分支：matesr

平台：Kubernetes

仓库地址：https://github.com/thx/rap2-delos

官方提供的是docker-compose部署，部署流程比较简单，因此部署到Kubernetes难度也不高，本篇文章主要是为了解决部署后的访问异常。

部署

后端

后端依赖数据库mysql和redis，部署这里就不赘述了。

rap2-backend-deploy.yaml

kind: Deployment
apiVersion: apps/v1
metadata:
  name: rap2-backend
  namespace: company
  labels:
    app: rap2-backend
spec:
  replicas: 1
  selector:
    matchLabels:
      app: rap2-backend
  template:
    metadata:
      labels:
        app: rap2-backend
    spec:
      containers:
        - name: rap2-backend
          image: 'rapteam/rap2-delos:latest'
          command:
            - /bin/sh
            - '-c'
            - node dispatch.js
          ports:
            - name: http-38080
              containerPort: 38080
              protocol: TCP
          env:
            - name: SERVE_PORT
              value: '38080'
            - name: MYSQL_URL
              value: xxx
            - name: MYSQL_PORT
              value: '3306'
            - name: MYSQL_USERNAME
              value: root
            - name: MYSQL_PASSWD
              value: xxx
            - name: MYSQL_SCHEMA
              value: rap2
            - name: REDIS_URL
              value: xxx
            - name: REDIS_PORT
              value: '6379'
            - name: REDIS_PWD
              value: xxx
            - name: MAIL_HOST
              value: smtp.exmail.qq.com
            - name: MAIL_USER
              value: xxx
            - name: MAIL_PASS
              value: xxx
            - name: MAIL_SENDER
              value: xxx
            - name: NODE_ENV
              value: production
          resources:
            limits:
              cpu: '5'
              memory: 5000Mi
            requests:
              cpu: 50m
              memory: 10Mi
          terminationMessagePath: /dev/termination-log
          terminationMessagePolicy: File
          imagePullPolicy: Always
      restartPolicy: Always
      terminationGracePeriodSeconds: 30
      dnsPolicy: ClusterFirst
      securityContext: {}
      imagePullSecrets:
        - name: harbor
      schedulerName: default-scheduler
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 25%
      maxSurge: 25%
  revisionHistoryLimit: 10
  progressDeadlineSeconds: 600

前端

rap2-front-deply.yaml

kind: Deployment
apiVersion: apps/v1
metadata:
  name: rap2-front
  namespace: company
  labels:
    app: rap2-front
spec:
  replicas: 1
  selector:
    matchLabels:
      app: rap2-front
  template:
    metadata:
      labels:
        app: rap2-front
    spec:
      containers:
        - name: rap2-front
          image: 'harbor.com/rapteam/rap2-dolores:v1'
          ports:
            - name: http-38081
              containerPort: 38081
              protocol: TCP
          resources:
            limits:
              cpu: 500m
              memory: 200Mi
            requests:
              cpu: 50m
              memory: 10Mi
          terminationMessagePath: /dev/termination-log
          terminationMessagePolicy: File
          imagePullPolicy: Always
      restartPolicy: Always
      terminationGracePeriodSeconds: 30
      dnsPolicy: ClusterFirst
      securityContext: {}
      imagePullSecrets:
        - name: harbor-sk8s
      schedulerName: default-scheduler
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 25%
      maxSurge: 25%
  revisionHistoryLimit: 10
  progressDeadlineSeconds: 600

数据库初始化

# 在后端服务中执行数据初始化
delos node scripts/init

问题

1. 浏览器访问前端地址xxx.com一直处于“拼命加载中”

   

打开浏览器开发者模式后可以看到info接口请求地址是http://xxx.com:38080/account/info

解决方法：该地址无法访问，如何变更后端接口地址。官方没有给出可变更的配置或者环境变量

重新编译会出现诸多问题，修改现有前端镜像中的js文件。

runtime-main.3d19d946.js

runtime-main.3d19d946.js.map

var a={serve:"https://rap2api.kreator-inc.com",

2. 后端启动失败

2025-02-13 16:18:02[2025-02-13 08:18:01.347] master#1 worker#13 has disconnected.
2025-02-13 16:18:02[2025-02-13 08:18:02.256] master#1 worker#13 died (SIGKILL). restarting...
2025-02-13 16:18:02[2025-02-13 08:18:02.782] master#1 worker#23 has disconnected.
2025-02-13 16:18:02[2025-02-13 08:18:02.782] master#1 worker#23 died (SIGKILL). restarting...
2025-02-13 16:18:03[2025-02-13 08:18:03.747] master#1 worker#109 has disconnected.
2025-02-13 16:18:03[2025-02-13 08:18:03.946] master#1 worker#109 died (SIGKILL). restarting...

解决方法：资源给太小导致，按照2C5G配置启动。