Systems and Application Security for SSCP

Sample Practice Questions

1. Question 1

   Which of the following is the most reliable, secure means of removing data from magnetic storage media such as a magnetic tape, or a cassette?
   1. A. Degaussing
   2. B. Parity Bit Manipulation
   3. C. Zeroization
   4. D. Buffer overflow
   Explanation

   The correct answer is: A. Degaussing.

   Degaussing is the most reliable secure-disposal method for magnetic media because it exposes the disk or tape to a strong alternating magnetic field that randomises the magnetic domains carrying the data, effectively returning the media to a blank or unreadable state. Properly executed degaussing makes recovery infeasible even with laboratory techniques. Parity bit manipulation merely flips redundancy bits used for error detection and does not erase the underlying recorded data. Zeroisation overwrites memory or storage with zeros and is used for solid-state memory and cryptographic keys; on magnetic media a single overwrite leaves recoverable remnants, which is why multi-pass standards exist. Buffer overflow is a software exploit category and has nothing to do with media sanitisation. For modern magnetic disks always pair degaussing or physical destruction with documented certificates of destruction; for SSDs use vendor secure-erase commands or shred.

2. Question 2

   What would be the name of a Logical or Virtual Table dynamically generated to restrict the information a user can access in a database?
   1. A. Database Management system
   2. B. Database views
   3. C. Database security
   4. D. Database shadowing
   Explanation

   The correct answer is: B. Database views.

   A database view is a logical or virtual table dynamically generated from the underlying base tables, restricting which columns and rows a user can see when they query the database. Views are the standard mechanism for fine-grained read access control in relational databases — the owner defines a view that exposes only the permitted subset, and end-users query the view rather than the underlying tables. A Database Management System (DBMS) is the software platform that hosts and manages databases overall (Oracle, SQL Server, PostgreSQL); it is not itself a virtual table. Database security is a broad discipline covering authentication, authorisation, encryption, and audit; it is not the specific mechanism the question describes. Database shadowing is a high-availability technique that replicates the entire database to a secondary location for resilience; it does not restrict access to information. Use views together with role-based grants and row-level security policies for layered fine-grained access control.

3. Question 3

   Which of the following logical access exposures INVOLVES CHANGING data before, or as it is entered into the computer?
   1. A. Data diddling
   2. B. Salami techniques
   3. C. Trojan horses
   4. D. Viruses
   Explanation

   The correct answer is: A. Data diddling.

   Data diddling is the technique of altering data before or during entry into a computer system — the change happens at the point of input, where validation may be weakest, before the data reaches its intended processing or storage. The attacker substitutes a different value than what was authorized, often for fraudulent gain, and the manipulated value then propagates as if it were legitimate. Salami techniques shave very small amounts off many transactions (the rounding-fraud classic); the manipulation happens in computation, not at data entry. A Trojan horse is malicious code hidden in an apparently useful program; it may carry out many actions but is not specifically about input alteration. A virus is self-replicating malware that attaches to host files; it propagates rather than tampering with data entry. Defend against data diddling with strong input validation, dual control on critical entries, and reconciliation against authoritative sources.

4. Question 4

   Why do buffer overflows happen? What is the main cause?
   1. A. Because buffers can only hold so much data
   2. B. Because of improper parameter checking within the application
   3. C. Because they are an easy weakness to exploit
   4. D. Because of insufficient system memory
   Explanation

   The correct answer is: B. Because of improper parameter checking within the application.

   Buffer overflows happen primarily because the application does not properly check the size, type, or content of parameters before placing them into a fixed-size buffer; with no validation, longer-than-expected input overruns the buffer and corrupts adjacent memory, opening the door for control-flow hijacking. The root cause is the missing validation. Saying buffers can only hold so much data restates the limitation but not the cause: every buffer is finite, yet most code never overflows because it validates its inputs. The argument that overflows happen because they are an easy weakness to exploit confuses cause with effect — exploitability is a downstream consequence of the underlying defect, not the reason the defect exists. Insufficient system memory describes a resource constraint, not a coding flaw. Defend against buffer overflows through safe string and array APIs, bounds checking, modern languages with built-in memory safety, and compiler protections such as stack canaries, ASLR, and DEP.

5. Question 5

   Which of the following statements pertaining to the Bell-LaPadula is TRUE if you are NOT making use of the strong star property?
   1. A. It allows "read up."
   2. B. It addresses covert channels.
   3. C. It addresses management of access controls.
   4. D. It allows "write up."
   Explanation

   The correct answer is: D. It allows "write up.".

   The Bell-LaPadula model enforces confidentiality through two main rules: the simple security property (no read up — a subject cannot read data at a higher classification) and the *-property, also called the star property (no write down — a subject cannot write to a lower classification). The strong star property additionally forbids writing up or to any level other than the subject's own, which prevents indirect leakage. Without the strong star property, write up is allowed (a subject at a lower level can write to a higher level), which means the model permits this seemingly unusual flow. The model does not allow read up under any variant because that would directly violate confidentiality. Bell-LaPadula is not specifically designed to address covert channels — that is its well-known weakness, since the rules cover overt access only. Management of access controls (administration of subjects and objects) is outside the model's scope; that concerns operational policy, not the formal flow rules.

Other SSCP domains

• Access Controls (203 questions)
• Cryptography (185 questions)
• Incident Response and Recovery (112 questions)
• Network and Communications Security (252 questions)
• Risk Identification, Monitoring, and Analysis (59 questions)
• Security Concepts and Practices (237 questions)