课堂记录：k8s滚动更新及数据管理（四）–技术流ken
滚动更新实践
实现效果当前生产环境httpd版本为2.2.31，在不停机以及中断服务的情况下，滚动更新到httpd：2.2.32版本
第一步：编写部署httpd的yaml文件
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: httpd31
spec:
replicas: 3
template:
metadata:
labels:
run: httpd31
spec:
containers:
- name: httpd31
image: httpd:2.2.31
imagePullPolicy: IfNotPresent
第二步：查看deploy
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 3/3 3 3 29s httpd31 httpd:2.2.31 run=httpd31
第三步：查看rs
[root@ken1 ~]# kubectl get rs -o wide
NAME DESIRED CURRENT READY AGE CONTAINERS IMAGES SELECTOR
httpd31-78c4cf6f 3 3 3 61s httpd31 httpd:2.2.31 pod-template-hash=78c4cf6f,run=httpd31
第四步：查看po
[root@ken1 ~]# kubectl get po -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
httpd31-78c4cf6f-2f2wp 1/1 Running 0 84s 10.244.2.110 ken3 <none> <none>
httpd31-78c4cf6f-cmhz7 1/1 Running 0 84s 10.244.2.111 ken3 <none> <none>
httpd31-78c4cf6f-qvpct 1/1 Running 0 84s 10.244.1.32 ken2 <none> <none>
滚动更新
第五步；复制一份上述yml文件并把版本改为2.2.32
[root@ken1 ~]# cp httpd31.yml httpd32.yml
[root@ken1 ~]# cat httpd32.yml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: httpd31
spec:
replicas: 3
template:
metadata:
labels:
run: httpd31
spec:
containers:
- name: httpd31
image: httpd:2.2.32
imagePullPolicy: IfNotPresent
第六步：执行该yml文件
[root@ken1 ~]# kubectl apply -f httpd32.yml
第七步：查看deploy
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 3/3 3 3 4m1s httpd31 httpd:2.2.32 run=httpd31
第八步：查看rs
[root@ken1 ~]# kubectl get rs -o wide
NAME DESIRED CURRENT READY AGE CONTAINERS IMAGES SELECTOR
httpd31-78c4cf6f 0 0 0 4m22s httpd31 httpd:2.2.31 pod-template-hash=78c4cf6f,run=httpd31
httpd31-7f8489457b 3 3 3 42s httpd31 httpd:2.2.32 pod-template-hash=7f8489457b,run=httpd31
第九步：查看po
[root@ken1 ~]# kubectl get po -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
httpd31-7f8489457b-8rccl 1/1 Running 0 77s 10.244.2.112 ken3 <none> <none>
httpd31-7f8489457b-j28z9 1/1 Running 0 77s 10.244.1.33 ken2 <none> <none>
httpd31-7f8489457b-p688p 1/1 Running 0 74s 10.244.1.34 ken2 <none> <none>
第十步：查看deploy详细信息
[root@ken1 ~]# kubectl describe deploy httpd31
Name: httpd31
Namespace: default
CreationTimestamp: Thu, 22 Aug 2019 09:38:12 +0800
Labels: run=httpd31
Annotations: deployment.kubernetes.io/revision: 2
kubectl.kubernetes.io/last-applied-configuration:
{"apiVersion":"extensions/v1beta1","kind":"Deployment","metadata":{"annotations":{},"name":"httpd31","namespace":"default"},"spec":{"repli...
Selector: run=httpd31
Replicas: 3 desired | 3 updated | 3 total | 3 available | 0 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 1 max unavailable, 1 max surge
Pod Template:
Labels: run=httpd31
Containers:
httpd31:
Image: httpd:2.2.32
Port: <none>
Host Port: <none>
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available True MinimumReplicasAvailable
OldReplicaSets: <none>
NewReplicaSet: httpd31-7f8489457b (3/3 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 5m45s deployment-controller Scaled up replica set httpd31-78c4cf6f to 3
Normal ScalingReplicaSet 2m5s deployment-controller Scaled up replica set httpd31-7f8489457b to 1
Normal ScalingReplicaSet 2m5s deployment-controller Scaled down replica set httpd31-78c4cf6f to 2
Normal ScalingReplicaSet 2m5s deployment-controller Scaled up replica set httpd31-7f8489457b to 2
Normal ScalingReplicaSet 2m2s deployment-controller Scaled down replica set httpd31-78c4cf6f to 1
Normal ScalingReplicaSet 2m2s deployment-controller Scaled up replica set httpd31-7f8489457b to 3
Normal ScalingReplicaSet 2m2s deployment-controller Scaled down replica set httpd31-78c4cf6f to 0
k8s回滚任意版本
实践：需要准备三份yml文件，版本分别为2.2.31 2.2.32 2.2.34, 三分yaml文件名称必须一致，只需要镜像版本不同就行
第一步：复制yaml文件
[root@ken1 ~]# cat httpd34.yml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: httpd31
spec:
replicas: 3
template:
metadata:
labels:
run: httpd31
spec:
containers:
- name: httpd31
image: httpd:2.2.34
imagePullPolicy: IfNotPresent
第二步：部署31版本yml文件
[root@ken1 ~]# kubectl apply -f httpd31.yml --record
deployment.extensions/httpd31 configured
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 3/3 3 3 12m httpd31 httpd:2.2.31 run=httpd31
第三步：部署32版本的yml文件
[root@ken1 ~]# kubectl apply -f httpd32.yml --record
deployment.extensions/httpd31 configured
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 3/3 3 3 13m httpd31 httpd:2.2.32 run=httpd31
第四步：部署34版本的yml文件
[root@ken1 ~]# kubectl apply -f httpd34.yml --record
deployment.extensions/httpd31 configured
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 2/3 2 2 14m httpd31 httpd:2.2.34 run=httpd31
实践回滚到31版本
第五步：查看命令执行的历史信息
[root@ken1 ~]# kubectl rollout history deploy httpd31
deployment.extensions/httpd31
REVISION CHANGE-CAUSE
3 kubectl apply --filename=httpd31.yml --record=true
4 kubectl apply --filename=httpd32.yml --record=true
5 kubectl apply --filename=httpd34.yml --record=true
第六步：实现回滚
[root@ken1 ~]# kubectl rollout undo deploy httpd31 --to-revision=3
deployment.extensions/httpd31 rolled back
[root@ken1 ~]# kubectl get deploy -o wide
NAME READY UP-TO-DATE AVAILABLE AGE CONTAINERS IMAGES SELECTOR
httpd31 3/3 3 3 18m httpd31 httpd:2.2.31 run=httpd31
指定版本记录的条数
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: httpd31
spec:
revisionHistoryLimit: 10
replicas: 3
template:
metadata:
labels:
run: httpd31
spec:
containers:
- name: httpd31
image: httpd:2.2.31
imagePullPolicy: IfNotPresent
~
k8ss数据管理
1.emptyDir
2.hostPath
3.NFS
4.pv/pvc
一、实践emptyDir
第一步：编写yml文件
apiVersion: v1
kind: Pod
metadata:
name: nginx
spec:
containers:
- name: nginx
image: nginx
imagePullPolicy: IfNotPresent
volumeMounts:
- name: ken
mountPath: /usr/share/nginx/html
volumes:
- name: ken
emptyDir: {}
第二步：执行yml文件
[root@ken1 ~]# kubectl apply -f nginx.yml
第三步：查看pod
[root@ken1 ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
nginx 1/1 Running 0 68s
第四步：在节点查看容器详细信息
[root@ken3 ken]# docker inspect 937208cf64ee
...
"Mounts": [
{
"Type": "bind",
"Source": "/var/lib/kubelet/pods/704e6443-9b31-45b5-b41d-39e8e84bacb2/volumes/kubernetes.io~empty-dir/ken",
"Destination": "/usr/share/nginx/html",
"Mode": "",
"RW": true,
"Propagation": "rprivate"
},
...
第五步：编写测试文件
[root@ken3 ~]# cd /var/lib/kubelet/pods/704e6443-9b31-45b5-b41d-39e8e84bacb2/volumes/kubernetes.io~empty-dir/ken
[root@ken3 ken]# echo "test emptyDir" > index.html
第六步：主节点测试
[root@ken1 ~]# curl 10.244.2.119
test emptyDir
二、实践hostPath
效果相当于执行： docker run -v /tmp:/usr/share/nginx/html
第一步：编写yml文件
apiVersion: v1
kind: Pod
metadata:
name: nginx
spec:
containers:
- name: nginx
image: nginx
imagePullPolicy: IfNotPresent
volumeMounts:
- name: ken
mountPath: /usr/share/nginx/html
volumes:
- name: ken
hostPath:
path: /tmp
第二步：执行yml文件
[root@ken1 ~]# kubectl apply -f nginx-hostp.yml
第三步：进行验证
[root@ken1 ~]# ls /tmp
ks-script-Kgah4x yum.log
systemd-private-1efe767de1b4430fb77c8d90bafce1de-vmtoolsd.service-nfl5a9 yum_save_tx.2019-08-21.17-03.Hc7vcy.yumtx
systemd-private-21c03f170cff40688b56b7b97b6d60d9-vmtoolsd.service-oR1NRh
[root@ken1 ~]# kubectl get po
NAME READY STATUS RESTARTS AGE
nginx 1/1 Running 0 31m
nginx-hostp 1/1 Running 0 31s
[root@ken1 ~]# kubectl exec nginx-hostp ls /usr/share/nginx/html
ks-script-Kgah4x
systemd-private-290df9f339b647a099c58e3b2d499278-vmtoolsd.service-2MNZBs
systemd-private-73ed2c6655a748c1b8190aa51cf682bb-vmtoolsd.service-nysa3U
yum.log
三、实践NFS
每个节点必须要安装nfs
主节点部署NFS服务器端
每个节点安装nfs-utils
第一步：部署nfs服务器端
[root@ken1 ~]# yum install nfs-utils rpcbind -y
[root@ken1 ~]# cat /etc/exports
/ken *(rw)
[root@ken1 ~]# chown -R nfsnobody: /ken
[root@ken1 ~]# systemctl restart nfs rpcbind
第二步：客户端下载nfs
[root@ken2 ~]# yum install nfs-utils -y
第三步：编写yml文件
apiVersion: v1
kind: Pod
metadata:
name: nginx
spec:
containers:
- name: nginx
image: nginx
imagePullPolicy: IfNotPresent
volumeMounts:
- name: ken
mountPath: /usr/share/nginx/html
volumes:
- name: ken
nfs:
path: /ken
server: 192.168.64.11
第四步：执行yml文件
[root@ken1 ~]# kubectl apply -f nginx-nfs.yml
第五步：验证
[root@ken1 ~]# ls /ken
index.html test
[root@ken1 ~]# cat /ken/index.html
test
[root@ken1 ~]# echo "nfs" >> /ken/index.html
[root@ken1 ~]# kubectl get po -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx 1/1 Running 0 38m 10.244.2.119 ken3 <none> <none>
nginx-hostp 1/1 Running 0 7m32s 10.244.1.41 ken2 <none> <none>
nginx-nfs 1/1 Running 0 32s 10.244.2.120 ken3 <none> <none>
[root@ken1 ~]# curl 10.244.2.120
test
nfs
四、实践pv/pvc
第一步：部署NFS
上同，略
第二步：编写pv的yml文件
apiVersion: v1
kind: PersistentVolume
metadata:
name: mypv
spec:
capacity:
storage: 1Gi
accessModes:
- ReadWriteMany
nfs:
path: /ken
server: 192.168.64.11
第三步：执行yml文件
[root@ken1 ~]# kubectl apply -f pv.yml
第四步：查看pv
[root@ken1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
mypv 1Gi RWX Retain Available 17h
注：
accessModes有三类
- ReadWriteOnce – 可以被单个节点进行读写挂载
- ReadOnlyMany – 可以被多个节点进行只读挂载
- ReadWriteMany – 可以被多个节点进行读写挂载
reclaim policy有三类
- retain – pvc被删除之后，数据保留
- recyle- 删除pvc之后会删除数据(被废弃)
- delete – 删除pvc之后会删除数据
创建pvc
第一步：编写yml文件
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: mypvc1
spec:
accessModes:
- ReadWriteMany
volumeName: mypv
resources:
requests:
storage: 1Gi
第二步：执行pvc
[root@ken1 ~]# kubectl apply -f pvc.yml
第三步：查看pvc
[root@ken1 ~]# kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
mypvc1 Bound mypv 1Gi RWX 19s
[root@ken1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
mypv 1Gi RWX Retain Bound default/mypvc1 17h
[root@ken1 ~]#
使用pvc
第一步：编写部署nginx的yml文件
apiVersion: v1
kind: Pod
metadata:
name: nginx-pvc
spec:
containers:
- name: nginx
image: nginx
imagePullPolicy: IfNotPresent
volumeMounts:
- name: ken
mountPath: /usr/share/nginx/html
volumes:
- name: ken
persistentVolumeClaim:
claimName: mypvc1
第二步：执行yml文件
[root@ken1 ~]# kubectl apply -f nginx-pvc.yml
第三步：测试
[root@ken1 ~]# kubectl exec -it nginx-pvc /bin/bash
root@nginx-pvc:/# ls
bin boot dev etc home lib lib64 media mnt opt proc root run sbin srv sys tmp usr var
root@nginx-pvc:/# cd /usr/share/nginx/i
bash: cd: /usr/share/nginx/i: No such file or directory
root@nginx-pvc:/# cd /usr/share/nginx/html/
root@nginx-pvc:/usr/share/nginx/html# ls
index.html test
root@nginx-pvc:/usr/share/nginx/html# touch ken
root@nginx-pvc:/usr/share/nginx/html# exit
exit
[root@ken1 ~]# ls /ken
index.html ken test
释放pv
第一步：先停掉pod
[root@ken1 ~]# kubectl delete -f nginx-pvc.yml
第二步：删除pvc
[root@ken1 ~]# kubectl delete pvc mypvc1
第三步：查看pv状态
[root@ken1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
mypv 1Gi RWX Retain Released default/mypvc1 17h
第四步：删除pv
[root@ken1 ~]# kubectl delete -f pv.yml
第五步：查看数据
[root@ken1 ~]# ls /ken
index.html ken test
第六步：再次创建pv
[root@ken1 ~]# kubectl apply -f pv.yml
persistentvolume/mypv created
[root@ken1 ~]# kubectl get pv
NAME CAPACITY ACCESS MODES RECLAIM POLICY STATUS CLAIM STORAGECLASS REASON AGE
mypv 1Gi RWX Retain Available 4s