【kubevirt】VirtualMachineInstanceReplicaSet(vmis)-扩缩容-弹性伸缩
@[TOC](文章目录)
## 概述/理解
VirtualMachineInstanceReplicaSet（vmis）确保指定数量的 VirtualMachineInstance（vmi） 副本在任何时候都在运行。
我们可以这样理解，vmis就是kubernetes（k8s）里面的控制器（DeployMent，ReplicaSet）管理我们pod的副本数，实现扩缩容、回滚等。也可以借助HorizontalPodAutoscaler（hpa）实现弹性伸缩。这里我们就说vmis控制器，在这里的vmis控制器，管理我们vmi虚拟机实例的副本数，也可以实现扩缩容，借助hpa实现弹性伸缩。所有我们的yaml文件写法原理都类似。
## 使用场景
当需要许多相同的虚拟机，并且不关心在虚拟机终止后任何磁盘状态时。
<br>
## 创建vmis
1. 编写vmis的yaml文件
```bash
[root@master vm]# cat vmis.yaml
apiVersion: kubevirt.io/v1alpha3
kind: VirtualMachineInstanceReplicaSet
metadata:
name: testreplicaset
spec:
replicas: 2
selector:
matchLabels:
myvmi: myvmi # 保持一致，选择
template:
metadata:
labels:
myvmi: myvmi # 保持一致，匹配
spec:
domain:
devices:
disks:
- name: containerdisk
disk:
bus: virtio
resources:
requests:
memory: 1024M
volumes:
- name: containerdisk
containerDisk:
image: centos7
imagePullPolicy: IfNotPresent
```
2. 使用kubectl命令创建vmis
```bash
[root@master vm]# kubectl apply -f vmis.yaml
virtualmachineinstancereplicaset.kubevirt.io/testreplicaset created
```
3. 查看运行状态
```bash
[root@master vm]# kubectl get vmis
NAME AGE PHASE IP NODENAME READY
testreplicaset6vm9s 42s Running 10.244.0.139 master False
testreplicaset8dshm 22s Scheduling False
testreplicasetbqxnb 22s Scheduling False
[root@master vm]# kubectl get vmis
NAME AGE PHASE IP NODENAME READY
testreplicaset8dshm 46s Running 10.244.0.141 master False
testreplicasetbqxnb 46s Running 10.244.0.140 master False
[root@master vm]# kubectl get pod
NAME READY STATUS RESTARTS AGE
virt-launcher-testreplicaset8dshm-nz7x2 2/2 Running 0 69s
virt-launcher-testreplicasetbqxnb-ljp2f 2/2 Running 0 70s
```
4. describe 查看详细信息
```bash
[root@master vm]# kubectl describe vmirs testreplicaset
Name: testreplicaset
Namespace: default
Labels: <none>
Annotations: kubevirt.io/latest-observed-api-version: v1
kubevirt.io/storage-observed-api-version: v1alpha3
API Version: kubevirt.io/v1
Kind: VirtualMachineInstanceReplicaSet
Metadata:
Creation Timestamp: 2022-05-02T13:50:05Z
Generation: 2
Managed Fields:
API Version: kubevirt.io/v1alpha3
Fields Type: FieldsV1
fieldsV1:
f:metadata:
f:annotations:
f:kubevirt.io/latest-observed-api-version:
f:kubevirt.io/storage-observed-api-version:
f:spec:
f:template:
f:metadata:
f:creationTimestamp:
f:status:
.:
f:labelSelector:
f:replicas:
Manager: Go-http-client
Operation: Update
Time: 2022-05-02T13:50:05Z
API Version: kubevirt.io/v1alpha3
Fields Type: FieldsV1
fieldsV1:
f:metadata:
f:annotations:
...
...
.:
f:memory:
f:volumes:
Manager: kubectl
Operation: Update
Time: 2022-05-02T13:50:05Z
Resource Version: 267261
Self Link: /apis/kubevirt.io/v1/namespaces/default/virtualmachineinstancereplicasets/testreplicaset
UID: 96d17d12-17b5-4df7-940a-fac7c6b820d2
Spec:
Replicas: 2
Selector:
Match Labels:
Myvmi: myvmi
Template:
Metadata:
Creation Timestamp: <nil>
Labels:
Myvmi: myvmi
Spec:
Domain:
Devices:
Disks:
Disk:
Bus: virtio
Name: containerdisk
Resources:
Requests:
Memory: 1024M
Volumes:
Container Disk:
Image: kubevirt/cirros-container-disk-demo
Image Pull Policy: IfNotPresent
Name: containerdisk
Status:
Label Selector: myvmi=myvmi
Replicas: 2
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal SuccessfulCreate 5m21s virtualmachinereplicaset-controller Started the virtual machine by creating the new virtual machine instance testreplicaseth6zsl
Normal SuccessfulCreate 5m21s virtualmachinereplicaset-controller Started the virtual machine by creating the new virtual machine instance testreplicasetw75s4
```
### 扩缩容
1. 查看vmis
```bash
[root@master vm]# kubectl get -f vmis.yaml
NAME DESIRED CURRENT READY AGE
testreplicaset 3 3 2m52s
```
2. 使用scale命令，设置副本数为5
```bash
[root@master vm]# kubectl scale vmirs testreplicaset --replicas 5
virtualmachineinstancereplicaset.kubevirt.io/testreplicaset scaled
```
3. 查看效果
```bash
[root@master vm]# kubectl get vmis
NAME AGE PHASE IP NODENAME READY
testreplicaset98x8d 5m29s Running 10.244.0.146 master False
testreplicasetddqc9 2m24s Running 10.244.0.148 master False
testreplicasetdss8l 5m29s Running 10.244.0.144 master False
testreplicasetmhm6x 5m29s Running 10.244.0.145 master False
testreplicasetv4dzs 2m24s Running 10.244.0.147 master False
[root@master vm]# kubectl get pod
NAME READY STATUS RESTARTS AGE
virt-launcher-testreplicaset98x8d-5p99p 2/2 Running 0 3m15s
virt-launcher-testreplicasetddqc9-6c2m4 2/2 Running 0 10s
virt-launcher-testreplicasetdss8l-9mv56 2/2 Running 0 3m15s
virt-launcher-testreplicasetmhm6x-r76wt 2/2 Running 0 3m15s
virt-launcher-testreplicasetv4dzs-bm4s8 2/2 Running 0 10s
```
### 弹性伸缩
#### 方法1
1. 使用Horizontal Pod Autoscaler（hpa），创建yaml文件
```bash
[root@master vm]# cat vmis-hpa.yaml
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
creationTimestamp: null
name: testreplicaset
spec:
maxReplicas: 5
minReplicas: 2
scaleTargetRef:
apiVersion: kubevirt.io/v1
kind: VirtualMachineInstanceReplicaSet
name: testreplicaset
```
2. 创建vmis-hpa
```bash
[root@master vm]# kubectl apply -f vmis-hpa.yaml
horizontalpodautoscaler.autoscaling/testreplicaset created
```
3. 查看状态
```bash
[root@master vm]# kubectl get hpa
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
testreplicaset VirtualMachineInstanceReplicaSet/testreplicaset <unknown>/80% 2 5 0 7s
```
#### 方法2
1. 使用`kubectl autoscale`命令
```bash
[root@master vm]# kubectl autoscale vmirs testreplicaset --min=2 --max=5
```
- 点赞
- 收藏
- 关注作者
评论（0）