1. State of Software Security 1 Jeff Ennis, CEH Solutions Architect Veracode

2. Agenda  Background – Metrics, Distribution of Applications  Security of Applications  Application Security - Industry Trends  Summary 2

3. Background – Basis for insights  For over three years, Veracode has been providing automated security analysis of software to large and small enterprises across various industry segments.  One of the residual effects is the wealth of security metrics derived from the anonymized data across varied industries and types of applications.  These metrics offer valuable insights on the quality of application security and issues related to the current state-of-practice and maturity of security in software.  Veracode was founded in 2006 by application security experts from @stake, Guardent, Symantec, and VeriSign.  Veracode provides automated security assessment capabilities in the cloud. Automated techniques include static binary analysis and dynamic analysis. Manual test data (if performed) is included in the analysis 3

4.  Enterprise  Industry vertical (enumerated)  Application  Application Supplier Type (internal, purchased, outsourced, open source)  Application Type (Web facing / Non-web)  Assurance Level (1 to 5)  Language (enumerated)  Platform (enumerated)  Scan  Scan Number  Scan Date  Lines of Code The Data Set + Metrics  Metrics  Flaw Count  FlawPercent  ApplicationCount  First Scan Acceptance Rate  Veracode Risk Adjusted Score  MeanTimeBetweenScans  Days to Remediation  Scans to Remediation  PCI pass/fail  SANS Top25 pass/fail  OWASP pass/fail  Two flavors: ’04 and ’07 4 1591 Applications and billions of lines of code

5. Sample Distribution 5

6. High Business Criticality does not drive all development projects “in-house.” More than 30% of all applications rated High or Very High in business criticality were sourced by Commercial software vendors

7. What is the distribution of languages in your enterprise? Do you have the same testing methodologies and practices across your application portfolio?

8. Security of Applications 8

9. Application Security – Scanning Results The majority of software (provided by customers for scanning) _______ Secure (Pass) _______ Insecure (Fail) 9

10. Majority of software is insecure 10 Pass: 42% Fail: 58% From all (self-selected) set of applications that were submitted to Veracode for assessment

11. Majority compliant with OWASP Top 10 or SANS Top 25 ? 11

12. Majority not compliant with OWASP Top 10 or SANS Top 25 12

13. Applications with the Best First-Scan Acceptance Rate 13 Outsourced Open Source Internally Developed Commercial

14. Internal Apps have Best First Scan Acceptance Rate 14

15. Most Common Issues in Applications (percent of application affected) 15 Cross-Site Scripting (XSS) Cryptographic Issues CRLF Injection Buffer Overflow SQL Injection

16. Cryptographic Issues Most Common in Applications 16

17. Most Prevalent Vulnerabilities 17 Cross-Site Scripting (XSS) Cryptographic Issues CRLF Injection Buffer Overflow SQL Injection Flaw Percent = Flaw Count / Total

18. This yields a very Different List Cross-site Scripting easy to fix but still most prevalent 18

19. Shortest Remediation Cycle 19 Outsourced Open Source Internally Developed Commercial

20. Commercial has longest remediation cycles while Open Source is shortest 20 Average Time to Remediate: 59 days

21. Higher percentage of “Very High” Severity Vulnerabilities: 21 Open Source Commercial Higher percentage of “High” Severity Vulnerabilities: Open Source Commercial

22. Open Source applications had an equivalent percentage of Very High severity vulnerabilities (Buffer Overflows, Numeric Errors), but a higher percentage of High Severity vulnerabilities (SQL Injection)

23. Most Dominant Vulnerability Across All Supplier Types 23 Cross-Site Scripting (XSS) Cryptographic Issues CRLF Injection Buffer Overflow SQL Injection Open Source/Outsourced/Commercial/Internally Developed

24. Vulnerability Distribution by Supplier

25. Most Dominant Vulnerability Across Languages 25 Cross-Site Scripting (XSS) Cryptographic Issues CRLF Injection Buffer Overflow SQL Injection Java.NET C/C++

26. Vulnerability Distribution by Language Flaw Type by Input

27. Application Security - Industry Trends 27

28. Industry with Best First Submission Rate  Finance-related  Government  Software-related  Other 28

29. Financial Services and Government fare best Software not so much 29

30. Most Dominant Vulnerability Across All Industries 30 Cross-Site Scripting (XSS) Cryptographic Issues CRLF Injection Buffer Overflow SQL Injection Financial-related/Government/Software-related

31. Vulnerability Distribution by Industry

32. Summary - Recommendations 32 1.Most software is indeed very insecure. Recommendation: Implement a comprehensive, risk-based application security program 2. Third-party software is a significant percentage of the enterprise software infrastructure, and third-party components are a significant percentage of most applications. Recommendation: Implement security acceptance criteria and policies for an approved list of third-party suppliers, and conduct security testing on third-party components prior to integration into the final application 3. Open source projects have comparable security, faster remediation times, and fewer potential backdoors than Commercial or Outsourced software. Recommendation: Test open source, outsourced, and commercial applications as rigorously as you would test internally developed code. Do not buy into FUD regarding the use of open source software in critical business applications. 4. A significant amount of Commercial and Open Source software is written in C/C++ making it disproportionately susceptible to vulnerabilities that allow attackers to gain control of systems. Recommendation: Apply the same review methodologies across all languages and platforms. Do not base your security review plan on ubiquity or complexity (or lack thereof).

33. Summary – Recommendations (continued) 33 5. The pervasiveness of easily remedied vulnerabilities indicates a lack of developer education on secure coding. Recommendation: Implement specific developer training initiatives as part of your overall security program 6. Software of all types from Finance and Government sectors was relatively more secure on first submission to Veracode for testing. Recommendation: Follow the lead of other organizations with high risk profiles; review the steps they took to implement operating controls in complex environments 7. Outsourced software is assessed the least, suggesting the absence of contractual security acceptance criteria. Recommendation: Pay particular attention to security requirements when contracting for Outsourced development. Insist upon the authority to perform independent security testing and set a minimum acceptance criteria. This way you are not charged/billed for reworking code due to security defects.

34. Sneak Preview – State of Software Security Volume 2 34 40% of an enterprise’s application inventory is comprised of 3 rd party applications 30 – 70% of what customers classify as “internally developed” is in fact 3 rd party components and libraries 40% 3 rd party applications + (30-70% 3 rd party libraries) Internal applications = A lot of 3 rd party code

35. Thank You Questions? 35