Kubernetes(K8S) Deployment 拉取阿里云镜像部署
原创
©著作权归作者所有：来自51CTO博客作者EicroSoft的原创作品，请联系作者获取转载授权，否则将追究法律责任
Docker Image 推到阿里云仓库，可以看 SpringBoot Docker 发布到 阿里仓库
1. 阿里镜像仓库加了授权，所以 K8S 拉之前要做下授权处理
[root@k8smaster ~]# kubectl create secret docker-registry registry-demo \
--namespace=default \
--docker-server=registry.cn-shanghai.aliyuncs.com \
--docker-username=hi帐户ID@aliyun.com \
--docker-password=vipsoft \
--docker-email=xxxxx@qq.com
secret/registry-demo created
[root@k8smaster ~]# kubectl get secrets
NAME TYPE DATA AGE
default-token-q7lps kubernetes.io/service-account-token 3 5d17h
registry-demo kubernetes.io/dockerconfigjson 1 2m53s
[root@k8smaster ~]# kubectl describe secrets registry-demo
Name: registry-demo
Namespace: default
Labels: <none>
Annotations: <none>
Type: kubernetes.io/dockerconfigjson
Data
====
.dockerconfigjson: 196 bytes
[root@k8smaster ~]#
解释说明：
[root@k8smaster ~]# kubectl create secret docker-registry registry-demo \ #创建类型为：docker-registry的secrets名称为registry-demo
--namespace=default \ # K8S 的命名空间，非阿里云仓库的命名空间，值 default 可不写，默认为 default， 不同Namespace,在创建secrets时需要指定Namespace,
--docker-server=registry.cn-shanghai.aliyuncs.com \ #阿里云镜像仓库，公网地址
--docker-username=hi帐户ID@aliyun.com \ #阿里云的帐号
--docker-password=vipsoft \ #仓库的密码，在 仓库管理-> 访问凭证中可设置
--docker-email=xxxx@qq.com # 用户邮箱
[root@k8smaster ~]# kubectl get secret registry-demo -o yaml #查看secret详细信息
2. 生成 yaml ,并修改yaml 配置 imagePullSecrets,否则后面会报 ImagePullBackOff，拉取需要登录 requested access to the resource is denied
K8S 拉取镜像 ImagePullBackOff pull access denied
[root@k8smaster ~]# kubectl create deployment javademo1 --image=registry.cn-shanghai.aliyuncs.com/vipsoft/vipsoft:1.0 --dry-run -o yaml > javademo1.yaml
W1018 18:49:33.112150 62765 helpers.go:535] --dry-run is deprecated and can be replaced with --dry-run=client.
[root@k8smaster ~]# vi javademo1.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
creationTimestamp: null
labels:
app: javademo1
name: javademo1
spec:
replicas: 1
selector:
matchLabels:
app: javademo1
strategy: {}
template:
metadata:
creationTimestamp: null
labels:
app: javademo1
spec:
containers:
- image: registry.cn-shanghai.aliyuncs.com/vipsoft/vipsoft:1.0
name: vipsoft
resources: {}
imagePullSecrets:
- name: registry-demo
status: {}
3. 创建容器
[root@k8smaster ~]# kubectl apply -f javademo1.yaml
deployment.apps/javademo1 created
[root@k8smaster ~]# kubectl get pods
NAME READY STATUS RESTARTS AGE
javademo1-84dd5c9485-8ckk6 0/1 ContainerCreating 0 6s
nginx-f89759699-5hkdw 1/1 Running 0 5d17h
[root@k8smaster ~]# kubectl get pods
NAME READY STATUS RESTARTS AGE
javademo1-84dd5c9485-8ckk6 1/1 Running 0 65s
nginx-f89759699-5hkdw 1/1 Running 0 5d17h
[root@k8smaster ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
javademo1-84dd5c9485-8ckk6 1/1 Running 0 111s 10.244.1.4 k8snode1 <none> <none>
nginx-f89759699-5hkdw 1/1 Running 0 5d17h 10.244.2.2 k8snode2 <none> <none>
[root@k8smaster ~]# kubectl scale deployment javademo1 --replicas=3 # 扩容3个服务
deployment.apps/javademo1 scaled
[root@k8smaster ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
javademo1-84dd5c9485-7vgkr 0/1 ImagePullBackOff 0 16s 10.244.2.6 k8snode2 <none> <none>
javademo1-84dd5c9485-8ckk6 1/1 Running 0 6m21s 10.244.1.4 k8snode1 <none> <none>
javademo1-84dd5c9485-8hfmd 0/1 ImagePullBackOff 0 16s 10.244.2.7 k8snode2 <none> <none>
nginx-f89759699-5hkdw 1/1 Running 0 5d17h 10.244.2.2 k8snode2 <none> <none>
[root@k8smaster ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
javademo1-84dd5c9485-7vgkr 1/1 Running 0 58s 10.244.2.6 k8snode2 <none> <none>
javademo1-84dd5c9485-8ckk6 1/1 Running 0 7m3s 10.244.1.4 k8snode1 <none> <none>
javademo1-84dd5c9485-8hfmd 1/1 Running 0 58s 10.244.2.7 k8snode2 <none> <none>
nginx-f89759699-5hkdw 1/1 Running 0 5d17h 10.244.2.2 k8snode2 <none> <none>
[root@k8smaster ~]# kubectl expose deployment javademo1 --port=8111 --target-port=8111 --type=NodePort # 对外暴露端口，不然不能访问
service/javademo1 exposed
[root@k8smaster ~]# kubectl get svc #查看 service 的运行状态
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
javademo1 NodePort 10.106.43.46 <none> 8111:31452/TCP 25s
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 5d17h
nginx NodePort 10.103.87.81 <none> 80:30339/TCP 5d17h
[root@k8smaster ~]#
4. 测试
5. 补充
CrashLoopBackOff
| 容器退出，kubelet正在将它重启
|
InvalidImageName
| 无法解析镜像名称
|
ImageInspectError
| 无法校验镜像
|
ErrImageNeverPul
| 策略禁止拉取镜像
|
ImagePullBackOff
| 正在重试拉取
|
RegistryUnavailable
| 连接不到镜像中心
|
ErrImagePull
| 通用的拉取镜像出错
|
CreateContainerConfigError
| 不能创建kubelet使用的容器配置
|
CreateContainerError
| 创建容器失败
|
m.internalLifecycle.PreStartContainer
| 执行hook报错
|
RunContainerError
| 启动容器失败
|
PostStartHookError
| 执行hook报错
|
ContainersNotInitialized
| 容器没有初始化完毕
|
ContainersNotReady
| 容器没有准备完毕
|
ContainerCreating
| 容器创建中
|
PodInitializing
| pod初始化中
|
DockerDaemonNotReady
| docker还没有完全启动
|
NetworkPluginNotReady
| 网络插件还没有完全启动
|
Evicted
| 即驱赶的意思，意思是当节点出现异常时，kubernetes将有相应的机制驱赶该节点上的Pod。多见于资源不足时导致的驱赶。
|
