Popping the Web Stack

Humans are a species that continuously evolve. A while ago, the sole source of written information was the libraries. And by libraries I mean those huge buildings stacked with long shelves filled with books. To access any piece of information, you had to visit the library in person, look for your resource in a catalog, identify the right location of the resource from the many shelves and finally begin to look for information from your resource. Do not get it twisted, I am a fan of libraries. My point is this very process that I have mentioned necessitated one of the most brilliant inventions of humans, The internet.

The Internet

The internet is a global network of interconnected devices around the world that work hand in hand to share information. These devices majorly include computers as the majority together with servers, routers, and switches which are majorly network devices. In other words, the internet is a community of computers.
One lovely thing about the Internet is that it is not owned by a single entity, making it free for anyone with an Internet connection. Isn't lovely? Nonetheless, every community has to have rules on how everyone engages and this is reflected on the internet too. The Internet is decentralized among a network of organizations including Internet service providers (ISPs), content creators, governments, and several agency bodies. All these organizations are tasked with regulations and making the internet safe.

The web

A lot of people think that the web and the internet are the same, but they are not. The web (or World Wide Web) is a subset of the internet that consists of globalized webpages, images, and videos that can be accessed from a web browser.
Then what is the difference between the two? The internet is the infrastructure upon which computers share information, while the web is an application or service that uses the internet to make hypertext documents accessible.

How does the web work?

The web works on a client-server model. This model differentiates the tasks between the client and the server. In this case, the server is the service or resource provider, while the client is the service or resource requestor.
This model is widely used because the data is centralized making it easy to maintain. Also, the capacities of the clients and servers are independent which brings flexibility to the model. Remember this makes it possible for you to access the internet using any device like a phone, tablet, or laptop which vary in terms of specifications.
This model forms the architecture which is the basis of our question, what happens when you type https://www.google.com in your browser and press Enter? It is the foundation of how the web works.

What happens when you type https://www.google.com in your browser and press Enter?

1. Type https://www.google.com on your web browser's address bar

Remember that the main application you can access the internet with from your device is the web browser. A web browser is just a software application that runs on your operating system, that is used to access, view and interact with websites. There are numerous web browsers which include google chrome, Mozilla Firefox, Apple's Safari, and Microsoft Edge. Of course, we had internet explorer back in time which ceased to exist.

The URL

What is keyed into the browser's address bar is known as the uniform resource locator (URL). We now understand that the internet is regulated and standardized, hence the use of URLs. A URL is a standardized way of specifying the location of a resource on the internet. It has the following format
protocol://host/path?query
The URL https://www.google.com does not include the path and query section of a standard URL, but here is its breakdown:

• Protocol - Specifies the communication protocol to be used to access the resource. It can be HTTP, HTTPS, FTP among other protocols. The protocol for this URL is HTTPS.
• Host - Specifies the domain name for the website. A domain name can be further split into the sub-domain, domain, and top-level domain:

1. www: This is the subdomain of the domain name google.com. The www subdomain is a common convention for web pages, but it is not strictly necessary for accessing the site.
2. Google: This is the primary domain name of the site, which identifies the organization or entity that owns it.
3. .com: This is the top-level domain (TLD) of the site, which indicates the type of organization or domain category. In this case, it is a commercial organization, as indicated by the "com" TLD. The rest of the sections of the URL can be described as:

• Path - Specifies the location of the resource
• Query - Specifies additional parameters used to specify the resource being searched.

2. DNS lookup

Computers only understand bits, that is 1s and 0s. Due to this reason, URLs need to be translated to the appropriate IP addresses of the respective host. This is done by DNS servers.

What is DNS?

Domain name system is the protocol that translates readable URLs into IP addresses. It is commonly referred to as the

phonebook
of the internet. Humans are poor at remembering numbers. This is why you have a phonebook that keeps names against cell numbers in your phone.
DNS is capable of doing IP mapping since it keeps records of hosts' domain names and respective IP addresses. These records include A, CNAME, MX, NS, SOA, and TXT records.

• An (Address) - Used to point a domain name to the associated IP address.
• CNAME (Canonical Name) - Used for creating aliases of domain names.
• MX (Mail Exchanger) - Used to deliver email to a specified address.
• NS (Name Server) - Used to specify an authoritative name server for a given host.
• SOA (Start of Authority) - Used to determine key information about a DNS zone.
• TXT - Used to verify domain ownership and hold SPF (Sender Policy Framework) data.

What is an IP address?

Remember that computers only understand bits. An IP address consists of a series of numbers separated by dots or colons. It is a unique identifier assigned to every computer in a network. There are two versions of IP addresses, IPv4 and IPv6. IPv4 is 32 bits long with the digits separated by dots, while IPv6 is 128 bits long with colons as the separators.

How is the IP address resolved (DNS lookup)

To resolve or get an IP address for google.com, a forward DNS lookup is done. DNS data is heavily cached, therefore, DNS lookup is first done locally. Let's say you've entered the URL using the Chrome browser, and a recursive query is done. The browser, which in this case is the DNS client, looks up the address from its browser cache, if it does not resolve it, it looks up in the operating system cache.
If the address is not resolved from the local DNS server, the local DNS server queries other DNS servers until a name resolution is found.

3. TCP connection

Once the server's IP address is resolved, the browser initiates a TCP connection. The Transmission Control Protocol (TCP) is a transport layer protocol according to the OSI model which is responsible for how network devices exchange data. Data is transmitted over the internet as packets. TCP is a reliable protocol, meaning any data packets lost or duplicated in transit are detected and corrected. TCP works on top of Internet Protocol (IP) that is responsible for routing data packets to the right location address.
TCP connection is established through a three-way handshake.
First, the client browser sends a SYN packet or segment to the server.
If the server has received the segment, it agrees to the connection by sending the SYN-ACK packet back to the client.
The client acknowledges the receipt of the SYN-ACK packet by sending its own ACK packet. At the same time, the client can already begin transmitting data packets to the server.

4. HTTP Request.

When we request a resource from Google, the client browser sends an HTTP request to the web server. HTTP stands for the Hypertext Transfer Protocol. It is used to request resources from the web.
HTTP requests sent to the web server are a combination of the following:
Request methods - This indicates the HTTP method used to fetch the resources. There are several methods like GET, HEAD, POST, PUT, DELETE, CONNECT, OPTIONS, and TRACE. GET is used to retrieve data from the web server.
Request URL - This is the URL of the resource requested. Remember URLs carry the path to the resource requested.
Request headers - This is the additional information such as the user-agent, which is the browser sending the HTTP requests.
Request body - This is optional in an HTTP request and is used for methods that send data to the web server like forms data.

HTTP with 's' (HTTPS)

HTTPS stands for Hypertext Transfer Protocol Secure. Up to this stage, we have explored HTTP requests that do not have a security mechanism within them. Dynamic websites are interactive and more than often it requires users or clients to send data in forms for services such as authentication. Let's say you want to log in to your favorite blog post, you have to key in your password. When HTTP protocol is used there is vulnerability to your data in this case password and username to anyone sniffing over your connection.
This has led to the use of a more secure protocol HTTPS. This protocol encrypts data as it is being exchanged between the client and the web server, therefore, providing secure communication. It achieves this by the use of SSL (Secure Sockets Layer) or TLS (Transport Layer Security) protocols. SSL/TLS uses public-key cryptography for its encryption. Public-key cryptography relies on two keys: a public key and a private key.
When a client connects to an HTTPS website, the website provides a public key to the client. This public key is also normally referred to as SSL or TSL certificate depending on the protocol in use. The client uses this public key to encrypt a symmetric encryption key and send it over to the web server. After the server receives it, it uses its private key to decrypt the symmetric encryption key. Once both the browser and the web server have the symmetric encryption key, they use it to encrypt data as they are transmitted. When the transmission is done, the symmetric encryption key is discarded.

5. Processing the requests and serving a response

Processing of HTTP requests is done by a web server. A web server is a software that runs on a physical server that serves web pages. It hosts website files that include HTML documents, CSS stylesheets, JavaScript files, images, and videos. Normally the web stack includes a database, which can be running on the same physical server as the web server or on a different server. The database is used to store user information for dynamic websites.
To fully process a request, the web server undergoes this series of steps before sending back a response.

• Parsing the request: The server first parses the incoming request to determine what action it is being asked to perform and what resources are being requested.
• Routing the request: Once the server has parsed the request, it determines which application or service should handle the request. This is typically done using a routing table or configuration file.
• Authenticating the request: Depending on the security requirements of the application or service handling the request, the server may need to authenticate the client before processing the request. This may involve validating credentials or checking permissions.
• Handling the request: The server then passes the request to the appropriate application or service to handle the requested resource. This may involve executing code, querying a database, or serving a static file.
• Generating a response: Once the requested resource has been processed, the server generates a response and sends it back to the client. The response typically includes an HTTP status code, headers, and a message body.
• Closing the connection: Finally, the server closes the connection with the client. Depending on the server configuration, the connection may be kept open for a certain period for subsequent requests or closed immediately.

6. Rendering the content

Rendering of responses refers to the process of converting a server's response into a visual representation that can be displayed to the user. When a user requests a server via their browser, the server sends a response containing HTML, CSS, and JavaScript files.

The browser then reads the response and begins to render the content. It starts by parsing the HTML and building the Document Object Model (DOM) tree, which represents the structure of the web page. Next, it parses the CSS to determine how the content should be styled, and it applies these styles to the appropriate elements in the DOM tree. Finally, it executes any JavaScript code included in the response to add interactivity and modify the content dynamically.

Monitoring and Managing traffic

So far, we have looked at the cycle that happens from when you enter a URL up to the point you receive a rendered page. This is a single instance of how that happens. The internet has made the whole world a village, in that sense so many people across the globe interact with the internet. It is a normal scenario to have numerous requests for the same resource. One funny thing about software engineering or engineering as a whole is that you design for a worst-case scenario. Yes, worst-case scenarios, remember Big O notations for algorithms? This is the same when it comes to web development. While designing the web infrastructure, you have to assume you will have numerous if not many concurrent requests to the same resource, and also someone will try and access your resources without authorization and maybe with malicious intent.

Load Balancing

Some of the world's leading giant corporates are web-based. Talk of Meta, Twitter, and LinkedIn among others. Have you ever wondered how you get your Twitter feed refreshed in a matter of seconds? How you can always have access to your feed at any time of the day? When do they maintain their systems? Then a load balancer is your answer.
A load balancer is a traffic manager that manages incoming traffic to a pool of servers. This is applicable when a distributed server system is used, where we have different web servers with the same resources and services distributed in different machine servers. Load balancing is a technique used in web infrastructure to distribute incoming network traffic across multiple servers. The main objective of load balancing is to prevent any single server from being overloaded with traffic, thereby ensuring the high availability and reliability of the web application.
When a user sends a request to access a web application, the request first goes to the load balancer, which examines the request and decides which server in the pool is best suited to handle the request. The load balancer can use various algorithms to make this decision, such as round-robin, least connections, IP hash, or other custom algorithms. Once the load balancer has selected a server, it forwards the request to that server, which then processes the request and sends the response back to the user through the load balancer. The load balancer can also monitor the health and performance of the servers in the pool and redirect traffic away from any servers that are experiencing issues or are overloaded. This is also important in server predictive maintenance which helps in preventing server breakdowns before they break.

Firewalls

Firewalls are used to provide security in networks and are used in the web infrastructure to protect an application from unauthorized access or malicious attacks. Remember that a lot of sensitive data like personal information, health records, and organization secrets among others is stored or shared across the web. This makes Firewalls a really important component in the web stack to ensure data does not fall into the wrong hands. A firewall network security system monitors both incoming and outgoing traffic on predetermined security rules.
In a web infrastructure, firewalls are typically placed between the internet and the web servers to provide an additional layer of security. The firewall acts as a barrier between the internet and the web servers, filtering and blocking incoming traffic that does not comply with the predefined security rules.