# Week 4 & 5

During this period, I was formulating user stories that aligned with my program objectives. Simultaneously, I've been dedicating time to studying security resources and configuration packages.

Over the past two weeks, I initially considered automating the packaging process through a pipeline. However, upon further reflection in subsequent weeks and following a call, I've realized that this approach is not suitable. It seems excessive for the current task at hand and diverts from the optimal path forward. Instead, I've come to understand that focusing on configuration and service packages should be the primary area of attention.

I discovered reprepro which makes repository management much more easier.

## User stories

**As a** node runner
**I want to** use apt-get to dowload the software
**So I can** save time and cognitive resources

**As a** node admin
**I want** to be offered defaults or configurations
**So I can** simplify the admin tasks

**As a** developer
**I want to** distribute my software much more easily
**So I can** focus on much more important tasks related to the software I'm developing

**As a** community member
**I want to** know where to point people who want to run a node
**So I can** say running a node is easy

**As a** security auditor
**I want to** run tests or audits against the packages and settings
**So I can** advise for best practices or flag security vulneberalities or fix security holes by modifying the package (debian)

## Automation of package creation (deprecated)

Automated process of creating debian packages
Per one package

1. Get the latest release
curl -L \
-H "Accept: application/vnd.github+json" \
-H "Authorization: Bearer <YOUR-TOKEN>" \
-H "X-GitHub-Api-Version: 2022-11-28" \
https://api.github.com/repos/OWNER/REPO/releases/latest
2. Compare if latest we already have
3. Package it if needed (minimal for now), add debian dir if needed (patch it)
4. Test it (skip for now)
5. inform me that there is a new package that needs to be released
upload it to the package server ?? maybe this step should be manual
to push into the testing packages

## Reprepro (path forward)

1. **Set Up `reprepro`**:

Install `reprepro` on your local machine or a server where you plan to manage your Debian repository. You can typically install it using your package manager (e.g., `apt` on Debian/Ubuntu).

2. **Create a Debian Repository**:

Set up the directory structure for your repository on your local machine or server. You'll need directories for "conf," "dists," "incoming," and "pool."

3. **Configure `reprepro`**:

Inside the "conf" directory, create a file named "distributions" with the configuration for your repository. This file defines repository metadata such as components, architectures, and signing information. Here's a simplified example:

   Codename: yourcodename
   Components: main
   Architectures: amd64

Adjust the settings based on your repository's specifics.

4. **Import Packages**:

Copy or move your Debian package files (`.deb` files) to the "incoming" directory. Then, use the `reprepro` command to process the incoming packages and add them to the repository. For example:

   reprepro -Vb /path/to/repository/incoming includedeb yourcodename /path/to/package/package.deb

5. **Generate Metadata**:

Generate repository metadata using the `reprepro` command:

   reprepro -Vb /path/to/repository export

8. **Access the Repository**:

## Resources
[Mastering Linux Security and Hardening: A practical guide to protecting your Linux system from cyber attacks, 3rd Edition](https://www.amazon.com/Mastering-Linux-Security-Hardening-protecting/dp/1837630518)