東川印記

一本東川，笑看爭龍斗虎；寰茫兦者，度橫佰昧人生。
senRsl

简单学习K8s

2022年7月12日星期二



学完k8s，就开始暂时停几天了。

近期学的有点多，需要消化消化。。。。

容器自动化部署技术 Kubernets

生产级别容器编排，按照一定目的依次排列，调度，安排。

Google 2014年容器编排引擎，go语言开发。

Go语言是Google2009年发布的开源编程语言。

K8S管理员认证CKA。

官方文档，很详细，还支持中文 https://kubernetes.io/zh-cn/docs/home/


1，整体架构

users      -> master ------------------------------------------> nodes

                ->api server                                               -> kubelet

                -> scheduler                                              -> kube proxy

                -> clusterState Sotre(ETCD数据库)          -> Pod(Container Runtime)

                -> Controller ManangerServer

 master是k8s集群控制节点，对集群进行调度管理，接收集群外用户去集群操作请求。

nodes是集群工作节点Worker Node，运行用户业务应用容器。


2，环境搭建方式

1）minikube

2) kind

3) kubeadm

4) 二进制包 github

5) yum

6）三方封装工具

7） 阿里云公有云平台K8s


3，使用kubeadm 部署 ks8

环境要求 centos7+、2G+2核、集群内通信，禁止swap分区。

Apple M1安不上centos，终于安上了ubuntu server....

1）安docker

有文档 https://docs.docker.com/engine/install/ubuntu/

senrsl@senrsl:~$ sudo apt-get remove docker docker-engine docker.io containerd runc

senrsl@senrsl:~$ sudo apt-get update

senrsl@senrsl:~$ sudo apt-get install \
>     ca-certificates \
>     curl \
>     gnupg \
>     lsb-release

senrsl@senrsl:~$ sudo mkdir -p /etc/apt/keyrings
senrsl@senrsl:~$ curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg
senrsl@senrsl:~$ echo \
>   "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu \
>   $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

senrsl@senrsl:~$ sudo apt-get update

senrsl@senrsl:~$ sudo apt-get install docker-ce docker-ce-cli containerd.io docker-compose-plugin

senrsl@senrsl:~$ docker --version
Docker version 20.10.17, build 100c701
senrsl@senrsl:~$

docker 有了。。。。

2）关防火墙

senrsl@senrsl:~$ sudo ufw status

Status: inactive
senrsl@senrsl:~$

默认就没开

3）切su权限

senrsl@senrsl:~/k8s$ sudo su -
root@senrsl:~# pwd
/root
root@senrsl:~# exit
logout
senrsl@senrsl:~/k8s$ sudo su
root@senrsl:/home/senrsl/k8s# docker images
REPOSITORY   TAG       IMAGE ID   CREATED   SIZE
root@senrsl:/home/senrsl/k8s#

4）再复制一台虚拟机

两台虚拟机，一台master、一台node

之前那台是 master 172.16.5.128

新复制的IP竟然一样。。。。断开网桥重启后再连接网桥

新复制的一台 node  172.168.5.129

5) 安kubeadm环境准备

Google 每周运行数十亿个容器，Kubernetes 基于与之相同的原则来设计，能够在不扩张运维团队的情况下进行规模扩展。

同样有文档 https://kubernetes.io/zh-cn/docs/setup/production-environment/tools/kubeadm/install-kubeadm/

修改主机名，主机名不能重复
root@senrsl:/home/senrsl/k8s# hostnamectl set-hostname senrslk8snode
root@senrsl:/home/senrsl/k8s# hostnamectl
   Static hostname: senrslk8snode
         Icon name: computer-vm
           Chassis: vm
        Machine ID: b6d041a2c6b04f0390bb4963c095b713
           Boot ID: 0830859f3e95471a905640d592904655
    Virtualization: vmware
  Operating System: Ubuntu 20.04.4 LTS
            Kernel: Linux 5.4.0-100-generic
      Architecture: arm64
root@senrsl:/home/senrsl/k8s#

6）设置iptables桥接流量

oot@senrsl:/home/senrsl# cat <<EOF | sudo tee /etc/modules-load.d/k8s.conf
> br_netfilter
> EOF
br_netfilter
root@senrsl:/home/senrsl# cat <<EOF | sudo tee /etc/sysctl.d/k8s.conf
> net.bridge.bridge-nf-call-ip6tables = 1
> net.bridge.bridge-nf-call-iptables = 1
> EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
root@senrsl:/home/senrsl# sysctl --system

7）安装kubeadm工具

root@senrsl:/home/senrsl# apt-get install -y apt-transport-https ca-certificates curl

root@senrsl:/home/senrsl#  curl -fsSLo /usr/share/keyrings/kubernetes-archive-keyring.gpg https://packages.cloud.google.com/apt/doc/apt-key.gpg
root@senrsl:/home/senrsl# echo "deb [signed-by=/usr/share/keyrings/kubernetes-archive-keyring.gpg] https://apt.kubernetes.io/ kubernetes-xenial main" | sudo tee /etc/apt/sources.list.d/kubernetes.list
root@senrsl:/home/senrsl# apt-get update

root@senrsl:/home/senrsl# apt-get install -y kubelet kubeadm kubectl

root@senrsl:/home/senrsl/k8s#  apt-mark hold kubelet kubeadm kubectl

8）reboot

重启后IP竟然又重复了，apple M1版的 vmware果然比inetl版的不稳定很多。。。。

之前复制虚拟机早了，应该在这里复制。。。。

9）确认安装

senrsl@senrsl:~$ kubelet --version
Kubernetes v1.24.2
senrsl@senrsl:~$

主要组件

kubelet：运行在所有cluster节点上，负责启动POD和容器；

kubeadm：用于初始化cluster的工具。

kubectl：是k8s的命令行工具，通过kubectl部署和管理应用。


10）部署k8s master主节点

之前的需要在所有机器运行，下面的就要分机器了。

master机器执行

root@senrsl:/home/senrsl/k8s# kubeadm init --kubernetes-version=1.24.2 --apiserver-advertise-address=172.16.5.128  --image-repository registry.aliyuncs.com/google_containers --service-cidr=10.96.0.0/12 --pod-network-cidr=10.244.0.0/16  --ignore-preflight-errors=Swap

。。。。成功

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join 172.16.5.128:6443 --token s9fwin.wyrz6fsm8swlvir0 \
        --discovery-token-ca-cert-hash sha256:7348dccbb862c71a41e6364b9db2f7d9012663ff87b171d61c7f8718b9b192b2
root@senrsl:/home/senrsl/k8s#

各种报错

Status from runtime service failed" err="rpc error: code = Unimplemented desc = unknown service runtime.v1alpha2.RuntimeService

root@senrsl:/home/senrsl/k8s# rm /etc/containerd/config.toml
root@senrsl:/home/senrsl/k8s# systemctl restart containerd

This error is likely caused by:
        - The kubelet is not running
        - The kubelet is unhealthy due to a misconfiguration of the node in some way (required cgroups disabled)

root@senrsl:/home/senrsl/k8s# systemctl status kubelet
● kubelet.service - kubelet: The Kubernetes Node Agent
     Loaded: loaded (/lib/systemd/system/kubelet.service; enabled; vendor preset: enabled)
    Drop-In: /etc/systemd/system/kubelet.service.d
             └─10-kubeadm.conf
     Active: activating (auto-restart) (Result: exit-code) since Mon 2022-07-11 07:39:59 UTC; 9s ago
       Docs: https://kubernetes.io/docs/home/
    Process: 5982 ExecStart=/usr/bin/kubelet $KUBELET_KUBECONFIG_ARGS $KUBELET_CONFIG_ARGS $KUBELET_KUBEADM_ARGS $KUBELET_EXTRA_ARGS (code=exit>
   Main PID: 5982 (code=exited, status=1/FAILURE)

root@senrsl:/home/senrsl/k8s#

禁用swap

root@senrsl:/home/senrsl/k8s# swapoff -a
root@senrsl:/home/senrsl/k8s# vi /etc/fstab  //注释最后一行
root@senrsl:/home/senrsl/k8s#

FileAvailable--etc-kubernetes-manifests-kube-scheduler.yaml]: /etc/kubernetes/manifests/kube-scheduler.yaml already exists

root@senrsl:/home/senrsl/k8s# kubeadm reset

终于成功了。。。。

master继续执行

root@senrsl:/home/senrsl/k8s# mkdir -p $HOME/.kube
root@senrsl:/home/senrsl/k8s# sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
root@senrsl:/home/senrsl/k8s# sudo chown $(id -u):$(id -g) $HOME/.kube/config
root@senrsl:/home/senrsl/k8s# kubectl get nodes
NAME     STATUS     ROLES           AGE     VERSION
senrsl   NotReady   control-plane   3m32s   v1.24.2
root@senrsl:/home/senrsl/k8s#

11）k8s node加入

同样的，需要关掉swap

root@senrslk8snode:/home/senrsl/k8s# kubeadm join 172.16.5.128:6443 --token s9fwin.wyrz6fsm8swlvir0 \>         --discovery-token-ca-cert-hash sha256:7348dccbb862c71a41e6364b9db2f7d9012663ff87b171d61c7f8718b9b192b2
[preflight] Running pre-flight checks
[preflight] Reading configuration from the cluster...
[preflight] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -o yaml'
[kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env"
[kubelet-start] Starting the kubelet
[kubelet-start] Waiting for the kubelet to perform the TLS Bootstrap...

This node has joined the cluster:
* Certificate signing request was sent to apiserver and a response was received.
* The Kubelet was informed of the new secure connection details.

Run 'kubectl get nodes' on the control-plane to see this node join the cluster.

root@senrslk8snode:/home/senrsl/k8s#

然后在master 可以看到这个node了。。。。

root@senrsl:/home/senrsl/k8s# kubectl get nodes
NAME            STATUS     ROLES           AGE     VERSION
senrsl          NotReady   control-plane   7m52s   v1.24.2
senrslk8snode   NotReady   <none>          38s     v1.24.2
root@senrsl:/home/senrsl/k8s#

12）master机器部署网络插件

root@senrsl:/home/senrsl/k8s# curl -O https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
root@senrsl:/home/senrsl/k8s# kubectl apply -f kube-flannel.yml

13）等待就绪
root@senrsl:/home/senrsl/k8s# kubectl get nodes
NAME            STATUS     ROLES           AGE     VERSION
senrsl          NotReady   control-plane   10m     v1.24.2
senrslk8snode   NotReady   <non

等状态由NotReady变成Ready

很快，两分钟

root@senrsl:/home/senrsl/k8s# kubectl get nodes
NAME            STATUS   ROLES           AGE     VERSION
senrsl          Ready    control-plane   11m     v1.24.2
senrslk8snode   Ready    <none>          4m45s   v1.24.2
root@senrsl:/home/senrsl/k8s#

14）查看pod

root@senrsl:/home/senrsl/k8s# kubectl get pods -n kube-system
NAME                             READY   STATUS    RESTARTS   AGE
coredns-74586cf9b6-5w4m4         1/1     Running   0          12m
coredns-74586cf9b6-6k9h9         1/1     Running   0          12m
etcd-senrsl                      1/1     Running   1          12m
kube-apiserver-senrsl            1/1     Running   1          12m
kube-controller-manager-senrsl   1/1     Running   1          12m
kube-proxy-nspjc                 1/1     Running   0          12m
kube-proxy-vrdkb                 1/1     Running   0          5m29s
kube-scheduler-senrsl            1/1     Running   1          12m
root@senrsl:/home/senrsl/k8s#

四点了，买菜做核酸。。。。

4，k8s部署容器化应用步骤

1）Dockerfile制作镜像，或从仓库拉取镜像；

2）通过控制器管理pod，镜像启动到一个容器，容器在pod里；

3）暴露应用，以使外界访问。

5，Pod部署Nginx

root@senrsl:/home/senrsl/k8s# kubectl create deployment nginx --image=nginx
deployment.apps/nginx created
root@senrsl:/home/senrsl/k8s# kubectl expose deploy nginx --port=80 --type=NodePort
service/nginx exposed

查看命令

root@senrsl:/home/senrsl/k8s# kubelctl get pod,svc
kubelctl: command not found
root@senrsl:/home/senrsl/k8s# kubectl get pod,svc
NAME                        READY   STATUS    RESTARTS   AGE
pod/nginx-8f458dc5b-ml78k   1/1     Running   0          43s

NAME                 TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
service/kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        4h19m
service/nginx        NodePort    10.96.126.74   <none>        80:30811/TCP   15s
root@senrsl:/home/senrsl/k8s# kubectl get node
NAME            STATUS   ROLES           AGE     VERSION
senrsl          Ready    control-plane   4h20m   v1.24.2
senrslk8snode   Ready    <none>          4h13m   v1.24.2
root@senrsl:/home/senrsl/k8s# kubectl get service
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        4h20m
nginx        NodePort    10.96.126.74   <none>        80:30811/TCP   68s
root@senrsl:/home/senrsl/k8s# kubectl get deployment
NAME    READY   UP-TO-DATE   AVAILABLE   AGE
nginx   1/1     1            1           116s
root@senrsl:/home/senrsl/k8s# kubectl get pod
NAME                    READY   STATUS    RESTARTS   AGE
nginx-8f458dc5b-ml78k   1/1     Running   0          2m3s

访问

http://172.16.5.128:30811/

http://172.16.5.129:30811/

其中128是master，刚才的操作在master上执行，129是node，node看来是自动同步了

6，pod部署tomcat

root@senrsl:/home/senrsl/k8s# kubectl create deployment tomcat --image=tomcat
deployment.apps/tomcat created
root@senrsl:/home/senrsl/k8s# kubectl expose deployment tomcat --port=8080 --type=NodePort
service/tomcat exposed
root@senrsl:/home/senrsl/k8s# kubectl get pod,svc
NAME                         READY   STATUS              RESTARTS   AGE
pod/nginx-8f458dc5b-ml78k    1/1     Running             0          10m
pod/tomcat-84c98b5bb-m629j   0/1     ContainerCreating   0          32s

NAME                 TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
service/kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP          4h29m
service/nginx        NodePort    10.96.126.74    <none>        80:30811/TCP     10m
service/tomcat       NodePort    10.104.63.100   <none>        8080:30239/TCP   11s
root@senrsl:/home/senrsl/k8s#

访问

http://172.16.5.128:30239/

http://172.16.5.129:30239/

执行之后需要等一会，部署需要时间

7，关系

master控制Node节点 -> service -> deployment控制器 -> pod -> docker

8，部署项目

1）命令部署

kubectl create deployment 项目名 --image=项目镜像 --dry-run -o yaml

-o 可以为yaml或json

kubectl create deployment 项目名 --image=项目镜像 --dry-run -o yaml > deploy.yaml

保存到本地yaml文件

2）yml部署

kubectl apply -f deploy.yaml

等于

kubectl create deplyment 项目名 --image=镜像名

其中yml文件中需要添加设置为从本地拉取，默认是从远程拉取

containers:

    imagePullPolicy:Never //设置为从本地拉取镜像


部署后，同样的暴露服务后访问

暴露服务也可以通过 --dry-run -o yaml 空运行生成文件

9，部署 dashboard

源码 https://github.com/kubernetes/dashboard

通过yml方式部署

root@senrsl:/home/senrsl/k8s# kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.6.0/aio/deploy/recommended.yaml
namespace/kubernetes-dashboard created
serviceaccount/kubernetes-dashboard created
service/kubernetes-dashboard created
secret/kubernetes-dashboard-certs created
secret/kubernetes-dashboard-csrf created
secret/kubernetes-dashboard-key-holder created
configmap/kubernetes-dashboard-settings created
role.rbac.authorization.k8s.io/kubernetes-dashboard created
clusterrole.rbac.authorization.k8s.io/kubernetes-dashboard created
rolebinding.rbac.authorization.k8s.io/kubernetes-dashboard created
clusterrolebinding.rbac.authorization.k8s.io/kubernetes-dashboard created
deployment.apps/kubernetes-dashboard created
service/dashboard-metrics-scraper created
deployment.apps/dashboard-metrics-scraper created
root@senrsl:/home/senrsl/k8s#

确认部署成功

root@senrsl:/home/senrsl/k8s# kubectl get pod -n kubernetes-dashboard
NAME                                        READY   STATUS    RESTARTS   AGE
dashboard-metrics-scraper-8c47d4b5d-mt2l8   1/1     Running   0          3m20s
kubernetes-dashboard-5676d8b865-rvm9v       1/1     Running   0          3m20s
root@senrsl:/home/senrsl/k8s#

然后发现这个地址只能在虚拟机里访问，外部访问不了。。。。

需要修改 recommonded.yaml

spec:
  type: NodePort
  ports:
    - port: 443
      targetPort: 8443
      nodePort: 30001
  selector:
    k8s-app: kubernetes-dashboard

然后重新应用

root@senrsl:/home/senrsl/k8s# vi recommended.yaml
root@senrsl:/home/senrsl/k8s# kubectl apply -f recommended.yaml

这时候再查看，发现暴露了30001端口

root@senrsl:/home/senrsl/k8s# kubectl get pod,svc  -n kubernetes-dashboard
NAME                                            READY   STATUS    RESTARTS   AGE
pod/dashboard-metrics-scraper-8c47d4b5d-mt2l8   1/1     Running   0          13m
pod/kubernetes-dashboard-5676d8b865-rvm9v       1/1     Running   0          13m

NAME                                TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)         AGE
service/dashboard-metrics-scraper   ClusterIP   10.100.134.163   <none>        8000/TCP        13m
service/kubernetes-dashboard        NodePort    10.103.234.165   <none>        443:30001/TCP   13m
root@senrsl:/home/senrsl/k8s#

访问 https://172.16.5.128:30001/

需要输入token

生成token

root@senrsl:/home/senrsl/k8s# kubectl create serviceaccount dashboard-admin -n kube-system
serviceaccount/dashboard-admin created
root@senrsl:/home/senrsl/k8s# kubectl create clusterrolebinding dashboard-admin --clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin
clusterrolebinding.rbac.authorization.k8s.io/dashboard-admin created
root@senrsl:/home/senrsl/k8s# kubectl describe secrets -n kube-system $(kubectl -n kube-system get secret | awk '/dashboard-admin/{print $1}')

执行完并没有生成token

看起来是 高版本改了，需要指定失效时间

root@senrsl:/home/senrsl/k8s# kubectl create token dashboard-admin -n kube-system --duration=87600h
eyJhbGciOiJSUzI1NiIsImtpZCI6IkJncDBwdTJWQURiRUl0YzdfMk1rWUhvT1pvd1FHMjg4VG9Vdl9peW9QUkUifQ.eyJhdWQiOlsiaHR0cHM6Ly9rdWJlcm5ldGVzLmRlZmF1bHQuc3ZjLmNsdXN0ZXIubG9jYWwiXSwiZXhwIjoxOTcyOTA3NzMxLCJpYXQiOjE2NTc1NDc3MzEsImlzcyI6Imh0dHBzOi8va3ViZXJuZXRlcy5kZWZhdWx0LnN2Yy5jbHVzdGVyLmxvY2FsIiwia3ViZXJuZXRlcy5pbyI6eyJuYW1lc3BhY2UiOiJrdWJlLXN5c3RlbSIsInNlcnZpY2VhY2NvdW50Ijp7Im5hbWUiOiJkYXNoYm9hcmQtYWRtaW4iLCJ1aWQiOiI3M2M2M2QxYS1jYmVlLTQ5MDctOGE3Yy03MmMzYTQyNGIwYTAifX0sIm5iZiI6MTY1NzU0NzczMSwic3ViIjoic3lzdGVtOnNlcnZpY2VhY2NvdW50Omt1YmUtc3lzdGVtOmRhc2hib2FyZC1hZG1pbiJ9.HCP1KeHFTeDh9Fw8Qg6kcaw0ycvTGajt6zfie0ZvyQcLhkzuvaed4FGN_tKwS5ydNwGs7ZCvsnVv3UjABUi3k3ssUECbs0FMNwWXpFgai4jCoKqT2hvgJ4ZenoIA5TZgSJHsDtSWrX5hOUSqpTFHyz2a23lFfcOdo79pt1WmofOaA4b2vO8zLRpKJNwiokAxFbGJ1SJHZynsHpQycKXQ8KtCINXUaYCQr5XtP0Pv_7glwiv3QNSB7E9X55XHdkjl01goHHZMx4V9YI55xiTNZL2tEsl3CF0IxsgfuyTE_PPcidRMjNAL03hd0yh52Xr9GPa45ziHcSILuwIXh_MXrQ
root@senrsl:/home/senrsl/k8s#

复制token，登录dashboard。

配图1

10，harbor

docker私服

11，暴露应用的几种方式

1）NodePort，每个节点产生30000以上端口

    只能供一个服务使用，只能使用30000-32767之间端口，IP变化后需要手动修改。


yaml文件中三种端口

nodePort：外部机器可以访问的端口。

targetPort:容器内端口，与制作容器是暴露端口一致。

port: 各服务之间访问的端口。


2）LoadBalancer

负载均衡器，深度耦合云平台

3）Ingress

相当于网关， 用户 -> ingress -> 多个service -> 多个pod

Ingress是k8s集群中一个API资源对象，相当于一个集群网关。需要单独安装。

12，Ingress Nginx

https://kubernetes.io/zh-cn/docs/concepts/services-networking/ingress/

https://github.com/kubernetes/ingress-nginx

1）环境确认

root@senrsl:/home/senrsl/k8s# kubectl delete service nginx
service "nginx" deleted
root@senrsl:/home/senrsl/k8s# kubectl delete deployment nginx
deployment.apps "nginx" deleted
root@senrsl:/home/senrsl/k8s# kubectl get pods
NAME                     READY   STATUS    RESTARTS   AGE
tomcat-84c98b5bb-m629j   1/1     Running   0          126m
root@senrsl:/home/senrsl/k8s# kubectl get node
NAME            STATUS   ROLES           AGE     VERSION
senrsl          Ready    control-plane   6h36m   v1.24.2
senrslk8snode   Ready    <none>          6h28m   v1.24.2
root@senrsl:/home/senrsl/k8s# kubectl get service
NAME         TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)          AGE
kubernetes   ClusterIP   10.96.0.1       <none>        443/TCP          6h36m
tomcat       NodePort    10.104.63.100   <none>        8080:30239/TCP   126m
root@senrsl:/home/senrsl/k8s# kubectl get deploy
NAME     READY   UP-TO-DATE   AVAILABLE   AGE
tomcat   1/1     1            1           127m
root@senrsl:/home/senrsl/k8s# kubectl get pod
NAME                     READY   STATUS    RESTARTS   AGE
tomcat-84c98b5bb-m629j   1/1     Running   0          127m
root@senrsl:/home/senrsl/k8s#

2）安nginx

root@senrsl:/home/senrsl/k8s# kubectl create deployment nginx --image=nginx
deployment.apps/nginx created
root@senrsl:/home/senrsl/k8s# kubectl expose deployment nginx --port=80 --target-port=80 --type=NodePort
service/nginx exposed
root@senrsl:/home/senrsl/k8s# 

3）安nginx ingress

https://kubernetes.github.io/ingress-nginx/deploy/

root@senrsl:/home/senrsl/k8s# kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.2.1/deploy/static/provider/cloud/deploy.yaml


补充

https://raw.githubusercontent.com/kubernetes/ingress-nginx/controller-v1.2.1/deploy/static/provider/cloud/deploy.yaml

需要修改

wait-shutdown上面标签为containers再上面标签spec

spec标签内添加hostNetwork:true

重新删除安装

删除service

root@senrsl:/home/senrsl/k8s# kubectl get service -n ingress-nginx
NAME                                 TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                      AGE
ingress-nginx-controller             LoadBalancer   10.99.134.145   <pending>     80:30663/TCP,443:30307/TCP   31m
ingress-nginx-controller-admission   ClusterIP      10.110.30.48    <none>        443/TCP                      31m
root@senrsl:/home/senrsl/k8s# kubectl delete service ingress-nginx-controller -n ingress-nginx
service "ingress-nginx-controller" deleted
root@senrsl:/home/senrsl/k8s# kubectl delete service ingress-nginx-controller-admission -n ingress-nginx
service "ingress-nginx-controller-admission" deleted
root@senrsl:/home/senrsl/k8s# kubectl get service -n ingress-nginx
No resources found in ingress-nginx namespace.
root@senrsl:/home/senrsl/k8s#

删除deployment

root@senrsl:/home/senrsl/k8s# kubectl get deploy -n ingress-nginx
NAME                       READY   UP-TO-DATE   AVAILABLE   AGE
ingress-nginx-controller   1/1     1            1           29m
root@senrsl:/home/senrsl/k8s# kubectl delete deploy ingress-nginx-controller -n ingress-nginx
deployment.apps "ingress-nginx-controller" deleted
root@senrsl:/home/senrsl/k8s#

删除pod

root@senrsl:/home/senrsl/k8s# kubectl get pod -n ingress-nginx
NAME                                   READY   STATUS      RESTARTS   AGE
ingress-nginx-admission-create-jntdr   0/1     Completed   0          32m
ingress-nginx-admission-patch-dmzlh    0/1     Completed   0          32m
root@senrsl:/home/senrsl/k8s# kubectl delete pod ingress-nginx-admission-create-jntdr -n ingress-nginx
pod "ingress-nginx-admission-create-jntdr" deleted
root@senrsl:/home/senrsl/k8s# kubectl delete pod ingress-nginx-admission-patch-dmzlh -n ingress-nginx
pod "ingress-nginx-admission-patch-dmzlh" deleted
root@senrsl:/home/senrsl/k8s# kubectl get pod -n ingress-nginx
No resources found in ingress-nginx namespace.
root@senrsl:/home/senrsl/k8s#


4）安装确认

root@senrsl:/home/senrsl/k8s# kubectl get service -n ingress-nginx
NAME                                 TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                      AGE
ingress-nginx-controller             LoadBalancer   10.99.134.145   <pending>     80:30663/TCP,443:30307/TCP   55s
ingress-nginx-controller-admission   ClusterIP      10.110.30.48    <none>        443/TCP                      55s
root@senrsl:/home/senrsl/k8s# kubectl get deploy  -n ingress-nginx
NAME                       READY   UP-TO-DATE   AVAILABLE   AGE
ingress-nginx-controller   0/1     1            0           73s
root@senrsl:/home/senrsl/k8s# kubectl get pod  -n ingress-nginx
NAME                                        READY   STATUS              RESTARTS   AGE
ingress-nginx-admission-create-jntdr        0/1     ContainerCreating   0          82s
ingress-nginx-admission-patch-dmzlh         0/1     ContainerCreating   0          82s
ingress-nginx-controller-778667bc4b-qbrvw   0/1     ContainerCreating   0          82s
root@senrsl:/home/senrsl/k8s#

这次的时间明显要长很多。。。。

重新确认环境

root@senrsl:/home/senrsl/k8s# kubectl get service -n ingress-nginx
NAME                                 TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                      AGE
ingress-nginx-controller             LoadBalancer   10.111.24.240   <pending>     80:30665/TCP,443:31988/TCP   25s
ingress-nginx-controller-admission   ClusterIP      10.109.227.12   <none>        443/TCP                      25s
root@senrsl:/home/senrsl/k8s# kubectl get deploy  -n ingress-nginx
NAME                       READY   UP-TO-DATE   AVAILABLE   AGE
ingress-nginx-controller   1/1     1            1           40s
root@senrsl:/home/senrsl/k8s# kubectl get pod  -n ingress-nginx
NAME                                       READY   STATUS    RESTARTS   AGE
ingress-nginx-controller-7ffc958f5-5nn4n   1/1     Running   0          46s
root@senrsl:/home/senrsl/k8s# kubectl get pod  -n ingress-nginx
NAME                                       READY   STATUS    RESTARTS   AGE
ingress-nginx-controller-7ffc958f5-5nn4n   1/1     Running   0          69s
root@senrsl:/home/senrsl/k8s#


5）应用ingress规则

root@senrsl:/home/senrsl/k8s# kubectl apply -f ingress_nginx_rule.yml
ingress.networking.k8s.io/k8s-ingress created
root@senrsl:/home/senrsl/k8s# kubectl get ing
NAME          CLASS    HOSTS         ADDRESS   PORTS   AGE
k8s-ingress   <none>   www.abc.com             80      6s
root@senrsl:/home/senrsl/k8s#


yml文件

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: k8s-ingress
spec:
  rules:
  - host: www.abc.com
    http:
      paths:
      - pathType: Prefix
        path: /
        backend:
          service:
            name: nginx
            port:
              number: 80

还是需要修改本机host....

看起来是把nginx service的端口80,绑定到了host上。。。。

然后通过get ingress应该可以看到实际指向地址才对。。。。。

文档 https://kubernetes.github.io/ingress-nginx/user-guide/basic-usage/

13，部署流程

1）项目打包 jar或war;

2) 制作项目镜像（Dockerfile文件）;

3) 用k8s部署镜像(命令方式或yml方式);

4)对外暴露服务(nodePort临时，Ingress统一暴露)；


14，写个简单的SpringCloudAlibaba项目

提供者provider，消费者consumer，网关gateway。

安装nacos

按照Spring Cloud Alibaba的版本说明，选用

Spring Cloud Alibaba 2.2.8.RELEASE

SpringCloud Hoxton.SR12

SpringBoot 2.3.12.RELEASE

Sentinel 1.8.4

Nacos 2.1.0

RocketMQ 4.9.3

Seata 1.5.1

1）安装nacos

下载2.1.0 https://github.com/alibaba/nacos/releases

解压，运行

root@senrsl:/home/senrsl/nacos/nacos/bin# java -version
openjdk version "1.8.0_312"
OpenJDK Runtime Environment (build 1.8.0_312-8u312-b07-0ubuntu1~20.04-b07)
OpenJDK 64-Bit Server VM (build 25.312-b07, mixed mode)
root@senrsl:/home/senrsl/nacos/nacos/bin# ./startup.sh -m standalone
/usr/lib/jvm/java-8-openjdk-arm64/bin/java -Djava.ext.dirs=/usr/lib/jvm/java-8-openjdk-arm64/jre/lib/ext:/usr/lib/jvm/java-8-openjdk-arm64/lib/ext  -Xms512m -Xmx512m -Xmn256m -Dnacos.standalone=true -Dnacos.member.list= -Xloggc:/home/senrsl/nacos/nacos/logs/nacos_gc.log -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCDateStamps -XX:+PrintGCTimeStamps -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 -XX:GCLogFileSize=100M -Dloader.path=/home/senrsl/nacos/nacos/plugins/health,/home/senrsl/nacos/nacos/plugins/cmdb,/home/senrsl/nacos/nacos/plugins/selector -Dnacos.home=/home/senrsl/nacos/nacos -jar /home/senrsl/nacos/nacos/target/nacos-server.jar  --spring.config.additional-location=file:/home/senrsl/nacos/nacos/conf/ --logging.config=/home/senrsl/nacos/nacos/conf/nacos-logback.xml --server.max-http-header-size=524288
nacos is starting with standalone
nacos is starting，you can check the /home/senrsl/nacos/nacos/logs/start.out
root@senrsl:/home/senrsl/nacos/nacos/bin#


访问 http://172.16.5.128:8848/nacos/index.html


2）内容提供者provider

Request URI does not contain a valid hostname

启用Ribbon的时候，服务名不能包含下划线。。。。屮

controller

@RefreshScope
@RestController
public class EchoController {

//    @Value("${spring.application.name}")
    @Value("${user.name}")  //这俩是配置在了nacos配置列表里
    private String name;

//    @Value("${server.port}")
    @Value("${user.age}")
    private int age;


    @RequestMapping("/echo")
    public String echo() {
        return "读取到的信息 " + name + "  " + age;
    }

}

bootstrap.yml

# 中文乱码
spring:
  application:
    name: test-provider
  cloud:
    nacos:
      # 配置
      config:
        server-addr: 172.16.5.128:8848
      # 发现
      discovery:
        server-addr: 172.16.5.128:8848

3）消费者consumer

controller

@Slf4j
@RefreshScope  //nacos配置中心更改了配置值，自动感知并更新到配置修改
@RestController
public class EchoController {

    @Value("${user.name}")
    private String name;

    @Value("${user.age}")
    private int age;

    @Value("${provider.server.name}")
    private String serviceName;
    @Autowired
    private RestTemplate restTemplate;


    //使用feign方式 feign 4-2
    @Autowired
    private EchoServiceFeignClient echoServiceFeign;

    @RequestMapping("/echo")
    public String echo() {
        log.info("读取到的 配置 {}  and {} and {}", name, age, serviceName);
        String resultAll = "本服务获取到的 " + name + "   " + age + "  " + serviceName + "  ";

        //restTemplate
        String url = "http://" + serviceName + "/echo";
//        String url = "http://test-provider/echo"; //服务名不能用下划线。。。。
        String result = restTemplate.getForObject(url, String.class);
        log.info("rest template is {}", result);


        //feign 4-3
        String result2 = echoServiceFeign.echo();

        resultAll = resultAll + " \n restTemp方式： " + result + " \n feign方式：" + result2;
        return resultAll;
    }

}

feign client

@FeignClient("test-provider")
public interface EchoServiceFeignClient {

    //或者这里写/echo，但是需要@FeignClient(name = "test-provider",url = "http://localhost:8080/test-provider") 这种格式写全路径
    @GetMapping("/echo")
    String echo();

}

启动

@SpringBootApplication
@EnableFeignClients //feign 4-4
public class TestConsumerApplication {

    public static void main(String[] args) {
        SpringApplication.run(TestConsumerApplication.class, args);
    }


    @Bean
    @LoadBalanced
    public RestTemplate getRestTemplate() {
        return new RestTemplate();
    }

}

bootstrap.yml

spring:
  application:
    name: test-consumer
  cloud:
    nacos:
      # 配置
      config:
        server-addr: 172.16.5.128:8848
        # 发现
      discovery:
        server-addr: 172.16.5.128:8848
server:
  port: 8090

4)gateway 网关

application.yml

server:
  port: 80

spring:
  application:
    name: test-gateway
  cloud:
    nacos:
      discovery:
        server-addr: 172.16.5.128:8848
    gateway:
      discovery:
        locator:
          enabled: true # 网关发现
      routes:
        - id: route111
          uri: lb://test-consumer
          predicates: #判断
            - Path=/**
            #- After=2020-08-15T22:35:25.335+08:00[Asia/Shanghai]
            #- Before=2020-09-15T17:35:25.335+08:00[Asia/Shanghai]
            #- Between=2020-08-13T17:35:25.335+08:00[Asia/Shanghai], 2020-08-14T17:35:25.335+08:00[Asia/Shanghai]
            #- Cookie=token, 123456
            #- Header=X-Request-Id, \d+
            #- Query=token
            #- Token=123456
            #- AccessTime=������6:00, ������6:00
          filters:
            - AddRequestHeader=X-Request-red, blue
            - AddRequestParameter=color, red


生成这三个的jar包，准备部署到k8s.....


15，手动部署这个项目

实际中都是绑定工具自动部署

k8s适合部署无状态服务。

一般nacos、mysql等属于有状态服务，一般直接安在机器上，不走k8s。


分别产生Dockerfile，部署到k8s.

在gateway上部署Ingress，统一入口。

编写Dockerfile_provider

FROM dcjz/jdk:1.8u202v5
# ARG JAR_FILE
ADD test_provider-0.0.1-SNAPSHOT.jar /
# EXPOSE 8001 9015
ENTRYPOINT ["java", "-jar","/test_provider-0.0.1-SNAPSHOT.jar"]

生成镜像

root@senrsl:/home/senrsl/k8s# docker build -t test-provider -f Dockerfile_provider .
root@senrsl:/home/senrsl/k8s# docker images
REPOSITORY      TAG         IMAGE ID       CREATED          SIZE
test-provider   latest      b9900bb4f3e0   20 seconds ago   489MB
dcjz/jdk        1.8u202v5   cf99cba415ec   4 days ago       439MB
root@senrsl:/home/senrsl/k8s#

k8s部署

生成yml文件

root@senrsl:/home/senrsl/k8s# kubectl create deployment test-provider --image=test-provider --dry-run=client -o yaml > test-provider.yaml

修改镜像为本地拉取

    spec:
      containers:
      - image: test-provider
        name: test-provider
        imagePullPolicy: Never

应用yml文件

root@senrsl:/home/senrsl/k8s# kubectl apply -f test-provider.yaml

查看部署状态为ErrImageNeverPull

从节点上也需要有镜像才可以。。。。不然分发到从节点，从节点找不到该镜像，pod就报这个错误。

确切说，应该是只部署在node节点就可以，在node节点工作。

后面还需要搭个镜像存储服务。。。。

provider不需要对外暴露，所以不需要对外expose端口

consumer、gateway也都要走上面流程

暴露端口


错误1：

vmware ubuntu 重启了一下不能启动了，研究了一下，应该是apt update的时候更新内核了

卡在 exiting boot service and installing virutal address map

在启动的时候，选择boot option，选低版本的内核5.4.0.100就能启动了。。。。

错误2：

The connection to the server localhost:8080 was refused - did you specify the right host or port?

因为node没有配置环境，之前也没配置。。。。

查看状态

root@senrsl:/home/senrsl/k8s# kubectl describe pod test-provider-6c6f7d6cd9-l8xc6
Name:         test-provider-6c6f7d6cd9-l8xc6
Namespace:    default
Priority:     0
Node:         senrslk8snode/172.16.5.129
Start Time:   Tue, 12 Jul 2022 14:15:30 +0000
Labels:       app=test-provider
              pod-template-hash=6c6f7d6cd9
Annotations:  <none>
Status:       Pending
IP:           10.244.1.21
IPs:
  IP:           10.244.1.21
Controlled By:  ReplicaSet/test-provider-6c6f7d6cd9
Containers:
  test-provider:
    Container ID:  
    Image:          test-provider
    Image ID:      
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       ErrImageNeverPull
    Ready:          False
    Restart Count:  0
    Environment:    <none>
    Mounts:
      /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-lxz64 (ro)
Conditions:
  Type              Status
  Initialized       True
  Ready             False
  ContainersReady   False
  PodScheduled      True
Volumes:
  kube-api-access-lxz64:
    Type:                    Projected (a volume that contains injected data from multiple sources)
    TokenExpirationSeconds:  3607
    ConfigMapName:           kube-root-ca.crt
    ConfigMapOptional:       <nil>
    DownwardAPI:             true
QoS Class:                   BestEffort
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
  ----     ------             ----              ----               -------
  Warning  FailedScheduling   2m35s             default-scheduler  0/2 nodes are available: 1 node(s) had untolerated taint {node-role.kubernetes.io/master: }, 1 node(s) had untolerated taint {node.kubernetes.io/disk-pressure: }. preemption: 0/2 nodes are available: 2 Preemption is not helpful for scheduling.
  Normal   Scheduled          61s               default-scheduler  Successfully assigned default/test-provider-6c6f7d6cd9-l8xc6 to senrslk8snode
  Warning  ErrImageNeverPull  6s (x6 over 61s)  kubelet            Container image "test-provider" is not present with pull policy of Never
  Warning  Failed             6s (x6 over 61s)  kubelet            Error: ErrImageNeverPull
root@senrsl:/home/senrsl/k8s#

错误3：

pod跟deploy全都不Ready了。

0/2 nodes are available: 1 node(s) had untolerated taint {node-role.kubernetes.io/master: }, 1 node(s) had untolerated taint {node.kubernetes.io/disk-pressure: }. preemption: 0/2 nodes are available: 2 Preemption is not helpful for scheduling.

查看是否被打污点

root@senrsl:/home/senrsl/k8s# kubectl describe node senrslk8snode | grep Taint
Taints:             <none>
root@senrsl:/home/senrsl/k8s# 

折腾半天，又变了

root@senrsl:/home/senrsl/k8s# kubectl describe node senrslk8snode | grep Taint
Taints:             node.kubernetes.io/disk-pressure:NoSchedule
root@senrsl:/home/senrsl/k8s#

污点值由三个

NoSchedule：一定不被调度
PreferNoSchedule：尽量不被调度【也有被调度的几率】
NoExecute：不会调度，并且还会驱逐Node已有Pod

奇怪，一会有一会没的。。。。


查看node状态

root@senrsl:/home/senrsl/k8s# kubectl describe node senrslk8snode
Name:               senrslk8snode
Roles:              <none>
Labels:             beta.kubernetes.io/arch=arm64
                    beta.kubernetes.io/os=linux
                    kubernetes.io/arch=arm64
                    kubernetes.io/hostname=senrslk8snode
                    kubernetes.io/os=linux
Annotations:        flannel.alpha.coreos.com/backend-data: {"VNI":1,"VtepMAC":"2a:0e:c7:c6:d6:4d"}
                    flannel.alpha.coreos.com/backend-type: vxlan
                    flannel.alpha.coreos.com/kube-subnet-manager: true
                    flannel.alpha.coreos.com/public-ip: 172.16.5.129
                    kubeadm.alpha.kubernetes.io/cri-socket: unix:///var/run/containerd/containerd.sock
                    node.alpha.kubernetes.io/ttl: 0
                    volumes.kubernetes.io/controller-managed-attach-detach: true
CreationTimestamp:  Mon, 11 Jul 2022 07:53:55 +0000
Taints:             <none>
Unschedulable:      false
Lease:
  HolderIdentity:  senrslk8snode
  AcquireTime:     <unset>
  RenewTime:       Tue, 12 Jul 2022 14:33:08 +0000
Conditions:
  Type                 Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----                 ------  -----------------                 ------------------                ------                       -------
  NetworkUnavailable   False   Tue, 12 Jul 2022 21:56:22 +0000   Tue, 12 Jul 2022 21:56:22 +0000   FlannelIsUp                  Flannel is running on this node
  MemoryPressure       False   Tue, 12 Jul 2022 14:33:13 +0000   Tue, 12 Jul 2022 13:56:27 +0000   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure         False   Tue, 12 Jul 2022 14:33:13 +0000   Tue, 12 Jul 2022 14:28:50 +0000   KubeletHasNoDiskPressure     kubelet has no disk pressure
  PIDPressure          False   Tue, 12 Jul 2022 14:33:13 +0000   Tue, 12 Jul 2022 13:56:27 +0000   KubeletHasSufficientPID      kubelet has sufficient PID available
  Ready                True    Tue, 12 Jul 2022 14:33:13 +0000   Tue, 12 Jul 2022 13:56:27 +0000   KubeletReady                 kubelet is posting ready status. AppArmor enabled
Addresses:
  InternalIP:  172.16.5.129
  Hostname:    senrslk8snode
Capacity:
  cpu:                2
  ephemeral-storage:  10255636Ki
  hugepages-1Gi:      0
  hugepages-2Mi:      0
  hugepages-32Mi:     0
  hugepages-64Ki:     0
  memory:             4013104Ki
  pods:               110
Allocatable:
  cpu:                2
  ephemeral-storage:  9451594122
  hugepages-1Gi:      0
  hugepages-2Mi:      0
  hugepages-32Mi:     0
  hugepages-64Ki:     0
  memory:             3910704Ki
  pods:               110
System Info:
  Machine ID:                 b6d041a2c6b04f0390bb4963c095b713
  System UUID:                68b84d56-8698-0d32-dd9b-8bc1f03e5072
  Boot ID:                    0981e7e5-1bdc-4f59-a858-1463bbc6b333
  Kernel Version:             5.4.0-100-generic
  OS Image:                   Ubuntu 20.04.4 LTS
  Operating System:           linux
  Architecture:               arm64
  Container Runtime Version:  containerd://1.6.6
  Kubelet Version:            v1.24.2
  Kube-Proxy Version:         v1.24.2
PodCIDR:                      10.244.1.0/24
PodCIDRs:                     10.244.1.0/24
Non-terminated Pods:          (8 in total)
  Namespace                   Name                                         CPU Requests  CPU Limits  Memory Requests  Memory Limits  Age
  ---------                   ----                                         ------------  ----------  ---------------  -------------  ---
  default                     nginx-8f458dc5b-8b68r                        0 (0%)        0 (0%)      0 (0%)           0 (0%)         9m56s
  default                     test-provider-6c6f7d6cd9-jpwnk               0 (0%)        0 (0%)      0 (0%)           0 (0%)         9m5s
  default                     tomcat-84c98b5bb-s4xdc                       0 (0%)        0 (0%)      0 (0%)           0 (0%)         10m
  ingress-nginx               ingress-nginx-controller-7ffc958f5-mpsss     100m (5%)     0 (0%)      90Mi (2%)        0 (0%)         9m57s
  kube-flannel                kube-flannel-ds-8zm8g                        100m (5%)     100m (5%)   50Mi (1%)        50Mi (1%)      30h
  kube-system                 kube-proxy-vrdkb                             0 (0%)        0 (0%)      0 (0%)           0 (0%)         30h
  kubernetes-dashboard        dashboard-metrics-scraper-8c47d4b5d-h29sb    0 (0%)        0 (0%)      0 (0%)           0 (0%)         23m
  kubernetes-dashboard        kubernetes-dashboard-5676d8b865-8wp7b        0 (0%)        0 (0%)      0 (0%)           0 (0%)         9m54s
Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.)
  Resource           Requests    Limits
  --------           --------    ------
  cpu                200m (10%)  100m (5%)
  memory             140Mi (3%)  50Mi (1%)
  ephemeral-storage  0 (0%)      0 (0%)
  hugepages-1Gi      0 (0%)      0 (0%)
  hugepages-2Mi      0 (0%)      0 (0%)
  hugepages-32Mi     0 (0%)      0 (0%)
  hugepages-64Ki     0 (0%)      0 (0%)
Events:
  Type     Reason                   Age                            From             Message
  ----     ------                   ----                           ----             -------
  Normal   Starting                 <invalid>                      kube-proxy      
  Normal   NodeHasDiskPressure      56m                            kubelet          Node senrslk8snode status is now: NodeHasDiskPressure
  Warning  FreeDiskSpaceFailed      52m                            kubelet          failed to garbage collect required amount of images. Wanted to free 633982156 bytes, but freed 0 bytes
  Warning  EvictionThresholdMet     52m (x6 over 56m)              kubelet          Attempting to reclaim ephemeral-storage
  Normal   NodeNotReady             47m                            node-controller  Node senrslk8snode status is now: NodeNotReady
  Warning  EvictionThresholdMet     31m                            kubelet          Attempting to reclaim ephemeral-storage
  Warning  FreeDiskSpaceFailed      22m                            kubelet          failed to garbage collect required amount of images. Wanted to free 552373452 bytes, but freed 72942770 bytes
  Normal   NodeHasNoDiskPressure    11m (x15 over <invalid>)       kubelet          Node senrslk8snode status is now: NodeHasNoDiskPressure
  Warning  FreeDiskSpaceFailed      7m46s                          kubelet          failed to garbage collect required amount of images. Wanted to free 476703948 bytes, but freed 357013411 bytes
  Warning  InvalidDiskCapacity      <invalid>                      kubelet          invalid capacity 0 on image filesystem
  Normal   NodeAllocatableEnforced  <invalid>                      kubelet          Updated Node Allocatable limit across pods
  Normal   Starting                 <invalid>                      kubelet          Starting kubelet.
  Normal   NodeHasSufficientPID     <invalid> (x7 over <invalid>)  kubelet          Node senrslk8snode status is now: NodeHasSufficientPID
  Normal   NodeHasSufficientMemory  <invalid> (x8 over <invalid>)  kubelet          Node senrslk8snode status is now: NodeHasSufficientMemory
root@senrsl:/home/senrsl/k8s#

错误4：

莫名其妙又变成了ErrImageNeverPull

尝试

从机

root@senrslk8snode:/home/senrsl/k8s# docker build -t test-provider:v1 -f Dockerfile_provider .

主机

root@senrsl:/home/senrsl/k8s# kubectl create deployment test-provider --image=test-provider:v1 --dry-run=client -o yaml > test-provider.yaml
root@senrsl:/home/senrsl/k8s# vi test-provider.yaml

修改imagePullPolicy为 IfNotPresent

然后应用

变成了 ErrImagePull 过一会又变成了 ImagePullBackOff

镜像拉取方式

以下列表包含了 imagePullPolicy 可以设置的值，以及这些值的效果：

IfNotPresent
只有当镜像在本地不存在时才会拉取。
Always
每当 kubelet 启动一个容器时，kubelet 会查询容器的镜像仓库， 将名称解析为一个镜像摘要。 如果 kubelet 有一个容器镜像，并且对应的摘要已在本地缓存，kubelet 就会使用其缓存的镜像； 否则，kubelet 就会使用解析后的摘要拉取镜像，并使用该镜像来启动容器。
Never
Kubelet 不会尝试获取镜像。如果镜像已经以某种方式存在本地， kubelet 会尝试启动容器；否则，会启动失败。 更多细节见提前拉取镜像。


特么的，不知道为啥。。。。


16，架构及组件介绍

https://kubernetes.io/zh-cn/docs/concepts/overview/components/

算了，够够的了。。。。

17，动态扩容

直接修改部署用的yml文件

spec -> replicas: 值，默认1，就是1个容器部署，修改值，改成N个容器部署

修改后直接 kubectl apply -f depoly.yml 即可重新应用部署。

2022年07月12日23:08:54

--
senRsl
2022年07月08日18:12:03
senRsl 23:09:00

没有评论 :

发表评论

订阅： 博文评论 ( Atom )