Kubernetes pod 无法启动并获得 CrashLoopBackOff答案

【问题标题】：Kubernetes pod wont start and get CrashLoopBackOffKubernetes pod 无法启动并获得 CrashLoopBackOff
【发布时间】：2022-07-11 23:02:34
【问题描述】：

当我尝试运行 kubectl apply -f frontend.yaml 时，我收到来自 kubectl get pods 和 kubectl describe pods 的以下响应

// frontend.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: malvacom-frontend
  labels:
    app: malvacom-frontend
spec:
  replicas: 1
  selector:
    matchLabels:
      app: malvacom-frontend
  template:
    metadata:
      labels:
        app: malvacom-frontend
    spec:
      containers:
        - name: malvacom-frontend
          image: docker.io/forsrobin/malvacom_frontend
          imagePullPolicy: IfNotPresent

          ports:
            - containerPort: 80
          resources:
            limits:
              memory: "128Mi"
              cpu: "200m"
          livenessProbe:
            httpGet:
              path: /index.html
              port: 80
            initialDelaySeconds: 15
            timeoutSeconds: 2
            periodSeconds: 5
            failureThreshold: 1
          readinessProbe:
            httpGet:
              path: /index.html
              port: 80
            initialDelaySeconds: 15
            periodSeconds: 5
            failureThreshold: 1
          command: [ "sleep" ]
          args: [ "infinity" ]

然后响应是

kubectl get pods
malvacom-frontend-8575c8548b-n959r   0/1     CrashLoopBackOff   5 (95s ago)   4m38s

kubectl describe pods
QoS Class:                   Guaranteed
Node-Selectors:              <none>
Tolerations:                 node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
                             node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
  Type     Reason     Age                From               Message
  ----     ------     ----               ----               -------
  Normal   Scheduled  17s                default-scheduler  Successfully assigned default/malvacom-frontend-8575c8548b-n959r to shoot--p1622--malvacom-web-xdmoi2-z1-54776-bpjpw
  Normal   Pulled     15s (x2 over 16s)  kubelet            Container image "docker.io/forsrobin/malvacom_frontend" already present on machine
  Normal   Created    15s (x2 over 16s)  kubelet            Created container malvacom-frontend
  Normal   Started    15s (x2 over 16s)  kubelet            Started container malvacom-frontend
  Warning  BackOff    11s (x4 over 14s)  kubelet            Back-off restarting failed container

正如我所理解的，pod 启动了，但是因为它没有继续执行的任务，kubernetes 删除/停止了 pod。我可以毫无问题地在本地运行图像，例如，如果我使用另一个图像thenetworkchuck/nccoffee:pourover ，它可以毫无问题地工作。这是我的 Dockerfile

FROM node:alpine as build
WORKDIR /app
ENV PATH /app/node_modules/.bin:$PATH
COPY ./package.json /app/
RUN yarn --silent
COPY . /app
RUN yarn build
FROM nginx:alpine
COPY --from=build /app/build /usr/share/nginx/html
RUN rm /etc/nginx/conf.d/default.conf
COPY nginx/nginx.conf /etc/nginx/conf.d
EXPOSE 80
CMD ["nginx", "-g", "daemon off;"]

【问题讨论】：

标签： reactjs docker nginx kubernetes networking

【解决方案1】：

你明确告诉 Kubernetes 不要运行它的普通服务器

command: [ "sleep" ]
args: [ "infinity" ]

但它应该通过 HTTP 健康检查

livenessProbe:
  httpGet:
    path: /index.html
    port: 80

由于sleep infinity 没有运行 HTTP 服务器，因此此探测将永远不会通过，这会导致您的容器被杀死并重新启动。

你不应该做人为的事情来“让容器保持活力”；删除 command: 和 args: 覆盖。（Dockerfile CMD 是正确的，但是您会从基础 nginx 映像中获得相同的 CMD，并且您不需要重复它。）

【讨论】：