图像管理应用提供了一个方便管理图片的平台,支持单机和Kubernetes集群部署。请确保您至少拥有一个MySQL数据库和一个Redis数据库,以及一个至少为Kubernetes 1.29版本的集群(如果选择集群部署)。
使用Docker拉取并运行image-manage:
docker run -p 8080:8080 \
-v 你的数据目录:/app\
-e CONFIG_ISCLUSTERMODEENABLED=false \
-e SPRING_DATASOURCE_URL=jdbc:mysql://192.168.0.254:3306/image_manage?useUnicode=true&characterEncoding=utf-8&useSSL=false&serverTimezone=Asia/Shanghai \
-e SPRING_DATASOURCE_USERNAME=root \
-e SPRING_DATASOURCE_PASSWORD=123456 \
-e SPRING_REDIS_HOST=192.168.0.254 \
-e SPRING_REDIS_PASSWORD=123456 \
wnzzer/image-manage:latest
kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
wget https://github.com/wnzzer/image-manage/releases/latest/download/image-manage-yaml
apiVersion: v1
kind: Namespace
metadata:
name: image-manage
---
# storeclass 声明,非常关键,用于动态分配卷
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: nfs-store-class
provisioner: image-manage-pv-provisioner # 替换为实际的卷插件,我这里使用的是nfs的自动供应器,可以采用公用云或者其他pv自动供应器
reclaimPolicy: Delete
parameters:
volumeSize: "1Gi" # 应用存储图片的空间
nfsServer: 192.168.0.254
nfsPath: /volume1/nfs
---
# configMap k8s权限配置到英雄
apiVersion: v1
kind: ConfigMap
metadata:
namespace: image-manage
name: k8s-role-config
data:
key1: ./admin.config
---
# 应用
apiVersion: apps/v1
kind: StatefulSet
metadata:
namespace: image-manage
name: image-manage
spec:
serviceName: "image-manage-headless-service"
replicas: 2
selector:
matchLabels:
app: image-manage
template:
metadata:
labels:
app: image-manage
spec:
containers:
- name: image-manage
image: wnzzer/image-manage:latest
ports:
- containerPort: 8080
env:
- name: SPRING_PROFILES_ACTIVE
value: "prod"
- name: CONFIG_ISCLUSTERMODEENABLED
value: "false"
- name: SPRING_DATASOURCE_URL
value: "jdbc:mysql://192.168.0.254:3306/image_manage?userUnicode=true&characterEncoding=utf-8&useSSL=false&serverTimezone=Asia/Shanghai"
- name: SPRING_DATASOURCE_USERNAME
value: "root"
- name: SPRING_DATASOURCE_PASSWORD
value: "123456"
- name: SPRING_REDIS_HOST
value: "192.168.0.254"
- name: SPRING_REDIS_PASSWORD
value: "123456"
volumeMounts:
- name: local-vol
mountPath: "/app"
- name: config-volume # 挂载ConfigMap作为卷
mountPath: "/etc/image-manage/config" # 指定挂载ConfigMap的路径
volumes:
- name: config-volume # 定义卷使用的ConfigMap
configMap:
name: k8s-admin-role-config # 指定ConfigMap的名称
volumeClaimTemplates:
- metadata:
name: local-vol
spec:
accessModes: [ "ReadWriteOnce" ]
storageClassName: "nfs-store-class"
resources:
requests:
storage: 1Gi
---
# 无头服务,用于应用的同步通信
apiVersion: v1
kind: Service
metadata:
name: image-manage-headless-service
namespace: image-manage
spec:
clusterIP: None
selector:
app: image-manage
ports:
- protocol: TCP
port: 8080
targetPort: 8080
---
# cluster service 用于输出api
apiVersion: v1
kind: Service
metadata:
name: image-manage-cluster-service
namespace: image-manage
spec:
type: ClusterIP
ports:
- port: 8080
targetPort: 8080
selector:
app: image-manage
kubectl apply -f image-manage.yaml
- 这里同样需要把redis,mysql修改成自己的配置
- 这里需要有自己的pv供应器,由于是pv动态创建,所以需要搭配自动供应器给statusfulSet创建pv,示例中使用的是nfs
- image-manage 需要搭配 k8s 配置进行对k8s api的访问,这里实例使用的是admin.conf,使用configmap挂载到容器里,如果想要更细致的权限划分,请将权限配置至少给予 image-manage级别的权限。
- 创建的k8s用户文件请将命名为admin.conf,因为应用里指定了k8s配置文件为admin.conf,其他名称会无法读取。
我们可以直接在k8s中部署nginx,反代 image-manage cluster ip,进行访问,
cd font-userui
npm i
npm run dev