Posted on | March 17, 2009 | 25 comments
A criminal hacker’s epiphany: Why not automate SQL inject attacks and use botnets to launch them?
That stroke of genius dawned on a criminal coder, possibly Chinese, a little less than a year ago. The result: in just 10 months, botnet-driven SQL injection attacks have been used to plant infections on multi-millions of webpages. These infections now lurk in wait for anyone who happens to click to what appears to be a reputable website.
Click on a tainted webpage and you won’t notice anything amiss. But here’s what happens next: A backdoor gets silently implanted on your harddrive. Through that backdoor the attacker will send coding that silently turns your machine into obedient “bot.” Your botted PC gets slotted into a bot network of 10,000 or more other bots.
As part of this botnet, your machine may be used to deploy spam, spread infections, steal data, launder stolen funds, participate in extortionist denial of service attacks or wage political warfare against small nations. And for good measure: a datastealer will also get installed on your harddrive; it will clean out your email address book and buddy lists, and thus converting all of your contacts into targets for similar infections. And henceforth any information you type at account logon pages, online banking forms, shopping carts — any webpage with a submit buttion — will will get summarily harvested.
History of SQL injection attacks
SQL injection attacks have been around for years. They require time and skill, and were traditionally done manually. A SQL attack involves querying the databases underlying a web page — until the database hiccups and accepts an injection of malicious code. The intruder then gains full access to the data, and a foothold to roam deeper. Socrates, the meth-addicted hacker I wrote about in Zero Day Threat, used a SQL hack to break into a Michigan uniform company’s website to steal 3,000 customer profiles, which he gave to his love interest, Hula Girl. This was back in 2005.
At some point in the spring of 2008, a bright hacker “had a Eureka moment,” says Ryan Barnett, Breach Security’s Director of Research. Instead of trying to steal data from one website at a time, why not use botnets to probe the Interent for webpages whose databases could be easily injected with a small bit of code — just enough to implant backdoor infections.
“It was pretty crafty,” says Barnett.
450,000 webpages attacked daily
LastWatchdog has been unable to get anyone to estimateÃ‚Â how many reputable websitesÃ‚Â have been tainted since then. But the number must beÃ‚Â easily into the multi-millions of webpages. Automated SQL attacks began surfacing last April when 100,000 webpages of the British civil service, United Nations and U.S. Environmental Protection Agency we so hacked.
“The current thinking is the bad guys, thought to be Chinese in origin, use search engines results to identify likely attack targets,” Jeremiah Grossman, White Hat Security founder and CTO told LastWatchdog in spring 2008.Ã‚Â “Next they blindly send in malicious SQL injection traffic to the target websites . . .When someone visits one of these hundreds of thousands of infected web pages, their browser is instructed to connect to a hacker controlled site behind the scenes that exploits their browser and loads their machine with malware.”
For the first five months of 2008 IBM ISS helped large corporations block about 5,000 SQL attacks a day. By mid-June, daily attacks spiked to 25,000; by October they topped 450,000 a day.Ã‚Â And keep this in mind: those blocks only protect corporations that retain IBM ISS to defend their networks. Websites that don’t pay for comparable protection are getting engulfed by the same wave of automated SQL injection attacks, says Holly Stewart, IBM ISS threat response manager.
Cyber crime gangs are “finding tons of sites that have a set of conditions that will allow them to inject malicious software which will infect anyone who visits that site,” says Jack Danahy, founder and CTO of security firm Ounce Labs.
Everything you should know about SQL injection attacks
Below is a FAQ on SQL injection attacks compiled by LastWatchdog with guidance from — and gratitude extended to — Breach Security, WhiteHat, IBM ISS, Guardium and Ounce Labs.
Q: What is a SQL injection attack?
A: SQL refers to the layer of databases underlying most websites.Ã‚Â SQL attacks involves an unauthorized party injecting coding into one of these databases — coding that should not be there. An intruder can can do this by typing coding into the browser URL address line, or into any box of any webform, such as those found on account logon pages or shopping carts.
Q: How are SQL injection attacks typically carried out?
A: Prior to the spring of 2008, SQL attacks were done manually. The hacker would try different database queries from the browser or from pages displayingÃ‚Â web forms, until he successfully injected code into the underlying database. These types of attacks are still done, but on a smaller scale compared to the automated SQL infection/attacks that have come on strong.
Major banks and online merchants are putting up strong defenses, says Phil Neray, Guardium vice-president of security strategy, at Guardium. But regional banks and credit unions, smaller online retailers, and many government agencies remain highly vulnerable to manual, targeted SQL attacks.
As cited above, someÃ‚Â 100,000 webpages of the British civil service, United Nations and U.S. Environmental Protection Agency we so hacked in spring 2008. More recently, Commerce Bank, a small Midwest bank that operates 360 branches in Missouri, Illinois and Kansas, Scarborough & Tweed, a New Hampshire-based company that sells corporate gifts online, and a Rhode Island government Web siteÃ‚Â got hit, according toÃ‚Â Neray.
Q: Are manuul SQL injections still a big concern?
A: Yes. Manual, one-off SQL injection hackers are still out there making a living. Not only can theseÃ‚Â intruders clean out a customer database, they can get a foothold inside of the corporate network serving up the company’s website.
“We’ve seen numerous instances in which attacks leveraged SQL vulnerabilities in order to get inside of corporate networks and gtet access to internal systems and information that was not supposed to be exposed to the Internet,” says Tom Cross, Manager, X-Force Advanced Research, at IBM ISS. “When we first started seeing this kind of attack occurring, it was pretty amazing how simple and straightforward it was, yet how deep the intruder could infiltrate the infrastructure and be relatively unseen.
“The bad guys are getting in and are not being detected,” Cross continued. “They’re finding and taking what they want and leaving, not bothering to clean up.”
Q: What was the breakthrough that enabled automated SQL attacks?
What’s worse, to date 74% of these recently revealed SQL security holes have no available security patch.
Keep in mind those metrics apply to garden-variety web applications. Many websites use custom made web applications; and these more sophisticated programs are even more susceptible to SQL attacks, says IBM’s Stewart.
A: First, the bad guys developed tools to search out SQL vulnerabilities in off-the-shelf and custom web applications being used by web sites all across the Internet. “Automated tools that search for SQL injection vulnerabilities are able to find these vulnerabilities in standard and custom web applications alike,” says IBM’s Stewart.
Q: What are the bad guys doing with all of this stolen personal data?
A: ID theft, of course, begins with stolen data. And millions of tainted websites lurking to silently infect visitors with data stealers appear to be swelling the gigantic pool of stolen identity data, along with viral spam campaigns, such as Waledec and KoobFace. And let’s not forget big data heists, such as the recent breach at Heartland Payments Systems.
It’s not a coincidence that identity theft is also rising. Nearly twice as many people — 7.5% of all U.S. adults — lost money as part of some sort of financial fraud in 2008 according to this survey by Gartner banking analyst Avivah Litan.Ã‚Â Last Fall, Litan surveyed 5,000 consumers. She found 70% had never been a victim of identity theft fraud; 14% had had their credit card information used to charge purchases or get money; 7% said their debit card was used; 6% said a new account had been opened in their name; 5% had money transferred out of their account; and 4% had had checks forged.
“It’s not getting better, it’s getting worse,” says Litan. “I think this coming year will be more severe. A lot of stolen data has yet to be used.”
Q: What can the average person do to avoid getting one of these infections?
A: Do your homework and be ready to give up convenience. There are numerous consumer tools designed to assess the goodness of the Web page you are about to click to, and tell you whether it’s safe. AVG LinkScanner, ScanSafe, McAfee SiteAdvisor,Ã‚Â Enigma SpyHunter andÃ‚Â Authentium SafeCentral are browser-based security tools worth checking out.
And here’s a tip: WinPatrol offers very powerful protection. It’s a terrific free tool, popular with techies since it was created 10 years ago by Bill Pytlovany, one of the original designers of AOL and a longtime open-source practitioner. The premier version, called WinPatrol Plus, costs just $30 for a lifetime subscription, which includes all updates, and is designed for the average consumer. WinPatrol takes a snapshot of your Windows run registry, and from then on blocks and alerts you to any new executable program, such as a malicious backdoor, that tries to install itself on your hard drive.
But that’s not enough. You must do all of your software updates promptly. Most SQL infectionsÃ‚Â work by exploiting long-ago discovered security flaws in your browser –Ã‚Â and in the programs that serve up Web-hosted video, music, photos, documents and work files. Keeping all of these web applications up to date will go a long way toward innoculating you.
This includes keeping current on updates for Internet Explorer, Firefox, Safari, Opera, Chrome, Adobe Flash, Adobe Reader, iTunes, QuickTime, Windows Media Player and RealPlayer. Microsoft and Mozilla do a credible job of alerting users to security updates for the IE and Firefox browsers, respectively.Ã‚Â But the rest of the software vendors don’t make it clear the updates increasingly include security patches.
–by Byron Acohido
Photos of Ryan Barnett, Obama, Avivah Litan, Jeremiah Grossman