(https://www.geeksforgeeks.org/computer-networks/tcp-ip-model/)

The course is generally overviewing the concept of what a network is and the different layers that are the focal point in this section. I will give a small overview, but if you require more attention, please check out page 324

• Networking is a small community of devices that can directly (or in IoT indirectly) connect with each other
• beyond the Data Link Layer (or link layer in the text book as referenced to OSI layer 2), it is necessary to use routers
• Network layer handles routing outside of individual networks (OSI 3)
• Application layer enables individual programs to connect with each other (OSI 7)
• packet switching : method the data is grouped up with a header, which contains information about where the information is coming from and where it’s going to.

{https://www.geeksforgeeks.org/computer-networks/difference-between-segments-packets-and-frames/)

• the actual content is contained in this frame and sent over the link to the recipient indicated on the frame’s header.
• The information in the textbook switches between frames and packets. Keep in mind that Data Link layer is the frame level while packets are at the Network Layer.
• frames would contain MAC addresses and error checking information
• packets contain IP addresses, protocol information, and other control data
• information moved from the physical objects within a computer, to the internal network where mac addresses would be necessary and then out to the beyond (transport layer would do this creating segments and adding source and destination ports, sequence numbers, and checksum)
• ip link show : displays a list of all the available network interfaces and their link layer addresses as well as some other information about the maximum packet size

(https://bluecatnetworks.com/blog/ipv4-vs-ipv6-whats-the-difference/)

Private Addresses

• 10.0.0.0/8

• 172.16.0.0/12

• 192.168.0.0/16

• Network Address Translation (NAT) : router maps internal addresses to a single external IP address aka masquerading. This allows any device on the inside network to establish new connections with any global IP address on the internet

• IP4 configuration

  • DHCP -automatic configuration of the Dynamic Host Configuration Protocol
    • as devices are connected to the network, they are assigned ip addresses
    • they can be manually added via ip addr add command
  • DNS - domain name system; is a distributed phone book that maps friendly and rememberable domain names to IP addresses.

  

  • DNS resolver will find out the address that correlates to the domain and then replies with that address and the web browser tries to reach the web server at that IP address (think of google.com or amazon.com we write out their readable names rather than their IP address)
  • /etc/resol.conf is where the DNS data configurations are located
  • resolver will check /etc/hosts file to see if it contains an address for the requested name.
  • used host command to perform a lookup in DNS
  • dig command Query DNS information and provide verbose information about DNS queries and responses (records can be checked such as -A)

• ping command is used for reachability of a device. some Linux Distros do not allow for ping to work with ipv6 which requires ping6 command

• Routing is the process in which a packet gets form the source network to the destination network.

  • when establishing a connection, the device looks up the target’s IP address in its routing table. If an entry matches the address, the packet is either sent to the respective link layer network or passed on to the router indicated in the table
  • Routers use routing tables to look up received packets to determine where to send the packet to its next router.

• ip route show command lists the current IPv4 table

• As the above image provides comparisons of IPv4 vs IPv6, it is important to note that IPv6 contains enough addresses due to 128 bit per address. The first 64 bits are known as the routing prefix (site prefix) and is used by routers to determine which network a device belongs to and therefore which path the data needs to be sent on.

  • subnetting occurs within this prefix and there are 8 or 16 available internal subnetting
  • local link address via fe80 allows to request additional addresses using automatic configuration and to find other computers on the network using the Neighbor Discovery Protocol

  

  (https://docs.oracle.com/cd/E18752_01/html/816-4554/ipv6-overview-10.html)

  Above we discussed what is on here known as site prefix and the subnet ID, the interface identifier is used to name devices one by one within a network.

  If the IPv6 is assigned manually then the interface ID is set as part of the address. automatic addressing either choose randomly or are derived from the devices link layer address.

  Automatic configuration for IPv6 can be through DHCPv6 or Stateless Address Autoconfiguration (SLAAC)

  

  (https://tolumichael.com/slaac-vs-dhcpv6-a-comprehensive-analysis/)

  The above is to provide a general overview in the difference between DHCPv6 and SLAAC and by no means required for this course. The textbook provides a general commentary about how Neighbor Discovery Protocol allows devices to find each other and request information regarding the network. The information is sent by the routers and then the device may use the IPv6 prefixes contained or combine them with an interface ID of their choice as long as it is not in use.

  DHCPv6 tends to have finer control like making sure the same address is handed out to a device every time, and sending more options to the device than SLAAC. Devices need explicit consent of DHCP in order to use an address. ( I said device but technically it is a client. although when I hear the word client I think of a business perspective not a computer)

  Sockets is a communication endpoint for two programs talking to each other. over the internet that would also refer to programs running on say a web browser on a laptop or a web server.

  Types of sockets

  • Unix sockets : connect processes running on the same device
  • UDP sockets: connect applications which are fast but not resilient
  • TCP sockets: more reliable than UDB ask it confirms the receipt of data

  

  (https://www.spiceworks.com/tech/networking/articles/tcp-vs-udp/#lg=1&slide=0)

  • When making connections via TCP/UDP, target ports are standardized while source ports are random
  • ss command allows an administrator to investigate all of the sockets on a linux computer including source address, destination, and type

Types of common commands that Udemy/Yt instructors go over for Linux:

1. ifconfig
   1. IP address information
2. route
   1. Kernel IP routing table
3. netstat (-tuna)
   1. Active Internet connections (servers and established)
4. host
   1. used for finding out a IP address based on an address (like google.com)