09:30 AM
Connect Directly

Researchers Create Legal Botnet Abusing Free Cloud Service Offers

Hack depends on scripts creating scores of unique email addresses and automating execution of email verification

Last week at the RSA Conference, a pair of researchers demonstrated how it was possible to legally create a botnet for free by abusing trial accounts made available by high-powered platform-as-a-service (PaaS) and infrastructure-as-a-service (IaaS) offerings.

RSA Conference 2014
Click here for more articles about the RSA Conference.

"We were curious if we could build a botnet out of freely available cloud services," said Rob Ragan, senior security associate for Bishop Fox, who has been experimenting on that premise for the past several years with his colleague Oscar Salazar, security associate at Bishop Fox. "We started getting all these emails and alerts of, 'Here's a free Amazon EC2 box, there's free storage space, here's a free platform to develop and host your code.' We thought, 'Wow! That is a lot of computing power for free.'"

The question they asked themselves was how hard would it be to automate the process of signing up for an unlimited number of free accounts from these sites and then developing a central control system from which an attacker could potentially launch malicious activities. The answer: not hard at all.

"We were really easily able to get hundreds of boxes on certain providers and have a central way to launch things like massive port scans," Ragan says. "We also did a proof-of-concept on cryptocurrency or Bitcoin mining. If you're getting this free computing power and don't have the power bill from it, why not use that to generate mining? That would be a huge motivation for malicious threat actors using these platforms." The project was made possible through the development of a process to automate the creation of unique email accounts on free email services, and then special scripting to automate the process of clicking on email verification links sent to those accounts.

The researchers initially refined that process during a penetration test of a sweepstakes website in which they showed an attacker could game the drawing and up the chances of winning through exactly such a method.

The process of developing the botnet came with the usual kinks. For example, the researchers told an amusing story about scrambling in the middle of the night to code a "stop" mechanism to their automated system once they turned it on and saw it worked.

"We had basically just coded the start button, and then we thought, 'Uh, oh! How do we stop it?'" Salazar said. "We had to quickly figure out how to stop our botnet from getting away from us."

Much of the abuse the pair tested was made possible due to poor verification of users during the trial account creation process. Even when some services tried to limit accounts by limiting Internet access, the researchers were easily able to employ quick workarounds. Of the 150 different PaaS and IaaS sites the duo tested, two-thirds were not doing any CAPTCHAs, SMS verification, or credit card verification beyond simple email account verification.

According to Ragan, while these extra steps can also be worked around, cloud services still need to embrace them in order to make it more expensive for potential attackers to do exactly what they did. Not only is it a huge vulnerability for these service providers, but it is opening them up to huge bills from Amazon because, in many instances, these services are launching free Amazon EC2 boxes on behalf of trial accounts. That is computing power they service providers will have to pay for in these abuse cases.

This is likely how attackers will try to work the system given that Amazon has implemented strict authentication controls to prevent exactly such kinds of abuse. "You could just register accounts. It didn't require SMS, it didn't require credit cards -- you just needed a valid email address, and then you'd have access to free Amazon EC2 boxes," Salazar said. "Over the last couple of years, Amazon has really increased its security, and now it requires a credit card and SMS and email verification. So Amazon is putting a bunch of protection around its services, and now these other companies starting up are basically leasing EC2 boxes, but they're not putting the same protections in place. So you're basically getting free Amazon boxes through them on their dime."

Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.

Ericka Chickowski specializes in coverage of information technology and business innovation. She has focused on information security for the better part of a decade and regularly writes about the security industry as a contributor to Dark Reading.  View Full Bio

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Current Issue
Flash Poll
Threat Intel Today
Threat Intel Today
The 397 respondents to our new survey buy into using intel to stay ahead of attackers: 85% say threat intelligence plays some role in their IT security strategies, and many of them subscribe to two or more third-party feeds; 10% leverage five or more.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
Published: 2014-08-31
The device-initialization functionality in the MSM camera driver for the Linux kernel 2.6.x and 3.x, as used in Qualcomm Innovation Center (QuIC) Android contributions for MSM devices and other products, enables MSM_CAM_IOCTL_SET_MEM_MAP_INFO ioctl calls for an unrestricted mmap interface, which all...

Published: 2014-08-31
Stack-based buffer overflow in the acdb_ioctl function in audio_acdb.c in the acdb audio driver for the Linux kernel 2.6.x and 3.x, as used in Qualcomm Innovation Center (QuIC) Android contributions for MSM devices and other products, allows attackers to gain privileges via an application that lever...

Published: 2014-08-31
app/aboot/aboot.c in the Little Kernel (LK) bootloader, as distributed with Qualcomm Innovation Center (QuIC) Android contributions for MSM devices and other products, allows attackers to overwrite signature-verification code via crafted boot-image load-destination header values that specify memory ...

Published: 2014-08-31
A certain Qualcomm Innovation Center (QuIC) patch to the NativeDaemonConnector class in services/java/com/android/server/ in Code Aurora Forum (CAF) releases of Android 4.1.x through 4.3.x enables debug logging, which allows attackers to obtain sensitive disk-encryption pas...

Published: 2014-08-31
The Qualcomm Innovation Center (QuIC) init scripts in Code Aurora Forum (CAF) releases of Android 4.1.x through 4.4.x allow local users to modify file metadata via a symlink attack on a file accessed by a (1) chown or (2) chmod command, as demonstrated by changing the permissions of an arbitrary fil...

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
This episode of Dark Reading Radio looks at infosec security from the big enterprise POV with interviews featuring Ron Plesco, Cyber Investigations, Intelligence & Analytics at KPMG; and Chris Inglis & Chris Bell of Securonix.