DNS and HTTP: the application layer

At the top of the stack, programs talk in application protocols. Two of them are behind every web page you open.

DNS: names to addresses

People remember example.com; packets need 93.184.216.34. The Domain Name System is a distributed database of names, spread over millions of servers:

1. Your computer asks a resolver (usually run by your ISP, or a public one like 1.1.1.1).
2. If the resolver does not know the answer, it asks a root server: "who handles .com?"
3. The .com TLD server answers: "ask the servers for example.com".
4. Those authoritative servers give the address.

Every answer has a TTL (time to live), and the resolver caches it that long, so most questions never reach the root at all.

authoritative = {"example.com": ("93.184.216.34", 300), "codebase.dev": ("76.76.21.21", 60)}
cache = {}                          # name -> (address, expires at)

def resolve(name, now):
    if name in cache and cache[name][1] > now:
        return cache[name][0] + " (cached)"
    address, ttl = authoritative[name]       # root -> TLD -> authoritative
    cache[name] = (address, now + ttl)
    return address + " (asked the servers)"

for now, name in [(0, "example.com"), (10, "example.com"), (100, "codebase.dev"), (200, "codebase.dev")]:
    print("t =", now, name, "->", resolve(name, now))

DNS mostly uses UDP port 53: one small question, one small answer.

HTTP: what browsers and servers say

HTTP is text (in versions 1.x): a request line, header lines, a blank line, and an optional body. The server answers with a status line, headers, a blank line and the page.

request = "GET /about.html HTTP/1.1\r\nHost: example.com\r\nAccept: text/html\r\nCookie: theme=dark\r\n\r\n"

head, body = request.split("\r\n\r\n", 1)
lines = head.split("\r\n")
method, path, version = lines[0].split(" ")
headers = {}
for line in lines[1:]:
    name, value = line.split(": ", 1)
    headers[name] = value

print("method:", method, " path:", path, " version:", version)
print("headers:", headers)
print("body:", repr(body))

Methods say what to do: GET reads, POST sends data (a form), PUT and DELETE change and remove. HTTP is stateless: each request stands alone, so websites use cookies (a header the browser sends back every time) to remember who you are.

HTTPS is HTTP inside TLS on port 443: the connection is encrypted, and the server proves who it is with a certificate. Today almost every site uses it. HTTP/2 and HTTP/3 keep the same requests and answers but send them in binary, many at once on one connection (HTTP/3 over UDP).

Mark lesson complete