Study guide
OSI and TCP/IP Models
CCST Networking Study Guide Exam
Track your progress as you read
Finished sections get a check mark.
1 Orientation, Official Sources, and Exam Strategy
Official Exam Profile and Source Hierarchy Registration, Delivery, Attempts, and Results Using the Six Official Topic Areas Building a Practical Study Plan Exam Strategy Without Unsupported Shortcuts Ethical Preparation and Path Forward
2 Network Standards, Models, and Common Protocols
Network Building Blocks OSI and TCP/IP Models Frames, Packets, Segments, and Data Flow Network Types and Scope Cloud, On-Premises, Protocols, and Ports Bandwidth, Throughput, Latency, and Jitter
3 Chapter 3: IPv4, IPv6 Addressing, and Subnet Formats
3.1 Public, Private, and Local Addressing 3.2 IPv4 Dotted Decimal and Binary Basics 3.3 Subnet Masks, CIDR, and Prefix Lengths 3.4 Subnet Logic, Host Ranges, and Gateways 3.5 IPv6 Address and Prefix Formats 3.6 MAC vs IP in Addressing Troubleshooting
4 Endpoints, Media, Cabling, and Client Connectivity
Endpoint Devices and Network Roles Network Interface Cards and Link Status Copper, Fiber, Coax, and Connector Basics Wired, Wi-Fi, and Cellular Access Methods Interference, Signal Quality, and Physical Reliability Client Connectivity Checks by Operating System
5 Infrastructure Devices, Ports, and Physical Installation
Infrastructure Device Roles Ports, Connectors, and Device Identification Cisco Device LEDs and Status Checks Network Diagrams and Cabling Instructions Power over Ethernet and Powered Devices Racks, Patch Panels, and Installation Basics
6 Switching, Routing, and Network Services
Basic Switching and MAC Learning VLANs and Local Boundaries Routing, Default Gateway, and ARP DNS, DHCP, and NAT Services Transport-Layer Services Small Network Traffic Flow
7 Troubleshooting Methodology, Ticketing, and Documentation
Help Desk Intake and Information Gathering Scope, Impact, and Prioritization Reproduce, Isolate, and Escalate Ticket Quality and Operational Documentation User Communication and Change History Physical, Data Link, and Network Layer Triage
8 Diagnostic Commands, Packet Capture, and Device Data
Command-Line Triage and Ping Path and Client Configuration Commands DNS and Socket Diagnostics Wireshark Capture and Save Device Access and Data Collection Basic Cisco Show Commands
9 Security, Firewalls, Wireless, and Safe Defaults
Firewall Filtering and Traffic Policy Least Privilege, AAA, Passwords, and MFA Wireless Security, WPA2, WPA3, and WPAx Guest Networks, Segmentation, and Shared Devices Common Threats and Safe User Practices Secure Home Router Baseline
10 Integrated Small Network Build and Repair Labs
Lab 1: Build a Basic Office LAN Lab 2: Repair No Internet After an Office Move Lab 3: Wireless Clients Cannot Reach Local Resources Lab 4: DHCP, DNS, and Default Gateway Symptoms Lab 5: Switch, Router, and Service Boundaries Lab 6: Escalation, Documentation, and Handoff
11 CCST to CCNA Career Path and Operational Readiness
Entry-Level Networking Roles and Expectations CCST vs CCNA and the Cisco Roadmap NetAcad 70-Hour Path and Study Plan Lab Habits for Real Operational Readiness Ethical Preparation, Portfolio, and Logbook Exam Scheduling Calendar and Readiness Check
12 Final Review, Exam Day, and Remediation Plan
Final Review Map to Official Objectives 50-Minute Exam Day Timing Unsupported Facts and Ethical Preparation Retake Planning Before the First Attempt Post-Fail Remediation Workflow Post-Pass Next Steps and Validity
OSI and TCP/IP Models
Key Takeaways Explain This Topic AI
The OSI model has seven layers used as a shared vocabulary for functions and troubleshooting boundaries.
The TCP/IP model collapses real Internet communication into four layers: network access, internet, transport, and application.
Encapsulation adds a header at each layer going down the stack; decapsulation strips them going up at the receiver.
Layered troubleshooting isolates whether a fault is physical, local-link, addressing, transport, name-resolution, or application related.
Last updated: June 2026
Why Models Matter
Models are not cables or commands; they are organized descriptions of what must happen for communication to work. The CCST Networking exam expects you to recognize both the seven-layer Open Systems Interconnection (OSI) reference model and the four-layer TCP/IP model, and to use them to localize faults. Both break a complicated event, like loading a secure web page, into smaller functions you can test independently.
The Seven OSI Layers
Memorize the layers in order, top to bottom or bottom to top, along with the keyword each one owns.
- # OSI layer Owns TCP/IP equivalent
- 7 Application User-facing services (HTTP, DNS, SSH) Application
- 6 Presentation Formatting, encryption, encoding Application
- 5 Session Conversation setup/teardown Application
- 4 Transport TCP/UDP, ports, reliability Transport
- 3 Network IP addressing, routing Internet
- 2 Data Link Frames, MAC, switching, VLANs Network access
- 1 Physical Cable, fiber, radio, voltage, link Network access
A common mnemonic bottom-up is "Please Do Not Throw Sausage Pizza Away." Layer 1 (Physical) covers connectors, pinouts, light, voltage, and link presence, so an unplugged cable, broken fiber, or weak signal starts here. Layer 2 (Data Link) handles local delivery with Ethernet/Wi-Fi frames and MAC addresses. Layer 3 (Network) handles logical addressing and routing: IPv4, IPv6, masks, prefixes, and default gateways. Layer 4 (Transport) provides process-to-process delivery using TCP or UDP port numbers ; TCP adds connection setup, sequencing, acknowledgments, and retransmission, while UDP stays lightweight.
Layers 5 through 7 are often discussed together at entry level. Layer 7 (Application) is where DNS, DHCP, HTTP/HTTPS, SMTP, IMAP, SSH, and SNMP operate; Layer 6 (Presentation) handles encryption and data representation; Layer 5 (Session) manages conversations.
The TCP/IP Model and Encapsulation
The TCP/IP model matches how real protocol suites are grouped. Network access combines OSI Layers 1-2 (Ethernet, Wi-Fi). Internet maps to IP addressing and routing. Transport maps to TCP and UDP. Application holds everything users touch.
Encapsulation adds layer-specific headers as data moves down the stack before transmission, and the protocol data unit (PDU) name changes at each step:
Application produces data .
Transport adds a TCP/UDP header to make a segment (or datagram for UDP).
Network adds an IP header to make a packet .
Data Link adds the frame header/trailer to make a frame .
Physical transmits bits as electrical, optical, or radio signals.
At the receiver, decapsulation strips each header in reverse so the application gets usable data.
Layered Troubleshooting in Practice
The layers turn vague tickets into an ordered checklist. If a desktop has no link light, testing DNS first wastes time, because Layer 1 is already suspect. If link is good and the host has a valid IP but cannot leave its subnet, focus on Layer 3 (gateway, routing). If a ping to an IP address succeeds but the same site fails by name , DNS at Layer 7 is the likely culprit. If DNS resolves but only one application fails, suspect port filtering, server availability, certificate trust, or the app itself.
Common trap: the exam loves the ping-by-IP-works-but-name-fails scenario; the answer is name resolution, not the cable. For CCST work, memorize each layer's purpose, learn the four-layer grouping, and practice mapping symptoms to layers. The point is never reciting a model in isolation; it is using the model to say where traffic is failing and what evidence supports your next move.
Mapping Tools and Symptoms to Layers
The real value of the models on the job is that each common tool and symptom lines up with a layer, so you can move methodically instead of randomly. A blank link light, a port stuck at 10 Mbps half-duplex, or a snapped fiber strand are Layer 1 problems. A switch port in the wrong VLAN, a MAC-address table that never learns the host, or a Wi-Fi client that cannot associate are Layer 2 problems. A wrong subnet mask, a missing or incorrect default gateway, or an unreachable remote network are Layer 3 problems. A blocked port, a refused connection, or a half-open TCP session are Layer 4 problems.
A failed name lookup, an expired certificate, or a server returning an error page are upper-layer (5-7) problems.
Two habits make this disciplined. Work bottom-up when the symptom is total loss of connectivity: confirm link and physical first, then addressing, then routing, then the application, because a low-layer fault makes every higher test meaningless. Work top-down when one specific application fails while others work: the lower layers are clearly fine, so start near the application. A technician who can say "ping to the gateway succeeds, ping to the remote IP succeeds, but name resolution fails, so this is a Layer 7 DNS issue" has produced exactly the kind of evidence the next support tier needs.
Finally, remember that the OSI model is a reference model and the TCP/IP model is the one the Internet actually implements; encapsulation and decapsulation happen the same way in both. Knowing the PDU name at each layer, segment, packet, frame, bits, lets you read a packet capture and instantly say which layer's header you are inspecting, which is the bridge into the next section.