使用 helm 部署 Metrics Server
第一种方式：
helm repo add bitnami https://charts.bitnami.com/bitnami
helm install bitnami/metrics-server 会有报错，执行以下命令
helm upgrade loopy-saola bitnami/metrics-server --set apiService.create=true
$ kubectl get pod 查看节点
loopy-saola-metrics-server-58796b4bc7-4mv4t 1/1 Running 0 16m
$ kubectl top nodes 不能获取资源指标，需要修改 deployment
$ kubectl get deployment loopy-saola-metrics-server -o yaml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "3"
creationTimestamp: "2019-12-06T10:18:00Z"
generation: 3
labels:
app.kubernetes.io/instance: loopy-saola
app.kubernetes.io/managed-by: Tiller
app.kubernetes.io/name: metrics-server
helm.sh/chart: metrics-server-4.1.0
name: loopy-saola-metrics-server
namespace: default
resourceVersion: "77963814"
selfLink: /apis/extensions/v1beta1/namespaces/default/deployments/loopy-saola-metrics-server
uid: ae24b7b2-1811-11ea-a9a8-b8ca3a614e64
spec:
progressDeadlineSeconds: 600
replicas: 1
revisionHistoryLimit: 10
selector:
matchLabels:
app.kubernetes.io/instance: loopy-saola
app.kubernetes.io/name: metrics-server
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
creationTimestamp: null
labels:
app.kubernetes.io/instance: loopy-saola
app.kubernetes.io/managed-by: Tiller
app.kubernetes.io/name: metrics-server
helm.sh/chart: metrics-server-4.1.0
spec:
containers:
- command:
- metrics-server
- --secure-port=8443
- --v=8 以下三行为修改的内容
- --kubelet-insecure-tls
- --kubelet-preferred-address-types=InternalIP
image: docker.io/bitnami/metrics-server:0.3.6-debian-9-r27
imagePullPolicy: IfNotPresent
name: metrics-server
ports:
- containerPort: 8443
hostPort: 8443 增加 host port
name: https
protocol: TCP
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
serviceAccount: loopy-saola-metrics-server
serviceAccountName: loopy-saola-metrics-server
terminationGracePeriodSeconds: 30
status:
availableReplicas: 1
conditions:
- lastTransitionTime: "2019-12-06T10:17:05Z"
lastUpdateTime: "2019-12-06T10:17:05Z"
message: Deployment has minimum availability.
reason: MinimumReplicasAvailable
status: "True"
type: Available
- lastTransitionTime: "2019-12-06T10:16:54Z"
lastUpdateTime: "2019-12-06T10:40:41Z"
message: ReplicaSet "loopy-saola-metrics-server-58796b4bc7" has successfully progressed.
reason: NewReplicaSetAvailable
status: "True"
type: Progressing
observedGeneration: 3
readyReplicas: 1
replicas: 1
updatedReplicas: 1
$ kubectl top node
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
k8s-master01.gdfsxxds.rjyun 423m 1% 28532Mi 44%
k8s-master02.gdfsxxds.rjyun 334m 1% 5932Mi 9%
k8s-master03.gdfsxxds.rjyun 361m 1% 4107Mi 6%
k8s-node01 242m 1% 10696Mi 8%
k8s-node02 560m 1% 12201Mi 18%
k8s-node03 1811m 5% 28904Mi 45%
$ kubectl top pod
NAME CPU(cores) MEMORY(bytes)
a9vg-project-v2-deployment-7bc4968b44-mzp5n 2m 273Mi
a9vg-static-deployment-7bb7576cd8-hcmgm 1m 26Mi
cm-acme-http-solver-x8tkd 1m 7Mi
成功获取资源指标数据
第二种方式：
helm fetch stable/metrics-server
tar -xvf metrics-server-2.7.1.tgz
cd metrics-server
helm install --name metric --namespace kube-system --set image.repository=gcr.azk8s.cn/google_containers/metrics-server-amd64 .
（报错的话可能是镜像下载不下来，docker pull mirrorgooglecontainers/metrics-server-amd64:v0.3.5，然后 docker tag 成报错的镜像名称）
kubectl get pods -n kube-system -l release=metric
$ kubectl get deployments.apps -n kube-system metric-metrics-server -o yaml
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
deployment.kubernetes.io/revision: "3"
creationTimestamp: "2020-04-06T07:48:50Z"
generation: 3
labels:
app: metrics-server
chart: metrics-server-2.8.8
heritage: Tiller
release: metric
name: metric-metrics-server
namespace: kube-system
resourceVersion: "227096"
selfLink: /apis/apps/v1/namespaces/kube-system/deployments/metric-metrics-server
uid: 3562767f-934f-4f37-a84f-eedd9765444f
spec:
progressDeadlineSeconds: 600
replicas: 1
revisionHistoryLimit: 10
selector:
matchLabels:
app: metrics-server
release: metric
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
creationTimestamp: null
labels:
app: metrics-server
release: metric
spec:
affinity: {}
containers:
- command:
- /metrics-server
- --cert-dir=/tmp
- --logtostderr
- --secure-port=8443
- --kubelet-insecure-tls 新增加两行： 不启用证书以及解析ip地址
- --kubelet-preferred-address-types=InternalIP
image: gcr.azk8s.cn/google_containers/metrics-server-amd64:v0.3.5
imagePullPolicy: IfNotPresent
livenessProbe:
failureThreshold: 3
httpGet:
path: /healthz
port: https
scheme: HTTPS
initialDelaySeconds: 20
periodSeconds: 10
successThreshold: 1
timeoutSeconds: 1
name: metrics-server
ports:
- containerPort: 8443
name: https
protocol: TCP
readinessProbe:
failureThreshold: 3
httpGet:
path: /healthz
port: https
scheme: HTTPS
initialDelaySeconds: 20
periodSeconds: 10
successThreshold: 1
timeoutSeconds: 1
resources: {}
securityContext:
allowPrivilegeEscalation: false
capabilities:
drop:
- all
readOnlyRootFilesystem: true
runAsGroup: 10001
runAsNonRoot: true
runAsUser: 10001
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
volumeMounts:
- mountPath: /tmp
name: tmp
dnsPolicy: ClusterFirst
restartPolicy: Always
schedulerName: default-scheduler
securityContext: {}
serviceAccount: metric-metrics-server
serviceAccountName: metric-metrics-server
terminationGracePeriodSeconds: 30
volumes:
- emptyDir: {}
name: tmp
status:
availableReplicas: 1
conditions:
- lastTransitionTime: "2020-04-06T08:04:03Z"
lastUpdateTime: "2020-04-06T08:04:03Z"
message: Deployment has minimum availability.
reason: MinimumReplicasAvailable
status: "True"
type: Available
- lastTransitionTime: "2020-04-06T07:49:04Z"
lastUpdateTime: "2020-04-06T08:04:03Z"
message: ReplicaSet "metric-metrics-server-69c486c888" has successfully progressed.
reason: NewReplicaSetAvailable
status: "True"
type: Progressing
observedGeneration: 3
readyReplicas: 1
replicas: 1
updatedReplicas: 1
部署 hpa
$ kubectl get pod |grep ssr 部署前查看是一个pod
tgbus-ssr-deployment-85df9c9c59-2s4lw 1/1 Running 0 23h
$ cat hpa.yaml
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
name: tgbus-ssr-deployment
namespace: default
spec:
maxReplicas: 15
minReplicas: 2
scaleTargetRef:
apiVersion: extensions/v1beta1
kind: Deployment
name: tgbus-ssr-deployment
targetCPUUtilizationPercentage: 50
$ kubectl create -f hpa.yaml
$ kubectl get pod |grep ssr
tgbus-ssr-deployment-85df9c9c59-2s4lw 1/1 Running 0 23h
tgbus-ssr-deployment-85df9c9c59-k75xh 1/1 Running 0 42s
发现已创建出新的pod 后续会根据负载情况动态增加减少pod 数量
报错：
[root@yzsjhl-suanfan-master01 ~]# kubectl top node
Error from server (NotFound): the server could not find the requested resource (get services http:heapster:)
解决：
替换镜像：k8s.gcr.io/metrics-server-amd64:v0.3.1 换为 ： registry.aliyuncs.com/google_containers/metrics-server-amd64:v0.3.1
git clone https://github.com/kodekloudhub/kubernetes-metrics-server.git
kubectl create -f kubernetes-metrics-server/
Kubernetes Metrics Server第三种在二进制K8s上部署
helm repo add metrics-server https://kubernetes-sigs.github.io/metrics-server/
helm upgrade --install metrics-server metrics-server/metrics-server
- 修改 deployments 关闭证书配置 --kubelet-insecure-tls
spec:
containers:
- args:
- --secure-port=4443
- --cert-dir=/tmp
- --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
- --kubelet-use-node-status-port
- --metric-resolution=15s
- --kubelet-insecure-tls
image: yz.xxxxxcom/base/metrics-server:v0.6.1
# kubectl logs metrics-server-75fff855cd-flcbl -f
I0712 07:51:20.127701 1 serving.go:342] Generated self-signed cert (/tmp/apiserver.crt, /tmp/apiserver.key)
I0712 07:51:20.524337 1 requestheader_controller.go:169] Starting RequestHeaderAuthRequestController
I0712 07:51:20.524362 1 configmap_cafile_content.go:201] "Starting controller" name="client-ca::kube-system::extension-apiserver-authentication::client-ca-file"
I0712 07:51:20.524401 1 shared_informer.go:240] Waiting for caches to sync for client-ca::kube-system::extension-apiserver-authentication::client-ca-file
I0712 07:51:20.524380 1 shared_informer.go:240] Waiting for caches to sync for RequestHeaderAuthRequestController
I0712 07:51:20.524470 1 configmap_cafile_content.go:201] "Starting controller" name="client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file"
I0712 07:51:20.524525 1 shared_informer.go:240] Waiting for caches to sync for client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file
I0712 07:51:20.525014 1 dynamic_serving_content.go:131] "Starting controller" name="serving-cert::/tmp/apiserver.crt::/tmp/apiserver.key"
I0712 07:51:20.525500 1 secure_serving.go:266] Serving securely on [::]:4443
I0712 07:51:20.525615 1 tlsconfig.go:240] "Starting DynamicServingCertificateController"
W0712 07:51:20.528374 1 shared_informer.go:372] The sharedIndexInformer has started, run more than once is not allowed
I0712 07:51:20.625632 1 shared_informer.go:247] Caches are synced for client-ca::kube-system::extension-apiserver-authentication::requestheader-client-ca-file
I0712 07:51:20.625833 1 shared_informer.go:247] Caches are synced for RequestHeaderAuthRequestController
I0712 07:51:20.625830 1 shared_informer.go:247] Caches are synced for client-ca::kube-system::extension-apiserver-authentication::client-ca-file
# kubectl top node
NAME CPU(cores) CPU% MEMORY(bytes) MEMORY%
yz-devk8s-master10011 153m 3% 4096Mi 52%
yz-devk8s-master10012 129m 3% 2478Mi 31%
yz-devk8s-master10013 129m 3% 2647Mi 33%
yz-devk8s-node10015 1730m 2% 15299Mi 5%
yz-devk8s-node10016 750m 0% 12179Mi 4%
