Getting Started: Your First Workflow

Create a runbook.yml, push it, and observe BlueGreen execute the workflow with live logs and a graph view.

Overview

This guide shows how to create, commit, and run a minimal BlueGreen Workflow, to which we’ll refer to as a “runbook”. Follow the steps to trigger a workflow and inspect its execution in the BlueGreen dashboard.

Prerequisites

• A repository on GitHub or GitLab
• A BlueGreen account (https://bluegreen.ci)
• A BlueGreen project connected to your repository

Minimal example

Create a runbook.yml at the repository root:

1
name: "Getting Started with BlueGreen"
2

3
steps:
4
- name: "Hello World"
5
  command: |
6
    echo "Hello, World from BlueGreen!"

Commit and push the file:

git add runbook.yml
git commit -m "Add initial BlueGreen runbook"
git push

Pushing the file will trigger a BlueGreen run for the repository.

Inspecting the run in the dashboard

Open the Projects dashboard and select your project’s Workflows view to see recent runs. Selecting a run opens the Run Details page and displays status (Queued / Running / Succeeded / Failed):

The Run Details view shows the execution graph and step dependencies:

Selecting an individual step reveals live-streamed logs and persisted logs for post-mortem analysis:

Extending the example

Add a dependent step and an environment variable:

1
name: "Getting Started with BlueGreen"
2

3
steps:
4
- name: "Hello World"
5
  command: |
6
    echo "Hello, World from BlueGreen!"
7

8
- name: "Greet"
9
  depends_on: ["Hello World"]
10
  env:
11
    GREETING: "Hi from BlueGreen"
12
  command: |
13
    echo $GREETING

What is a workflow?

A workflow is a declarative directed graph of steps defined in runbook.yml. Each step is a node and depends_on entries form edges. This model expresses ordering and parallelism explicitly.

A runbook describes:

• Steps to execute
• Inter-step dependencies
• Commands run for each step
• Environment variables that affect execution
• Optional global configuration

Example: expanded

Here is a slightly larger example demonstrating dependencies and environment variables:

1
name: "Getting Started with BlueGreen"
2

3
steps:
4
- name: "Hello World"
5
  command: |
6
    echo "Hello, World!"
7

8
- name: "Later Gator"
9
  depends_on:
10
    - "Hello World"
11
  env:
12
    GREETING: "See you later, alligator!"
13
  command: |
14
    echo $GREETING

Behavior summary

• Hello World executes first.
• Later Gator waits for the Hello World step (depends_on).
• GREETING is injected as an environment variable for the dependent step.
• Steps run in isolated containers and stream logs to the UI.

1
name: "CI Pipeline"
2

3
global:
4
  env:
5
    GO111MODULE: "on"
6

7
steps:
8
- name: build
9
  command: go build ./...
10

11
- name: lint
12
  command: golangci-lint run
13

14
- name: test
15
  depends_on: [build]
16
  command: go test ./... -v
17

18
- name: quality-gate
19
  depends_on: [lint, test]
20
  command: echo "All checks passed!"

Step properties

Each step supports the following properties:

Global environment variables

1
global:
2
  env:
3
    GREETING: "Hello from BlueGreen!"

Step-level env entries override global variables.