Case Introduction

You have been assigned on a project as a build engineer. Project is a solution for embedded systems. Your team members are strong C++ developers who are focused on development of the solution. At the moment they are having difficulties with integration of developed modules. Team has to waste a lot of time on manual activities like checking code style, building, testing and so on. Team decided to try the Continuous Integration (CI) approach and to automate some everyday activities. Your team isn't familiar with CI and they have asked you to prepare proposal which should include documentation and minimum viable product (MVP).

Tasks

1. You have to prepare development environment which will be based on docker technology and include the next instances:

• Jenkins master;
• Jenkins Agent;
• SonarQube*;
• Artifactory
• Git (In this case you don't need to run additional container with git, just use github)

NOTE: Environment is configured and doesn't require additional manual steps from user. (exclude the seed job)

Configuration details:

• Authorization is disabled. If it's impossible to disable authentication use admin user with admin password. Admin is usually available for all services;
• All necessary plugins are installed;
• С/С++ rules are configured in SonarQube*;
• Quality Gates for C/C++ project are configured in SonarQube**;
• I can specify the number of Jenkins agents***;
• On-demand Jenkins agents****.

2. Create Jenkins items.

Seed item:

• Configure the seed job manually. Seed job is Jenkins item which is based on JobDSL plugin;
• Prepare script which will create this via Jenkins API*;
• The seed job is created automatically (using either shell script or Makefile) and available after running Jenkins. In case of using Makefile the job should be created by calling make seed-job command*;
• The seed job should monitor the repository, take DSL files from ./job folder, generate Jenkins items and execute one after generation**.

Pipeline item: Create DSL declaration for Jenkins pipeline item which will work with Jenkinsfile from a repository. Jenkins pipeline DSL declaration.

Finally you will get two Jenkins items:

• seed item;
• pipeline item.

3. Automate build process for a project.

• Find C/C++ project at Github (check a license of a project before you start using one). For example, you can take the curl;
• Implement declarative or scripted pipeline for you project. This pipeline should contain the following steps: -- checkout project repository -- build -- execute unit tests -- execute sonar-scanner and push result into SonarQube -- prepare build artifact -- check quality gates in SonarQube; fail job if quality gates are not passed -- push the artifact into artifactory;
• build results and links to sonar and artifactory should be available on build dashboard in Jenkins.

4. Prepare documentation.

All documentation for your project has to be put into README.md file and written in markdown

Acceptance criteria:

• File of docker-compose is prepared and located in git repository;
• The environment is launched via docker-compose. Environment is available and doesn't need additional manual configuration;
• Seed job creation is well-documented in the README.md file. (seed job is available*);
• Seed job after its run creates Pipeline job and triggers it;
• Pipeline job takes project from github and builds, tests, checks and etc;
• Project analysis results are displayed on SonarQube dashboard;
• There's a binary artifact available in Artifactory;
• Build results and links to SonarQube and Artifactory are available on build dashboard in Jenkins;
• Documentation is available.

Jenkins auto build with creating seed job, and reading Jenkinsfile

Description of installing and running a project.

1. Clone a git repository use a command: "git clone https://github.com/AndreyNovik1993/ci_test-task.git"
2. Install docker and docker-compose to your enviroment
3. Create folder for an artifactory /opt/artifactory/ with perrmission 777.
4. Run command "docker-compose -f docker-compose.yml up -d --build" in the directory where a project was cloned
5. After will be complited docker-compose command wait several minutes until all docker enviroment start
6. If you have to auto create seed_job will run script ./add_job.sh and approve a script on the jenkins configure (it has to do manually)
7. if you have to run jobs are configured in the jenkins automaticly, will connect to jenkins http://localost:8080 and run these jobs manually.

P.S If you have problems with running pipeline on the stage (SonarQube), please will wait more time and restart script ./add_job.sh again. Warning!!! Run tasks in the docker containers on the host or VM with memory > 4Gb.

If you have to see resoults a static code analizy in the sonarqube will visit page http://localhost:9000, artifactory will visit a page http://localhost:8081.

All jobs are configured to use jenkins _as_a_code plugin. Pipelines are described on the Declarative language.