How should I get started with DevOps & Docker?

08 Tem 18:17

If I were to compare Docker to something, I'd say it's like a "portable program box." For example, all the settings, libraries, and even Python itself required for your Python project are packaged inside Docker, so you can run it anywhere without needing to install anything else. Basically, the first thing I did was write a Dockerfile and then learn the `docker build` and `docker run` commands. For DevOps, besides Docker, it might also be helpful to look into Linux commands, CI/CD tools (like GitHub Actions or Jenkins), and maybe even Kubernetes. No matter where you start, always aim for the next step.

08 Tem 20:09

The best way to start with Docker is to first fundamentally understand what they call a "container" in your mind. Simply put, Docker packages an application along with the necessary files, libraries, and dependencies, much like a crate. When this crate is run on another Linux system, everything works the same way, eliminating environment issues. At first, I also wondered, "How does everyone run code without installing the same system?" You write a file called a Dockerfile to "build" your application, then you can push this image to Docker Hub or your own server. Following a similar logic, I started running multiple containers and setting up networks between them, which led me to microservices architecture—the foundation of DevOps. After Docker, the most natural step in your DevOps journey is to transition to Kubernetes, but first, make sure you fully grasp Docker. In my early projects, I used Docker Compose to run several services together, which taught me a lot. Don't be afraid of mistakes while learning—I also made silly errors, like copying the entire directory in my first Dockerfile with "COPY . /app." Other important things besides Docker include CI/CD pipelines (Jenkins, GitHub Actions), monitoring (Prometheus, Grafana), log management (ELK stack), and, of course, mastering Linux commands. Just like me, you might feel confused at first, but after using Docker in a few simple projects, everything starts to make sense.

08 Tem 22:30

Docker is really exciting to get into, but it can be confusing at first—don’t worry, you’re not alone, buddy. If I were to explain Docker in the simplest way possible, you can think of it as "lightweight virtual machines." For example, you're using Fedora and want to run another Linux distribution inside it, but you don’t want to set up a full VM. That’s where Docker comes in: when you need something that runs quickly while isolating resources at the OS level, Docker saves the day. The first thing you need to focus on during the learning process is really understanding container logic. To get how it works, you should practice commands like `docker run`, `docker ps`, and `docker images` until they’re second nature. At first, I was like, "What’s the point of this?" but once I started running and stopping apps with those simple commands, I saw how useful it was. Apart from Docker, some of the DevOps tools I focus on include CI/CD tools (GitHub Actions, GitLab CI), IaC (Terraform, Ansible), and monitoring (Prometheus, Grafana). As you dive deeper into this field, these technologies will start to feel like second nature. For example, the other day, when I Dockerized a Python app, I realized how practical writing a `Dockerfile` was. Being able to run my app the same way everywhere with just a few lines was a huge relief. If you start with small projects and push them to GitHub, you’ll not only get practice but also build up your portfolio. Starting small is key—like spinning up an Nginx container. After that, everything gets easier, trust me.