27. HPA
27.1 Pod伸缩简介
根据当前pod的负载，动态调整pod副本数量，业务高峰期自动扩容pod的副本数以尽快响应pod的请求
在业务低峰期对pod进行缩容，实现降本增效的目的
公有云支持node级别的弹性伸缩
27.2 Scale命令扩容
#当--replicas的值设置为比原来的pod数，k8s会杀掉一些pod，下面3个变成1个
root@k8s-master1:~/k8s-data/yaml/dubbo# kubectl get deployments.apps -n demo
NAME READY UP-TO-DATE AVAILABLE AGE
nginx-deployment 1/1 1 1 6d21h
tomcat-app1-deployment 1/1 1 1 6d22h
root@k8s-master1:~/k8s-data/yaml/dubbo# kubectl scale deployment -n demo nginx-deployment --replicas=1
27.3 动态伸缩控制器类型
基于pod资源利用率横向调整pod副本数量
基于pod资源利用率，调整对单个pod的最大资源限制，不能与HPA同时使用
基于集群中node资源使用情况，动态伸缩node节点，从而保证有CPU和内存资源用于创建pod
27.4 HPA控制器简介
# kube-controller-manager --help | grep initial-readiness-delay
Horizontal Pod Autoscaling(HPA)控制器，根据预定义好的阈值及pod当前的资源利用率，自动控制在k8s集群中运行的pod数量（自动弹性水平自动伸缩）
--horizontal-pod-autoscaler-sync-period #默认每隔15s也可以通过这个参数进行修改，查询metrics的资源使用情况
--horizontal-pod-autoscaler-downscale-stabilization #缩容间隔周期，默认5分钟
--horizontal-pod-autoscaler-sync-period #HPA控制器同步pod副本数的间隔周期
--horizontal-pod-autoscaler-cpu-initialization-period #初始化延迟时间，在此时间内pod的CPU资源指标将不会生效，默认为5分钟
--horizontal-pod-autoscaler-initial-readiness-delay #用于设置pod准备时间，在此时间内的pod统统被认为未就绪及不采集数据，默认为30秒
--horizontal-pod-autoscaler-tolerance #HPA控制器能容忍的数据差异（浮点数，默认为0.1）即新的指标要与当前的阈值差异在0.1或以上即要大于1+0.1=1.1, 或小于1-0.1=0.9，比如阈值为CPU利用率50%，当前为80%，那么80/50=1.6 > 1.1则会触发扩容，反之会缩容。
触发条件：avg(CurrentPodsConsumption)/Target >1.1 或 <0.9=把N个pod的数据相加后根据pod的数量计算出平均数除以阈值，大于1.1就扩容，小于0.9就缩容
计算公式：TargetNumOfPods = ceil(sum(CurrentPodsCPUUtilization)/Target) #ceil是一个向上取整的目的pod整数
指标数据需要部署metrics-server，即HPA使用metrics-server作为数据源
在k8s 1.1引入HPA控制器，早期使用Heapster组件采集pod指标数据，在k8s 1.11版本开始使用Metrices Server完成数据采集，然后将采集到的数据通过API，例如：metrics.k8s.io、custom.metrics.k8s.io、external.metrics.k8s.io、然后再把数据提供给HPA控制器进行查询，以实现基于某个资源利用率对pod进行扩缩容的目的
27.5 metrics-server部署
metrics-server 是Kubernetes内置的容器资源指标来源
metrics-server 从node节点上的kubelet收集资源指标，并通过Metrics API在 Kubernetes apiserver中公开指标数据，以供Horizontal Pod Autoscaler和Vertical Pod Autoscaler使用，也可以通过访问kubectl top node/pod 查看指标数据
root@k8s-master1:~/20220814/metrics-server-0.6.1-case# cat metrics-server-v0.6.1.yaml
apiVersion: v1
kind: ServiceAccount
metadata:
labels:
k8s-app: metrics-server
name: metrics-server
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
labels:
k8s-app: metrics-server
rbac.authorization.k8s.io/aggregate-to-admin: "true"
rbac.authorization.k8s.io/aggregate-to-edit: "true"
rbac.authorization.k8s.io/aggregate-to-view: "true"
name: system:aggregated-metrics-reader
rules:
- apiGroups:
- metrics.k8s.io
resources:
- pods
- nodes
verbs:
- get
- list
- watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
labels:
k8s-app: metrics-server
name: system:metrics-server
rules:
- apiGroups:
- ""
resources:
- nodes/metrics
verbs:
- get
- apiGroups:
- ""
resources:
- pods
- nodes
verbs:
- get
- list
- watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: RoleBinding
metadata:
labels:
k8s-app: metrics-server
name: metrics-server-auth-reader
namespace: kube-system
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: Role
name: extension-apiserver-authentication-reader
subjects:
- kind: ServiceAccount
name: metrics-server
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
labels:
k8s-app: metrics-server
name: metrics-server:system:auth-delegator
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: system:auth-delegator
subjects:
- kind: ServiceAccount
name: metrics-server
namespace: kube-system
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
labels:
k8s-app: metrics-server
name: system:metrics-server
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: system:metrics-server
subjects:
- kind: ServiceAccount
name: metrics-server
namespace: kube-system
---
apiVersion: v1
kind: Service
metadata:
labels:
k8s-app: metrics-server
name: metrics-server
namespace: kube-system
spec:
ports:
- name: https
port: 443
protocol: TCP
targetPort: https
selector:
k8s-app: metrics-server
---
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
k8s-app: metrics-server
name: metrics-server
namespace: kube-system
spec:
selector:
matchLabels:
k8s-app: metrics-server
strategy:
rollingUpdate:
maxUnavailable: 0
template:
metadata:
labels:
k8s-app: metrics-server
spec:
containers:
- args:
- --cert-dir=/tmp
- --secure-port=4443
- --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
- --kubelet-use-node-status-port
- --metric-resolution=15s
#更改了镜像名字
#image: k8s.gcr.io/metrics-server/metrics-server:v0.6.1
image: harbor.nbrhce.com/demo/metrics-server:v0.6.1
imagePullPolicy: IfNotPresent
livenessProbe:
failureThreshold: 3
httpGet:
path: /livez
port: https
scheme: HTTPS
periodSeconds: 10
name: metrics-server
ports:
- containerPort: 4443
name: https
protocol: TCP
readinessProbe:
failureThreshold: 3
httpGet:
path: /readyz
port: https
scheme: HTTPS
initialDelaySeconds: 20
periodSeconds: 10
resources:
requests:
cpu: 100m
memory: 200Mi
securityContext:
allowPrivilegeEscalation: false
readOnlyRootFilesystem: true
runAsNonRoot: true
runAsUser: 1000
volumeMounts:
- mountPath: /tmp
name: tmp-dir
nodeSelector:
kubernetes.io/os: linux
priorityClassName: system-cluster-critical
serviceAccountName: metrics-server
volumes:
- emptyDir: {}
name: tmp-dir
---
apiVersion: apiregistration.k8s.io/v1
kind: APIService
metadata:
labels:
k8s-app: metrics-server
name: v1beta1.metrics.k8s.io
spec:
group: metrics.k8s.io
groupPriorityMinimum: 100
insecureSkipTLSVerify: true
service:
name: metrics-server
namespace: kube-system
version: v1beta1
versionPriority: 100
27.6 HPA实现
root@k8s-master1:~/20220814/metrics-server-0.6.1-case# cat tomcat-app1.yaml
kind: Deployment
#apiVersion: extensions/v1beta1
apiVersion: apps/v1
metadata:
labels:
app: tomcat-app1-deployment-label
name: tomcat-app1-deployment
namespace: demo
spec:
replicas: 2
selector:
matchLabels:
app: tomcat-app1-selector
template:
metadata:
labels:
app: tomcat-app1-selector
spec:
containers:
- name: tomcat-app1-container
#image: harbor.magedu.local/magedu/tomcat-app1:v7
#image: tomcat:7.0.93-alpine
#这个是压测的镜像运行两个CPU直接100%好实现目标
image: lorel/docker-stress-ng
args: ["--vm", "2", "--vm-bytes", "256M"]
##command: ["/apps/tomcat/bin/run_tomcat.sh"]
imagePullPolicy: IfNotPresent
##imagePullPolicy: Always
ports:
- containerPort: 8080
protocol: TCP
name: http
env:
- name: "password"
value: "123456"
- name: "age"
value: "18"
resources:
limits:
cpu: 1
memory: "512Mi"
requests:
cpu: 500m
memory: "512Mi"
---
kind: Service
apiVersion: v1
metadata:
labels:
app: tomcat-app1-service-label
name: tomcat-app1-service
namespace: demo
spec:
type: NodePort
ports:
- name: http
port: 80
protocol: TCP
targetPort: 8080
nodePort: 40003
selector:
app: tomcat-app1-selector
#注意关联的是哪个名字哪个控制器
#添加最小副本minReplicas: 3
#添加最大副本maxReplicas: 10 最多就扩容到10
# targetCPUUtilizationPercentage 这个是CPU阈值 也就是触发的机制
root@k8s-master1:~/20220814/metrics-server-0.6.1-case# cat hpa.yaml
#apiVersion: autoscaling/v2beta1
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
namespace: demo
name: tomcat-app1-podautoscaler
labels:
app: tomcat-app1
version: v2beta1
spec:
scaleTargetRef:
apiVersion: apps/v1
#apiVersion: extensions/v1beta1
kind: Deployment
name: tomcat-app1-deployment
minReplicas: 3
maxReplicas: 10
targetCPUUtilizationPercentage: 60
#metrics:
#- type: Resource
# resource:
# name: cpu
# targetAverageUtilization: 60
#- type: Resource
# resource:
# name: memory
#这样显示就代表扩容了
root@k8s-master1:~/20220814/metrics-server-0.6.1-case# kubectl get hpa -n demo
NAME REFERENCE TARGETS MINPODS MAXPODS REPLICAS AGE
tomcat-app1-podautoscaler Deployment/tomcat-app1-deployment 199%/60% 3 10 10 15m
