Cloud
3/29/2013
01:34 AM
Connect Directly
RSS
E-Mail
50%
50%

Should Cloud Providers Secure Their Outbound Traffic?

As attackers focus on using hosted or virtual servers to power their denial-of-service attacks, calls for a cleaner cloud may become louder

Discerning between malicious traffic and legitimate traffic in real time is challenging for companies targeted by distributed denial-of-service attacks, but the task is made more difficult when the attacks come from reputable Internet properties that cannot easily be filtered.

The attacks on U.S. financial institutions, for example, have used compromised publishing platforms to target banks with a variety of attack traffic since last September. A key factor in the success of those distributed denial-of-service (DDoS) attacks is the use by attackers of compromised, but reputable hosts. While attacks from hosted and cloud platforms have been uncommon so far, they will likely become a larger problems in the future, says Carl Herberger, vice president of security solutions at network security firm Radware.

Providers need to take some responsibility for obvious abuses of acceptable-use policy, such as packet floods and attacks on applications, he says.

"They need to come to terms with the fact that, if one of their hosted clients is attacking a victim on the Internet, they play a role," Herberger says.

Just as Internet service providers have started taking a more involved role in detecting, warning, and mitigating compromised systems in their network that are owned by consumers, cloud service providers may need to take a more hands-on approach to detecting and helping mitigate compromises that turn their customers' hosted systems into an attack.

[Outages at CloudFlare and Microsoft's Azure in the past month underscore that widespread chaos can be the result of a weak point in cloud infrastructure. See Cloud Providers Work To Disperse Points Of Failure.]

Yet for cloud providers, any sort of filtering could cause a disruption to their customers' services, an unacceptable risk. Some cloud providers that focus on secure hosting, such as FireHost, do monitor outbound traffic for signs of malicious behavior. But adapting that generally to public clouds may not work, says Kurt Hagerman, director of information security for FireHost.

"When you look at public cloud infrastructure in general, most of those come with little to no outbound security," Hagerman says. "They rely on the customers to add security."

Defending against DDoS attacks is expensive, making the stakes higher. One recent survey found that companies pay up to $6,500 an hour to recover from the attacks.

Such damages could make legal liability a problem. Despite common carrier statutes and the Communications Decency Act, which protects providers from being held responsible for the acts of their subscribers, distributed denial-of-service attacks coming from hosted, or cloud, servers could expose a provider to risk.

"If I was running a bank, and I was repeatedly getting DDoSed from the same space, and I was reporting that to the provider and trying to work with them, but it seemed like it just didn't matter, might I try to sue them? Sure, I might," says Hagerman. "It might get their attention and convince them to do something to stop the attacks."

Not all massive DDoS attacks come from issues with hosted servers and cloud providers. The recent distributed denial-of-service attack that topped 300Gbps originated from a different, although arguably related, problem of using misconfigured infrastructure--in this case, domain-name system (DNS) recursive resolvers--to redirect and amplify the control of a much smaller amount of bandwidth into a much larger attack.

If cloud providers run their infrastructure like a gated community, then attackers should not be able to get away with such damaging attacks, says Dan Holden, director of security research for Arbor Networks. Paying attention to outbound traffic could give attackers less incentive to use cloud resources for their attacks.

"It comes down to a return-on-investment issue for the attacker," Holden says. "I think it would come down to whether they thought the success of the attack would be better using the cloud or not."

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.

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Cartoon
Current Issue
Dark Reading Must Reads - September 25, 2014
Dark Reading's new Must Reads is a compendium of our best recent coverage of identity and access management. Learn about access control in the age of HTML5, how to improve authentication, why Active Directory is dead, and more.
Flash Poll
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2012-5485
Published: 2014-09-30
registerConfiglet.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via unspecified vectors, related to the admin interface.

CVE-2012-5486
Published: 2014-09-30
ZPublisher.HTTPRequest._scrubHeader in Zope 2 before 2.13.19, as used in Plone before 4.3 beta 1, allows remote attackers to inject arbitrary HTTP headers via a linefeed (LF) character.

CVE-2012-5487
Published: 2014-09-30
The sandbox whitelisting function (allowmodule.py) in Plone before 4.2.3 and 4.3 before beta 1 allows remote authenticated users with certain privileges to bypass the Python sandbox restriction and execute arbitrary Python code via vectors related to importing.

CVE-2012-5488
Published: 2014-09-30
python_scripts.py in Plone before 4.2.3 and 4.3 before beta 1 allows remote attackers to execute Python code via a crafted URL, related to createObject.

CVE-2012-5489
Published: 2014-09-30
The App.Undo.UndoSupport.get_request_var_or_attr function in Zope before 2.12.21 and 3.13.x before 2.13.11, as used in Plone before 4.2.3 and 4.3 before beta 1, allows remote authenticated users to gain access to restricted attributes via unspecified vectors.

Best of the Web
Dark Reading Radio
Archived Dark Reading Radio
In our next Dark Reading Radio broadcast, we’ll take a close look at some of the latest research and practices in application security.