ECF-GSF Core flashcards
ECF-GSF Core · Build recall with a focused review session.
34 cards. Read the term and the definition here. The deck at the top of this page is optional.
- ALE formula
- Annualized Loss Expectancy = Single Loss Expectancy (asset value times exposure factor) multiplied by Annualized Rate of Occurrence. It estimates expected yearly loss for a given risk and supports cost-benefit comparison of controls.
- Residual risk
- The risk that remains after controls are applied. It equals inherent risk minus mitigated risk and must be formally accepted by an accountable asset or business owner, never silently ignored.
- Defense in depth
- Layering independent preventive, detective, and corrective controls across people, process, and technology so that failure of any single control does not expose the protected asset.
- Risk treatment options
- Four responses to identified risk: mitigate by adding controls, transfer via insurance or contract, avoid by discontinuing the activity, or accept with documented approval from an accountable owner.
- Threat modeling STRIDE
- A structured technique enumerating Spoofing, Tampering, Repudiation, Information disclosure, Denial of service, and Elevation of privilege against each component, then defining mitigations for each identified threat.
- Stateful vs stateless firewall
- A stateful firewall tracks connection state and allows return traffic automatically; a stateless ACL evaluates each packet independently and requires explicit rules for both directions of a flow.
- DMZ screened subnet
- A network segment hosting internet-facing services between two filtering points, so a compromised public host cannot directly reach internal systems without crossing an additional enforced boundary.
- IPsec tunnel mode
- Mode that encrypts the entire original IP packet and encapsulates it in a new packet, used for gateway-to-gateway VPNs and remote access; transport mode protects only the payload.
- 802.1X authentication
- Port-based network access control where a supplicant authenticates through an authenticator switch to a RADIUS server before the port grants normal LAN access.
- Zero trust architecture
- A model that removes implicit trust based on network location and verifies every request using identity, device posture, and context at policy decision and enforcement points.
- SYN flood mitigation
- A denial-of-service attack exhausting half-open TCP connection tables. SYN cookies and connection rate limits mitigate it by avoiding state allocation until the handshake completes.
- MFA factor types
- Authentication factors are knowledge (password), possession (token, certificate), and inherence (biometric). Multi-factor authentication requires two or more distinct types, not two instances of one type.
- RBAC vs ABAC
- Role-based access control assigns permissions to roles and users to roles. Attribute-based access control evaluates user, resource, and environment attributes in policy, allowing finer, context-aware decisions.
- Password hashing choice
- Store passwords with salted, deliberately slow algorithms such as bcrypt, scrypt, Argon2, or PBKDF2. Fast hashes like MD5 or SHA-1 enable rapid offline cracking and are unsuitable.
- Kerberos clock skew
- Kerberos rejects tickets whose timestamps differ from the domain controller beyond tolerance, commonly five minutes. Time synchronization is therefore a prerequisite for successful domain authentication.
- Single sign-on federation
- SSO lets one authentication grant access to multiple applications; federation extends this across trust domains using protocols such as SAML or OpenID Connect with an identity provider's assertions.
- Three states of data
- Data exists at rest in storage, in transit across networks, and in use in memory or processing. Each state requires distinct controls such as encryption, TLS, and confidential computing.
- Tokenization vs encryption
- Tokenization replaces sensitive data with a non-sensitive token mapped in a secured vault, reducing compliance scope. Encryption transforms data reversibly with a key and preserves its format and length.
- 3-2-1 backup rule
- Maintain three copies of data on two different media types with one copy offsite. Immutable or offline copies additionally protect against ransomware encryption of connected backups.
- Crypto-shredding
- Secure destruction by deleting the encryption keys protecting stored data, rendering ciphertext unrecoverable. It is effective for cloud and distributed storage where physical destruction is impractical.
- Authenticated encryption modes
- Modes such as AES-GCM and CCM provide confidentiality and integrity in one operation. ECB leaks plaintext patterns and must not protect real data; CBC requires a random IV plus a MAC.
- Digital signature purpose
- A hash signed with the sender's private key and verified with the public key. It proves integrity, origin authenticity, and non-repudiation but does not by itself provide confidentiality.
- Forward secrecy
- Property achieved with ephemeral Diffie-Hellman key exchange, ensuring compromise of a long-term private key does not allow decryption of previously recorded sessions.
- PKI trust chain
- A self-signed root CA signs an intermediate CA, which signs leaf certificates. Clients validate the path and check revocation through CRL or OCSP before trusting the endpoint.
- HMAC
- A keyed hash construction combining a shared secret with a hash function to provide message authentication and integrity, detecting tampering without requiring asymmetric cryptography.
- EDR versus antivirus
- Endpoint detection and response continuously collects process, file, registry, and network telemetry for behavioral detection and remote response; signature-based antivirus matches known malware patterns only.
- SAST versus DAST
- Static application security testing analyzes source or bytecode without executing it; dynamic testing probes a running application from outside. Both are complemented by dependency scanning.
- Cloud shared responsibility
- The provider secures the infrastructure of the cloud, including physical and hypervisor layers; the customer secures data, identity, configuration, and access in the cloud.
- Application allowlisting
- Permitting only explicitly approved executables to run, which blocks novel and unknown malware. It is stronger than blocklisting, which stops only already-known bad software.
- NIST incident response phases
- Preparation; detection and analysis; containment, eradication, and recovery; and post-incident activity. Lessons learned from the final phase feed back into improved preparation and detection.
- Chain of custody
- Documented, unbroken record of who collected, handled, and stored evidence, including hashes and timestamps, so findings remain admissible and defensible in legal or disciplinary proceedings.
- Order of volatility
- Forensic collection priority from most to least volatile: CPU registers and cache, memory, network state, disk, remote logs, archival media. Collect volatile evidence before it disappears.
- GDPR breach notification
- Controllers must notify the supervisory authority of a personal data breach within 72 hours of awareness unless unlikely to result in risk, and inform affected data subjects when risk is high.
- Policy hierarchy
- Policy states management intent; standards define mandatory requirements; procedures give step-by-step instructions; guidelines are recommendations. Lower documents must align with higher ones.