排障现场记不住？收藏这篇就够，或者直接 AI 对话式学习 🤪

0️⃣ TL;DR 一图速查

步骤 场景 常用命令
观察 看 Pod 状态 / 资源用量 kubectl get po -n <ns>
kubectl top po/nodes
操作 CRUD / 扩缩容 kubectl apply/delete -f
kubectl scale deploy ...
排障 进盒子 / 测连通 kubectl exec -it ...
kubectl port-forward ...
运维 版本回滚 / 集群切换 kubectl rollout undo ...
kubectl config use-context ...

1️⃣ 观察｜先搞清楚发生了什么

1
2
3
4
kubectl get po -n <ns>          # 列出 Pod，快速看状态
kubectl describe po <name> -n <ns>   # 详细事件 / 条目
kubectl logs <po/rs/deploy> -c <container> -n <ns>   # 日志（加 -f 实时跟）
kubectl top po/nodes            # CPU / 内存用量（需装 metrics-server）

2️⃣ 操作｜对象的 CRUD

1
2
3
4
kubectl apply -f <file|dir>     # 创建 / 更新一切（IaC 友好）
kubectl delete -f <file|dir>    # 按清单删除
kubectl edit <kind> <name> -n <ns>   # 临时在线热改（会起默认编辑器）
kubectl scale deploy <name> --replicas=N -n <ns>   # 手动扩缩容

3️⃣ 排障｜进盒子、测连通

1
2
kubectl exec -it <po> -c <container> -- /bin/sh   # SSH 替代，进容器
kubectl port-forward svc/<svc> 8080:80 -n <ns>    # 本地调试

4️⃣ 运维｜版本演进 & 环境切换

1
2
kubectl rollout history|status|undo deploy <name> -n <ns>   # 版本管理三件套
kubectl config get-contexts / use-context <ctx>             # 多集群切换

常用通配符备忘

  • n <ns> 指定命名空间，默认 default
  • o yaml|json 输出可直接回 apply
  • -all-namespaces 横扫全集群

牢牢记住 “get → describe/logs → exec” 这条排障黄金链路，再加上 apply / delete 两极操作，你就具备了在生产现场看得见、改得掉、救得了的核心功力。

有意思的，用了下 Warp 这个终端，问下它：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
I'll tell you about the core kubectl commands that are essential for managing Kubernetes clusters:

Core kubectl Commands

Cluster and Context Management
•  kubectl config view - View kubeconfig settings
•  kubectl config current-context - Show current context
•  kubectl config use-context <context> - Switch between contexts
•  kubectl cluster-info - Display cluster information

Resource Management
•  kubectl get <resource> - List resources (pods, services, deployments, etc.)
•  kubectl describe <resource> <name> - Show detailed information about a resource
•  kubectl create -f <file> - Create resources from a file
•  kubectl apply -f <file> - Apply configuration changes
•  kubectl delete <resource> <name> - Delete a resource

Pod Operations
•  kubectl get pods - List all pods
•  kubectl logs <pod-name> - View pod logs
•  kubectl exec -it <pod-name> -- <command> - Execute commands in a pod
•  kubectl port-forward <pod-name> <local-port>:<pod-port> - Forward ports

Deployment Management
•  kubectl create deployment <name> --image=<image> - Create a deployment
•  kubectl scale deployment <name> --replicas=<count> - Scale deployment
•  kubectl rollout status deployment/<name> - Check rollout status
•  kubectl rollout history deployment/<name> - View rollout history

Service and Networking
•  kubectl expose deployment <name> --port=<port> - Expose deployment as service
•  kubectl get services - List services
•  kubectl get ingress - List ingress resources

Debugging and Troubleshooting
•  kubectl top nodes - Show node resource usage
•  kubectl top pods - Show pod resource usage
•  kubectl get events - View cluster events
•  kubectl describe node <node-name> - Get node details

Common Resource Types
•  pods (po) - Running containers
•  services (svc) - Network services
•  deployments (deploy) - Deployment configurations
•  replicasets (rs) - Replica management
•  configmaps (cm) - Configuration data
•  secrets - Sensitive data
•  nodes (no) - Cluster nodes
•  namespaces (ns) - Resource isolation

Useful Flags
•  -n <namespace> - Specify namespace
•  --all-namespaces - Show resources across all namespaces
•  -o yaml/json - Output in YAML or JSON format
•  --watch - Watch for changes
•  --dry-run=client - Preview changes without applying

Would you like me to demonstrate any of these commands or help you with specific kubectl operations?