IP Addressing

Question 1

What is the 1st octet range, default subnet mask and max hosts per network of classes A to C?

Answer 1

Class A: Range: 1 - 127, Subnet: 255.0.0.0, Hosts: 16.7M
Class B: Range: 128 - 191, Subnet: 255.255.0.0, Hosts: 65,536
Class C: Range: 192 - 223, Subnet: 255.255.255.0, Hosts 256

Question 2

What is special about Class D addresses and whats the range of the 1st octet?

Answer 2

Range: 224 - 239
They arent used to represent specific hosts but they are used for multi casting, meaning they represent a group of hosts, if a packet is sent to them, its multicasted to this whole group.

Question 3

What is special about Class E addresses, whats the range of the 1st octet and how many hosts include ALL those addresses?

Answer 3

Range: 240 - 255, Hosts: 268M
They are “reserved” and cannot be used in any production environments.

Question 4

What are the Class A to C private IP ranges and the max hosts per network?
Like stated in RFC 1918

Answer 4

Class A: 10.0.0.0 - 10.255.255.255, Max: 16.7M
Class B: 172.16.0.0 - 172.31.255.255, Max: 1.05M
Class C: 192.168.0.0 - 192.168.255.255, Max: 65,536

Question 5

What is the Automatic Private IP Addresses (APIPA) range and when is it used?

Answer 5

Its used by the OS when the networks DHCP server is unavailable and the machines doesnt have a static IP address assigned yet.
Range: 169.254.0.0 - 169.254.255.255

Question 6

Whats the loopback (localhost) address and theoretical range and what its for?

Answer 6

Its for internal testing of the devices network protocols.
Localhost: 127.0.0.1
Theoretical Range: 127.0.0.1 - 127.255.255.255

Question 7

What does CIDR stand for and whats it for? And whats CIDR notation?

Answer 7

Classless Inter-Domain Routing:
Its used to borrow bits from the host portion of an IP address for the network portion to create smaller subnets.
CIDR Notation: The number of bits assigned to the network portion are written behind the IP address with a / like 127.0.0.1/8

Question 8

What are the types of IPv4 data flows?

Answer 8

Unicast: from a single source to a single destination
Multicast: from a single source to multiple specific destinations
Broadcast: from a single source to all devices on a destination network

Question 9

What are the 4 methods for IP assignment?

Answer 9

• BOOTP (Legacy)
• DHCP (Standard today, modern BOOTP replacement)
• APIPA
• ZeroConf (Newer based on APIPA)

Question 10

What are the 4 settings, DHCP assigns?

Answer 10

1. IP Address
2. Subnet Mask
3. Default Gateway
4. DNS/WINS Server

Question 11

What is the crucial problem with APIPA configured devices?

Answer 11

APIPA assigned devices cannot communicate outside the LAN (no Internet) and cannot communicate with non-APIPA devices in the LAN

Question 12

What are alternative implementations of ZeroConf and what are its features?

Answer 12

It has the basic APIPA features and additionally:
- Resolve computer names to IPs without DNS using mDNS
- Service discovery on a network
Apple: Bonjour
Windows: Link-Local Multicast Name Resolution (LLMNR)
Linux: SystemD

Question 13

A Class C network for example has 256 possible addreses, why are only 254 hosts possible? (Applies for all Networks)

Answer 13

Network Address: If all host bits are 0, for example the Class C address 192.168.1.0, this IP address represents the network itself

Broadcast Address: If all host bits are 1, for example the Class C address 192.168.1.255, this IP address is used to send to all hosts in that network

Question 14

What are the number of possible subnets and number of IPs per subnet for the CIDR notations /24 to /30

Answer 14

Number of usable IPs is always IPs - 2, because of Network & Boradcast ID

/24: Subnets: 1; IPs: 256
/25: Subnets: 2; IPs: 128
/26: Subnets: 4; IPs: 64
/27: Subnets: 8; IPs: 32
/28: Subnets: 16; IPs: 16
/29: Subnets: 32; IPs: 8
/30: Subnets: 64; IPs: 4

For /23 downwards its again Subnets X 2 for each step, but now IPs is also IPs X 2 instead of IPs / 2, Example:

/23: Subnets: 2; IPs: 512
/22: Subnets: 4; IPs: 1024
/21: Subnets: 8; IPs: 2048 etc.

Question 15

What is Variable Length Subnet Mask? What Routing protocols support VLSM?

Answer 15

Variable-Length Subnet Mask:
Allows subnets of various sizes. Instead of 4x 64 IPs one can be 64 IPs, one split by 2 into 2x 32 IPs etc. Esentially its subnetting of subnets.
Routing Protocols supporting VLSM:
- RIP
- OSPF
- IS-IS
- EIGRP
- BGP

Question 16

What rules exist to shorten IPv6 addresses?

Answer 16

If an octet only contains 0s (0000) it can be first shorted to one 0. Multiple octets with 0, can be shortened to a double : (::) once.

So: 1234:0000:0000:0000:3423:3847:AB30
= 1234:0:0:0:3423:3847:AB30
= 1234::3423:3847:AB30

Question 17

What IPv6 Address types exist? What are they for?

Answer 17

• Unicast (Used to identify a single Interface)
• MultiCast (Used to identify a set of interfaces, beginds with FF)
• AnyCast (Used to identify a set of interfaces so packets can be sent to any member of the set)

Question 18

What is special about IPv6 in terms of network interfaces compared to IPv4?

Answer 18

A single network interface, like a network card can be assigned multiple different IPv6 addresses.

Question 19

What types of IPv6 Unicast addresses exist and how to identify them?

Answer 19

1. Globally routed: They are like public IPv4 addresses and they always have the 1st octet between 2000 to 3999
2. Link-Local / Local Use: Like private IPv4 addresses and they always have the 1st octet being FE80

Question 20

How are Link-Local IPv6 addresses are assigned?

Answer 20

Through Stateless Address Autoconfiguration (SLAAC), which doesnt need to obtain addresses or other info from a central server like DHCP does.

Question 21

How does SLAAC work?

Answer 21

1. Take the MAC address of the network card.
2. Split OUI and NIC of the MAC and insert “FFFE” between them, this gives us a 64bit address called EUI
3. Then SLAAC reads the network part of the IPv6 address of the network the device is on, which is also 64bit and puts it in front of the EUI
4. Done! We now have a working 128bit IPv6 address

Question 22

What is NDP and how does it work?

Answer 22

Neighbour Discovery Protocol (NDP):
It learns the Layer2 (MAC) Addresses of other devices on the network through Router Solicitation/Advertisement and Neighbour Solicitation/Advertisement.
It can also redirect traffic to another router if its faster.

Question 23

Whats the difference between Broadcasting in IPv4 and Anycasting in IPv6?

Answer 23

Anycast is a way to assign the same IP address to multiple devices like for example DNS servers. When a request is then send to this address, the protocol automatically routes the request to the nearest server with that IP and not to all of them.

Question 24

What is Dual Stack?

Answer 24

It allows for IPv4 and IPv6 to coexist on the same network for gradual migration. Favors IPv6 by default, falling back to IPv4 if 6 is unavailable.

Question 25

What is Tunneling (in IPv4/IPv6 Architecture)?

Answer 25

Tunneling allows IPv6 packets to get encapsulated so they can traverse through IPv4 infrastructure.

Question 26

What is NAT64?

Answer 26

An address translation mechanism allowing IPv6 only devices to speak to IPv4 devices. Crucial for environments where DualStack is not feasible.