摘要：介紹架構(gòu)圖首先從獲取集群中所有的信息，然后通過(guò)這些上的獲取有用數(shù)據(jù)，而本身的數(shù)據(jù)則是從得到。所有獲取到的數(shù)據(jù)都被推到配置的后端存儲(chǔ)中，并還支持?jǐn)?shù)據(jù)的可視化。部署實(shí)施下載鏡像，上傳文件就可以了。

本篇文章，所使用的任何鏡像和yaml我都會(huì)發(fā)一個(gè)網(wǎng)盤(pán)鏈接，供大家下載學(xué)習(xí)！
鏈接:https://pan.baidu.com/s/1inmW... 密碼:92ua
github：https://github.com/heyangguang
有任何問(wèn)題可以直接聯(lián)系我的Email：heyangev@cn.ibm.com

Heapster是一個(gè)收集者，將每個(gè)Node上的cAdvisor的數(shù)據(jù)進(jìn)行匯總，然后導(dǎo)到第三方工具(如InfluxDB)。

架構(gòu)圖：

Heapster首先從K8S Master獲取集群中所有Node的信息，然后通過(guò)這些Node上的kubelet獲取有用數(shù)據(jù)，而kubelet本身的數(shù)據(jù)則是從cAdvisor得到。所有獲取到的數(shù)據(jù)都被推到Heapster配置的后端存儲(chǔ)中，并還支持?jǐn)?shù)據(jù)的可視化。現(xiàn)在后端存儲(chǔ) + 可視化的方法，如InfluxDB + grafana。

下載heapster鏡像，上傳heapster.yaml、heapster-mod.yaml文件apply就可以了。

k8smaster:
[root@k8smaster ~]# docker pull registry.cn-hangzhou.aliyuncs.com/google_containers/heapster-amd64:v1.5.4
Trying to pull repository registry.cn-hangzhou.aliyuncs.com/google_containers/heapster-amd64 ...
v1.5.4: Pulling from registry.cn-hangzhou.aliyuncs.com/google_containers/heapster-amd64
91380601464e: Pull complete
f351486260ed: Pull complete
Digest: sha256:c4a8d9c0007abb73a1b9e4f9c8bfb044e475aae2b4e6276ab2f8b13959cf6949
Status: Downloaded newer image for registry.cn-hangzhou.aliyuncs.com/google_containers/heapster-amd64:v1.5.4
[root@k8smaster ~]# ls
anaconda-ks.cfg  heapster-mod.yaml  heapster.yaml  images.tar  kube-flannel.yml
[root@k8smaster ~]# kubectl apply -f heapster.yaml
serviceaccount/heapster created
clusterrolebinding.rbac.authorization.k8s.io/heapster created
deployment.apps/heapster created
service/heapster created
[root@k8smaster ~]# kubectl apply -f heapster-mod.yaml
Warning: kubectl apply should be used on resource created by either kubectl create --save-config or kubectl apply
clusterrole.rbac.authorization.k8s.io/system:heapster configured

查看狀態(tài)：
[root@k8smaster ~]# kubectl get pods -n kube-system
NAME                                READY   STATUS    RESTARTS   AGE
coredns-86c58d9df4-kmfct            1/1     Running   0          23m
coredns-86c58d9df4-qn2k2            1/1     Running   0          23m
etcd-k8smaster                      1/1     Running   0          23m
heapster-569b679494-rktzf           1/1     Running   0          18s
kube-apiserver-k8smaster            1/1     Running   1          22m
kube-controller-manager-k8smaster   1/1     Running   0          22m
kube-flannel-ds-amd64-9rmfz         1/1     Running   0          19m
kube-flannel-ds-amd64-vnwtf         1/1     Running   0          15m
kube-flannel-ds-amd64-x7q4s         1/1     Running   0          15m
kube-proxy-7zl9n                    1/1     Running   0          22m
kube-proxy-t2sx9                    1/1     Running   0          23m
kube-proxy-txsfr                    1/1     Running   0          22m
kube-scheduler-k8smaster            1/1     Running   0          23m

稍等一會(huì)，使用kubectl top node查看集群狀態(tài)即可：
[root@k8smaster ~]# kubectl top node
NAME        CPU(cores)   CPU%   MEMORY(bytes)   MEMORY%
k8smaster   122m         3%     2848Mi          8%
k8snode-1   27m          0%     603Mi           1%
k8snode-2   26m          0%     582Mi           1%

到這里heapster收集集群數(shù)據(jù)教程就完成了！

希望大家可以給我指出問(wèn)題，我們一起前進(jìn)！

謝謝大家！