Hybrid analysis of vulnerable information flows

• US 9,177,155 B2
• Filed: 09/13/2013
• Issued: 11/03/2015
• Est. Priority Date: 12/31/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Beskrovny, Evgeny, Sharabani, Adi, Tripp, Omer
• Primary Examiner(s): Pwu, Jeffrey
• Assistant Examiner(s): Nipa, Wasika
• Application Number:
  US14/025,874
• Publication Number:
  US 20140189875A1
• Time in Patent Office: 781 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 21/577 Assessing vulnerabilities and evaluating computer system security
• CPC Class Codes: H04L 63/1433 Vulnerability analysis

Abstract

Arrangements described herein relate to analyzing vulnerable information flows in an application. A black-box scan of the application can be performed to record a call-tree representation of call stacks arising in the application due to test inputs provided during the black-box scan. For each path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, a static analysis can be performed to determine at least one parameter value that, when abstracted, drives execution of the application, via the path, to flow to the at least one security sink. A security report can be generated identifying at least one of the paths in the call-tree representation that does not constitute the vulnerable information flow during the black-box scan, but flows to the at least one security sink when the at least one parameter value is abstracted.

Claims

1. A method of analyzing vulnerable information flows in an application, the method comprising:

performing, using a hardware processor, a black-box scan of the application;

based on the black-box scan, recording a call-tree representation of a plurality of call stacks arising in the application due to test inputs provided during the black-box scan, the call tree representation indicating paths between the plurality of call stacks;

performing, for each of the paths in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, a static analysis, and by the static analysis determining at least one parameter value that, when abstracted, drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink; and

generating a security report identifying at least one of the paths in the call-tree representation that does not constitute the vulnerable information flow during the black-box scan, but flows to the at least one security sink when the at least one parameter value is abstracted.

2. The method of claim 1, further comprising:

determining whether abstracting a single parameter value drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the security sink;

responsive to determining that abstracting the single parameter value does not drive execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the security sink, abstracting at least two parameter values, and determining whether abstracting at least two parameter values drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the security sink.

3. The method of claim 2, further comprising:

responsive to determining that abstracting the two parameter values does not drive execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the security sink, abstracting at least three parameter values, and determining whether abstracting at least three parameter values drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the security sink.

4. The method of claim 1, further comprising:

determining, for each path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, a security confidence level of the path according to a number of parameter values that need to be abstracted to drive execution of the application, via the path, to flow to the at least one security sink.

5. The method of claim 1, further comprising:

determining whether abstracting a first parameter value drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink;

responsive to determining that abstracting the first parameter value does not drive execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink, abstracting at least a second parameter value, and determining whether abstracting the second parameter value drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink.

6. The method of claim 1, wherein performing the static analysis to determine at least one parameter value that, when abstracted, drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink comprises:

abstracting a parameter value for each respective parameter the application is configured to receive from a user.

7. The method of claim 1, wherein the security report further indicates the at least one parameter that, when its parameter value is abstracted, drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink.

8. The method of claim 7, wherein the security report further indicates a parameter value, or range of parameter values, for the parameter that drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink.

9. A method of analyzing vulnerable information flows in an application, the method comprising:

performing a black-box scan, using a hardware processor, of the application;

based on the black-box scan, recording a call-tree representation of a plurality of call stacks arising in the application due to test inputs provided during the black-box scan, the call tree representation indicating paths between the plurality of call stacks;

performing, for each of the paths in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, a static analysis, and by the static analysis determining at least one parameter value that, when abstracted, drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink;

responsive to determining that abstracting the first parameter value does not drive execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink, abstracting at least a second parameter value, and determining whether abstracting both the first and second parameter values drives execution of the application, via the path in the call-tree representation that does not constitute a vulnerable information flow during the black-box scan, to flow to the at least one security sink; and

generating a security report that identifies at least one of the paths in the call-tree representation that does not constitute the vulnerable information flow during the black-box scan, but flows to the at least one security sink when both the first and second parameter values are abstracted.