CompSci-Notes/Computer Network Notes at master · theCodeBear/CompSci-Notes · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
Computer Network Notes


Let's break down a Uniform Resource Locator (URL) into its parts:

    URL:                http://www.toddkronenberg.com/SocialFit/index.html
    Hostname:           www.toddkronenberg.com
    File Location:      SocialFit/index.html


Some terms to know:
    IP - Internet Protocol - the addressing protocol of all addresses on the internet
    ISP - Internet Service Provider - the company/organization that provides your internet service
    DNS - Domain Name System - network of name servers that can query each other to find
                               the location of any host's IP address on the internet


How a computer connects to a website:
    You type a URL into a browser and the browser attempts to make a connection with the
    server where the document you are looking for lives. It must find the network location
    of the host (aka: network host, host machine). The corresponding location is an IP
    address. But to get the IP address of the host you are looking your computer must query
    a program called a name server. This name server is most likely in your ISP's network,
    so your machine contacts your ISP's name server for the IP address of the host you are
    looking for. The name server will find this out by querying other name servers in a
    worldwide network called the Domain Name System (DNS). The task of finding the IP
    address of a hostname is called "resolving hostnames". Once the information gets back
    to your nameserver, so that you have the IP address, your machine will go to that IP.

Domain Name System (DNS):
    The DNS is the whole network of programs and databases that cooperate together to
    translate hostnames to IP addresses for any system on the internet that is trying to
    locate an address.
    It is made up of DNS servers (nameservers).
    Internet hostnames are composed of parts separated by dots. A domain is a collection
    of machines that share a common name suffix. Domains can live as subdomains inside of
    other domains.
    Each domain is defined by an authoratative name server that knows the IP addresses of
    the other machines in the domain. The authoritative, or primary, name server may have
    backups in case it goes down, which are referred to as secondary name servers or
    secondary DNS. The secondaries usually refresh their information from the primaries
    every few hours to keep up to date.
    The key is that nameservers for a domain don't have to konw the locations of all the
    machines in other domaines, including their own subdomains. They only have to know the
    location of the nameservers. The whole network is a tree, with several root DNS servers.
    All machines on the internet know the addresses of the root servers. The root servers
    know the addresses of the servers on the next level down: the top-level domains, which
    are things like .org, .com, etc. Each top-level domain server knows where the nameservers
    for the domains directly beneath it are, and etc on down until there is a DNS server
    that knows the IP address for a particular host.
    The Domain Name System has been designed so that each machine can get by with a minimum
    amount of knowledge and use the system to get locations of web hosts instead of having
    each nameserver contain an enormous amount of information. Local changes to subtrees
    in the system can be made by just changing one authoritative server's database of the
    name-to-IP address mapping.

    So to locate www.toddkronenberg.com on the internet, a computer would as its authoritative
    nameserver to find the host name's location. That nameserver would query a root server
    to find hostname, that root server would query the hostname to the .com top-level domain
    server, that server would query the toddkronenberg.com domain server with the host name,
    and that name server would look in its database to find the IP address of the host name.
    Notice that a dot in the URL separates each level in the Domain Name System.

    Once your authoritative nameserver resolves a hostname to its IP address it caches that
    information for some time period so that it doesn't have to go throught the whole Domain
    Name System when people continually go to the same address. Nameservers do a lot of
    caching to save time and resources. It is important that nameservers only cache IP
    mappings for a time because addresses change and the nameservers need to keep up to date
    and not have old and invalid information cached. The cached IP address for a site is
    also deleted if the host is unreachable when a computer tries to connect to it.

Packets and Routers:
    A browser sends commands to hostnames as packets. A packet is a block of bits that has
    three important things: the source IP address (of your machine), the destination IP
    address (of the target machine) and the port number to use (80 is standard for HTTP
    that indicates it is a world wide web request).
    The packet is sent to your router, the router has a map of the internet in its memory
    that completely describes the local network neighborhood so that it can send the packet
    to routers on neighboring networks. A packet will travel through several routers on its
    way to its destination. Routers keep track of how long it takes for other routers to
    acknowledge having received a packet, and use this information to direct traffic over
    the fastest links in the network. They also use this knowledge to know when a router
    has dropped off the network. So the internet is a decentralized, distributed network
    of tons of routers all around the world, so failures tend to be very localized and the
    network can route around them.

    Once your packet reaches its destination, that machine uses the port number to feed the
    packet to the web server. The web server replies to the packet's source address. The
    web server returns the document to the source address as a number of packets.

TCP / IP:
    TCP/IP are the main protocols of the internet. TCP (Transmission Control Protocol)
    exists on the Transport Layer of the OSI model. IP (Internet Protocol) exists on the
    Network/Internet Layer of the OSI model.
    The lower level, IP, is responsible for labeling individual packets with the source
    address and destination address of the two computers exchanging information over a
    network.
    The upper level, TCP, gives you reliability. When two computer negotiate an TCP
    connection, using IP, the receiver knows to send acknowledgements of the packets it
    sees back to the sender. If the sender doesn't see an acknowledgement for a packet within
    some time period, it resends the packet. This way there is no loss of data (unlike
    another protocol: UDP). The sender also gives each TCP packet a sequence number that
    the receiver can use to reassemble the packet in case they show up out of order, which
    can often happen). TCP/IP packets also contain a checksum to enable detection of data
    corruption.

HTTP:
    HTTP (HyperText Transfer Protocol) is the main protocol at the Application layer of
    the OSI network model. The Application layer is the layer on which web browsers and
    servers speak to each other. This runs on top of the TCP/IP protocol. The application
    layer protocol, HTTP, is a way to pass strings of yes back and forth. Two commands
    that use HTTP are GET and POST. So the packets of data are sent through TCP/IP but the
    actual information in those packets is text sent through HTTP. A typical group of
    HTTP headers for a response to a request for a webpage will look something like this:
        HTTP/1.1 200 OK
        Date: Tue, 24 Jul 2014 14:27:48 GMT
        Server: Apache/1.2.6 Red Hat
        Last-Modified: Wed, 22 Apr 2013 04:49:11 GMT
        Content-Length: 4928
        Content-Type: text/html
    The headers shown directly above would be followed by a blank line and then the text
    of the web page. After all this information is sent the connection is dropped. The
    browser displays the page; the headers tell it how to display the page.


OSI Layer Model:
    The OSI (Open Systems Interconnection) is a conceptual model that standardizes the
    communication system between machines on the internet. There is a 7 layer model and a
    4 layer model, which merely merges some of the 7 layers.

    OSI 7 Layer Model:                      OSI 4 Layer Model:

        Application Layer                       Application Layer
        Presentation Layer
        Session Layer
        Transport Layer                         Tranport Layer
        Network Layer                           Internet Layer
        Datalink Layer                          Network Layer
        Physical Layer