kubernetes中Deployment配置-云计算-互联网-天达云

kubernetes中Deployment配置
更新：HHH 时间：2023-1-7

Deployment对象，顾名思义，是用于部署应用的对象。它使Kubernetes中最常用的一个对象，它为ReplicaSet和Pod的创建提供了一种声明式的定义方法，从而无需像前两篇文章中那样手动创建ReplicaSet和Pod对象（使用Deployment而不直接创建ReplicaSet是因为Deployment对象拥有许多ReplicaSet没有的特性，例如滚动升级和回滚）。

Deployment:
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
name: skx
spec:
replicas: 3
template:
metadata:
labels:
app: skx_server
spec:
containers:

name: httpd-test
image: 192.168.1.10:5000/httpd:v1
ports:
- containerPort: 80

PS：注意，在Deployment资源对象中，可以添加Port字段，但此字段仅供用户查看，并不实际生效。
如下：
ports:

containerPort: 80

SERVICE

kind: Service
apiVersion: v1
metadata:
name: skx-svc
spec:
selector:
app: skx_server
ports:

protocol: TCP
port: 80
targetPort: 80

关联主机：
kind: Service
apiVersion: v1
metadata:
name: skx-svc
spec:
type: NodePort
selector:
app: skx_server
ports:

protocol: TCP
port: 80
targetPort: 80
nodePort: 30123
更改pod访问界面：
[root@master ~]# kubectl exec -it skx-694cc5db89-45nvk /bin/bash
root@skx-694cc5db89-45nvk:/usr/local/apache2# echo no.1 > htdocs/index.html
root@skx-694cc5db89-45nvk:/usr/local/apache2# exit
查看规则：
[root@master ~]# iptables-save
SNAT：Source NAT(源地址转换) DNAT：Destnation(目标地址转换) MASQ：动态的源地址转换
Service实现的负载均衡使用的时iptables规则。IPVS
10.107.64.232
KUBE-SVC-QDLMDMK46RWAY7QJ

查找对应的防火墙规则
[root@master ~]# kubectl get svc

[root@master ~]# iptables-save | grep 10.107.64.232

[root@master ~]# iptables-save | grep KUBE-SVC-QDLMDMK46RWAY7QJ

[root@master ~]# iptables-save | grep KUBE-SEP-YPYQNHI3JGSZCBF5

与
[root@master ~]# kubectl get pod -o wide

ip一样

回滚到指定版本：
删除之前创建的资源：
[root@master ~]# kubectl delete -f skx-svc.yaml
service "skx-svc" deleted
[root@master ~]# kubectl delete -f skx.yaml
deployment.extensions "skx" deleted

[root@master ~]# kubectl get deployments.
No resources found.
[root@master ~]# vim skx.yaml
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
name: skx
spec:
revisionHistoryLimit: 10 //添加
replicas: 3
template:
metadata:
labels:
app: skx_server
spec:
containers:

name: httpd-test
image: 192.168.1.10:5000/httpd:v1
ports:
- containerPort: 80
  准备三个版本所使用的私有镜像，来模拟每次升级不同的镜像。
  [root@master ~]# mv skx.yaml skx1.yaml
  [root@master ~]# cp skx1.yaml skx2.yaml
  [root@master ~]# cp skx1.yaml skx3.yaml

[root@master ~]# vim skx1.yaml

15行：

    image: 192.168.1.10:5000/httpd:v1

[root@master ~]# vim skx2.yaml

15行：

    image: 192.168.1.10:5000/httpd:v2

[root@master ~]# vim skx3.yaml

15行：

    image: 192.168.1.10:5000/httpd:v3

此处3个yaml文件指定不同版本的镜像。
运行一个服务，并记录一个版本信息。
[root@master ~]# kubectl apply -f skx1.yaml --record
deployment.extensions/skx created
[root@master ~]# kubectl apply -f skx2.yaml --record
deployment.extensions/skx configured
[root@master ~]# kubectl apply -f skx3.yaml --record
deployment.extensions/skx configured
查看有哪些版本信息
[root@master ~]# kubectl rollout history deployment skx
deployment.extensions/skx
REVISION CHANGE-CAUSE
1 kubectl apply --filename=skx1.yaml --record=true
2 kubectl apply --filename=skx2.yaml --record=true
3 kubectl apply --filename=skx3.yaml --record=true

运行并升级Deployment资源，并记录版本信息。
[root@master ~]# kubectl apply -f skx2.yaml --record
deployment.extensions/skx configured
此时可以运行一个关联的Service自愿去验证，升级是否成功
[root@master ~]# kubectl apply -f skx-svc.yaml
service/skx-svc created

[root@master ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
httpd-svc NodePort 10.97.81.154 <none> 80:31194/TCP 43h
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 5d17h
skx-svc NodePort 10.96.174.199 <none> 80:30123/TCP 16s

[root@master ~]# curl 10.96.174.199
<h2>songkaixiong | test-web | httpd | v3<h2>

回滚到指定版本。
[root@master ~]# kubectl rollout undo deployment skx --to-revision=1
deployment.extensions/skx rolled back
curl 10.96.174.199
<h2>songkaixiong | test-web | httpd | v1<h2>

用label控制Pod的位置
给node03打上一个标签
[root@master ~]# kubectl label nodes node03 disk=ssd
node/node03 labeled
指定查看node03标签：
[root@master ~]# kubectl get nodes --show-labels | grep node03
node03 Ready <none> 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disk=ssd,kubernetes.io/arch=amd64,kubernetes.io/hostname=node03,kubernetes.io/os=linux
不指定查看所有节点标签：
[root@master ~]# kubectl get nodes --show-labels
NAME STATUS ROLES AGE VERSION LABELS
master Ready master 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=master,kubernetes.io/os=linux,node-role.kubernetes.io/master=
node02 Ready <none> 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node02,kubernetes.io/os=linux
node03 Ready <none> 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disk=ssd,kubernetes.io/arch=amd64,kubernetes.io/hostname=node03,kubernetes.io/os=linux
删除skx1资源：
[root@master ~]# kubectl delete -f skx1.yaml
deployment.extensions "skx" deleted

[root@master ~]# kubectl delete -f skx-svc.yaml
service "skx-svc" deleted

修改skx1.yaml配置文件：
[root@master ~]# vim skx1.yaml
kind: Deployment
apiVersion: extensions/v1beta1
metadata:
name: skx
spec:
revisionHistoryLimit: 10
replicas: 3
template:
metadata:
labels:
app: skx_server
spec:
containers:

name: httpd-test
image: 192.168.1.10:5000/httpd:v1
ports:
- containerPort: 80
  nodeSelector: //添加节点选择器
  disk: ssd //和标签内容一致
  [root@master ~]# kubectl apply -f skx1.yaml
  deployment.extensions/skx created
  三台都在node03上运行：
  [root@master ~]# kubectl get pod -o wide
  NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
  skx-55c4dc6dbc-7ztl9 1/1 Running 0 73s 10.244.2.27 node03 <none> <none>
  skx-55c4dc6dbc-jsms7 1/1 Running 0 73s 10.244.2.28 node03 <none> <none>
  skx-55c4dc6dbc-rfss7 1/1 Running 0 73s 10.244.2.26 node03 <none> <none>

删除标签：
查看node03的标签：
[root@master ~]# kubectl get nodes --show-labels node03
NAME STATUS ROLES AGE VERSION LABELS
node03 Ready <none> 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,disk=ssd,kubernetes.io/arch=amd64,kubernetes.io/hostname=node03,kubernetes.io/os=linux
删除node03标签
[root@master ~]# kubectl label nodes node03 disk-
node/node03 labeled
再次查看node03标签：
[root@master ~]# kubectl get nodes --show-labels node03
NAME STATUS ROLES AGE VERSION LABELS
node03 Ready <none> 5d17h v1.15.0 beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node03,kubernetes.io/os=linux

删除资源：
[root@master ~]# kubectl delete deployments. skx
deployment.extensions "skx" deleted
重新生成：
[root@master ~]# kubectl apply -f skx1.yaml
deployment.extensions/skx unchanged
会找不到，起不来：
[root@master ~]# kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
skx-55c4dc6dbc-8rpl6 0/1 Pending 0 11s <none> <none> <none> <none>
skx-55c4dc6dbc-c2blp 0/1 Pending 0 11s <none> <none> <none> <none>
skx-55c4dc6dbc-zk7gw 0/1 Pending 0 11s <none> <none> <none> <none>

[root@master ~]# kubectl get pod -n kube-system
NAME READY STATUS RESTARTS AGE
coredns-5c98db65d4-52cdm 1/1 Running 3 5d17h
coredns-5c98db65d4-sl96w 1/1 Running 4 5d17h
etcd-master 1/1 Running 3 5d17h
kube-apiserver-master 1/1 Running 3 5d17h
kube-controller-manager-master 1/1 Running 3 5d17h
kube-flannel-ds-amd64-9vnsc 1/1 Running 4 5d17h
kube-flannel-ds-amd64-tdzrm 1/1 Running 2 5d17h
kube-flannel-ds-amd64-tvl2q 1/1 Running 5 5d17h
kube-proxy-492jr 1/1 Running 2 5d17h
kube-proxy-gccnb 1/1 Running 3 5d17h
kube-proxy-klznh 1/1 Running 2 5d17h
kube-scheduler-master 1/1 Running 4 5d17h

查看master日志：
[root@master ~]# kubectl logs -n kube-system kube-scheduler-master
查看详细描述信息
[root@master ~]# kubectl describe pod skx-55c4dc6dbc-8rpl6
查看日志
[root@master ~]# kubectl logs -n kube-system kube-scheduler-master
查看kubectl日志
[root@master ~]# less /var/log/messages | grep kubelet


返回云计算教程...

新手上路

产品管理

支付方式

关于我们