Argo Workflows on Kubernetes: CI/CD Guide

💡 Quick Answer: deployments

The Problem

Engineers need production-ready guides for these essential Kubernetes ecosystem tools. Incomplete documentation leads to misconfiguration and security gaps.

The Solution

Install Argo Workflows

kubectl create namespace argo
kubectl apply -n argo -f https://github.com/argoproj/argo-workflows/releases/latest/download/install.yaml

# Access UI
kubectl port-forward -n argo svc/argo-server 2746:2746
# Open https://localhost:2746

Simple Workflow

apiVersion: argoproj.io/v1alpha1
kind: Workflow
metadata:
  generateName: hello-world-
spec:
  entrypoint: main
  templates:
    - name: main
      steps:
        - - name: build
            template: build-step
        - - name: test        # Sequential after build
            template: test-step
        - - name: deploy-staging    # Parallel
            template: deploy
            arguments:
              parameters: [{name: env, value: staging}]
          - name: deploy-prod
            template: deploy
            arguments:
              parameters: [{name: env, value: production}]

    - name: build-step
      container:
        image: golang:1.22
        command: [go, build, -o, /output/app, .]
      outputs:
        artifacts:
          - name: binary
            path: /output/app

    - name: test-step
      container:
        image: golang:1.22
        command: [go, test, ./...]

    - name: deploy
      inputs:
        parameters:
          - name: env
      container:
        image: kubectl:latest
        command: [kubectl, apply, -f, "deploy-{{inputs.parameters.env}}.yaml"]

DAG Workflow

spec:
  entrypoint: pipeline
  templates:
    - name: pipeline
      dag:
        tasks:
          - name: build
            template: build-step
          - name: unit-tests
            template: test
            dependencies: [build]
          - name: integration-tests
            template: test
            dependencies: [build]
          - name: deploy
            template: deploy
            dependencies: [unit-tests, integration-tests]

Cron Workflow

apiVersion: argoproj.io/v1alpha1
kind: CronWorkflow
metadata:
  name: nightly-pipeline
spec:
  schedule: "0 2 * * *"
  timezone: "Europe/Rome"
  concurrencyPolicy: Replace
  workflowSpec:
    entrypoint: main
    templates:
      - name: main
        container:
          image: my-pipeline:v1
          command: ["/run.sh"]

graph TD
    A[Build] --> B[Unit Tests]
    A --> C[Integration Tests]
    B --> D[Deploy Staging]
    C --> D
    D --> E[Smoke Tests]
    E --> F[Deploy Production]

Frequently Asked Questions

Argo Workflows vs Tekton?

Argo Workflows has a better UI, DAG support, and artifact management. Tekton is more Kubernetes-native with reusable Tasks. Argo is better for complex data pipelines; Tekton for simpler CI/CD.

Argo Workflows vs Argo CD?

Different tools! Workflows runs jobs/pipelines (CI). Argo CD does GitOps deployment (CD). They work well together.

Best Practices

• Start with default configurations and customize as needed
• Test in a non-production cluster first
• Monitor resource usage after deployment
• Keep components updated for security patches

Key Takeaways

• This tool fills a critical gap in the Kubernetes ecosystem
• Follow the principle of least privilege for all configurations
• Automate where possible to reduce manual errors
• Monitor and alert on operational metrics