J'ai procastiné Docker pendant un mois... Voici ce qui m'a débloquée

In early July, the author embarked on a DevOps mentorship with Endah. Initially motivated and resolved, she found herself unproductive from July 21st to August 22nd as she delved into the documentation, architectural diagrams, and the reasoning behind Docker's existence. The turning point came when she stopped seeking theoretical understanding and embraced hands-on learning.

The author recounts a pivotal moment where she created a container named "boxeur," an Alpine-based image that ran a simple Bash script and logged its activities. This project allowed her to understand Docker's fundamental concepts - images, containers, communication between them, and data management - step by step. By coding, launching, and observing errors, she progressed from passive learning to actively constructing her comprehension of Docker.

One of the challenges she faced was a 403 Forbidden error during her first Git push. To tackle this, she learned how to generate a Personal Access Token on GitHub, gaining a deeper understanding of GitHub's authentication process. Another challenge involved logs not appearing in the terminal, as her script wrote logs inside the container, not to standard output. She resolved this by reading logs directly from within the container using 'docker exec.'

The author also highlights the importance of using the correct package manager for each base image. For example, trying to install Nginx with 'apt-get' in Alpine led to an immediate failure. She learned that each base image has its own package manager, an essential distinction when working with container images.

Docker Hub's naming conventions for repositories were another lesson learned. The author discovered that Docker Hub does not accept uppercase letters in repository names, a crucial detail that led to a straightforward fix once understood. The project provided practical insight into Docker's core concepts, such as the distinction between images and containers, Docker Compose as an orchestration tool, volumes for data persistence, and the impact of choosing the right base image on build and deployment speed.

The author emphasizes that Endah's mentorship played a crucial role in helping her transition from passive learning to active, hands-on understanding.