Comptia Network+ (N10-008) course notes

Introduction

And another one in the box. Learned a lot of new concepts and it was also a great was to refresh my networking knowledge. This course is given by Keith Barker on CBT nuggets. The purpose of this document is to summarize some of the key learnings I got from

OSI Model

Physical: copper, fibre, wifi, bluetooth, DSL, optical cable, coax cable
Data Link: MAC address found with ARP
Network: IP address
Transport: Protocols include UDP, TCP, GRE
Session: In the TCP/IP stack it is part of the Application layer 7
Presentation: In the TCP/IP stack it is part of the Application layer 7
Application

Network Topologies

Bus: coax cable connecting computers at 10Mb/s. 1 Frequency and max 200 meters. 1 device can speak at a time by sending a broadcast
Hub: Layer 1 device that repears the frames. It has the same disadvantage as a Bus because there is 1 collision domain. Each computer can either send or receive. Max speed should be 0.5 duplex speed
L2 Switch: smarter than a hub. One computer on the switch can communicate with another computer in the domain and there is no collision with other frames.

Network Types

Peer to peer: emule, napster, uTorrent Client to server: web server Personal area network: NFC, bluetooth, infrared Campus area network: 3 tier architecture Metropolitan area network

cables and connectors

RJ45 has two standard: T568A(represented as A in the diagram below) and T568B(represented as B in the diagram below)

PC – L2 SW A – A

PC – L2 SW B – B

PC – PC A – B

When pin 1 on one side goes to pin 1 on the other side we call it a straight through cable. Otherwise it’s a cross over cable

Internet protocol

3 public ipv4 addresses defined in the RFC 1918

10.0.0.0 – 10.255.255.255 /8 172.16.0.0 – 172.31.255.255 /12 192.168.0.0 – 192.168.255.255 /16

private ipv4 addresses are either handed out by DHCP or they are configured manually on the machine. To make private ip addresses accessible over the internet you’ll either have to :

Do port address translation
Do network address translation NAT. You define a destination nat ip address and a service e.g. counter strike server
Use a reverse proxy

Source NAT typically occurs when browsing the internet as a private ipv4 address makes its way to the internet. This generates a NAT table on the router so that it can properly route the response back to the user. Load balancers also do SNAT and swap internal addresses so that we make sure the replies go through him. The inital flow is the one that counts for us to distinguish SNAT and DNAT

Unicast

DNS is an example of unicast request. You receive an ip from your dns request and there is only one recipient for that DNS request

Multicast

is a method of communication where data is sent from one sender to multiple specific recipients simultaneously. Examples include streaming, online gaming and multi cast DNS. mDNS is useful to resolve IP address from hostnames when DNS is not working.

Broadcast

sent to everyone. ARP is an example. It is typically sent to the 255.255.255.255 address

Anycast

Multiple servers (usually in different geographic locations) are assigned the same IP address. e.g. google’s DNS 8.8.8.8

Subnet exercise 1

Give me the bit version of an ipv4 address 192.168.1.1

128 64 32 16 8 4 2 1

1 1 0 0 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1

answer: 11000000 10101000 00000001 00000001

Subnet exercise 2

I want 4 new subnets from my 10.1.0.0/16, what are their network addresses?

2^1 = 2 not enough 2^2 = 4 not enough 2^3 = 6 is enough

16+3= /19

Number of host bits: 32 - 19 = 13 bits Number of possible IPs = 2¹³ = 8,192 addresses Usable hosts = 8,192 - 2 = 8,190 hosts

8 192 / 256 = 32 blocks

10.1.0.0/19 10.1.32.0/19 10.1.32.1 - 10.1.32.254 10.1.64.0/19 10.1.128.0/19

0+256 = 255 so why is the last address at 10.1.32.254? Because 255 is used for the broadcast address

Subnet exercise 3

I have 150 hosts, what subnet do I need?

2^8 = 256 256-2 = 254 32-8 = 23

192.168.0.0/23

Subnet exercise 4

How many hosts can I fit in a /27 network?

32-27 = 5 bits for the host 2^5 = 32 32-2 = 30 hosts -2 because we need one address for the broadcast and the other for the network

ipv6

addresses that start with 2 xxx or 3 xxx are global unicast addresses Link local addresses start with FE80 Multicast address groups start with FF There is no DHCP and no ARP just neighbor sollicitation and neighbor adverstisements

Application level services

SMTP over StartTLS 587 IMAP over StartTLS 143 SMTP over SSL(TLS) 465 IMAP over SSL(TLS) 993 DHCP: discover, offer, request, acknowledgment the DHCP process goes through a three way handshake: syn, syn/ack, ack It is usually a local service but you can use a DHCP relay in case there is no DHCP server on the LAN

Network architectures

Access layer has L2 switches

3tier arch

ip routing

There are 3 ways routers learn about networks

By being connected directly
Static route: either default route or actual routing
dynamic routing protocol: depending on your switch

Distance vector method: RIP
Link state: OSPF
IGP: interior gateway protocol
RIP

uses udp/520. I’m not sure how to configure this in the unifi envronment that we have but you basically advertise the networks that you can reach so that other routers know how to reach it as well.

routers send multicast to this address 224.0.0.9

OSPF

uses its own protocol #89. routers send multicast to this address 224.0.0.5 or 224.0.0.6

switching

connecting a bunch of computers to a switch does not create a network by itself. You’ll need a router to hand out IP addresses. In a 5 port L2 switch all of the broadcasts are sent to each physical interface. Each link is a full duplex connection which means we have 5 routes/collision domains.

ARP requests in this case are sent as a broadcast to everyone. PC1 for instance could advertise that it is looking for a mac address and a router that holds all of the mac addresses will then reply to that request.

Virtual Lans

VLANs are layer two broadcast domains that use an 802.1Q tag (header) to signify to your router that the layer 2 frame belongs to another network.

There are two primary ways to set up VLANs depending on whether the segmentation is handled at the switch level or the host (OS/network interface)

VLANs on Switches (Port-Based or Tagged VLANs)

Access Ports: Assigned to a single VLAN. The device connected to the port is unaware of the VLAN. All frames are untagged.
Trunk Ports: Carry traffic for multiple VLANs. Frames are tagged using 802.1Q so the receiving device knows the VLAN ID.

How it works:

VLANs are defined on the switch.
Ports are assigned to VLANs.
Trunk ports connect to other switches or routers, carrying traffic for multiple VLANs.

VLANs on Host Interfaces (Software/OS-Level VLAN Tagging)

Overview: The host machine (Linux, Windows, etc.) is configured to tag network traffic itself using virtual interfaces.

In Linux, this is often done using tools like ip, vconfig, nmcli, or /etc/network/interfaces.
Each VLAN gets a virtual interface (e.g., eth0.10 for VLAN 10).

How it works:

The host sends/receives tagged frames directly on a trunk port.
The switch port the host connects to must be a trunk port and allow the required VLANs.

Use case: Servers or VMs that need to be on multiple VLANs without needing separate physical NICs.

Spanning tree protocol

STP uses bridge id to decide which routes have priority and block parallel paths and avoid broadcast storms/loops You’ll need to setup root guard/or bpdu filter to prevent hackers from claiming they have a new switch that has lower bridge port ids. There are other software security controls you can put in place to avoid users being able to trick an access port to become a trunk port or to block rogue DHCP servers.

QoS

Quality of service is a tag you set on a switch. Once the switch is under heavy load it will start to look at this tag to make forwarding decisions.

Redundancy

you can have a virtual ip address that acts as a load balancer for your gateway. Your windows client thinks he’s talking to one server but it’s actually a load balancer. This can be achieved with FHRP First Hop Redundancy Protocol or VRRP (virtual router protocol)

Security

skipped

Troubleshooting

when troubleshooting connectivity problems start by:

Start by pinging the default gateway if that works, ping the address you’re trying to reach.

ping by default on linux uses udp but you can use a flag to use icmp packets instead.

Is the network path going through a VPN tunnel? Or an SDN? In that case I would verify the MTU on those network adapters and make sure the values match along the tunnel.