Monday, October 13. 2014
In the past year we have seen several high-profile breaches of brick and mortar retailers. Estimates range in the tens of millions of credit cards stolen in each case. For the most part, these retailers have weathered the storm with virtually no ill effects. In fact, it seems the same increase in stock price that TJ Maxx saw after their breach still rings true today. A sad fact indeed.
Regardless, the recent slew of breaches has finally prompted the credit card industry to act. They have declared that 2015 will be the year that chip and pin becomes the standard for all card-present transactions. And while chip and pin isn't a silver bullet, and attackers will eventually find new and innovative ways to circumvent it, it has proven to be quite effective in Europe where it has been the standard for years.
Chip and pin changes how the credit card information is transmitted to the processor. Instead of the credit card number being read, in plain text, off of the magnetic strip, the card reader initiates an encrypted communication between the chip on the card and the card reader. The card details are encrypted and sent, along with the user's PIN, to the card processor for verification. It is this encrypted communication between the card and, ultimately, the card processor that results in increased security. In short, the attack vectors used in recent breaches is difficult, if not impossible to pull off with these new readers. Since the information is not decrypted until it hits the card processor, attackers can't simply skim the information at the card reader. There are, of course, other attacks, though these have not yet proven widespread.
At it's heart, though, chip and pin only "fixes" one type of credit card transaction, card-present transactions. That is, transactions in which the card holder physically scans their card via a card reader. The other type of transaction, card-not-present transactions, are unaffected by chip and pin. In fact, the move to chip and pin may result in putting online transactions at greater risk. With brick and mortar attacks gone, attackers will move to online retailers. Despite the standard SSL encryption used between shoppers and online retailers, there are plenty of ways to steal credit card data. In fact, one might argue that a single attack could net more card numbers in a shorter time since online retailers often store credit card data as a convenience for the user.
It seems that online fraud, though expected, is being largely ignored for the moment. After all, how are we going to protect that data without supplying card readers to every online shopper? Online solutions such as PayPal, Amazon Payments, and others mitigate this problem slightly, but we still have to rely on the security they've put in place to protect cardholder data. Other solutions such as Apple Pay and Google Wallet seemingly combine on and offline protections, but the central data warehouse remains. The problem seems to be the security of the card number itself. And losing this data can be a huge burden for many users as they have to systematically update payment information as the result of a possible breach. This can often lead to late payments, penalties, and more.
One possible alternative is to reduce the impact a single breach can cause. What if the data that retailers stored was of little or no value to an attacker while still allowing the retailer a way to simplify payments for the shopper? What if a breach at a retailer only affected that retailer and resulted in virtually no impact on the user? A solution like this may be just what we need.
Instead of providing a retailer your credit card number and CVV, the retailer is provided a simple token. That token, coupled with a private retailer-specific token should be all that is needed to verify a transaction. Tokens can and should be different for each retailer. If a retailer is compromised, new tokens can be generated, reducing the impact on the user significantly. Attackers who successfully breach a retailer can only submit transactions if they can obtain both the private retailer token as well as the user token. And if processors put simple access-control lists in place, it increases the difficulty an attacker encounters when trying to push through a fraudulent transaction.
Obtaining tokens can be handled by redirecting a user to a payment gateway for their initial transaction. The payment gateway verifies the user and their credit card data, and then passes the generated token back to the retailer. This is similar to how retailers using existing online payment processors such as Paypal and Amazon Payments already handle payments. The credit card data never passes through the retailer network. The number of locations credit card data is stored reduces significantly as well. This, in turn, means that attackers have fewer targets and while this increases the risk a payment processor network incurs, one can argue that these networks should already have significant defenses in place.
This is only one possible solution for online payments. There are many other solutions out there being presented by both security and non-security folks. But there seems to be no significant movement on an online solution. Will it take several high-profile online breaches to convince credit card companies that a solution is needed? Or will credit card companies move to protect retailers and card holders ahead of attackers redirecting their efforts? If history is any indication, get used to having your card re-issued several times a year for the foreseeable future.
Thursday, April 10. 2014
Unless you've been living under a rock the past few days, you've probably heard about the Heartbleed vulnerability in OpenSSL that was disclosed on Monday, April 7th. Systems and network administrators across the globe have spent the last few days testing for this vulnerability, patching systems, and probably rocking in the corner while crying. Yes, it's that bad. What's more, there are a number of reports that intelligence agencies may have known about this vulnerability for some time now.
The quick and dirty is that a buffer overflow bug in the code allows an attacker to remotely read memory of an affected system in 64k chunks. The only memory accessible to an attacker would be memory used by the process being connected to, but, depending on the process, there may be a LOT of useful data in there. For instance, Yahoo was leaking usernames and passwords until late Tuesday evening.
The fabulous web comic, xkcd, explains how the attack works in layman's terms. If you're interested in the real nitty gritty of this vulnerability, though, there's an excellent write-up on the IOActive Labs blog. If you're the type that likes to play, you can find proof-of-concept code here. And let's not forget about the client side, there's PoC code for that as well.
OpenSSL versions 1.0.1 through 1.0.1f as well as the 1.0.2 beta code are affected. The folks at OpenSSL released version 1.0.1g on Monday which fixed the problem. Or, at least, the current problem. There's a bit of chatter about other issues that may be lurking in the OpenSSL codebase.
Now that a few days have passed, however, what remains to be done? After all, everyone has patched their servers, right? Merely patching doesn't make the problem disappear, though. Vulnerable code is out there and mistakes can be made. For the foreseeable future, you should be regularly scanning your network for vulnerable systems with something like Nmap. The Nmap NSE for Heartbleed scanning is already available. Alternatively, you can use something like Nagios to regularly check your existing servers.
Patching immediately may not have prevented a breach, either. Since Heartbleed doesn't leave much of a trace beyond some oddities that your IDS may have seen, there's virtually no way to know if anything has been taken. The best way to deal with this is to just go ahead and assume that your private keys are compromised and start replacing them. New keys, new certs. It's painful, it's slow, but it's necessary.
For end users, the best thing you can do is change your passwords. I'm not aware of any "big" websites that have not patched by now, so changing passwords should be relatively safe. However, that said, Wired and Engadget have some of the best advice I've seen about this. In short, change your passwords today, then change them again in a few weeks. If you're really paranoid, change them a third time in about a month. By that time, any site that is going to patch will have already patched.
Unfortunately, I think the fun is just beginning. I expect we'll start seeing a number of related attacks. Phishing attacks are the most likely in the beginning. If private keys were compromised, then attackers can potentially impersonate websites, including their SSL certificates. This would likely involve a DNS poisoning attack, but could also be accomplished by compromising a user's local system and setting a hosts file entry. Certificate revocation is a potential defense against this, but since many browsers have CRL checks disabled by default, it probably won't help. Users will have to watch what they click, where they go, and what software they run. Not much different from the advice given already.
Another possible source of threats are consumer devices. As Bruce Schneier put it, "An upgrade path that involves the trash, a visit to Best Buy, and a credit card isn't going to be fun for anyone." What he's referring to are the many embedded devices we use on a daily basis that may never receive updates to protect the end user. In other words, that router you purchased from the discount store? That may be affected and unless you replace it, you'll continue to be vulnerable. Fortunately, most of these devices aren't configured, by default, to face the Internet, so there may yet be hope.
The Heartbleed vulnerability is a serious contender for the worst security vulnerability ever released. I'm not sure of another vulnerability that exposes so many systems to such a degree as this one. Network and systems administrators will be cleaning up after this one for a while.
Friday, March 28. 2014
Let's get this out of the way. One of the primary reasons I'm writing this is in response to a request by John Carmack for coherent commentary about the recent acquisition of Oculus VR by Facebook. My hope is that he does, in fact, read this and maybe drop a comment in response. <fanboy>Hi John!</fanboy> I've been a huge Carmack fan since the early ID days, so please excuse the fanboyism.
And I *just* saw the news that Michael Abrash has joined Oculus as well, which is also incredibly exciting. Abrash is an Assembly GOD. <Insert more fanboyism here />
Ok, on to the topic a hand. The Oculus Rift is a VR headset that got its public start with a Kickstarter campaign in September of 2012. It blew away it's meager goal of $250,000 and raked in almost $2.5 Million. For a mere $275 and some patience, contributors would receive an unassembled prototype of the Oculus Rift. Toss in another $25 and you received an assembled version.
But what is the Oculus Rift? According to the Kickstarter campaign :
Oculus Rift is a new virtual reality (VR) headset designed specifically for video games that will change the way you think about gaming forever. With an incredibly wide field of view, high resolution display, and ultra-low latency head tracking, the Rift provides a truly immersive experience that allows you to step inside your favorite game and explore new worlds like never before.
In short, the Rift is the culmination of every VR lover's dreams. Put a pair of these puppies on and magic appears before your eyes.
For myself, Rift was interesting, but probably not something I could ever use. Unfortunately, I suffer from Amblyopia, or Lazy Eye as it's commonly called. I'm told I don't see 3D. Going to 3D movies pretty much confirms this for me since nothing ever jumps out of the screen. So as cool as VR sounds to me, I would miss out on the 3D aspect. Though it might be possible to "tweak" the headset and adjust the angles a bit to force my eyes to see 3D. I'm not sure if that's good for my eyes, though.
At any rate, the Rift sounds like an amazing piece of technology. In the past year I've watched a number of videos demonstrating the capabilities of the Rift. From the Hak5 crew to Ben Heck, the reviews have all been positive.
And then I learned that John Carmack joined Oculus. I think that was about the time I realized that Oculus was the real deal. John is a visionary in so many different ways. One can argue that modern 3D gaming is largely in part to the work he did in the field. In more recent years, his visions have aimed a bit higher with his rocket company, Armadillo Aerospace. Armadillo started winding down last year, right about the time that John joined Oculus, leaving him plenty of time to deep dive into a new venture.
For anyone paying attention, Oculus was recently acquired by Facebook for a mere $2 Billion. Since the announcement, I've seen a lot of hatred being tossed around on Twitter. Some of this hatred seems to be Kickstarter backers who are under some sort of delusion that makes them believe they have a say in anything they back. I see this a lot, especially when a project is taking longer than they believe it should.
I can easily write several blog posts on my personal views about this, but to sum it up quickly, if you back a project, you're contributing to make something a reality. Sometimes that works, sometimes it doesn't. But Kickstarter clearly states that you're merely contributing financial backing, not gaining a stake in a potential product and/or company. Nor are you guaranteed to receive the perks you've contributed towards. So suck it up and get over it. You never had control to begin with.
I think Notch, of Minecraft fame, wrote a really good post about his feeling on the subject. I think he has his head right. He contributed, did his part, and though it's not working out the way he wanted, he's still willing to wish the venture luck. He may not want to play in that particular sandbox, but that's his choice.
VR in a social setting is fairly interesting. In his first Oculus blog post, Michael Abrash mentioned reading Neal Stephenson's incredible novel, Snow Crash. Snow Crash provided me with a view of what virtual reality might bring to daily life. Around the same time, the movie Lawnmower Man was released. Again, VR was brought into the forefront of my mind. But despite the promises of books and movies, VR remained elusive.
More recently, I read a novel by Ernest Cline, Ready Player One. Without giving too much away, the novel centers around a technology called the OASIS. Funnily enough, the OASIS is, effectively, a massive social network that users interact with via VR rigs. OASIS was the first thing I thought about when I heard about the Facebook / Oculus acquisition.
For myself, my concern is Facebook. Despite being a massively popular platform, I think users still distrust Facebook quite a bit. I lasted about 2 weeks on Facebook before having my account deleted. I understand their business model and I have no interest in taking part. Unfortunately, I'm starting to miss out on some aspects of Internet life since some sites are requiring Facebook accounts for access. Ah well, I guess they miss out on me as well.
I have a lot of distrust in Facebook at the moment. They wield an incredible amount of information about users and, to be honest, they're nowhere near transparent enough for me to believe what they say. Google is slightly better, but there's some distrust there as well. But more than just the distrust, I'm afraid that Facebook is going to take something amazing and destroy it in a backwards attempt to monetize it. I'm afraid that Facebook is the IOI of this story. (It's a Ready Player One reference. Go read it, you can thank me later)
Ultimately, I have no stake in this particular game. At least, not yet, anyway. Maybe I'm wrong and Facebook makes all the right moves. Maybe they become a power for good and are able to bring VR to the masses. Maybe people like Carmack and Abrash can protect Oculus and fend off any fumbling attempts Facebook may make at clumsy monetization. I'm not sure how this will play out, only time will tell.
How will we know how things are going? Well, for one, watching his Facebook interacts with this new property will be pretty telling. I think if Facebook is able to sit in the shadows and watch rather than kicking in the front door and taking over, maybe Oculus will have a chance to thrive. Watching what products are ultimately released by Oculus will be another telling aspect. While I fully expect that Oculus will add some sort of Facebook integration into the SDK over time, I'm also hoping that they continue to provide an SDK for standalone applications.
I sincerely wish Carmack, Abrash, and the rest of the Oculus team the best. I think they're in a position where they can make amazing things happen, and I'm eager to see what comes next.
Wednesday, February 19. 2014
I was debugging an odd network issue lately that turned out to have a pretty simple explanation. A client on the network was intermittently experiencing significant delays in accessing the network. Upon closer inspection, it turned out that prior to the delay, the client was being left idle for long periods of time. With this additional information it was pretty easy to identify that there was likely a connection between the client and server that was being torn down for being idle.
So in the end, the cause of the problem itself was pretty simple to identify. The fix, however, is more of a conundrum. The obvious answer is to adjust the timers and prevent the connection from being torn down. But what timers should be adjusted? There are the keepalive timers on the client, the keepalive timers on the server, and the idle teardown timers on the firewall in the middle.
TCP keepalive handling varies between operating systems. If we look at the three major operating systems, Linux, Windows, and OS X, then we can make the blanket statement that, by default, keepalives are sent after two hours of idle time. But, most firewalls seem to have a default TCP teardown timer of one hour. These defaults are not conducive to keeping idle connections alive.
The optimal scenario for timeouts is for the clients to have a keepalive timer that fires at an interval lower than that of the idle tcp timeout on the firewall. The actual values to use, as well as which devices should be changed, is up for debate. The firewall is clearly the easier point at which to make such a change. Typically there are very few firewall devices that would need to be updated as compared to the larger number of client devices. Additionally, there will likely be fewer firewalls added to the network over time, so ensuring that timers are properly set is much easier. On the other hand, the defaults that firewalls are generally configured with have been chosen specifically by the vendor for legitimate reasons. So perhaps the clients should conform to the setting on the firewall? What is the optimal solution?
And why would we want to allow idle connections anyway? After all, if a connection is idle, it's not being used. Clearly, any application that needed a connection to remain open would send some sort of keepalive, right? Is there a valid reason to allow these sorts of connections for an extended period of time?
As it turns out, there are valid reasons for connections to remain active, but idle. For instance, database connections are often kept for longer periods of time for performance purposes. The TCP handshake can take a considerable amount of time to perform as opposed to the simple matter of retrieving data from a database. So if the database connection remains established, additional data can be retrieved without the overhead of TCP setup. But in these instances, shouldn't the application ensure that keepalives are sent so that the connection is not prematurely terminated by an idle timer somewhere along the data path? Well, yes. Sort of. Allow me to explain.
When I first discovered the source of the network problem we were seeing, I chalked it up to lazy programming. While it shouldn't take much to add a simple keepalive system to a networked application, it is extra work. As it turns out, however, the answer isn't quite that simple. All three major operating systems, Windows, Linux, and OS X, all have kernel level mechanisms for TCP keepalives. Each OS has a slightly different take on how keepalive timers should work.
There's a pretty good reference page with information on how to set these parameters that can be found here.
We still haven't answered the question of optimal settings. Unfortunately, there doesn't seem to be a correct answer. The defaults provided by most firewall vendors seem to have been chosen to ensure that the firewall does not run out of resources. Each connection through the firewall must be tracked. As a result, each connection uses up a portion of memory and CPU. Since both memory and CPU are finite resources, administrators must be careful not to exceed the limits of the firewall platform.
There is some good news. Firewalls have had a one hour tcp timeout timer for quite a while. As time has passed and new revisions of firewall hardware are released, the CPU has become more powerful and the amount of memory in each system has grown. The default one hour timer, however, has remained in place. This means that modern firewall platforms are much better prepared to handle an increase in the number of connections tracked. Ultimately, the firewall platform must be monitored and appropriate action taken if resource usage becomes excessive.
My recommendation would be to start by setting the firewall tcp teardown timer to a value slightly higher than that of the clients. For most networks, this would be slightly over two hours. The firewall administrator should monitor the number of connections tracked on the firewall as well as the resources used by the firewall. Adjustments should be made as necessary.
If longer lasting idle connections are unacceptable, then a slightly different tactic can be used. The firewall teardown timer can be set to a level comfortable to the administrator of the network. Problematic clients can be updated to send keepalive packets at a shorter interval. These changes will likely only be necessary on servers. Desktop systems don't have the same need as servers for long-term establishment of idle connections.
Thursday, February 13. 2014
SSL, as I mentioned in a previous blog entry, has some issues when it comes to trust. But regardless of the problems with SSL, it is a necessary part of the security toolchain. In certain situations, however, it is possible to overcome these trust issues.
Commercial providers are not the only entities that are capable of being a Certificate Authority. In fact, anyone can become a CA and the tools to do so are available for free. Becoming your own CA is a fairly painless process, though you might want to brush up on your openSSL skills. And lest you think you can just start signing certificates and selling them to third parties, it's not quite that simple. The well-known certificate authorities have worked with browser vendors to have their root certificates added as part of the browser installation process. You'll have to convince the browser vendors that they need to add your root certificate as well. Good luck.
Having your own CA provides you the means to import your own root certificate into your browser and use it to validate certificates you use within your network. You can use these SSL certificates for more than just websites as well. LDAP, RADIUS, SMTP, and other common applications use standard SSL certificates for encrypting traffic and validating remote connections. But as mentioned above, be aware that unless a remote user has a copy of your root certificate, they will be unable to validate the authenticity of your signed certificates.
Using certificates signed by your own CA can provide you that extra trust level you may be seeking. Perhaps you configured your mail server to use your certificate for the POP and IMAP protocols. This makes it more difficult for an attacker to masquerade as either of those services without obtaining your signing certificate so they can create their own. This is especially true if you configure your mail client such that your root certificate is the only certificate that can be used for validation.
Using your own signed certificates for internal, non-public facing services provides an even better use-case. Attacks such as DNS cache poisoning make it possible for attackers to trick devices into using the wrong address for an intended destination. If these services are configured to only use your certificates and reject connection attempts from peers with invalid certificates, then attackers will only be able to impersonate the destination if they can somehow obtain a valid certificate signed by your signing certificate.
Sound good? Well, how do we go about creating our own root certificate and all the various machinery necessary to make this work? Fortunately, all of the necessary tools are open-source and part of most Linux distributions. For the purposes of this blog post, I will be explaining how this is accomplished using the CentOS 6.x Linux distribution. I will also endeavor to break down each command and explain what each parameter does. Much of this information can be found in the man pages for the various commands.
OpenSSL is installed as part of a base CentOS install. Included in the install is a directory structure in /etc/pki. All of the necessary tools and configuration files are located in this directory structure, so instead of reinventing the wheel, we'll use the existing setup.
To get started, edit the default openssl.cnf configuration file. You can find this file in /etc/pki/tls. There are a few options you want to change from their defaults. Search for the following headers and change the options listed within.
Once the openssl.cnf file is set up, the rest of the process is painless. First, switch into the correct directory.
Next, use the CA command to create a new CA.
And that's about it. The root certificate is located in /etc/pki/CA/cacert.pem. This file can be made public without compromising the security of your system. This is the same certificate you'll want to import into your browser, email client, etc. in order to validate and certificates you may sign.
Now you can start signing certificates. First you'll need to create a CSR on the server you want to install it on. The following command creates both the private key and the CSR for you. I recommend using the server name as the name of the CSR and the key.
openssl req -newkey rsa:4096 -keyout www.example.com.key -out www.example.com.csr
Once you have the CSR, copy it over to the server you're using to sign certificates. Unfortunately, the existing tools don't make it easy to merely name the CSR you're trying to sign, so we need to create our own tool. First, create a new directory to put the CSRs in.
Next, create the sign_cert.sh script in the directory we just created. This file needs to be executable.
That's all you need to start signing certificates. Place the CSR you transferred from the other server into the csr directory and use script we just created to sign it.
The script automatically renamed the newly signed certificate. In the above example, the signed certificate is in www.example.com.2014.crt. Transfer this file back to the server it belongs on and you're all set to start using it.
That's it! You're now a certificate authority with the power to sign your own certificates. Don't let all that power go to your head!
(Page 1 of 52, totaling 260 entries) » next page
"The column of truth has a hole in it"