CS 3106N / Cram / Set 1
Cram
Only what the quizzes ask, module by module. Each section gives a line of context, the facts to know, the traps that lose marks, and the NetAcad quiz questions on it.
8 modules · 233 facts · 137 questions
- 1Read a section: the context line, then the facts.
- 2Answer its questions in your head before you look.
- 3Finish a module, then drill it as multiple choice.
Module 8 · 28 facts · 14 questions
Linux Overview
Linux is the operating system most SOC work runs on: open source, driven from the command line and built for networks. This module covers the shell, servers and ports, logs, file permissions, links, processes and rootkits.
- 8.1Linux Basics
- 8.2Working in the Linux Shell
- 8.3Linux Servers and Clients
- 8.4Basic Server Administration
- 8.5The Linux File System
- 8.6Working with the Linux GUI
- 8.7Working on a Linux Host
8.1Linux Basics
Why a SOC runs on Linux, what the root user can do, and the tool types an analyst works with.
- Linux
- open source OS (1991) anyone may modify and recompile; distros bundle the kernel with tools: Debian, Red Hat, Ubuntu, CentOS, SUSE.
- Security Onion
- open source suite of network security analysis tools; Sguil is its analyst console.
- SOC tools
- Wireshark captures packets, IDS acts on rule matches, firewall allows or blocks, SIEM analyzes IDS and firewall logs live, plus ticketing.
- PenTesting
- finding vulnerabilities by attacking; tools are packet generators, port scanners, exploits. Kali Linux bundles them.
Quiz questions3
- 01
Which type of tool is used by a Linux administrator to attack a computer or network to find vulnerabilities?
PenTesting
- 02
What is a benefit of Linux being an open source operating system?
Its source code can be modified and then recompiled.
- 03
Which operating system source code can be downloaded and modified by any person or company?
Linux
8.2Working in the Linux Shell
The commands the quiz names by job: man, pwd, ps, chmod, sudo and the rest.
- man
- shows a command's documentation. The path is the list of directories the shell searches for a typed command.
- Basic commands
- pwd: current directory, ps: running processes, chmod: permissions, chown: owner, dd: copy input to output, cat: show file.
- su vs sudo
- su simulates logging in as another user or superuser; sudo runs one command as superuser (superuser do).
- ifconfig
- shows or sets NIC config, deprecated for ip address; iwconfig does wireless; passwd changes password; grep searches text.
- File commands
- ls lists, cd changes dir, mkdir creates dir, cp copies, mv moves or renames, rm removes.
- nano
- keyboard-only CLI editor, vital over SSH; CTRL+O saves, CTRL+W searches, CTRL+G help.
- Configuration files
- text files of option = value settings, # for comments; edits need superuser and a service restart. /etc/hosts maps names to IPs.
Trap: su and sudo are not the same. su switches you into another user (or root); sudo runs a single command as superuser by default and then drops you back. Likewise chmod changes permissions, chown changes the owner.
Quiz questions2
- 01
In the context of a Linux operating system, which command can be used to display the syntax and parameters for a specific command?
man
- 02
Which Linux command can be used to display the name of the current working directory?
pwd
8.3Linux Servers and Clients
Servers listen on ports and clients talk to them. Learn the well-known port numbers.
- Ports
- 20/21 FTP, 22 SSH, 23 Telnet, 25 SMTP, 53 DNS, 67/68 DHCP, 69 TFTP, 80 HTTP, 110 POP3, 123 NTP, 143 IMAP, 161/162 SNMP, 443 HTTPS.
- Client vs server
- the server provides a service and listens on a port; the client (e.g. a web browser) requests it.
Quiz questions2
- 01
An author is uploading one chapter document from a personal computer to a file server of a book publisher. What role is the personal computer assuming in this network model?
Client
- 02
A technician has captured packets on a network that has been running slowly when accessing the internet. Which port number should the technician look for within the captured material to locate HTTP packets?
80
8.4Basic Server Administration
Services read config files, devices get hardened, and every event lands in a log file.
- Hardening
- physical security, fewer packages, disable unused services, SSH without root login, updates, no USB auto-detect, strong rotated passwords.
- Daemon
- background process that needs no user interaction; e.g. SSSD handles remote access and single sign-on.
- Log files
- /var/log/messages: generic activity (syslog on Debian); auth.log: Debian/Ubuntu logins; secure: RedHat/CentOS sudo and SSH logins.
- More logs
- boot.log: startup; dmesg: kernel ring buffer, hardware and drivers; kern.log: kernel; cron: scheduled jobs; mysqld.log or mysql.log.
Quiz questions2
- 01
Which method can be used to harden a device?
Force periodic password changes.
- 02
What is a daemon?
A background process that runs without user interaction
8.5The Linux File System
File system types, reading rwx permissions from ls -l, and hard versus symbolic links.
- ext2 / ext3 / ext4
- ext2 no journal, best for flash; ext3 adds a journal for crash recovery, 32 TB max file; ext4 extends ext3, can run journal-less.
- Other file systems
- NFS: network, open standard; CDFS: optical discs; swap: disk used when RAM runs out; HFS+ and APFS: Apple (APFS current).
- MBR
- first sector of a partitioned disk, holds file system layout and hands control to the OS loader. Mounting assigns a directory to a partition.
- ls -l permissions
- type (- file, d dir) then User, Group, Other; -rwxrw-r-- is owner rwx, group rw-, others r--. Only root can override.
- Octal permissions
- r=4, w=2, x=1: 7 rwx, 6 rw-, 5 r-x, 4 r--; chmod 754 gives rwxr-xr--.
- Hard link
- ln; same inode as the original, survives deleting it, cannot cross file systems or link to directories.
- Symbolic link
- ln -s; shows name -> target, can cross file systems and link directories, breaks if the original is deleted.
Trap: Do not swap the failure behaviour. Deleting the original breaks a symlink, not a hard link. Symlinks still win overall: they show their target in ls -l, cross file systems and can link directories.
Quiz questions2
- 01
Consider this ls -l output: -rwxrw-r-- sales staff 1028 May 28 15:50 analyst.txt. What are the group file permissions assigned to the analyst.txt file?
Read, write
- 02
Which file system is the primary file system used by Apple in current Macintosh computers?
APFS
8.6Working with the Linux GUI
X Window draws the desktop; Gnome 3 is the default Ubuntu interface.
- X Window System (X11)
- GUI framework that draws windows and can show a remote app locally; window managers like Gnome and KDE define the look.
8.7Working on a Linux Host
Installing and updating packages, managing processes, and finding rootkits.
- Package managers
- pacman (Arch), dpkg and apt (Debian/Ubuntu); apt-get update refreshes the package list, apt-get upgrade updates all installed.
- Processes
- fork is the only way to make one (child: new PID, same code); ps lists, top lists live (q quits), kill removes, restarts or pauses.
- Rootkit
- gains hidden access by changing kernel code; detect by booting trusted media, fix by reinstalling the OS; scan with chkrootkit.
Trap: apt-get update only refreshes the list of available packages from the repository. apt-get upgrade is what updates the installed software.
Quiz questions3
- 01
A system administrator issues the command ps on a server that is running the Linux operating system. What is the purpose of this command?
To list the processes currently running
- 02
A system administrator issues the apt-get upgrade command on a Linux operating system. What is the purpose of this command?
Every installed application updates to its latest version.
- 03
Why would a rootkit be used by a hacker?
To gain access to a device without being detected
Module 9 · 28 facts · 23 questions
System & Endpoint Protection
Every endpoint is a way into the network, so protection layers host-based tools (antimalware, host firewall, HIDS/HIPS) on top of network-based ones. This module covers OS hardening, patching, antimalware, host intrusion detection, boot integrity and shrinking the attack surface.
- 9.1Defending Systems and Devices
- 9.2Antimalware Protection
- 9.3Host-Based Intrusion Prevention
- 9.4Application Security
9.1Defending Systems and Devices
Hardening an OS: patches, antimalware, host-based tools, encryption, secure boot and locks.
- OS hardening
- remove unneeded programs and services, patch promptly, compare to a baseline; updates: monitor, evaluate, plan, install by a documented plan.
- Antimalware types
- antivirus: quarantines viruses; adware: popup ads; phishing: blocks phishing IPs; spyware: keyloggers; trusted sources: unsafe sites.
- Rogue antivirus
- fake Windows-style warning popup; clicking it installs malware instead of cleaning.
- Fileless malware
- runs in memory via legitimate tools like PowerShell, leaves no footprint, ends on reboot. Scripts (Python, Bash, VBA) can be malware too.
- Patches
- code updates against new malware; a service pack combines patches and upgrades. Also patch third-party apps (Acrobat, Java, Chrome).
- Automated patch service
- admins approve or decline, force by date, get reports, serve from a local server; users cannot disable updates.
- Host encryption
- EFS encrypts files, folders or drives; BitLocker does full disk encryption with a TPM; BitLocker To Go encrypts removable drives, no TPM.
- TPM
- motherboard chip storing encryption keys, certificates and passwords; enabled in BIOS for BitLocker. Self-encrypting drives encrypt all data.
- BIOS vs UEFI
- BIOS is the first program at power-on; UEFI is the newer standard interface and preferred because it runs in 64-bit mode.
- Secure Boot vs Measured Boot
- Secure Boot checks each boot component's signature; Measured Boot logs each in the TPM for remote verification, stronger.
- Apple security
- Secure Enclave AES chip; Find My Mac locates, erases; XProtect blocks by signature; MRT removes malware; Gatekeeper allows only notarized apps.
- Device threat countermeasures
- password and lockout policy, screen lock, no admin rights, patching, disable CD/DVD/USB, content filtering, approved app list.
- Physical protection
- cable locks, locked telecom rooms, Faraday cages; RFID tags track assets with radio waves and need no battery.
- Door locks
- keyed entry lock is most common but easy to force; deadbolt adds security; cipher lock uses a button code, time limits and an access record.
Quiz questions3
- 01
What can you do to ensure that network operating software remains secure? (Choose three.)
Install patches regularly, test software before launch, and set an update policy
- 02
Which type of technology can prevent malicious software from displaying unwanted popup ads on a device?
Adware protection
- 03
What type of lock is recommended to secure an office door?
Keyed entry lock
9.2Antimalware Protection
Why endpoints are the weak spot, and the network and host tools that protect them.
- Endpoint
- any host that can access or be accessed by other hosts, including IoT devices and VPN clients; cloud and IoT widen the challenge.
- Internal LAN elements
- endpoints (malware attacks) and network infrastructure such as switches (MAC table overflow, spoofing, DHCP, STP, VLAN attacks).
- Detection approaches
- signature-based: known malware characteristics; heuristics-based: general shared features; behavior-based: suspicious behavior.
- Agent-based vs agentless
- agent-based (host-based) AV runs on every machine; agentless scans from a central system, suits VMs (VMware vShield).
- Host-based security suite
- antivirus, anti-phishing, safe browsing, HIPS and firewall; telemetry is its logging, sent to a central location.
- Network-based protection
- AMP: endpoint malware; ESA: filters spam; WSA: website filtering and blocklisting; NAC: only compliant devices join.
Quiz questions9
- 01
How does the course define endpoints?
Hosts that can access or be accessed by other hosts on the network
- 02
Which type of antimalware software recognizes various characteristics of known malware files?
Signature-based
- 03
Which type of endpoint protection includes iptables and TCP Wrappers?
Host-based firewall
- 04
What provides filtering of websites and blocklisting to prevent endpoints from accessing malicious web pages?
Web security appliance
- 05
Which type of endpoint protection permits only authorized and compliant devices to connect to the network?
Network admission control
- 06
Which device in a LAN infrastructure is susceptible to MAC address table overflow and spoofing attacks?
Switch
- 07
In most host-based security suites, which function provides robust logging of security-related events and sends logs to a central location?
Telemetry
- 08
Which technology might increase the security challenge to the implementation of IoT in an enterprise environment?
Cloud computing
- 09
Which statement describes agentless antivirus protection?
Antivirus scans are performed on hosts from a centralized system.
9.3Host-Based Intrusion Prevention
Host firewalls and HIDS: how they log, detect by signature or anomaly, and enforce policy.
- Windows Defender Firewall profiles
- Public: restrictive, public networks; Private: behind a home router; Domain: trusted business network.
- Linux firewalls
- iptables: Netfilter kernel rules; nftables: its successor, kernel virtual machine; TCP Wrappers: rule-based access control and logging.
- HIDS
- agent-based; antimalware plus firewall; signatures miss zero-day and polymorphic malware. Products: Cisco AMP, AlienVault USM, Tripwire, OSSEC.
- Anomaly-based vs policy-based
- anomaly compares to a learned baseline of normal behavior (many false results); policy acts on predefined rule violations.
- OSSEC
- open source HIDS; central manager plus agents on Mac, Windows, Linux, Solaris; reads syslog, checks file integrity, detects rootkits.
Trap: Signature-based detection fails against zero-day exploits (no known patch or signature) and polymorphic malware (it keeps changing to evade signatures). That gap is why HIDS adds anomaly and policy detection.
Quiz questions7
- 01
What allows Linux system administrators to configure network access rules that are part of the Linux kernel Netfilter modules?
iptables
- 02
Which of the following uses a profile-based approach to configuring firewall functionality?
Windows Firewall
- 03
Which of the following is a rule-based access control and logging system for Linux?
TCP Wrappers
- 04
Which of the following uses a learned baseline model?
Anomaly-based strategy
- 05
Which HIDS is an open-source based product?
OSSEC
- 06
In Windows Firewall, when is the Domain profile applied?
When the host is connected to a trusted network such as an internal business network
- 07
What is a host-based intrusion detection system (HIDS)?
It combines the functionalities of antimalware applications with firewall protection.
9.4Application Security
Shrinking the attack surface with allowlists and blocklists, and testing files in a sandbox.
- Attack surface
- sum of vulnerabilities reachable by an attacker. SANS: network (protocols, IoT), software (web, cloud, apps), human (social engineering).
- Blocklist vs allow list
- blocklist names apps or sites that may not run; allow list names the only ones that may, based on a security baseline.
- Sandboxing
- run suspicious files safely to study behavior. Cisco Threat Grid Glovebox, Cuckoo (free, local), VirusTotal, Joe Sandbox, ANY.RUN.
Trap: A blocklist and an allow list are opposites. A blocklist forbids named programs and allows the rest; an allow list permits only named programs and forbids the rest.
Quiz questions4
- 01
As described by the SANS Institute, which attack surface includes the exploitation of vulnerabilities in wired and wireless protocols used by IoT devices?
Network attack surface
- 02
Which statement describes the term attack surface?
It is the total sum of vulnerabilities in a system that is accessible to an attacker.
- 03
As described by the SANS Institute, which attack surface includes the use of social engineering?
Human attack surface
- 04
Which security endpoint setting would be used by a security analyst to determine if a computer has been configured to prevent a particular application from running?
Blocklisting
Module 10 · 28 facts · 12 questions
Cybersecurity Principles, Practices & Processes
Module 10 organises everything in cybersecurity into one model, the cybersecurity cube: what you protect (goals), where the data is (states) and how you protect it (countermeasures). Expect to name all three dimensions and sort examples into them.
10.1The Three Dimensions
The cybersecurity cube: principles (CIA), data states, and the safeguards that protect them.
- Cube, 1st dimension
- goals (CIA). Availability — accessible by authorized users when needed; a DoS on a website attacks it.
- Confidentiality
- prevents disclosure to unauthorized people, resources or processes. Integrity — accuracy, consistency, trustworthiness of data.
- Cube, 2nd dimension
- the three data states: in transit, at rest (in storage), in process. All three must be safeguarded.
- Cube, 3rd dimension
- safeguards: technology; policy and practices; education, training and awareness in people.
- Tokenization
- confidentiality without encryption: a random value with no mathematical relationship replaces the data; can keep its type and length.
- DRM vs IRM
- both use encryption. DRM protects copyrighted music, films, books; IRM lets an owner control access to shared email and files.
- Sensitive information
- personal (traceable to a person: medical, card numbers), business (acquisition plans, trade secrets), classified.
- Privacy enhancement technologies
- anonymization, data minimization, tokenization. Anonymization makes private data irreversibly anonymous.
- Integrity methods
- hashing, data validation checks, data consistency checks, access controls; a bank values integrity more than a social media channel.
- Not causes of system failure
- equipment maintenance and system backups; they maintain availability despite short planned outages.
- Ensuring availability
- maintenance, updates and patches, backup testing, disaster planning, new tech, activity monitoring, availability testing.
- Equipment maintenance
- component replacement, cleaning, alignment. Availability testing — port scans, vulnerability scans, penetration tests.
Trap: The first dimension is the goals (the security principles), not the safeguards. Safeguards are the third dimension.
Trap: Equipment maintenance and backups are not causes of system failure. They help availability, even when they cause short planned downtime.
Quiz questions5
- 01
What is identified by the first dimension of the cybersecurity cube?
Goals
- 02
What type of cybersecurity laws protect you from an organization that might want to share your sensitive data?
Privacy
- 03
Which two methods help to ensure data integrity?
Hashing and data consistency checks
- 04
Which of the following are types of sensitive information?
Personal, business and classified
- 05
Which of the following are foundational principles of the cybersecurity domain?
Confidentiality, integrity and availability
10.2States of Data
Data at rest, in transit and in process, where each lives, and how each gets attacked.
- Data at rest
- in storage, no user or process accessing, requesting or amending it; NAS and SAN hold stored data. Includes idle backups.
- DAS
- direct-attached storage (hard drive, USB flash drive) connected to one computer; not shared by default; vulnerable to attacks on the local host.
- RAID
- many disks seen as one. NAS — network device, central store. SAN — high-speed links many servers to one disk repository.
- Sneaker net
- carrying removable media; never fully eliminated. Wireless — radio waves; BYOD guests grow the attack surface.
- Data in transit countermeasures
- confidentiality: VPN, SSL, IPsec, encryption; integrity: hashing, data redundancy; availability: mutual authentication.
- Mutual authentication
- user authenticates to the server and server to the user; counters a rogue device posing as a wireless access point.
- Data in process
- data during initial input, modification, computation or output; input threats: data entry, scanned forms, file uploads, sensors.
- Data modification
- encoding, compression, encryption are intentional; data corruption (unreadable) comes from failed equipment or malicious code.
Trap: Data in process means being input, modified, computed or output. It is not data in storage and not data on the wire, so do not put a file on a SAN or a packet on a VPN here.
Quiz questions6
- 01
Customer information has been modified in transit. It was not manual error and no equipment failed, but accounting files with bank transaction details have been corrupted. What could have been the cause?
Malicious code
- 02
What name is given to a storage device connected to a network?
NAS
- 03
What is a method of sending information from one device to another using removable media?
Sneaker net
- 04
Which data state is maintained in NAS and SAN services?
Stored data
- 05
Which type of networks poses increasing challenges to cybersecurity specialists due to the growth of BYOD on campus?
Wireless networks
- 06
An organization allows employees to work from home two days a week. Which technology should be implemented to ensure data confidentiality as data is transmitted?
VPN
10.3Cybersecurity Countermeasures
Technologies, awareness training and policies, standards, guidelines and procedures.
- Software safeguards
- software firewalls, network and port scanners, protocol analyzers, vulnerability scanners (assess weaknesses), host-based IDS.
- Protocol analyzer
- a.k.a. signature analyzer; examines traffic to find performance problems, misconfigurations, misbehaving apps, baselines.
- Hardware safeguards
- firewalls, proxy servers (one org-wide IP, act for the client), biometric hardware access control, network switches.
- Security policy tasks
- shows commitment, sets expected behavior, ensures consistency, defines legal consequences, gives staff management backing.
- Policy types
- identification and authentication, password, AUP (what users may and may not do), remote access, maintenance, incident handling.
- Standards
- mandatory, for consistency; e.g. passwords of 8+ mixed-case alphanumeric with a special character, changed every 30 days, history of 12.
- Guidelines
- flexible, usually not mandatory suggestions (best practices); sources: NIST Computer Security Resource Center, NSA guides, Common Criteria.
- Procedures
- longer, more detailed than standards and guidelines: step-by-step instructions with graphics, e.g. how to change your password.
Trap: Standards are required consistency; guidelines are optional advice. If a question says flexible or not usually mandatory, the answer is guideline.
Quiz questions1
- 01
What tasks are accomplished by a comprehensive security policy?
Sets rules for behavior, defines legal consequences of violations, backs security staff with management
Module 11 · 28 facts · 10 questions
Understanding Defense
Module 11 is about defending an organisation on purpose: know what you own, know what can hurt it, then stack layers so one failure does not expose everything. The exam weights defense-in-depth (onion versus artichoke, the five strategies), asset management and the policy types.
- 11.1Defense-in-Depth
- 11.2Cybersecurity Operations Management
- 11.3Security Policies, Regulations, and Standards
11.1Defense-in-Depth
Know your assets, threats and vulnerabilities, then layer defenses: onion versus artichoke.
- Identify first
- assets (anything of value; data is the greatest), vulnerabilities (weakness in a system or design), threats (any potential danger).
- Attack surface
- all devices and info the organization owns or manages. Asset management — inventory them, then set policies to protect them.
- Asset classification
- groups resources by common traits so the most critical get the most protection; labels rate value and sensitivity.
- Classification steps
- 1 identification category (information, software, physical, services); 2 accountability (owners); 3 criteria; 4 schema.
- Classification criteria
- confidentiality, value, time, access rights, destruction. Schema — consistent labeling for uniform protection.
- Asset standards
- the specific hardware and software products used, so a failed part is replaced fast; non-standard setups cost more to maintain.
- Asset lifecycle
- procurement (bought, inventoried), deployment (inspected, tested, barcoded, put in use), utilization, maintenance, disposal.
- Utilization
- longest stage: performance checks, patches, upgrades, licenses, audits. Maintenance — modify or upgrade. Disposal — wipe all data.
- Threat identification
- list of likely threats: what are the vulnerabilities, who may exploit them, what are the consequences if assets are lost.
- E-banking threats
- internal system compromise, stolen customer data, phony transactions, insider attack, data input errors, data center destruction.
- Defense-in-depth
- multiple layers at the edge, inside the network and on endpoints; the failure of one safeguard does not affect the others.
- Edge router
- first line of defense; rules allow or deny traffic, passes LAN-bound connections to the firewall.
- Firewall
- second line; tracks connection state, denies connections started from untrusted outside, can do authentication proxy for remote users.
- Internal router
- applies final filtering before the destination. Other layers: IPS, AMP (Advanced Malware Protection), web/email content security, NAC.
- Security onion
- attacker peels defenses layer by layer. Security artichoke — borderless networks; remove only certain leaves (e.g. mobile devices).
- Layering
- multiple defenses, each harder to beat. Limiting — each user gets only the access the job needs (file permissions, CCTV rooms).
- Diversity
- different layers, vendors, encryption and authentication (one vendor's swipe card, another's fingerprint reader).
- Obscurity
- reveal no OS or equipment details, e.g. generic error messages. Simplicity — complexity is not security: simple inside, complex outside.
Trap: A vulnerability is the weakness; a threat is the danger. A threat actor exploits a vulnerability to reach an asset. Do not swap the two terms.
Trap: The security onion here is an analogy for visualising defense-in-depth. It is not the Security Onion suite of network security tools.
Quiz questions5
- 01
In a defense-in-depth approach, which three items must be identified to effectively defend a network against attacks?
Assets, threats to assets, and system vulnerabilities
- 02
Which device is usually the first line of defense in a layered defense-in-depth approach?
Edge router
- 03
Which device in a layered defense-in-depth approach denies connections initiated from untrusted networks to internal networks, but allows internal users within an organization to connect to untrusted networks?
Firewall
- 04
With the evolution of borderless networks, which vegetable is now used to describe a defense-in-depth approach?
Artichoke
- 05
What is the benefit of a defense-in-depth approach?
The effectiveness of other security measures is not impacted when a security mechanism fails.
11.2Cybersecurity Operations Management
Configuration management and the logs that show what happened.
- Configuration management
- identifying, controlling and auditing changes to a system's baseline, the template configuration for all similar systems.
- Log management
- generate, transmit, store, analyze, dispose. Audit log — authentication attempts; access log — requests for specific files.
- OS log
- OS events: logins, failed logins, usage. Application security log — security software flags malicious activity (Word going online).
- Packet analyzer
- packet sniffer on wired and wireless; traffic logging, problem analysis, misuse and intrusion detection, isolating exploited systems.
Quiz questions1
- 01
Which tool can be used to gather information about the different types of traffic that exist in a network?
Protocol analyzer
11.3Security Policies, Regulations, and Standards
Business and security policies, the AUP, BYOD rules, and the laws an organization must follow.
- Business policies
- company (conduct, attendance, dress code), employee (HR: salary, benefits, vacations), security (objectives, rules, requirements).
- Security policy parts
- identification and authentication (who may access), password, AUP, remote access, network maintenance (updates), incident handling.
- AUP
- acceptable (appropriate) use policy: explicitly what users may and may not do, allowed traffic; all staff sign it, kept for the employment term.
- BYOD policy goals
- program goals, which employees and devices, access level, security staff rights on the device, regulations, safeguards if compromised.
- BYOD best practices
- unique passwords, Wi-Fi/Bluetooth off when unused, updates, backups, Find my Device with remote wipe, antivirus, MDM software.
- Compliance regulations
- define what organizations must provide and their liability if they fail; depend on organization type and data handled.
Quiz questions4
- 01
What component of a security policy explicitly defines the type of traffic allowed on a network and what users are allowed and not allowed to do?
Acceptable use policies
- 02
Which section of a security policy is used to specify that only authorized individuals should have access to enterprise data?
Identification and authentication policy
- 03
An administrator discovers that a user is accessing a newly established website that may be detrimental to company security. What action should the administrator take first in terms of the security policy?
Revise the AUP immediately and get all users to sign the updated AUP.
- 04
What three goals does a BYOD security policy accomplish?
Identify who may bring devices, describe security staff's rights on the device, and set safeguards if one is compromised
Module 12 · 30 facts · 14 questions
System & Network Defense
Module 12 is the widest in the unit: it hardens a system from the front door to the data centre, covering physical security, applications, network services, segmentation, wireless and mobile devices, resilience and embedded systems. The exam leans on RAID, availability and five nines, the DMZ zones, and wireless security.
- 12.1Physical Security
- 12.2Application Security
- 12.3Network Hardening: Services and Protocols
- 12.4Network Hardening: Segmentation
- 12.5Hardening Wireless and Mobile Devices
- 12.6Cybersecurity Resilience
- 12.7Embedded and Specialized Systems
12.1Physical Security
Fences, biometrics, badges, guards and cameras: keeping people away from the hardware.
- Type I vs. Type II error
- Type I: false rejection of an authorized user; Type II: false acceptance of an imposter, the worst for access control.
- Crossover error rate (CER)
- point where false acceptance and false rejection rates meet; the common measure of overall biometric accuracy.
Quiz questions2
- 01
Cybercriminals already on the premises plan to enter the server room while the surveillance cameras are down to steal client information. Which three physical security measures would be effective at preventing this attack?
Fingerprint reader, swipe cards and guards
- 02
A company implementing biometric access to its data center is worried about people being falsely accepted as legitimate users. What type of error is false acceptance?
Type II
12.2Application Security
Secure development, input validation, and the threats an application faces.
- Normalization
- converts input to its simplest form so each string has a unique binary representation and malicious input is spotted.
- Stored procedure
- precompiled SQL statements in the database taking input parameters; less network traffic, faster results.
- Obfuscation vs. camouflage
- obfuscation hides data with random characters; camouflage swaps in realistic fictional data. Both deter reverse engineering.
- Validation rule criteria
- size (character count), format, consistency (codes in related items), range (min/max), check digit (extra calculation).
- Integrity check
- hashes a snapshot of data to prove it is unchanged (checksum); hash a download to verify it. MD5, SHA-1, SHA-256, SHA-512.
- Code signing vs. secure cookies
- code signing proves software is authentic and unchanged; secure cookies store browsing data safely for later requests.
Trap: Obfuscation and camouflage are not the same. Obfuscation scrambles data with random characters; camouflage swaps in believable fake data.
Quiz questions1
- 01
A user wants to distribute a new program to everyone in the company and ensure it is not changed while in transit. What can the user do?
Create a hash to verify the file after download
12.3Network Hardening: Services and Protocols
Closing unneeded ports and replacing insecure protocols with secure ones.
- DHCP snooping vs. DNSSEC
- snooping blocks rogue DHCP servers by validating untrusted messages; DNSSEC digitally signs DNS, which maps names to IPs.
- ICMP, RIP, NTP
- ping uses ICMP to test reachability; RIP routes by hop count, max 15 hops; NTP syncs clocks for logs and certificates.
- SSH vs. Telnet
- SSH encrypts remote sessions on TCP 22; Telnet sends credentials in plaintext on TCP 23. SCP copies files over SSH.
- Secure protocols
- SNMPv3 adds cryptography; HTTP port 80, TLS replaces SSL (HTTPS); FTPS adds TLS/SSL; POP 110, IMAP 143; S/MIME signs and encrypts mail.
Quiz questions3
- 01
An administrator of a small data center wants a flexible, secure method of remotely connecting to servers. Which protocol would be best to use?
Secure Shell
- 02
Which service will resolve a specific web address into an IP address of the destination web server?
DNS
- 03
Which utility uses the Internet Control Messaging Protocol (ICMP)?
ping
12.4Network Hardening: Segmentation
VLANs, the DMZ and security zones, and how traffic flows between them.
- VLAN
- logical (not physical) grouping of switch ports; limits broadcasts and isolates sensitive hosts. Trunks carry multiple VLANs between switches.
- DMZ
- small network between the trusted LAN and the Internet holding web and mail servers, so outsiders reach them without touching the internal network.
- Zones of risk
- LAN low, extranet medium-low, DMZ medium-high, Internet high. Zero Trust trusts no one automatically and monitors every user.
Trap: A DMZ is not the trusted LAN. It holds the public-facing servers and sits at medium-high risk, so keep internal data out of it.
Quiz questions1
- 01
What is the purpose of a DMZ?
It allows external hosts to access specific company servers while maintaining the security restrictions for the internal network.
12.5Hardening Wireless and Mobile Devices
Wireless standards and their security, and keeping phones and tablets under control.
- WEP → WPA
- WPA replaced WEP: 256-bit keys (WEP 64/128), MIC to detect altered data, TKIP key handling; WPA-PSK is the most common setup.
- WPA2 vs. WPA3
- WPA2 (2006) made AES mandatory and replaced TKIP with CCMP; WPA3 keeps strong crypto and improves key exchange.
- WPS
- joins devices with a PIN that can be brute-forced; disable it.
- Open system vs. shared key
- open lets any device connect; shared key authenticates and encrypts client-to-AP. Best choice: WPA2 with EAP.
- EAP methods
- EAP-TLS: client and server certs, high security, hard; PEAP and EAP-TTLS: server cert only, medium; EAP-FAST: no certs, easiest.
- Mutual authentication
- both ends authenticate each other before connecting; detects rogue access points and stops man-in-the-middle attacks.
- Jailbreaking vs. rooting vs. sideloading
- jailbreak lifts Apple's app limits; rooting gives full Android admin access; sideloading installs unapproved apps.
Quiz questions3
- 01
You have set up a wireless network device in the office for employees. What is the best way to secure it?
Use WPA2 with EAP authentication
- 02
Why is WPA2 better than WPA?
mandatory use of AES algorithms
- 03
Mutual authentication can prevent which type of attack?
man-in-the-middle
12.6Cybersecurity Resilience
Staying up: five nines, redundancy, RAID, backups and disaster recovery.
- Five nines
- 99.999% availability, under 5.26 min downtime a year. HA principles: no single points of failure, reliable crossover, detect failures.
- N+1 redundancy
- every N components have at least one spare (four tires plus a spare); not a fully redundant system.
- RAID
- fault tolerance for drive failure; data survives one lost disk except in RAID 0. Mirroring duplicates, striping spreads, parity adds checksums.
- Spanning Tree Protocol (STP)
- blocks redundant switch paths to prevent Layer 2 loops; unblocks them if a link or switch fails.
- Power terms
- spike: momentary high, surge: prolonged high; fault: momentary loss, blackout: complete loss; sag/dip: momentary low, brownout: prolonged low.
Trap: RAID 0 is faster but has no redundancy: losing any one disk loses the data. It is the exception to RAID protecting against disk loss.
Quiz questions2
- 01
An audit finds several servers using single drives for their operating systems. Which solution best provides fault tolerance during a drive failure?
RAID
- 02
A user wants to build redundancy into the network in case of a switch failure but prevent Layer 2 looping. What should the user implement?
Spanning Tree Protocol
12.7Embedded and Specialized Systems
SCADA, IoT, VoIP and other devices that are hard to patch and easy to attack.
- IoT first steps
- make sure firmware can be patched and change default admin credentials, which are publicly known.
- VoIP protection
- encrypt voice packets against eavesdropping, SSH for gateways and switches, change default passwords, VoIP-aware firewalls.
- Deception technologies
- honeypot: exposed decoy server that logs attackers; honeynet: a set of honeypots; DNS sinkhole: blocks resolving malicious URLs.
Quiz questions2
- 01
Which technology can be used to protect VoIP against eavesdropping?
encrypted voice messages
- 02
What two steps should be taken before connecting any IoT device to a home or business network?
Patch the firmware and change the default admin credentials
Module 13 · 30 facts · 17 questions
Access Control
Access control decides who gets in (authentication), what they may do (authorization) and what they did (accounting), under least privilege and zero trust. Expect questions on control types, the access control models, the authentication factors and RADIUS versus TACACS+.
13.1Access Controls
Physical, logical and administrative controls, and the steps of AAA: identify, authenticate, authorize, account.
- Physical access controls
- actual barriers: guards, fences, motion detectors, laptop locks, swipe cards, cameras, mantraps, alarms.
- Logical access controls
- hardware and software: encryption, smart cards, passwords, biometrics, ACLs, protocols, firewalls, routers, IDS, clipping levels.
- Administrative access controls
- policies and procedures focused on personnel and business practices, e.g. security training.
- Authentication factors
- something you know (password, PIN), something you have (token, card, fob), something you are (fingerprint).
- Federated identity management
- same credentials across several enterprises (single sign-on to the web); protect it by tying login to an authorized device.
- Physiological vs. behavioral biometrics
- physiological: fingerprint, DNA, face, hand, retina, ear; behavioral: gestures, voice, gait, typing rhythm.
- Multi-factor authentication
- at least two different factor types (ATM card plus PIN); 2FA is MFA with exactly two. Best: PIN, fob and fingerprint.
Trap: A swipe card at a door is a physical control, but a smart card (a card with an embedded microchip) used to log in is a logical one. Ask whether it is a barrier in the building or a technology that checks identity.
Trap: Two passwords is still one factor. MFA needs different categories (know, have, are), not two items from the same category. An ATM is the simple example: you must have the card and know the PIN.
Quiz questions3
- 01
A member of the accounting team at @Apollo left their security key fob on the train. Which multi-factor authentication combination would best prevent such a security breach if it happened again?
Fingerprint, PIN and security fob
- 02
Which three processes are examples of logical access controls? (Choose three.)
Firewalls, biometrics and IDS
- 03
An organization plans to implement security training to educate employees about security policies. What type of access control is the organization trying to implement?
Administrative
13.2Access Control Concepts
Zero trust, and the access control models: MAC, DAC, RBAC, ABAC and rule-based.
- Zero trust
- never trust, always verify; each access decision point is a perimeter. Pillars: workforce (people), workloads (apps), workplace (devices, IoT).
- DAC
- discretionary, least restrictive: owners control access to their own data, often with ACLs.
- MAC
- mandatory, strictest, military and mission critical: objects carry security labels, users need matching clearance.
- RBAC (role-based)
- access by role, job classification or responsibilities; also called non-discretionary access control.
- ABAC
- attribute-based: attributes of the object, the subject and the environment (e.g. time of day).
- Rule-based vs. time-based
- rule-based: admin-set rules in ACLs (IPs, protocols, no payroll after hours), users cannot change them; TAC: by time and day.
- Least privilege vs. privilege escalation
- give only the minimum access needed; escalation exploits flaws to gain more privilege than allowed.
- NAC
- network access control: checks device compliance by user type, device type and OS before granting access; makes BYOD and IoT control practical.
Trap: "RBAC" is used for two models. Role-based access control is non-discretionary and follows job roles. Rule-based access control (sometimes written Rule Based RBAC) follows admin-set rules. Read the scenario for roles versus rules or conditions.
Quiz questions6
- 01
Which access control model is based on attributes of the object (resource) to be accessed, the subject (user) accessing the resource, and environmental factors regarding how the object is to be accessed, such as time of day?
Attribute-based access control
- 02
Which access control model is based on an individual's roles and responsibilities within the organization?
Non-discretionary access control
- 03
Which access control model applies the strictest access control and is typically used in military or mission critical applications?
Mandatory access control
- 04
A security audit finds that new users have access to network resources beyond their normal job roles, and users who move to different positions keep their prior permissions. What kind of violation is occurring?
Least privilege
- 05
Which pillar of the zero trust security approach focuses on the secure access of devices, such as servers, printers, and other endpoints, including devices attached to IoT?
Workplace
- 06
What is used to scan a BYOD device to verify that it is compliant with company security policies before the device is permitted to access the network?
NAC
13.3Account Management
Account types, permissions, password and lockout policies, and authentication protocols.
- Account types
- user (day-to-day), service (interacts with the OS), guest (temporary); replace default accounts and require passwords on all.
- Privileged account practices
- reduce their number, least privilege, secure password storage, no shared accounts, rotate passwords every 30 or 60 days, MFA.
- File permissions
- Full control: all; Modify: change/delete, not create; Read and execute: view and run; Write: create and change; Read: view and open.
- Deny and propagation
- a deny overrides every other permission, even for admins. Only a move within the same volume keeps original permissions.
- Local Security Policy
- secpol.msc, for PCs not in a domain; sets password and account lockout policies (e.g. lock 30 min after 5 failed tries).
- SSO, OAuth, password vault, KBA
- one login for many apps; third-party use of account info; one password guards stored credentials; question-based reset.
- HMAC
- hash plus secret key to prove integrity and authenticity; IPsec VPNs use it on every packet (e.g. with SHA-1).
- PAP vs. CHAP
- PAP sends username and password in plaintext; CHAP compares one-way hashes and re-verifies the client during the session.
- Kerberos
- mutual authentication by tickets: a ticket-granting ticket, then a service ticket; tickets are timestamped and expire.
- Hash choice
- use SHA-256 or higher; avoid SHA-1 and MD5. Hashing alone catches accidental, not deliberate, changes.
Trap: Only data moved within the same volume keeps its permissions. A copy always inherits new ones, even on the same volume.
Quiz questions5
- 01
An @Apollo employee wants their team to view existing project documents in a shared folder but not edit, create or delete any documents. What permission level should the team be granted?
Read and execute
- 02
Which access control model assigns security privileges based on the position, responsibilities, or job classification of an individual or group within an organization?
Role-based
- 03
A user is implementing IPsec for inbound external connections using SHA-1 and wants to ensure the integrity and authenticity of the connection. What security tool can the user use?
HMAC
- 04
After a security audit, multiple accounts were found to have privileged access to systems and devices. Which three best practices for securing privileged accounts should be in the audit report? (Choose three.)
Reduce privileged accounts, secure password storage and enforce least privilege
- 05
What Windows utility should be used to configure password rules and account lockout policies on a system that is not part of a domain?
Local Security Policy tool
13.4AAA usage and operation
Local versus server-based AAA, RADIUS versus TACACS+, and what accounting records.
- AAA
- authentication: who are you (passwords, challenge-response, token cards); authorization: what you may do (FTP only); accounting: what you did.
- Local vs. server-based AAA
- local (self-contained) suits small networks; centralized server suits medium-to-large and is preferred, more scalable.
- TACACS+
- Cisco, TCP 49; separates all three A's; encrypts the whole packet body; per-user router command authorization; limited accounting.
- RADIUS
- open standard, UDP 1812/1813 (or 1645/1646); combines authentication and authorization; encrypts only the password; extensive accounting.
- Accounting types
- network (PPP sessions), connection (outbound, e.g. SSH), EXEC (user shells), system (reboots), command (shell commands per privilege level).
Trap: Do not swap the details. TACACS+ is TCP 49 and encrypts everything; RADIUS is UDP and encrypts only the password. RADIUS has the extensive accounting; TACACS+ has the finer authorization.
Quiz questions3
- 01
What is the purpose of the network security accounting function?
To keep track of the actions of a user
- 02
Because of implemented security controls, a user can only access a server with FTP. Which AAA component accomplishes this?
Authorization
- 03
Which AAA component can be established using token cards?
Authentication
Module 14 · 29 facts · 22 questions
Access Control Lists
An ACL is an ordered list of permit and deny statements a router checks top-down, first match wins, with a hidden deny at the end. Expect wildcard masks, standard versus extended, placement and writing the IOS commands.
- 14.1Introduction to Access Control Lists
- 14.2Wildcard Masking
- 14.3Configure ACLs
- 14.4Named Standard IPv4 ACL Syntax
- 14.5Implement ACLs
- 14.6Mitigate Attacks with ACLs
- 14.7IPv6 ACLs
- +Across the module
14.1Introduction to Access Control Lists
What an ACL is, standard versus extended, numbered versus named, inbound versus outbound.
- ACL vs ACE
- an ACL is a sequential list of permit or deny statements; each statement is an ACE. A router has no ACLs by default.
- Standard vs extended ACL
- standard filters at Layer 3 on source IPv4 only; extended adds destination, protocol and TCP/UDP ports (L3 and L4).
- ACL number ranges
- standard 1–99 and 1300–1999; extended 100–199 and 2000–2699.
- Named ACLs
- preferred method; can be standard or extended; alphanumeric, case sensitive, unique, no spaces or punctuation; entries can be added or deleted.
- Inbound ACL
- filters before routing, saving a routing lookup for discarded packets; best when one inbound network is the only source to examine.
- Outbound ACL
- filters after routing, toward the destination; best when one filter covers packets from several inbound interfaces leaving one interface.
- Implicit deny
- hidden last ACE of every ACL blocks anything unmatched; an ACL with no permit statement denies all traffic. First match wins, rest skipped.
Trap: Forgetting permit any (standard) or permit ip any any (extended). An ACL of only deny statements blocks everything, because the implicit deny catches all the traffic you did not mention.
Trap: Wrong direction. Decide in or out from the router's point of view: a packet entering the interface is in, a packet leaving is out.
Quiz questions5
- 01
Which ACL is capable of filtering based on TCP port number?
Extended ACL
- 02
Which statement about ACLs is true?
Named ACLs can be standard or extended
- 03
In applying an ACL to a router interface, which traffic is designated as outbound?
Traffic leaving the router toward the destination host
- 04
What type of ACL offers greater flexibility and control over network access?
Extended
- 05
Which scenario would cause an ACL misconfiguration and deny all traffic?
Apply an ACL that has only deny ACEs
14.2Wildcard Masking
0 means match and 1 means ignore. Work out masks from subnets, and use host and any.
- Wildcard mask
- 32-bit; bit 0 = match the address bit, bit 1 = ignore it (reverse of a subnet mask). Also used by OSPF.
- Common wildcards
- 0.0.0.0 one host; 0.0.0.255 a /24; 0.0.255.255 a /16; 255.255.255.255 all hosts.
- Wildcard calculation
- 255.255.255.255 minus the subnet mask: /28 = 0.0.0.15, /23 = 0.0.1.255, /20 = 0.0.15.255; 172.16–19.0.0 = 0.3.255.255.
- host and any
- host replaces the 0.0.0.0 mask (one address); any replaces 255.255.255.255 (every address).
Trap: Using the subnet mask instead of the wildcard. For a /24 write 0.0.0.255, not 255.255.255.0. In a wildcard, 0 is match and 1 is ignore, the reverse of a subnet mask.
Quiz questions6
- 01
Which wildcard mask would permit only host 10.10.10.1?
0.0.0.0
- 02
Which wildcard mask would permit only hosts from the 10.10.0.0/16 network?
0.0.255.255
- 03
Which wildcard mask would permit all hosts?
255.255.255.255
- 04
Which wildcard mask would permit all hosts from the 192.168.10.0/24 network?
0.0.0.255
- 05
What wildcard mask will match networks 172.16.0.0 through 172.19.0.0?
0.3.255.255
- 06
Which two keywords can be used in an access control list to replace a wildcard mask or an address and wildcard mask pair?
host and any
14.3Configure ACLs
The commands for standard and extended ACLs, ports, and the established keyword.
- Standard ACE defaults
- if the source wildcard is omitted, 0.0.0.0 is assumed; remove a numbered ACL with no access-list number.
- remark and log
- remark documents an ACL (max 100 characters); log messages on the first matched packet, is CPU intensive, troubleshooting or attacks only.
- Extended ACE fields
- protocol (ip = all IP protocols, tcp, udp, icmp), source and destination, port operators lt, gt, eq, neq; eq matches one application.
- Protocol and port numbers
- ICMP 1, TCP 6, UDP 17; SSH (22) and HTTPS (443) have no port name, so enter the number; www = 80.
- established
- TCP only; matches returning segments with ACK or RST set, letting replies to inside traffic back in; a basic 1st generation stateful firewall.
- Named ACL commands
- ip access-list standard NAME or ip access-list extended NAME enters ACL sub-configuration mode; no need to repeat the command per ACE.
Trap: A general statement above a specific one. permit 192.168.10.0 0.0.0.255 before deny host 192.168.10.10 means the deny never matches. Put the most specific first.
Trap: Removing the ACL from the interface does not delete it. It stays configured on the router until you run no access-list. Do both: no ip access-group on the interface, then no access-list globally.
Quiz questions2
- 01
Which operator is used in an ACL statement to match packets of a specific application?
eq
- 02
Refer to the exhibit. Which two actions occur if this access list is placed inbound on a router Gigabit Ethernet port with address 192.168.10.254?
Exhibit access-list 100 permit ip host 192.168.10.1 any access-list 100 deny icmp 192.168.10.0 0.0.0.255 any echo access-list 100 permit ip any any
Telnet or SSH from 192.168.10.0 to the router is allowed, and its devices can reply to pings
14.4Named Standard IPv4 ACL Syntax
Editing an ACL by its sequence numbers instead of retyping it.
- Text editor method
- copy ACL, fix it, remove the old ACL from the router, then paste; pasting without removing only appends ACEs.
- Sequence number method
- remove one ACE quickly with no plus its sequence number; numbers show in show access-lists, not show running-config.
Trap: Typing a new line with an existing sequence number does not replace it. Run no 10 first. Also new ACEs without a number go to the bottom, which may be below a statement that already catches the traffic.
Quiz questions1
- 01
What is the quickest way to remove a single ACE from a named ACL?
Use no with the ACE's sequence number
14.5Implement ACLs
Applying ACLs to interfaces and VTY lines, and where to place standard versus extended ACLs.
- ACL guidelines
- most specific ACE first; one ACL per interface, protocol and direction; new ACEs go last; outbound ACLs skip router-generated packets.
- Applying ACLs
- ip access-group in|out on an interface; ip access-class in|out on vty lines. no ip access-group detaches; no access-list deletes it.
- ACL placement
- standard close to the destination; extended close to the source. Factors: organizational control, network bandwidth, ease of configuration.
Trap: access-group filters traffic through an interface; access-class filters remote management sessions (Telnet/SSH) on the vty lines. Using access-group on the VTY lines is a classic lost mark.
Trap: Standard near the destination, extended near the source. Swap them and a standard ACL near the source blocks that source from everywhere, not just the one destination.
Quiz questions4
- 01
Where should a standard ACL be placed?
As close to the destination as possible
- 02
Where should an extended ACL be placed?
As close to the source as possible
- 03
Refer to the exhibit. A standard ACL must stop network 1 traffic from reaching the Research and Development network. On which router interface and in which direction should it be applied?
Exhibit Network 1 — switch — Gi0/0 R1 S0/0/0 ——— S0/0/0 R2 Gi0/0 — switch — Research and Development
R2 Gi0/0 outbound
- 04
Which two statements describe appropriate general guidelines for configuring and applying ACLs?
No permit statements denies all traffic, and enter the most specific ACEs first
14.6Mitigate Attacks with ACLs
Stopping spoofing, allowing only needed services, and filtering ICMP.
- Anti-spoofing sources to deny inbound
- all zeros, broadcast, 127.0.0.0/8, APIPA 169.254.0.0/16, RFC 1918 private, multicast 224.0.0.0/4.
- ICMP to block inbound
- echo (ping sweeps, DoS floods) and redirect (alters host routing tables).
- ICMP to allow in
- echo reply, source quench, unreachable. Allow out — echo, parameter problem, packet too big (MTU discovery), source quench.
- SNMP attacks
- filter SNMP with interface ACLs, but the most effective prevention is disabling it with no snmp-server.
Quiz questions1
- 01
Which ICMP message type should be stopped inbound?
echo
14.7IPv6 ACLs
IPv6 ACLs are all named and extended, and have their own implicit rules.
- Dual stack and Teredo risks
- attackers use IPv4 to exploit IPv6 on dual-stack hosts; Teredo tunnels IPv6 in IPv4 UDP through NAT unseen by IPv4 firewalls.
- IPv6 ACLs
- no standard type (extended-like only), must be named; created with ipv6 access-list NAME, applied with ipv6 traffic-filter; any = ::/0.
- IPv6 implicit rules
- implicit deny ipv6 any any plus implicit permits for NDP nd-na and nd-ns; with an explicit deny, permit nd-na/nd-ns first.
Quiz questions3
- 01
Refer to the exhibit. Which statement describes the function of the two permit ACEs?
Exhibit permit icmp any any nd-na permit icmp any any nd-ns deny ipv6 any any
They allow IPv6 neighbor discovery traffic
- 02
What method is used to apply an IPv6 ACL to a router interface?
The ipv6 traffic-filter command
- 03
Refer to the exhibit. The IPv6 ACL Deny_WEB should let hosts on 2001:DB8:CAFE:10::/64 reach remote web servers, except PC1. Yet a user on PC1 can still reach the web server PC2. Why?
Exhibit PC1 ::A/64 — Fa0/2 S1 Fa0/1 — Fa0/1 R1 LAN 2001:DB8:CAFE:10::/64 R1 S0/1 ——— S0/1 R2 link 2001:DB8:CAFE:12::/64 PC2 ::A/64 — Fa0/2 S2 Fa0/1 — Fa0/1 R2 LAN 2001:DB8:CAFE:30::/64 R1# show ipv6 access-list IPv6 access list Deny_WEB permit tcp 2001:DB8:CAFE:10::/64 any eq www sequence 10 permit tcp 2001:DB8:CAFE:10::/64 any eq 443 sequence 20 deny tcp host 2001:DB8:CAFE:10::A any eq www sequence 30 deny tcp host 2001:DB8:CAFE:10::A any eq 443 sequence 40 R1# R1# show running-config <output omitted> interface Serial0/1 ipv6 address 2001:DB8:CAFE:12::1/64 ipv6 traffic-filter Deny_WEB outDeny_WEB permits all web traffic before PC1 is blocked
+Across the module
Points from the labs and summary that span several sections.
Trap: Packet Tracer grades named ACLs by exact name (case-sensitive) and statement order. Copy the names from the instructions and keep the order they give.
Trap: These five cost marks every time: no permit any, wrong interface or direction, subnet mask for wildcard, general above specific, access-group on VTY. Run through them before you submit.
Module 18 · 32 facts · 25 questions
Cryptography
Cryptography protects data on the wire: encryption hides it, hashes and signatures prove it was not changed and who sent it, and PKI says whose public key you are holding. The exam asks you to pick the right tool for each goal, and to do RSA and Diffie-Hellman by hand with small numbers.
- 18.1Confidentiality
- 18.2Obscuring Data
- 18.3Integrity and Authenticity
- 18.4Using Hashes
- 18.5Public Key Cryptography
- 18.6Authorities and the PKI Trust System
- 18.7Applications and Impacts of Cryptography
18.1Confidentiality
Classic ciphers, then modern symmetric and asymmetric encryption, and how keys are managed.
- Symmetric vs asymmetric
- symmetric: one pre-shared key, 40–256 bits, fast, bulk data (VPN). Asymmetric: public + private key pair, 512–4096 bits, slow.
- Symmetric algorithms
- DES (legacy, do not use), 3DES (avoid), AES (recommended; 128/192/256-bit keys), SEAL (stream), RC4 (insecure). Use 128+ bit keys.
- Asymmetric (public-key) algorithms
- the encrypting and decrypting keys differ and neither can be worked out from the other; slow, so used for signatures and key exchange.
- Asymmetric key lengths
- DH 512–4096; RSA 512–2048; DSA 512–1024; ElGamal 512–1024; elliptic curve 224+. 2048+ bits can be trusted; 1024 or less is insufficient.
- RSA
- Rivest, Shamir, Adleman; security rests on how hard it is to factor very large numbers; the first algorithm fit for both signing and encryption; used in e-commerce.
- RSA step 1: n
- pick two primes p and q, then n = p × q; n is part of both keys. Example: p = 3, q = 11, so n = 33.
- RSA step 2: φ(n)
- φ(n) = (p − 1)(q − 1), kept secret and used only to find d. Example: (3 − 1)(11 − 1) = 2 × 10 = 20.
- RSA step 3: e
- public exponent with 1 < e < φ(n) and gcd(e, φ(n)) = 1 (no shared factor). Example: e = 3, since gcd(3, 20) = 1.
- RSA step 4: d
- private exponent with e × d mod φ(n) = 1. Example: 3 × 7 = 21 and 21 mod 20 = 1, so d = 7. Public key (3, 33), private key (7, 33).
- RSA step 5: encrypt
- C = Mᵉ mod n, with M smaller than n. Example: M = 4, so 4³ = 64 and 64 mod 33 = 31. C = 31.
- RSA step 6: decrypt
- M = Cᵈ mod n. Example: 31⁷ mod 33; 31 ≡ −2, (−2)⁷ = −128, and −128 mod 33 = 4. The message 4 is back.
- RSA signing
- swap the keys: sign with the private key (S = Mᵈ mod n), verify with the public key (Sᵉ mod n = M). Only the key owner could have signed.
- Confidentiality
- public key (encrypt) + private key (decrypt). Alice encrypts with Bob's public key; only Bob's private key opens it. Leaked private key = new key pair.
- Authentication
- private key (encrypt) + public key (decrypt). Alice encrypts with her private key; anyone can decrypt with her public key, which proves she sent it.
- Diffie-Hellman (DH)
- asymmetric algorithm that lets two computers that never talked before generate an identical shared secret; the key itself is never sent.
- DH is key exchange only
- DH encrypts no data and authenticates nobody; the agreed key feeds a symmetric cipher (AES, 3DES) that encrypts the bulk traffic.
- Where DH is used
- data exchanged over an IPsec VPN (through IKE) and SSH sessions. Asymmetric protocols overall: IKE, SSL/TLS, SSH and PGP.
- DH colour analogy
- common yellow (public, not secret) + secret red or blue (private keys) = public orange and green, swapped; each mix gives the same brown (shared key).
- DH steps
- agree public prime p and base g; pick private a and b; Alice sends A = gᵃ mod p, Bob sends B = gᵇ mod p; key K = Bᵃ mod p = Aᵇ mod p.
- DH example: public values
- p = 23, g = 5, a = 6, b = 15. Alice sends A = 5⁶ mod 23 = 8; Bob sends B = 5¹⁵ mod 23 = 19. Both travel in the open.
- DH example: shared key
- Alice: 19⁶ mod 23 = 2. Bob: 8¹⁵ mod 23 = 2. Same K = 2, never transmitted; an eavesdropper sees only p, g, A and B.
- DH security
- raising a number to a power is easy, finding which power was used is hard (discrete logarithm); a 1024-bit DH number is about 309 decimal digits.
- DH groups
- 1 = 768, 2 = 1024, 5 = 1536, 14 = 2048, 15 = 3072, 16 = 4096 bits; 19, 20, 24 are elliptic curve. Higher group = more secure, slower.
- RSA vs DH
- RSA encrypts, signs and exchanges keys (factoring); DH only agrees a key (discrete log) and needs signatures or certificates against man-in-the-middle.
Trap: Confidentiality is public key encrypts, private key decrypts. Authentication is the reverse: private key encrypts, public key decrypts. Swap them and you lose the mark; encrypting with your own private key proves who you are but hides nothing, since anyone has your public key.
Trap: Use φ(n), not n, when finding d: the condition is e · d ≡ 1 (mod φ(n)). Check with the example: 3 × 7 = 21 is 1 more than a multiple of 20, not of 33.
Trap: DH encrypts nothing and authenticates nobody. An attacker in the middle can run one exchange with Alice and another with Bob and neither notices (a man-in-the-middle attack), unless the exchange is combined with signatures or certificates.
Quiz questions4
- 01
True or false? In asymmetric encryption, encryption and decryption use the same key.
False, they use different keys
- 02
What is an example of a symmetric encryption algorithm?
Advanced Encryption Standard (AES)
- 03
Which algorithm provides asymmetric encryption?
Diffie-Hellman (DH)
- 04
What is the purpose of the DH algorithm?
To generate a shared secret between two hosts that have not communicated before
18.2Obscuring Data
Masking, steganography and obfuscation: hiding data instead of encrypting it.
- Masking vs steganography
- masking swaps real data for a realistic fake (substitution, shuffling, nulling out); steganography hides data in an image, audio or text file.
Quiz questions2
- 01
What term is used to describe the technology that replaces sensitive information with a nonsensitive version?
Masking
- 02
What term is used to describe concealing data in another file such as a graphic, audio, or other text file?
Steganography
18.3Integrity and Authenticity
Hashes prove data did not change, and HMAC proves who sent it.
- Secure communications
- integrity (SHA-2/SHA-3 hashes), origin authentication (HMAC), confidentiality (encryption), non-repudiation (sender cannot deny).
- Hashes
- one-way, fixed-length digest, not reversible. MD5 128-bit and SHA-1 160-bit are legacy; SHA-2 224–512; SHA-3 is newest and most secure.
- HMAC
- a hash with a secret key as input: integrity plus origin authentication, so a man-in-the-middle cannot redo it. Used in IPsec, SSH, SSL and OSPF.
Trap: Plain hashing only catches accidental changes. Anyone can compute a hash, so an attacker in the middle can alter the message, recompute the hash and append it, and the receiver will validate against whatever hash arrives. Integrity against deliberate tampering needs a secret key (HMAC) or a signature.
Quiz questions4
- 01
Which hash function is the most secure?
SHA-3
- 02
Which statement is a feature of HMAC?
HMAC uses a secret key as input to the hash function, adding authentication to integrity assurance.
- 03
Which requirement of secure communications is ensured by the implementation of MD5 or SHA hash generating algorithms?
Integrity
- 04
What is a strength of using a hashing function?
It is a one-way function and not reversible.
18.4Using Hashes
Hashing passwords with salt, and the attacks on them.
Quiz questions2
- 01
John sends you an important data file and shares its SHA-512 hash over a secure trusted channel. Malory captures the file and modifies some records. When you compare hashes, they do not match. What does this indicate?
The file was tampered with and should not be used
- 02
An investigator finds a USB drive at a crime scene, creates a forensic image of it, and hashes both the original device and the image. What is the investigator attempting to prove when the evidence is submitted in court?
The image is an exact copy and nothing was altered by the process
18.5Public Key Cryptography
Public and private key pairs, digital signatures, and Diffie-Hellman key exchange.
- Digital signature
- the signer's private key encrypts the hash: authenticity, integrity, non-repudiation. DSS: DSA (original), RSA (common), ECDSA (newest, small).
- Code signing
- proves downloaded code is authentic from the publisher, unmodified since it left them, and undeniably published by them (non-repudiation).
Quiz questions2
- 01
What technology allows users to verify the identity of a website and to trust code that is downloaded from the Internet?
Digital signature
- 02
What two assurances does digital signing provide about code that is downloaded from the Internet? (Choose two.)
It is authentic from the publisher and unmodified since it left them
18.6Authorities and the PKI Trust System
Certificate authorities, X.509 certificates, and how PKI builds trust.
- Certificate classes
- 0 testing; 1 email; 2 organizations; 3 servers, software signing; 4 business-to-business; 5 private orgs, government (most trusted).
- Linking and revoking
- large PKIs link CAs as cross-certified or hierarchical; revoke a certificate with a CRL (polled list) or OCSP (online status query).
Quiz questions6
- 01
Which certificate is the most trusted?
Class 5
- 02
On large networks, which two methods are used to link PKI CAs?
Cross-certified and hierarchical CA topologies
- 03
Which of the following is NOT an X.509 v3 application?
LDAP
- 04
Which two methods are used to revoke a digital certificate?
OCSP and CRL
- 05
Which two statements correctly describe certificate classes used in the PKI? (Choose two.)
Class 0 is for testing; class 4 is for business-to-business transactions
- 06
Which statement describes the use of certificate classes in the PKI?
A class 5 certificate is more trustworthy than a class 4 certificate.
18.7Applications and Impacts of Cryptography
Where cryptography is used, and how it hides traffic from security monitoring.
Quiz questions5
- 01
True or false? Instant messaging can be encrypted using PKI.
True, secure instant messaging is a PKI use
- 02
True or false? PKI can be used by the enterprise to secure USB storage devices.
True, PKI can secure USB storage
- 03
What must be specified by the X.509v3 validity date range? (Choose two.)
Not before and not after
- 04
Which of the following is not a key component of the cipher suite?
Authorization algorithm
- 05
In which way does the use of HTTPS increase the security monitoring challenges within enterprise networks?
HTTPS traffic enables end-to-end encryption.