kubernetes-orchestration

Files (1)

• SKILL.md

SKILL.md (first lines)

## When to Use

- Deploying containerized applications to Kubernetes clusters.
- Writing or reviewing Deployment, Service, Ingress, or other K8s manifests.
- Configuring health probes, resource limits, autoscaling, or rolling updates.
- Troubleshooting pod scheduling, networking, or configuration issues.
- Setting up Kustomize overlays for multi-environment deployments.

## Critical Patterns

- **Resource Limits Are Mandatory:** ALWAYS set both `requests` and `limits` for CPU and memory.
  Without them, a single pod can starve the entire node.
- **Health Probes on Every Workload:** ALWAYS define `readinessProbe` and `livenessProbe`. Without
  readiness probes, traffic routes to unready pods. Without liveness probes, stuck containers are
  never restarted.
- **Startup Probes for Slow Init:** Use `startupProbe` for applications with long initialization (
  JVM, large model loading) to avoid premature liveness kills.
- **Never Use `latest` Tag:** Pin image tags to immutable versions or digests. `latest` causes
  unpredictable deployments and breaks rollback.
- **Namespaces for Isolation:** Separate workloads by team, environment, or domain using namespaces.
  Apply ResourceQuotas per namespace.
- **Secrets Are Not Encrypted at Rest by Default:** Enable encryption at rest or use external secret
  managers (Vault, Sealed Secrets, External Secrets Operator).
- **Pod Disruption Budgets:** Always set PDBs for production workloads to prevent all replicas from
  being evicted during node maintenance.

## Code Examples

### Pod Spec with Complete Health Probes

```yaml
apiVersion: v1
kind: Pod
metadata:
  name: api-server
  labels:
    app: api-server
    version: v1.2.0
spec:
  containers:
    - name: api
      image: myregistry/api-server:1.2.0
      ports:
        - containerPort: 8080
          protocol: TCP
      resources:
        requests:
          cpu: 100m
          memory: 128Mi
        limits:
          cpu: 500m
          memory: 512Mi
      startupProbe:
        httpGet:
          path: /healthz
          port: 8080
        failureThreshold: 30
        periodSeconds: 10
      readinessProbe:
        httpGet:
          path: /ready
          port: 8080
        initialDelaySeconds: 5
        periodSeconds: 10
        timeoutSeconds: 3
        failureThreshold: 3
      livenessProbe:
        httpGet:
          path: /healthz
          port: 8080
        initialDelaySeconds: 15
        periodSeconds: 20
        timeoutSeconds: 5
        failureThreshold: 3
      env:
        - name: DATABASE_URL
          valueFrom:
            secretKeyRef:
              name: db-credentials
              key: url
        - name: LOG_LEVEL
          valueFrom:
            configMapKeyRef:
              name: app-config
              key: log-level
  restartPolicy: Always
```

### Production Deployment with Rolling Updates

```yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: api-server
  namespace: production
  labels:
    app: api-server
spec:
  replicas: 3
  revisionHistoryLimit: 5
  selector:
    matchLabels:
      app: api-server
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 0
  template:
    metadata:
      labels:
        app: api-server
        version: v1.2.0
    spec:
      terminationGracePeriodSeconds: 60
      containers:
        - name: api
          image: myregistry/api-server:1.2.0

From dallay/agents-skills, MIT.

More from dallay/agents-skills