Study guide
2.1 Networking Fundamentals
JNCIA-Junos Juniper Networks Certified Associate
Track your progress as you read
Finished sections get a check mark.
1 Introduction & JNCIA-Junos Exam Overview
1.1 JNCIA-Junos Exam Facts 1.2 Juniper Certification Track 1.3 Exam Blueprint & Study Strategy
2 Networking & Junos OS Fundamentals
2.1 Networking Fundamentals 2.2 Junos OS Architecture 2.3 Transit Traffic & Packet Flow
3 User Interfaces & Configuration Basics
3.1 Junos CLI & J-Web 3.2 Configuration Model 3.3 Initial Configuration & Interfaces
4 Operational Monitoring & Maintenance
4.1 Monitoring with Operational Commands 4.2 Logging & Traceoptions 4.3 Device Maintenance
5 Routing Fundamentals
5.1 Routing & Forwarding Tables 5.2 Static & Default Routing 5.3 Route Preference & Dynamic Routing Overview
6 Routing Policy, Firewall Filters & Final Review
6.1 Routing Policy 6.2 Stateless Firewall Filters 6.3 High-Yield Review & Test Strategy
2.1 Networking Fundamentals
Key Takeaways Explain This Topic AI
The OSI model has 7 layers and the TCP/IP model has 4; both describe how data is encapsulated and delivered, but TCP/IP is the practical model real networks run on
A switch forwards Layer 2 Ethernet frames using MAC addresses; a router forwards Layer 3 packets using IP addresses and the longest-prefix-match rule
Each switch port is its own collision domain, but all ports in a VLAN share one broadcast domain; routers break up broadcast domains
IPv4 uses 32-bit addresses written as dotted-decimal; IPv6 uses 128-bit addresses written as eight hextets, removing the need for NAT scarcity workarounds
TCP is connection-oriented and reliable with handshake, sequencing, and retransmission; UDP is connectionless, lightweight, and used for latency-sensitive traffic
Last updated: May 2026
Why Networking Fundamentals Matter for JNCIA-Junos
Quick Answer: JNCIA-Junos is a Junos OS exam, but it assumes you already understand general networking. The Networking Fundamentals objective tests the OSI (Open Systems Interconnection) model, the TCP/IP (Transmission Control Protocol/Internet Protocol) model, switching versus routing, collision and broadcast domains, IPv4 and IPv6 addressing, subnetting, and TCP versus UDP (User Datagram Protocol). Candidates who already work in networking often lose points here on small details, not on Junos.
Juniper publishes Networking Fundamentals as one of seven objective domains. It is foundational: if you cannot reason about a subnet boundary or which device breaks a broadcast domain, you will struggle later when routing tables, longest-match selection, and firewall filters build on these ideas. Treat this section as the layer everything else stands on.
The OSI and TCP/IP Models
The OSI model is a 7-layer conceptual reference for how data moves across a network. The TCP/IP model is the 4-layer model that real protocols actually implement. Junos, like every modern network OS, operates on the TCP/IP stack, but the OSI vocabulary ("a Layer 2 problem", "a Layer 3 device") is used constantly in practice.
- OSI Layer OSI Name TCP/IP Layer Example Protocols / Units
- 7 Application Application HTTP, SSH, DNS, SNMP
- 6 Presentation Application TLS, encoding, encryption
- 5 Session Application session setup/teardown
- 4 Transport Transport TCP, UDP (segments/datagrams)
- 3 Network Internet IPv4, IPv6, ICMP (packets)
- 2 Data Link Link Ethernet, ARP (frames)
- 1 Physical Link cables, optics, electrical/optical signals
Encapsulation is the key idea: application data is wrapped in a transport header, then an IP header, then an Ethernet header as it moves down the stack, and unwrapped in reverse on the receiving side. The unit name changes per layer — segment (Transport), packet (Network), frame (Data Link).
Switching vs Routing
A switch is a Layer 2 device. It learns source MAC (Media Access Control) addresses, builds a forwarding table, and forwards Ethernet frames within the same broadcast domain. A switch does not change the IP header.
A router is a Layer 3 device. It forwards IP packets between different networks by consulting a routing table and applying the longest-prefix-match rule: when multiple routes could reach a destination, the route with the most specific (longest) matching prefix wins. This rule is central to Junos route selection later in the exam. Host A ---- Switch ---- Router ---- Switch ---- Host B
(same VLAN/L2) (L3 hop) (same VLAN/L2)
Same subnet, same VLAN: traffic is switched (Layer 2).
Different subnets: traffic is routed (Layer 3), and the default gateway is the router.
Collision and Broadcast Domains
These two terms are tested directly and are easy to confuse.
Concept Definition Boundary device
Collision domain A network segment where frames can collide on shared media Each switch port is its own collision domain
Broadcast domain The set of devices that receive a Layer 2 broadcast frame A router (or a VLAN boundary) ends a broadcast domain
Key facts to memorize:
A modern switch gives every port its own collision domain, so full-duplex switched links are effectively collision-free.
All ports in the same VLAN (Virtual LAN) form one broadcast domain.
A router does not forward broadcasts , so each router interface bounds a broadcast domain.
An 8-port switch with no VLANs = 8 collision domains, 1 broadcast domain.
IPv4 and IPv6 Addressing
IPv4 addresses are 32 bits, written as four dotted-decimal octets (for example 192.0.2.10 ). The prefix length (such as /24 ) defines how many leading bits identify the network. Address types include unicast, broadcast, and multicast. RFC 1918 private ranges are 10.0.0.0/8 , 172.16.0.0/12 , and 192.168.0.0/16 .
IPv6 addresses are 128 bits, written as eight 16-bit hextets in hexadecimal (for example 2001:db8::1 ). Leading zeros can be omitted and one run of all-zero hextets can be compressed to :: . IPv6 has no broadcast; it uses unicast, multicast, and anycast, and link-local addresses ( fe80::/10 ) on every interface. The vastly larger address space removes the IPv4 scarcity that drove NAT (Network Address Translation) workarounds.
Subnetting Essentials
Subnetting divides one IP network into smaller networks by borrowing host bits for the network portion. For IPv4, remember:
Usable hosts per IPv4 subnet = 2^(32 − prefix) − 2 (subtract network and broadcast addresses).
/24 = 256 addresses, 254 usable. /30 = 4 addresses, 2 usable (common for point-to-point links). /31 is also used for point-to-point links in modern designs.
A subnet mask in binary is a run of 1s (network) followed by 0s (host); /26 = 255.255.255.192 .
Be fluent converting decimal to binary for the four mask boundary values: 128, 192, 224, 240, 248, 252, 254, 255. The exam expects quick prefix-to-mask math.
TCP vs UDP
Both are Transport-layer protocols, but they behave very differently.
- Property TCP UDP
- Connection Connection-oriented (3-way handshake) Connectionless
- Reliability Reliable: sequencing, ACKs, retransmission Best-effort, no retransmission
- Ordering Guaranteed in-order delivery No ordering guarantee
- Overhead Higher (larger header, state) Lower (8-byte header)
- Typical use HTTP/HTTPS, SSH, file transfer DNS queries, VoIP, streaming, SNMP traps
The exam may describe a traffic type and ask which protocol fits. Latency-sensitive, loss-tolerant traffic implies UDP; must-arrive, in-order traffic implies TCP.