability and allows an attacker to build an attack profile (excessive meaning say, more than 3-5 checkins daily). As one friend put it “updating foursquare 24/7″ = bad. Foursquare is not “HELPING” the problem – yes they are “CONTRIBUTING” to it, but they are not “THE” problem.
  • This is not a “new” attack vector. Foursquare is not the first application to allow one to publish ones whereabouts (if you REALLY wanna crap your pants, have a look at lattitude. If you think foursquare is bad your head will fall off)
  • No, in this context, knowing if you’re in a building or in a certain room to a building is irrelevant. The point here is you’re “leaving your home vulnerable”. Personal security is a different subject entirely, and I prefer to stay on topic. The site that was mentioned was “Please rob me”, inferring “come to my home and rob it while I’m not there”. If people would like to have a healthy discussion about personal security, I’d be happy to be a part of it – however this is not it. This discussion is about the home.
  • It is less likely that an ACTUAL home-invader will use foursquare over any other social/web2.0 site. Standard usage dictates one has to click an accept button to allow someone to view their checkins (unless they’re published to facebook/twitter, then it’s moot anyway). I’ve had friends that have had their homes burglarized and in every case the attacker was not what any of us would consider an “advanced enough” computer user to utilize foursquare as a prelude to a burglary. It was always something like “we saw them packing up to leave on a ski trip” -visual, in person. If an attacker is enlightened enough to employ the use of attacks like CSRF and social engineering methodology they’re going to go after what you have in the bank, in investments, carbon credits (a new one!) and other things that are far more valuable than your television.
  • In this context its foursquare that’s being thrown under the bus. Their ‘fault’ in this case was to take an already popular idea (dodgeball) and make it more popular. It’s the “in” thing to do rightnow – overshare. Some people do it, other people don’t – people manage their own risk. Telling twitter you’re going to the bar, versus checking in on foursquare AT the bar, versus gowalla, or a facebook update – its all the same thing: You’re telling the internet you’re not home. The problem is the behavior, not the “tool used”.

The last line of the last post I wrote is more or less the overall point I’m trying to make. Somehow, or for some reason the masses have decided to have an epiphany where they throw their hands in the air and declare foursquare unsafe.

Agreed, they have a valid point. I won’t argue that, but its synonymous with walking into the burn ward at a hospital, walking past rows and rows of disfigured and suffering individuals, stopping at one random person then exclaiming to the world how THIS PARTICULAR PERSON is suffering and needs medical attention and oh-woe-is-me-what-a-world.

Generally speaking, the same people who have ‘come to this realization now’ are guilty of using many other applications that “tell people they are not home”.

My point, reconstituted without snark is: You’ve been doing it for years, and you JUST NOW realized it? THATS the problem. Not foursquare. The very same author of the blogpost I linked to is guilty of frequently publishing their location using a variety of applications. At best I can only speculate, but my speculation is that it was done for the readership and stir the pot – not to actually provide any real warning.

State of the pwnion.

Thursday, August 6th, 2009
message begins
Personal details were revealed, emails, chat logs – pretty scary stuff – and very sobering. A clear demonstration that things like cross site scripting and the spreading of malware (likely for the use of cascading spam or addition to botnets) is the least of our problems. Also clear proof that people who consider themselves security folks have to be very wary of using creature comforts such as reusing passwords or even operating a wordpress blog (3 updates in a month?! and 2.8.2 is vulnerable? gaw!).
The textfile the group distributed was called zf05.txt and after skimming it’s abundantly clear that wordpress played a huge part in these folks getting rooted. Almost every example was sort of an ‘all in one’ server that was used for ‘whatever’. Its also become clear that jam packing one server with a bunch of services makes it more vulnerable to compromise. Ever heard of KISS? “Keep it simple, stupid”. It’s used very commonly among engineers, computer people – you name it. Anyone that has to build things or design things. The minute you start adding complexity for no reason the proverbial altimeter begins its decline.
People who fake tech exacerbate things. There are groups that call themselves “tech” when in reality they are simply PR or Marketing. The Web 2.0 craze has hypnotized people into putting almost everything they think and do ‘behind the scenes’. They let someone else worry about it. Some ruby programmers I’ve met are incapable of manually issuing a sql query. Others are incapable of interacting with sql unless they have phpmyadmin. These folks generate a requirement to artificially make systems more complex and less secure entirely to suit their evergrowing hatred of looking things up themselves or actually learning anything about the technology they use every day. The easiest way to think about it is this: Think of some people. Now think of these people all owning cars. Think of these people now requiring something as simple as an oil change, a tire change, or a simple tune up. Now think of these people taking their cars to a shop to get work done – for whatever reason: maybe they lack the tools, maybe their HOA doesn’t allow them to perform work on their cars on the grounds (those HOA people desperately need to be stabbed in the lungs, by the way) or maybe they just don’t know how. Now lets imagine these people have the work done, and are talking to the mechanics as they are preparing the invoice behind the counter. The mechanic begins to explain how their oil was changed, and these people abjectly refuse to learn or understand how this works even from a top-level non-technical aspect – they plug their ears and yell “NO! NO! AAALALALALA!! NOT LISTENING NO NOOOO! ALLALAAAAAA!”.
These people strongly support a fancy new term. “Cloud Computing”. Cloud computing will make this worse for everyone.
Let me jump away for a moment. I’d like to point out a fact. The attackers that distributed zf05.txt made a valid point – a point I’ve tried to make to peers, friends and clients alike – If your site/data are on shared hosting and you consider them secure that may mitigate some amount of risk. But if the other people hosting their data are vulnerable and your data is on the same system, you’re still vulnerable.
Now we have some ingredients – lets make a stew. Lets take these bits of information and put them all together and let it simmer.
- Non technical people whos requirements and behavior are insecure and promote systems being rooted
- Systems with lots of various services running on them
- A new trend of mashing these systems together to form giant systems that do the same thing, ending up being bigger and more powerful
- Commonly used software being exploited within a week of a patch.
Mix in a bowl with a wisk until creamy. Add a teaspoon of extra virgin olive oil to a cast-iron skillet. Add a bit of freshly cracked pepper to the oil and some freshly pressed/minced garlic. Let simmer until the pepper and garlic begin to bubble, then pour the mixture from the bowl into the skillet and add a squeeze of fresh key lime if you wish. Cook until firm or golden brown, flip once, then serve! Let stand for 10 minutes to cool. What do you get? What does it smell like? (Well if people actually taste of chicken then that may make one hell of a breakfast omlette). We dont know. Here’s why we don’t know:
- “Business people” like the idea of getting rid of systems administrators and IT overhead
- “Cloud Computing” does not have a security model yet
- There are no standards – this stuff is too new
- Far too many people are comfortable being hacked, and say “oh there’s nothing important on that sit/box”
.. Really, guys? You don’t use that same wordpress password everywhere? For your bank, for gmail, for your car insurance or your mobile provider to login? If a blackhat gets that password you’re really okay with it? If thats the case, I’d like you to kindly leave the internet, never to return. Please – do us all a favor, for the people that like keeping their privates private and their secrets secret, go away.
So we’re going to take all of these insecurities, vulnerabilities and holes – package them up with non-technical people demanding insecure practices so that they don’t have to learn or think and we’re going to replicate this ad nauseum and store the results in one gigantic computer grid system? Awesome. Maybe I should trade in my whitehat for a black one – since thats obviously where all the focus, media, fear and money are going to be. Or maybe I’ll just make my white hat bigger – perhaps people will come to their senses and listen to fact and reason. Perhaps not. I guess we’ll see.
I’m not the only one, either…
http://darkreading.com/securityservices/security/app-security/showArticle.jhtml?articleID=218102139&cid=RSSfeed
http://www.sensepost.com/blog/3706.html – open the ppt, this was the defcon talk. they pwned amazon ec2.
http://evilpacket.net/ – see the ‘theft of a rackspace cloud api key’. These guys got root on the rackspace/mosso cloud.

I was late to hear – by a day. Thats 10 years in internet time, we all know. If you’re not in InfoSec you probably didn’t hear. Maybe you heard somewhere, irc, twitter, other bits of the intarnets that Kevin Mitnick got hacked. Everyone chuckled. As it turns out a whole bunch of people got compromised. People I know personally who I consider friends. Rob Fuller, Dan Kaminsky, the Hak5 group and a handful of others, including Kevin Mitnick.

Personal details were revealed, emails, chat logs – pretty scary stuff – and very sobering. A clear demonstration that things like cross site scripting and the spreading of malware (likely for the use of cascading spam or addition to botnets) is the least of our problems. Also clear proof that people who consider themselves security folks have to be very wary of using creature comforts such as reusing passwords or even operating a wordpress blog (3 updates in a month?! and 2.8.2 is vulnerable? gaw!).

(more…)

Heads Up!

Friday, July 10th, 2009

First, click this, and read.

There’s a group out there now called ‘Anti-Sec’ and they’re angry about how full disclosure works.

They decided that to make their point, they’d hack imageshack – All of it. They replaced every image on imageshack with a jpeg of their manifesto statement.

In it there are claims of threats – aimed at security blogs and other sites that publish exploits.

It’s probably a good idea to get an audit right now – as contradictory as that sounds.

If theres a hacking group traveling the internet touting “everyone and everything will get hacked” and their first tent-stake in the ground was imageshack?

It’s time for a perimeter check.

We’ve been doing free 15 minute audits all week – so we’re going to extend that through next week as well. Come see us if you’d like one!

Post Mortem

Wednesday, November 26th, 2008

So security101 went fairly well – people didn’t show up until later, and I had spent too much time screwing aroung with ettercap and MITM attacks to have enough battery to complete the entirety of the talk with all the examples I had hoped for.

Some of the attendees ended up asking lots of questions so the ‘flow’ I had envisioned sort of went out the window – but I’d much rather have people interested and actively asking me questions: It shows interest. I’d rather have interest then have them all silent while I blather on and on.

We all ended up at my place afterwards and I was giving short demos on MITM dns tomfoolery, rewriting all queries for microsoft.com to linux.com, and doing SSL MITM attacks against hotmail using ettercap. Pretty fun stuff!

I’ll be holding the class again for anybody that missed it the first time and wants to have it again, but I haven’t chosen a date yet.

If you’re interested in a date, please leave a comment! I’d like to hold the class when more people can attend.