Argo Workflows Patterns

When to Use This Skill

Production patterns for Argo Workflows: reusable templates, error handling, concurrency control, workflow composition, and scheduled automation.

---

Implementation

1. Define WorkflowTemplates - Create reusable, tested building blocks
2. Add Error Handling - Configure retry strategies for transient failures
3. Control Concurrency - Use mutexes or semaphores for shared resources
4. Compose Workflows - Chain templates into complex pipelines
5. Schedule Automation - Run workflows on cron schedules

---

Techniques

Why Argo Workflows?

Kubernetes provides primitives (Pods, Jobs, CronJobs), but building complex automation from primitives is painful. You end up with shell scripts that check Pod status in loops, cleanup logic scattered across multiple places, and debugging that requires correlating logs from dozens of sources.

Argo Workflows provides higher-level abstractions designed for automation. Define workflows declaratively. Let the controller handle scheduling, retries, and cleanup. Visualize execution in a purpose-built UI. Focus on what the automation does, not how to orchestrate it.

---

Pattern Categories

Category	Description
WorkflowTemplate Patterns	Reusable workflow definitions with error handling, volumes, and RBAC
Concurrency Control	Mutex synchronization, semaphores, and TTL strategies
Workflow Composition	Parent/child workflows, orchestration, and cross-workflow communication
Scheduled Workflows	CronWorkflow patterns and GitHub integration

---

Quick Start

1. Define WorkflowTemplates - Create reusable, tested building blocks
2. Add Error Handling - Configure retry strategies for transient failures
3. Control Concurrency - Use mutexes or semaphores for shared resources
4. Compose Workflows - Chain templates into complex pipelines
5. Schedule Automation - Run workflows on cron schedules

---

Troubleshooting

Workflow Stuck in Pending

1. Check service account permissions: kubectl describe rolebinding -n argo-workflows
2. Verify resource quotas: kubectl describe quota -n argo-workflows
3. Check node resources: kubectl top nodes
4. Look for mutex waits: kubectl get workflows -l workflows.argoproj.io/sync-id

Workflow Failed with RBAC Error

1. Verify ServiceAccount exists in workflow namespace
2. Check ClusterRoleBinding subjects match namespace
3. Use kubectl auth can-i to test permissions:

kubectl auth can-i patch deployments \
  --as=system:serviceaccount:argo-workflows:my-sa \
  -n target-namespace

Mutex Deadlock

1. Find workflows waiting on mutex: kubectl get workflows -l workflows.argoproj.io/sync-id
2. Identify the workflow holding the lock
3. Check if the holding workflow is stuck or failed
4. Terminate stuck workflows to release mutex

---

Prerequisites

Argo Workflows must be installed in your cluster. See the official installation guide for setup instructions.

---

Related

• Argo Events Setup - EventSource, EventBus, and Sensor configuration
• ConfigMap as Cache - Volume mounts for zero-API reads
• Event-Driven Deployments - The journey to zero-latency automation

Troubleshooting

See troubleshooting.md for common issues and solutions.

Related Patterns

• Argo Events Setup
• ConfigMap as Cache
• Event-Driven Deployments

References

• Source Documentation
• AEL Patterns