Maintainer: Hoeghh

Important : STACI repository will be deleted, as it is no longer being developed on. Please see our ASK project instead. https://github.com/Praqma/ask

Support Tracking And Continuous Integration

STACI lets you run Atlassian products with easy command line scripts. It uses Docker to create the environment and adds features for integrating the services.

STACI consists of:

• Jira
• Confluence
• Bamboo
• Bitbucket Server
• Crowd
• Crucible
• MySQL

This repository is maintained by www.praqma.com

Quick Start

STACI uses properties files to configure the stack. You get started by following these steps:

• Create configuration files
  • Copy the .template files in the conf folder to files with just a .properties extension
  • Edit your .properties files
  • If you are not using Linux, you probably want to set provider_type:virtualbox in staci.properties
• Start docker
• Run ./staci.sh install to install and start the stack

Commands

• ./staci.sh install : Create Docker host and images, and start Docker containers for the chosen Atlassian tools
• ./staci.sh stop : Stop the Docker containers
• ./staci.sh start : Start existing Docker containers
• ./staci.sh delete : Delete the containers

There are currently no STACI commands to manage the Docker host.

Access Services

The install command creates a SystemInfo.html in the root directory. It contains links to all the services started.

Requirements

• Docker
  • Docker version 1.10
  • Docker-compose version 1.6.0
  • Docker-machine 0.6.0
• Bash
  • The STACI scripts are written in bash and have been tested on Linux and Mac
• VirtualBox (optional)
  • On a Mac, an easy way to get started is to use Docker with VirtualBox
• Ubuntu Server 16.x
  • Download the latest iso from: https://www.ubuntu.com/download/server and install/spin up VM
  • Install all the prerequesits for Staci:
    • clone this repo and run bin/install_on_ubuntu_server_16.x.sh. It installs all the needed tools and setting described above
  • Follow the instructions..

Preparing Volume folder (only used when running locally)

• Create a folder structure in /data/atlassian where a user with uid 1000 is owner and has full read-write access.
• Create a jira, atlassiandb, confluence, bamboo, crowd, crucible, bitbucket and backup folder inside /data/atlassian with same rights.
• You can specify another folder in ./conf/staci.properties instead of /data/atlassian (look for volume_dir=/data)

Getting started

Turn off SELinux or AppAmour, if applicable
Create data directory (See Preparation section)
pull the repository (git clone https://github.com/Praqma/staci.git)
cd staci
cp conf/staci.properties.template conf/staci.properties
vim conf/staci.properties
./staci.sh install

If you want to change the behaviour of STACI, edit the file ./conf/staci.properties

Taking backup of local containers (not working with external providers yet)

The containers have consistent data in /data/atlassian/ (default, edit staci.properties). You can take a backup by executing the script ./bin/backup.sh. This will tar-gz the volumes to /data/atlassian/backup/[date].

Continuous delivery pipeline

The pipeline folder contains an implementation of the pragmatic workflow for this project: A Jenkins setup which uses the Pretested Integration plugin to merge changes from ready branches into master.

It is run like this:

• Start Jenkins in Docker using the scripts in pipeline/docker
• Manually add GitHub credentials in Jenkins, with the ID github
• Watch the seed job run once

Notice that the generated job has no trigger. You need to manually build it to pick up changes on a ready branch.

A pragmatic workflow

Praqma has a workflow that encourages feature branching and a continuous delivery pipeline. It is described in this blog post: http://www.praqma.com/stories/a-pragmatic-workflow/

Once you have it set up with ghi and the git aliases, here is how you can work on an issue:

• ghi open - Create an issue
• ghi list - List open issues to see the issue number you will fix, say 48
• git work-on 48 - Creates a feature branch for you, assigns the issue to you
• Hack hack hack - fix the issue
• git wrapup - Commits changes and closes the issue
• git deliver - Pushes to a ready branch, adds delivered prefix to your local feature branch
• Trigger the pipeline - it should pick up the ready branch and integrate it to master
• git purge-all-delivered - To delete your local feature branch that you no longer need