majiayu000 Storage & Networking
Executive Summary
Production-grade Kubernetes storage and networking covering persistent storage patterns, CSI driver configuration, CNI plugins, service discovery, and ingress routing. This skill provides deep expertise in building reliable, high-performance data and network infrastructure.
Core Competencies
1. Storage Architecture
Storage Stack
┌─────────────────────────────────────────────────┐
│ APPLICATION POD │
│ Volume Mount: /data │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ PERSISTENT VOLUME CLAIM (PVC) │
│ Namespace-scoped storage request │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ PERSISTENT VOLUME (PV) │
│ Cluster-wide resource │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ CSI DRIVER │
│ aws-ebs-csi, csi-driver-nfs, etc. │
└─────────────────────────────────────────────────┘
Production StorageClass
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: fast-ssd
provisioner: ebs.csi.aws.com
parameters:
type: gp3
iops: "5000"
throughput: "250"
encrypted: "true"
reclaimPolicy: Retain
allowVolumeExpansion: true
volumeBindingMode: WaitForFirstConsumer
---
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: shared-efs
provisioner: efs.csi.aws.com
parameters:
provisioningMode: efs-ap
fileSystemId: fs-abc123
reclaimPolicy: Retain
VolumeSnapshot for Backup
apiVersion: snapshot.storage.k8s.io/v1
kind: VolumeSnapshotClass
metadata:
name: ebs-snapclass
driver: ebs.csi.aws.com
deletionPolicy: Retain
---
apiVersion: snapshot.storage.k8s.io/v1
kind: VolumeSnapshot
metadata:
name: db-backup
spec:
volumeSnapshotClassName: ebs-snapclass
source:
persistentVolumeClaimName: postgresql-data-0
2. Networking Architecture
Network Stack
┌─────────────────────────────────────────────────┐
│ EXTERNAL TRAFFIC │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ LOAD BALANCER (ALB/NLB) │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ INGRESS CONTROLLER / GATEWAY API │
│ TLS termination, routing │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ KUBERNETES SERVICE │
│ ClusterIP, NodePort, LoadBalancer │
└─────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────┐
│ CNI PLUGIN │
│ Cilium, Calico, AWS VPC CNI │
└─────────────────────────────────────────────────┘
Service Configuration
apiVersion: v1
kind: Service
metadata:
name: api-server
namespace: production
spec:
type: ClusterIP
selector:
app.kubernetes.io/name: api-server
ports:
- name: http
port: 80
targetPort: 8080
- name: grpc
port: 9090
targetPort: 9090
3. Ingress & Gateway API
Production Ingress
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
name: api-ingress
annotations:
nginx.ingress.kubernetes.io/ssl-redirect: "true"
nginx.ingress.kubernetes.io/limit-rps: "100"
cert-manager.io/cluster-issuer: "letsencrypt-prod"
spec:
ingressClassName: nginx
tls:
- hosts:
- api.example.com
secretName: api-tls
rules:
- host: api.example.com
http:
paths:
- path: /v1
pathType: Prefix
backend:
service:
name: api-v1
port:
number: 80
Gateway API
apiVersion: gateway.networking.k8s.io/v1
kind: Gateway
metadata:
name: production-gateway
spec:
gatewayClassName: istio
listeners:
- name: https
hostname: "*.example.com"
port: 443
protocol: HTTPS
tls:
mode: Terminate
certificateRefs:
- name: wildcard-tls
---
apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
name: api-routes
spec:
parentRefs:
- name: production-gateway
hostnames:
- "api.example.com"
rules:
- matches:
- path:
type: PathPrefix
value: /api/v1
backendRefs:
- name: api-v1
port: 80
weight: 90
- name: api-v1-canary
port: 80
weight: 10
4. Network Policies
Zero-Trust Architecture
# Default deny all
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: default-deny-all
namespace: production
spec:
podSelector: {}
policyTypes:
- Ingress
- Egress
---
# Allow DNS
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: allow-dns
namespace: production
spec:
podSelector: {}
policyTypes:
- Egress
egress:
- to:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: kube-system
podSelector:
matchLabels:
k8s-app: kube-dns
ports:
- protocol: UDP
port: 53
---
# API server policy
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: api-server-policy
namespace: production
spec:
podSelector:
matchLabels:
app: api-server
policyTypes:
- Ingress
- Egress
ingress:
- from:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: ingress-nginx
ports:
- protocol: TCP
port: 8080
egress:
- to:
- podSelector:
matchLabels:
app: postgresql
ports:
- protocol: TCP
port: 5432
5. CNI Comparison
┌─────────────┬─────────────┬─────────────┬─────────────┐
│ Feature │ Cilium │ Calico │ AWS VPC CNI │
├─────────────┼─────────────┼─────────────┼─────────────┤
│ Performance │ Excellent │ Very Good │ Excellent │
│ L7 Policy │ ✓ (native) │ Via Envoy │ ✗ │
│ eBPF │ ✓ │ ✓ (option) │ ✗ │
│ Encryption │ WireGuard │ WireGuard │ VPC native │
│ Observ. │ Hubble │ Basic │ CloudWatch │
└─────────────┴─────────────┴─────────────┴─────────────┘
Integration Patterns
Uses skill: cluster-admin
- Node storage configuration
- CNI deployment
Coordinates with skill: security
- Network policy enforcement
- mTLS configuration
Works with skill: monitoring
- Network metrics
- Storage monitoring
Troubleshooting Guide
Decision Tree: Storage Issues
Storage Problem?
│
├── PVC Pending
│ ├── Check StorageClass exists
│ ├── Check provisioner running
│ └── WaitForFirstConsumer → Schedule pod
│
├── Pod can't mount
│ ├── Already attached → Force detach
│ ├── Permission denied → Check fsGroup
│ └── Filesystem error → Resize PVC
│
└── Performance issues
├── Check IOPS limits
└── Use faster StorageClass
Decision Tree: Network Issues
Network Problem?
│
├── Service not reachable
│ ├── No endpoints → Selector mismatch
│ ├── DNS not resolving → CoreDNS
│ └── Timeout → NetworkPolicy
│
├── Ingress not working
│ ├── 404 → Path mismatch
│ ├── 502 → Backend not ready
│ └── TLS error → Certificate
│
└── Pod-to-pod fails
├── Check NetworkPolicy
└── Check CNI pods
Debug Commands
# Storage
kubectl get pv,pvc -A
kubectl describe pvc <name>
kubectl get storageclass
# Network
kubectl get svc,endpoints,ingress -A
kubectl run debug --rm -it --image=nicolaka/netshoot -- nslookup <svc>
kubectl get networkpolicy -A
Common Challenges & Solutions
| Challenge | Solution |
|---|---|
| PVC Pending | Check StorageClass, provisioner |
| Volume timeout | Check node health, force detach |
| Ingress 502 | Check backend health |
| DNS failures | Verify CoreDNS, egress policy |
Success Criteria
| Metric | Target |
|---|---|
| PVC provision time | <30s |
| Storage availability | 99.99% |
| Service latency | <10ms |
| Network policy coverage | 100% |