CIA
- confidentiality –> only auth subjects access objects
- integrity –> no modif w/o authorization
- availability –> it’s there when it’s needed
ISC2 code of ethics (short version)
- Protect society, commonwealth and the infrastructure
- Act honorably, honestly, justly, responsibly and legally
- Provide diligent & competent service to principals
- Advance & protect the profession
DUE DILIGENCE –> practicing the activities that maintain due care before decision
- research, planning, evaluation
- best practices, ind standards, laws & regulations
- INCREASES understanding, REDUCES risk
DUE CARE –> do what a reasonable person would do after decision
- implementation, upkeep, measures
- report sec incidents, sec awareness training, disable access
- “prudent man” rule
Security planning
- STRATEGIC –> long term, CISO
- TACTICAL –> medium term, IT manager
- OPERATIONAL –> short term, IT engineer
4 levels of security policy development
- acceptable use policy –> assign roles & responsibilities
- baseline –> minimum levels
- guideline –> recommendaations
- procedure –> step-by-step description
Response to risk
- ACCEPT –> do nothing
- MITIGATION –> implement countermeasure & accept residual risk
- ASSIGNMENT –> transfer to 3rd party (insurance)
- AVOID –> mitigating costs > benefits, let it go …
- DETERRENCE –> implement deterrents to would-be violators (video surveillance)
- REJECTION –> aka ignore, DON’T DO IT!!!
Risk mgmt frameworks
- NIST 800-37 –> the only one relevant for the exam
- OCTAVE - Operationally Critical Threat, Asset and Vuln Evaluation
- FAIR - Factor Analysis of Information Risk
- TARA - Threat Agent Risk Assessment
NIST 800-37 steps
- prepare to execute the RMF
- categorize IS
- select security controls
- implement sec controls
- assess sec controls if work correctly
- authorize the system
- monitor sec controls
!!! NOT EVERY RISK CAN BE MITIGATED --> mgmt decides how the risk is handled --> HUMAN LIFE = HIGHEST IMPORTANCE
legal issues are involved --> "call an attorney"
Types of risk
- RESIDUAL –> remains after safeguards - AFTER safeguards
- INHERENT –> newly identified, not yet addressed (absence of controls) - BEFORE safeguards
- TOTAL –> amount of risk if no safeguards were implemented - NO safeguards
controls gap –> amount of risk reduced by implementing safeguards
threats X vulnerabilities X asset value = total risk
rusk = threat * vulnerability
total risk - controls gap = residual risk
Risk analysis
- QUANTITATIVE –> assign $ value to evaluate effectiveness of countermeasures; works with VALUES; is objective
- six steps in quantitative risk analysis
- inventory assets & assign a value –> AV = asset value
- identify threats –> calculate EF & SLE - EF - Exposure Factor (percentage destroyed), SLE = Single Loss Expectancy
- perform a threat analysis –> likelihood of each threat within a year - ARO = Annual Rate of Occurence (expected frequency)
- estimate the potential loss –> ALE = annualized loss expectancy (possible yearly cost)
- research countermeasures for threats & calculate their improvements
- perform cost/benefit analysis –> see if the value is positive
- six steps in quantitative risk analysis
- QUALITATIVE –> uses a scoring systems to rank threats & effectiveness of countermeasures; is subjective - Delphi technique –> anonymous feedback/response to arrive at a consensus
LOSS POTENTIAL –> what would be lost if the threat agent is successful
DELAYED LOSS –> amount of loss that can occur over time
SAFEGUARD EVALUATION –> mitigate risk, cost effective & difficult to bypass
SLE = AV * EF
ALE = SLE * ARO
ALE before safeguard - ALE after safeguard - annual cost of safeguard = VALUE OF SAFEGUARD
SUPPLY CHAIN –> chain of multiple entities delivering a service; must be secure, reliable, trustworthy
supply chain evaluation
- on-site evaluation
- document exchange & review
- process/policy review
- third-party audit
Threat modeling
security process where potential threats are id, categorised and analysed.
goal: eradicate/reduce threats
can be focused on:
- assets –> use asset evaluation results to id threats
- attackers –> based on attacker’s goals
- software –> potential threats against the sw
Threat modeling methodologies
- STRIDE (Microsoft)
- Spoofing
- Tampering
- Repudiation
- Information disclosure
- DoS
- Elevation of privilege
- PASTA –> focuses on controls relative to asset value
- VAST –> based on Agile
- Visual
- Agile
- Simple
- Threat
- TRIKE –> focuses on a risk-based approach
- DREAD –> ?
- Damage potential
- Reproducibility
- Exploitability
- Affected users
- Discoverability
- Diagramming potential attacks
- Reduction analysis –> - trust boundary –> any location where the level of trust/security changes - data flow paths –> movement of data - input points –> external input - privileged operations –> greater privs than of a standard user
SECURITY CONTROLS –> safeguard = proactive, countermeasure = reactive
categories:
- technical (hw, sw)
- administrative (policies, procedures)
- physical (can touch)
types:
- deterrent - discourage
- preventative - stop
- detective - discover/detect
- compensating - aid other controls
- corrective - modify env to return systems to normal after smth has happened
- recovery - extension of corrective w/ more capabilities
- directive - direct/confine/control actions of subjects to force/encourage compliance
Legal and regulatory
- criminal law
- civil law
- administrative law
LAWS
- CFAA - first US cybercrime-specific
- FISMA - infosec operations for federal agencies
- DMCA - literary/musical works
Intellectual property
- trademark
- patent
- trade secret
- copyright
Licensing
- contractual –> written contract
- shrink-wrap –> written outside of the sw package
- click-through –> click button to agree
- cloud services –> link & checkbox for terms/conditions
privacy laws
- HIPAA - health
- HITECH - health
- Gramm-Leach-Bliley –> financial institutions
- COPPA –> Children’s Online Privacy Protection Act
- ECPA –> Electronic Communications Privacy Act
- CALEA –> Comms Assistance for Law Enforcement Act
awareness = WHAT
sec training = HOW
sec education = WHY