M6: Switches & Routers

M6 CISCO: Switches and Routers. Describe Cisco Switches and Routers.

Switch

Layer 2 device. It connects devices within the same LAN using MAC addresses. It forwards frames based on MAC addresses. It builds a MAC address table. It creates separate collision domains. It does not route between networks.

Router

Layer 3 device. It connects multiple networks using IP addresses. It forwards packets based on IP addresses. It performs routing, NAT, DHCP, ACLs. It is required for WAN connectivity.

Access Point

Is an essential LAN device that provides wireless connectivity. It is used in enterprise deployments for mobility. Often centrally managed.

Firewall

Is an essential LAN device that protect the network. It enforce security policies. It provides threat detection and containment. Often sit at the network edge.

Home router

It is an all-in-one device for small networks.

Uplink port

Is a switch port that is used to connect to another switch or router. It prevents bottlenecks by giving the switch a fast path to the rest of the network.

Access-layer connectivity

It is the point where end devices are physically connected to the network. It refers to Switch ports that connect to end devices.

Major Physical Components of Switches

1. LAN Access Ports

2. Uplink Ports

3. Status LEDs

4. Console Ports

5. Storage Port

LAN Access Ports

They connect end devices. They support PoE/UPoE. It can power devices directly over the Ethernet cable.

Status LEDs

Lights on switches that provide quick visual feedback. Flashing green = activity; amber = problem; off = no link.

Console Port

It is a port use for initial configuration and local management. If a switch has no IP address yet, you must use this port.

Storage Port

This port is not used for normal network traffic. It is used for transferring IOS images, backing up configurations, and loading files during recovery.

2 Ports use for device management

1. Console port

2. Storage port

Frame Forwarding Methods

1. Store and Forward Switching

2. Cut-through Switching

Store-and-Forwarding Switching

Frame forwarding method were the switch receives the entire frame. It performs a CRC (Cyclic Redundancy Check) to detect errors. If the CRC is valid, the switch forwards the frame. If the CRC is invalid, the switch drops the frame.

Cut-Through Switching

Frame forwarding method were the switch begins forwarding the frame as soon as it reads the destination MAC address. It does not wait for the entire frame. It does not check the CRC.

First 6 bytes of the Ethernet Frame

It is where the destination MAC address is located.

2 Variants of Cut-Through Switching

1. Fast-Forward Switching

2. Fragment-Free Switching

Fast-Forward Switching

It is pure Cut-Through type of switching. It works by forwarding the frame immediately after reading the destination MAC. Lowest possible latency. It does not check for errors. Latency measured from first bit in → first bit out. This is the default cut-through method.

Fragment-Free Switching

It is a modified Cut-Through switching. It works were a switch stores the first 64 bytes of the frame. Then forwards the frame. Checks for errors only in the first 64 bytes. It is a compromise between speed and reliability.

Memory Buffering on Switches

Switches sometimes need to store frames temporarily before forwarding them. This happens when: The destination port is busy, The switch must handle traffic bursts, and Ports operate at different speeds (asymmetric switching).

Two Main Buffering Methods

1. Port-based memory buffering

2. Shared memory buffering

Port-Based memory buffering

Memory buffering were each port has its own dedicated queue. Frames arriving on a port are stored in that port’s buffer. Frames leave the buffer in order, one after another.

Shared Memory Buffering

Memory buffering were all frames go into a common memory pool shared by all ports. Memory is allocated dynamically based on need. Frames are linked to their destination port without moving between queues. It supports asymmetric switching.

2 Fundamental Settings of Switch Ports

1. Speed (bandwidth)

2. Duplex mode

Full Duplex

Duplex mode were both ends can send and receive simultaneously. No collisions. It is required for Gigabit Ethernet and above.

Half Duplex

Duplex mode were only one side can send at a time. Collisions can occur. Used only on legacy hubs or misconfigured links.

Actual Link Speed

It is the highest common speed supported by both devices.

Autonegotiation

It allows two devices to automatically choose the highest common speed and the best duplex mode.

Duplex Mismatch

A common real-world problem which occurs when one side is full-duplex and the other side is half-duplex. The full duplex sends anytime it wants while the half-duplex device must wait for a clear link. It results in collisions.

Straight-Through Cable

Cable that connects unlike devices. Switch → host, switch → router

Crossover Cable

Cable that connects like devices. Switch → switch, router → router, host → host. router → host

Auto-MDIX

It’s an automatic cable detection. Modern switches support this. It automatically detects whether the cable is straight-through or crossover. It automatically adjusts the port’s transmit/receive pairs. You do not have to worry about cable type - the switch handles it if this is enabled.

3 Basic Steps to Power up Switch

1. Check the Components

2. Connect the Cables

3. Power up the Switch

2 Critical Files loads when Cisco Device boots

1. The IOS image file

2. The startup configuration file

IOS Image file

Is the software that controls: switching and routing functions, hardware operation, CLI commands, network protocols, and security features. It is stored in Flash memory.

Startup-Configuration File

It contains hostname, interface configurations, VLANs, IP addressing, Passwords, Security protocols, and any saved configuration commands. It is stored in NVRAM. It determines how the device behaves after reboot.

Running-configuration File

Is the active configuration stored in RAM. Any changes made in the CLI take effect immediately, but they are not saved unless you manually save them.

Boot Process Summary

1. Loads the IOS image from Flash → RAM

2. Loads the startup-config from NVRAM → RAM

3. Creates the running-config

4. Begins normal operation

Cisco Routers

Regardless of size, model, or purpose - are essentially computers designed for networking. It requires: OS, CPU, RAM, ROM, NVRAM, Flash memory.

CPU

A router’s component. Its purpose is to execute the IOS instructions, it handles routing decisions, it manages switching functions, it performs system initialization, and it controls internal processes. It is the brain of the router.

IOS

A router’s component that provides: routing and switching functions, security features, CLI interface, protocol support and device management.

RAM

A router’s component that stores the running-config, stores the IOS during operation, holds routing tables, ARP tables, buffers, and act as a temporary working memory. It is volatile and the contents are lost when the router reboots.

ROM

A router’s component that stores the bootstrap program, runs POST (Power-On Self-Test), provides ROMMON mode for recovery, and contains a minimal IOS version for emergencies. It is non-volatile and cannot be erased during normal operation.

NVRAM

A router’s component that stores the startup-config. It retains configuration even when powered off. It is non-volatile, so it keeps the configuration across reboots.

Flash Memory

A router’s component that stores the IOS image file, stores system files, backups, and additional IOS versions. It is non-volatile storage. It is similar to a computer’s SSD.

Router’s Key Interface Ports

1. Console Ports

2. LAN Interfaces

3. Network Interface Module Slots

4. USB Port

5. Management Interface

6. Auxiliary Port

Console Ports

A router’s interface used for initial configuration. Two types are RJ-45 and USB Type-B. It is for local, out-of-band management. Used when the router has no IP address or is misconfigured.

RJ-45 Console Port

It is one of the two console ports that has traditional console connection. It is used with a rollover cable, and it provides CLI access for setup and troubleshooting.

USB Type-B Console Port

It is a console port which is a modern alternative to RJ-45. It is easier for laptops without serial ports. It performs the same function as the RJ-45 console.

LAN Interfaces

It is the router’s interface that is used for connecting LAN switches, connecting to WAN circuits (Ethernet handoff), and routing between networks. It is also known as Gigabit Ethernet Ports. GE 0/0/0 supports two physical media types: RJ-45 copper and SFP.

NIM Slots

A router’s interface that provide modularity and allow the router to support: serial interfaces, DSL connections, additional ethernet ports, switch modules, wireless modules, T1/E1 circuits, and Specialized WAN technologies. It allows the router to adapt to different network environments without replacing the entire device.

USB Port

A router’s interface that is used for file transfers. It can store IOS images, configs, or logs. It is useful for upgrades and recovery.

MGMT Port

A router’s interface that is dedicated out-of-band management. It is used when GE interfaces are down or misconfigured. It provides remote access without affecting production traffic.

AUX Port

A router’s interface which is dedicated as legacy port for connecting a dial-up modem. Used for remote CLI access when network connectivity is unavailable. Rarely used in modern network.

3 Interface that can manage router

1. AUX

2. MGMT

3. Console

Routers Management Ports

1. Console Port

2. SSH

3. AUX Port

Out-of-band Management

One of two method of accessing the CLI of a router. It does not require network connectivity.

In-band Management

One of two method of accessing the CLI of a router. It requires network connectivity.

Console Port

Router management port which is an Out-of-Band Management. This port is used for initial configuration. It is used when the router has no IP address. It is used when the router is misconfigured or inaccessible over the network.

SSH

Router management port which is an In-Band management . This port provides remote CLI access over the network. Encrypted and secure. The router must have active network interface. It has configured IP address. A default gateway if remote. Configured Domain name, RSA keys, vty lines.

AUX Port

Router management port which is a Legacy Remote Access. It provides remote CLI access using a dial-up modem. It is used when network connectivity is unavailable. Is historically used for out-of-band access in remote sites.

5 Major Stages of Router Boot Process

1. POST

2. Bootstrap Program Loads

3. Locate and Load the IOS Image

4. Locate and Load the Startup Configuration

5. Interfaces Initialize & Routing Begins

Power-On Self-Test

The first stage of Router boot process. The router powers on. The CPU runs diagnostics stored in ROM. Hardware components are tested.

Bootstrap Program Loads

The 2nd stage of Router boot process. This is loaded from ROM. It locates the IOS image, it load the IOS into RAM. It reads the configuration register. It decide where to boot from (Flash, TFTP, ROMMON)

Bootstrap

It is a traffic controller that decides how the router boots.

Locate and Load the IOS Image

The third stage of Router boot process. The bootstrap now looks for the IOS image. Search order: Flash memory, TFTP server, and then ROMMON mini-IOS.

Locate and Load the Startup Configuration

The 4th stage of Router boot process. Once IOS is running, the router loads the startup-configuration from NVRAM.

Interfaces Initialize & Routing Begins

The 5th stage of Router boot process. After loading the configuration, Interfaces are enabled; IP addresses applied, routing protocols start; and routing begins forwarding packets.