Pipeline represents a projects integration and delivery flow.
Pipelines comprises:
- A list of steps.
- Step - defines what to do. For example, a step that builds image from your code.
klovercloud/pipeline/pipeline.yaml file contains pipeline configuration.
Steps are jobs or tasks with successors that make a Pipeline. Pipeline steps are defined in the pipeline.yaml file. It includes the following:
- Build step: - Build step is used to build images from source code and push into registry.
Example:
- name: build # Name of the step. This must be unique across the whole pipeline file.
type: BUILD # Type of the step. Case sensitive.
trigger: AUTO # Trigger can be Auto or Manual, case sensitive. Trigger Auto means step will run without manual interactions.
params: # Parameters of the step.
repository_type: git # Type of the scm.
revision: master # Revision/commit id/branch of the repository. If provided, this will be image tag, otherwise commit id will be default. $BRANCH as variable will replace revision by branch name.
service_account: test-sa # Service account that references source code clone and image push secret.
images: zeromsi2/test-dev,zeromsi2/test-pro # List of images to build.
args_from_configmaps: tekton/cm-test # Comma separated list of configmaps, convention is namespace/name of configmap. Pass arguments using configmap.
args: key3:value1,key4:value2 # Pass argument as key and value pairs.
env: init # name of environment
allowed_branches: dev,master # If commits are not from these branches, pipeline won't get triggered.
dockerfile_path: "" # Dockerfile path from base directory of the project.
next:
- interstep # Name of the next steps.- Intermediary Step: - Intermediary step is used to run container.
Example:
- name: interstep # Name of the step. This must be unique across the whole pipeline file.
type: INTERMEDIARY # Type of the step. Case sensitive.
trigger: AUTO # Trigger can be Auto or Manual, case sensitive. Trigger Auto means step will run without manual interactions.
params: # Parameters of the step.
revision: latest # Revision of the image.
service_account: test-sa # Service account to use.
images: ubuntu # Container image to run.
envs_from_configmaps: tekton/cm-test # Comma separated list of configmaps, convention is namespace/name of configmap. Mount environment variables from configmaps.
envs_from_secrets: tekton/cm-test # Comma separated list of secrets, convention is namespace/name of secret. Mount environment variables from secrets.
envs: key3:value1,key4:value2 # Mount environment variables as key and value pair.
command: echo "Hello World" # Command to run.
command_args: Hello World # Command arguments.
script: echo "Hello zeromsi" # Script to run.
script_from_configmap: default/configmap_name # Run Script from configmap
env: build # name of environment
next:
- deployDev[N:B] For mounting script from config map in interstep, follow namespace/configmap_name convention for script_from_configmap params.
Script is expected to be in data.script path:
apiVersion: v1 # Version of the config map.
kind: ConfigMap # Kind of the config map.
metadata:
name: script # Name of the config map.
data: # Data of the config map.
script: | # Script to run.
curl www.google.com- Deployment Step: - Deployment step is used to deploy the descriptors and modify kubernetes work loads.
Example:
- name: deployDev # Name of the step. This must be unique across the whole pipeline file.
type: DEPLOY # Type of the step. Case sensitive.
trigger: AUTO # Trigger can be Auto or Manual, case sensitive. Trigger Auto means step will run without manual interactions.
params: # Parameters of the step.
agent: local_agent # Agent represents your cluster. Every cluster will have at least one agent running. This is the name of the agent.
name: ubuntu # Name of the resource.
namespace: default # Namespace of the resource.
type: deployment # Type of the resource.
revision: master # If the commit is for master branch, deploy job will run.
trunk_based: enabled # If Enabled, image revision will be replaced by commit id.
rollout_restart: true # If true, after updating resource, it will rollout restart.
env: dev # Environment to use.
images: zeromsi2/test-dev # List of images to use.
next: # List of next stages.
- jenkins[N:B] As the environment name is dev ,all payloads inside klovercloud/pipeline/configs/dev.yaml file will be applied first.
- Jenkins Job Step: - Jenkins job step is used to run a jenkins job.
Example:
- name: jenkins # Name of the stage.
type: JENKINS_JOB # Type of the step. Case sensitive.
trigger: AUTO # Trigger can be Auto or Manual, case sensitive. Trigger Auto means step will run without manual interactions.
params: # Parameters of the stage.
url: http://jenkins.default.svc:8080 # URL of the Jenkins server.
job: test_jenkins_job # Name of the Jenkins job.
secret: jenkins-credentials # Name of the secret to use.
params: id:123,verbosity:high # List of parameters to pass to the Jenkins job.
env: post # name of environment
next: null # List of next stages.A process is an object of a Pipeline. When a Pipeline is triggered a new Process starts.
name: test
_metadata:
allowed_branches: dev,master
steps:
- name: build
type: BUILD
trigger: AUTO
params:
repository_type: git
revision: master
service_account: test-sa
images: zeromsi2/test-dev,zeromsi2/test-pro
args_from_configmaps: tekton/cm-test
env: init
next:
- interstep
- name: interstep
type: INTERMEDIARY
trigger: AUTO
params:
revision: latest
service_account: test-sa
images: ubuntu
envs_from_secrets: tekton/cm-test
command: echo "Hello World"
command_args: Hello World
env: init
next:
- deployDev
- name: deployDev
type: DEPLOY
trigger: AUTO
params:
agent: local_agent
name: ubuntu
namespace: default
type: deployment
revision: master
env: dev
images: zeromsi2/test-dev
next:
- jenkins
- name: jenkins
type: JENKINS_JOB
trigger: AUTO
params:
url: http://jenkins.default.svc:8080
job: test_jenkins_job
secret: jenkins-credentials
params: id:123,verbosity:high
env: post
next: null
Figure: Pipeline diagram.