For more info on a specific newsletter, click the title. Details will be displayed in a new window.
Subscribe to
Computerworld 40 years of the most authoritative source of news and information for IT leaders.
Ten questions to ask about application security systems
or
Other Security Stories
|
|
|
|
November 22, 2004 (Computerworld) -- Robust application security is necessary to ensure Web site availability and to protect sensitive customer and corporate data and application-enabled revenue. However, there's growing confusion about what constitutes application security and how it's achieved.
The following 10 questions will help you evaluate whether a product delivers true application protection.
1. Does it inspect application communications or just packets?
To reliably identify application-layer threats, a security appliance must "see" the same communication stream as the application it is protecting. This means the security device must perform a full deconstruction of the HTML data payload, as well as track the state of each application session. It is technologically impossible to analyze application behavior through simple inspection of IP packets, either individually or reassembled into their original sequence.
2. Does it detect and defeat encrypted application attacks?
Virtually all Web applications that process confidential customer or corporate information use Secure Sockets Layer encryption to protect both the confidentiality and integrity of data while in transit. However, SSL also provides hackers with a useful tool to evade detection because it's impossible to detect attacks that are strongly encrypted. Therefore, application-layer security can be performed only if SSL-encrypted traffic is decrypted into its original clear text form prior to inspection.
3. Does it protect the application infrastructure and users?
Application security involves protecting all elements of an application infrastructure (e.g., server operating system, application program and back-end databases), as well as users of the application. Protecting the application program and application data isn't sufficient. Trust relationships with users must also be closely guarded to ensure the continuing business viability of the application. For example, an application security product should thwart the hijacking of user sessions.
4. Does it defeat zero-day attacks?
Zero-day attacks come in two varieties: attacks exploiting vulnerabilities in custom applications, and attacks aimed at vulnerable packaged applications for which a patch has not yet been released. An application security solution must detect and defeat all forms of zero-day attacks. Using attack signatures or event correlation will fail to protect against zero-day attacks. A positive security model that understands and enforces correct application behavior in real time is the only viable defense for zero-day attacks.
5. Does it cloak application infrastructure elements?
Many attacks against Web applications are custom-crafted and designed to exploit security vulnerabilities in the application infrastructure. Effective reconnaissance can help the hacker focus his attack methodology, target a smaller number of potential security weaknesses and craft more damaging attacks. At a minimum, a Web application protection solution should provide the following cloaking capabilities to mask details of the application infrastructure.
- Remove all unnecessary server response headers
- Rewrite all application URLs
- Remove HTML comments
- Encrypt cookie names and values, URLs and hidden form fields
6. Does it prevent the leakage of sensitive corporate or customer data?
The objective of many Web application attacks is to steal sensitive customer data such as credit card numbers, account IDs and Social Security numbers. An organization can never be completely assured that all avenues to application-layer attacks are blocked. Preventing the leakage of sensitive data in application server responses requires the following capabilities:
- Inspecting the entire stream for the presence of sensitive data objects
- Precision matching of data objects. For example, credit card numbers should be tested for validity before real-time action is taken.
- Transforming matching data objects. For example, reflect only the last four digits of a customer's credit card number for verification purposes.
7. Does it block benign traffic?
Web application security is typically deployed to protect an organization's most business-critical systems. False positives that block benign and legitimate traffic are intolerable in these environments. For example, mistakenly blocking a brokerage transaction for a large trading partner may cost an organization tens of thousands of dollars, or more. To eliminate false positives, an application security system must:
- Perform true application-layer traffic inspection
- Maintain full communications context
- Possess a semantic understanding of all application data
8. Does it rationalize the Web infrastructure?
Application security is typically deployed in a Web demilitarized zone (DMZ), directly fronting protected application servers. In many enterprises, the DMZ may already be heavily populated with numerous single-function devices, such as load balancers, SSL acceleration devices, application proxies and TCP connection offload appliances. A well-designed application security solution should subsume the capabilities of point products residing in the DMZ. Reducing overall box count improves application performance and infrastructure reliability. Typical functions that should be performed by an application gateway include:
- SSL acceleration
- Noncaching application proxy (i.e., rewriting application URLs)
- Web I/O acceleration
- Web site defacement prevention
9. Can it deploy consistent security for all applications?
Common threats to applications, such as SQL injection, can be handled by "global" security settings. However, there is often a need to define per-application security rules. For example, a global security policy that prohibits any modification to data in a hidden form field may break an application that uses client-side JavaScript to legally modify form field data within the client's browser. In this scenario, different policies would be necessary. An application security system should virtualize security policies, accommodating both global and per-application attributes.
10. Does it adapt policies for dynamic application environments?
Many Web applications use client-side Java scripting to generate dynamic content within a client's browser. Prime examples of application content that may be legally modified on the client include URL parameters, application session cookies and form field data. A positive security model that is unable to accurately anticipate and accommodate dynamically generated content from the client will potentially block legitimate application traffic. Look for an application security solution that can learn correct application behavior and automatically generate policy recommendations so that legitimate dynamic content is recognized and permitted by the security system.
Conclusion
Internet threats have shifted from network-layer exploits to attacks against Web applications. Use this checklist to evaluate the depth of protection provided by application security products, and select the most appropriate technology to protect your vital online resources.
Abhishek Chauhan is chief technology officer and co-founder of Teros in Santa Clara, Calif. Prior to coming to Teros, he helped design scalable network services and distributed programs at Sun Microsystems Inc. While at Sun, he was a contributor to the Application Programming Model for the J2EE platform, and he worked on Web access optimization techniques for the Java Web Server and Java Server Pages specifications. Chauhan also co-founded VXtreme, which was bought by Microsoft Corp. in 1997, where he was lead architect on the design of its streaming server.
|
Print this Story |
|
Send Us Feedback |
|
E-mail this Story |
|
Digg this Story |
|
Slashdot this Story |
|
|
|
|
|
 |
|
All Zones Application Performance Zone Enterprise-Class Security Zone Enterprise Solutions Zone The File Data Management Zone Grid Computing on Windows Zone Security Management Zone ITIL Best Practices Zone The SAS Zone Storage Virtualization Zone The Data Center Management Zone |
 See your link here
|
 |
||||||||
| ||||||||
| ||||||||
| ||||||||
|
Why SaaS is Vital to Email and Web Security
Download this Executive Bulletin (a $49.95 value) for free, compliments of MessageLabs.
Get this white paper now!
Read up on the latest ideas and technologies from companies that sell hardware, software and services.
- Mozilla launches Firefox 3.0 RC1 early
- Microsoft: Don't misunderstand UAC, other Vista features
- HP confirms XP SP3 endless reboot snafu, promises patch
- More top stories
Security Management ZoneSecurity management is the process of developing a comprehensive data protection plan. It takes into account all potential threats, the existing network environment, the future needs of the organization, and lays out a multi-tiered blueprint to integrate the security technology needed to combat these threats. CDW can help keep your network and data secure. Visit the CDW Security Management Zone now See All Zones
|
 Fired up about IT? Join Sharkbait and share your true tales of IT. SharkBait is the place for you to sound off about everything IT – the good, the bad, and the rest of the weird stuff you deal with every day.New baits 
|
"Security Directions" virtual trade show2008's Code-Red Security Issues for Protecting the EnterpriseWebcasts, white papers, demos, and more. Presented in a unique 3-d environment. Enter our show right now! Click here to enter
|
 In SecurityThere's plenty of talk about how to behave during a Customs search of your computer and gear, but Jon Espenschied's got tips for securing your data (and privacy) before you reach the border. Click here to read the latest column by Jon Espenschied |
 |
 |
Layered Security Solutions
Although basic network security issues have changed very little over the past decade, the
network security landscape has changed dramatically. Today's IT professionals still have the
primary responsibility of protecting the confidentiality of corporate information, preventing
unauthorized access, and defending the network against attacks. Security experts and analysts agree that a security solution comprised of multiple layers is the best defense against today's increasingly sophisticated attacks.Download this white paper
|
Universal Threat Management - Because Conventional UTM is Not Enough!
This white paper, written by Mark Bouchard of Missing Link Security Services, examines the challenges confronting today's enterprises with respect to managing threats on a network. It also discusses the need for "Universal Threat Management", which is a security solution approach for all physical locations within an enterprise that require threat protection.Download this white paper |
Selecting the Right Threat Management Solution
This short demo will guide you through key considerations for selecting a solution to manage threats on a network. Learn about the popularity of Unified Threat Management (UTM), and how it fits into an overall security solution. Explore critical elements of a network-wide solution for multisite and large network-size deployments and identify the four key features of a threat management solution.View this demo
|
CIO
Computerworld
CSO
DEMO
GamePro
Games.net
IDG Connect
IDG World Expo
Infoworld
The Industry Standard
ITworld
JavaWorld
LinuxWorld
MacUser
Macworld
Network World
PC World
Playlist
Copyright © 2008 Computerworld Inc. All
rights reserved. Reproduction in whole or in part in any form or medium
without express written permission of Computerworld Inc. is prohibited.
Computerworld and Computerworld.com and the respective logos are
trademarks of International Data Group Inc.
|
