If it can be network-enabled, it should be discussed and reported!

Telecom Innovation

Subscribe to Telecom Innovation: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get Telecom Innovation: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

Cloud computing environments are just as suited to illegitimate use as legitimate use. Do providers need a way to separate the chaff from the wheat to reassure enterprise-class customers that they’re doing everything they can to eliminate the hijacking of cloud computing resources for nefarious purposes?

One of the negatives of being the technology darling du jour is that every misstep, problem, and outage is immediately jumped on and reported everywhere. Amazon is particularly susceptible to such coverage, being recognized as one of the leaders in public cloud computing. Last week Amazon suffered yet another outage, true, but more interesting may be the discovery that it had been infected by the Zeus bot, a password-stealing banking Trojan.

blockquote On Wednesday, security researchers for CA found that a variant of the infamous password-stealing Zeus banking Trojan had infected client computers after hackers were able to compromise a site on EC2 and use it as their own C&C (command and control) operation.image

The Zeus bot has been loose for quite some time and Amazon is certainly not the first – nor likely the last – organization to be infected by this nasty little trojan. In October social  networking giant Facebook was targeted by miscreants attempting to spread some Zeus-bot love around as well. The bot is a Windows-specific trojan that, like so many others, attempts to lure its victims into installing it via phishing and drive-by attacks.

The bot itself is nothing new, nor is the targeting of Windows-specific machines, or the use of phishing. Neither is the attempt to leverage the large scale nature of specific services on the Internet as a means to spreading a virus around. And as Carl Brooks pointed out via Twitter, the use of cloud computing as on-demand bot-net farms is no surprise to the security community at large. But what isn’t being discussed – and probably needs to be – is what can be done about the situation? Is there a solution at all or will we just have to live with it?

It may be that the only remediation available is the establishment of enterprise-class clouds.


credit_cards One of the hallmarks of “open” clouds like Amazon is “anyone with a credit card” can use them. The saddest part about that is it that “anyone” includes miscreants and it might be your credit card they’re using to turn the inherent elasticity of cloud computing into a giant, on-demand botnet capable of wreaking havoc not just for some external site, but the provider itself.

blockquote With Amazon FPS, you can build innovative payment applications with a new level of flexibility in how you execute payments. Amazon FPS supports the processing of payments using credit cards, bank accounts and Amazon Payments account balances to send or receive money.

According to an article in the Register, “"We believe this was a legitimate service that was purchased and compromised via a vulnerability" such as a weak password, Don DeBolt, CA's director of threat research, told The Reg. "It could have been any vulnerable system on the internet."

It could have been any vulnerable system on the Internet. True, very true. What makes this incident particularly troubling is that the compromise of a single machine inside a public cloud computing environment necessarily raises questions about the fact that the vulnerability was likely shared either across all similar virtual images or the management system itself. That doesn’t mean all like systems were infected, but it does mean that if they shared the vulnerability they could have been infected. The compromise of even a single machine inside the environment necessarily leads to the possibility that every other machine like it inside the environment could be compromised – and possibly by the original infection.

Public cloud computing environments are like the human body; if a virus gets inside it can easily and rapidly multiply, infecting other like machines (cells) with relative ease. The efficiency for the provider comes in part from offering base “images”; these images are the same across all customers, which includes any possible vulnerabilities in the operating system and software deployed via the image. And like the body, it’s often not noticeable until the patient has a fever. It doesn’t really matter whether this infection was, in fact, introduced via the purchase of a legitimate service. What the possibility that it was has done – or should do – is raise the question whether “open” public cloud computing is too open, at least for enterprise-class use.


It may be the case that the awareness of the risk introduced by “open” public cloud computing, i.e. anyone with a credit card can use, will result in a re-evaluation of service offerings and the reassurances needed by organizations that similar scenarios have been mitigated – somehow – in public cloud computing offerings. Whether it requires more than a credit-card, i.e. some sort of proof of legitimate use/need, or it forces public cloud computing offerings to be more transparent with the ways in which they mitigate and manage such risks, the infection of Amazon should be a wake-up call for every provider.

Providers are likely to respond in a manner similar to Amazon’s response to this event, which is to say “we take these things seriously, as soon as we know we do something about it, yada, yada, yada.” The problem is that these responses are just that – reactive responses to something that’s already occurred. What the enterprise needs – and deserves – is some proactive measures to guard against the legitimate use of cloud computing resources for illegitimate purposes. The negative to employing stricter control on utilization is, of course, that it removes the ability to “immediately begin using” a cloud computing instance. Without the ability to simply “swipe your card here” providers may be concerned that fewer people and organizations will take advantage of its offerings. But in many cases the additional layer of security added by authenticating legitimate use in some way – as far as that can be authenticated – may reassure organizations that the cloud computing environment in question is safer than perhaps one that continues to require no more authentication than your local gas station.

Vulnerable systems would remain vulnerable, yes, but removing the ease with which a miscreant can drive in, deposit a virus, and leave would certainly not hurt the overall security posture of the environment. If “open” public clouds cannot or will not address the issue, it is likely that someone will see the need and fill it with more “enterprise-class” cloud computing offerings that do.

Follow me on Twitter    View Lori's profile on SlideShare  friendfeed icon_facebook

AddThis Feed Button Bookmark and Share



Related blogs & articles:

Read the original blog entry...

More Stories By Lori MacVittie

Lori MacVittie is responsible for education and evangelism of application services available across F5’s entire product suite. Her role includes authorship of technical materials and participation in a number of community-based forums and industry standards organizations, among other efforts. MacVittie has extensive programming experience as an application architect, as well as network and systems development and administration expertise. Prior to joining F5, MacVittie was an award-winning Senior Technology Editor at Network Computing Magazine, where she conducted product research and evaluation focused on integration with application and network architectures, and authored articles on a variety of topics aimed at IT professionals. Her most recent area of focus included SOA-related products and architectures. She holds a B.S. in Information and Computing Science from the University of Wisconsin at Green Bay, and an M.S. in Computer Science from Nova Southeastern University.