pod-management

Pod Management and Debugging

Overview

Skills such as Pod lifecycle management, log viewing, and exec debugging.

Pod Viewing

# 列出 Pod
kubectl get pods
kubectl get pods -o wide                    # 详细信息
kubectl get pods -A                         # 所有命名空间
kubectl get pods -n namespace
kubectl get pods -l app=nginx               # 按标签过滤
kubectl get pods --field-selector status.phase=Running

Pod Details

kubectl describe pod pod-name kubectl get pod pod-name -o yaml kubectl get pod pod-name -o jsonpath='{.status.phase}'

## 日志查看

```bash
# 查看日志
kubectl logs pod-name
kubectl logs pod-name -c container-name     # 指定容器
kubectl logs pod-name --all-containers      # 所有容器

# 实时跟踪
kubectl logs -f pod-name
kubectl logs -f pod-name --since=1h         # 最近1小时
kubectl logs -f pod-name --tail=100         # 最后100行

# 上一个容器的日志（崩溃后）
kubectl logs pod-name --previous

# 多 Pod 日志
kubectl logs -l app=nginx                   # 按标签
kubectl logs -l app=nginx --max-log-requests=10

Container Debugging

exec into Container

# 进入容器
kubectl exec -it pod-name -- /bin/bash
kubectl exec -it pod-name -- sh
kubectl exec -it pod-name -c container-name -- /bin/bash

# 执行命令
kubectl exec pod-name -- ls -la
kubectl exec pod-name -- cat /etc/hosts
kubectl exec pod-name -- env

Debugging Container

# 使用 debug 容器（K8s 1.18+）
kubectl debug pod-name -it --image=busybox --target=container-name

# 复制 Pod 进行调试
kubectl debug pod-name -it --copy-to=debug-pod --container=debug --image=busybox

# 节点调试
kubectl debug node/node-name -it --image=busybox

Port Forwarding

# 转发到本地
kubectl port-forward pod-name 8080:80
kubectl port-forward pod-name 8080:80 --address 0.0.0.0

# 后台运行
kubectl port-forward pod-name 8080:80 &

Pod Lifecycle

Create Pod

# pod.yaml
apiVersion: v1
kind: Pod
metadata:
  name: nginx-pod
  labels:
    app: nginx
spec:
  containers:
  - name: nginx
    image: nginx:latest
    ports:
    - containerPort: 80
    resources:
      requests:
        memory: "64Mi"
        cpu: "250m"
      limits:
        memory: "128Mi"
        cpu: "500m"
    livenessProbe:
      httpGet:
        path: /
        port: 80
      initialDelaySeconds: 10
      periodSeconds: 5
    readinessProbe:
      httpGet:
        path: /
        port: 80
      initialDelaySeconds: 5
      periodSeconds: 3

kubectl apply -f pod.yaml
kubectl create -f pod.yaml

Delete Pod

kubectl delete pod pod-name
kubectl delete pod pod-name --force --grace-period=0   # 强制删除
kubectl delete pods -l app=nginx                        # 按标签删除
kubectl delete pods --all -n namespace                  # 删除所有

Pod Status

# 查看 Pod 状态
kubectl get pod pod-name -o jsonpath='{.status.phase}'

# 等待 Pod 就绪
kubectl wait --for=condition=Ready pod/pod-name --timeout=60s

# 查看事件
kubectl get events --field-selector involvedObject.name=pod-name

Resource Management

View Resource Usage

# Pod 资源使用
kubectl top pods
kubectl top pods -n namespace
kubectl top pod pod-name --containers

# 按资源排序
kubectl top pods --sort-by=cpu
kubectl top pods --sort-by=memory

Resource Configuration

resources:
  requests:
    memory: "64Mi"
    cpu: "250m"
  limits:
    memory: "128Mi"
    cpu: "500m"

Common Scenarios

Scenario 1: Troubleshooting Pod Startup Failure

# 1. 查看 Pod 状态
kubectl get pod pod-name -o wide

# 2. 查看事件
kubectl describe pod pod-name | grep -A 20 Events

# 3. 查看日志
kubectl logs pod-name
kubectl logs pod-name --previous    # 如果容器重启

# 4. 检查镜像
kubectl get pod pod-name -o jsonpath='{.spec.containers[*].image}'

Scenario 2: Troubleshooting CrashLoopBackOff

# 1. 查看退出原因
kubectl describe pod pod-name | grep -A 5 "Last State"

# 2. 查看上一次日志
kubectl logs pod-name --previous

# 3. 检查资源限制
kubectl describe pod pod-name | grep -A 10 "Limits"

# 4. 检查探针配置
kubectl get pod pod-name -o yaml | grep -A 10 "livenessProbe"

Scenario 3: Temporarily Running a Debugging Pod

# 运行临时 Pod
kubectl run debug --rm -it --image=busybox -- sh
kubectl run debug --rm -it --image=nicolaka/netshoot -- bash

# 在特定节点运行
kubectl run debug --rm -it --image=busybox --overrides='{"spec":{"nodeName":"node-name"}}' -- sh

Scenario 4: Copying Files

# 从 Pod 复制到本地
kubectl cp pod-name:/path/to/file ./local-file
kubectl cp namespace/pod-name:/path/to/file ./local-file

# 从本地复制到 Pod
kubectl cp ./local-file pod-name:/path/to/file

Troubleshooting

Status	Possible Cause	Troubleshooting Method
Pending	Insufficient Resources/Scheduling Issues	kubectl describe pod View events
ImagePullBackOff	Image Pull Failure	Check image name and registry authentication
CrashLoopBackOff	Application Crash	kubectl logs --previous
OOMKilled	Memory Limit Exceeded	Increase memory limits
Evicted	Insufficient Node Resources	Check node resources, clean up Pods
Unknown	Node Not Ready	Check node status