1、k8s中namespace的作用?
答：Namespace命名空间，是kubernetes系统中的另一个非常重要的概念，Namespace在很多情况下用于实现多租户的资源隔离，不同的业务可以使用不同的namespace进行隔离。
2、k8s中创建namespace。
1 [root@k8s-master ~]# kubectl create namespace biehl 2 namespace "biehl" created 3 [root@k8s-master ~]#
3、k8s中查看namespace。
1 [root@k8s-master ~]# kubectl get namespace 2 NAME STATUS AGE 3 biehl Active 38s 4 default Active 17d 5 kube-system Active 17d
可以查看指定的namespace，或者查看所有的namespace的内容，如下所示：
1 [root@k8s-master ~]# kubectl get all --namespace=kube-system 2 NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE 3 deploy/kube-dns 1 1 1 1 4d 4 deploy/kubernetes-dashboard-latest 1 1 1 1 3d 5 6 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 7 svc/kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 4d 8 svc/kubernetes-dashboard 10.254.12.102 <none> 80/TCP 3d 9 10 NAME DESIRED CURRENT READY AGE 11 rs/kube-dns-778415672 1 1 1 4d 12 rs/kubernetes-dashboard-latest-3333846798 1 1 1 3d 13 14 NAME READY STATUS RESTARTS AGE 15 po/kube-dns-778415672-q23st 4/4 Running 4 58m 16 po/kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 3d 17 [root@k8s-master ~]# kubectl get all --all-namespaces 18 NAMESPACE NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE 19 kube-system deploy/kube-dns 1 1 1 1 4d 20 kube-system deploy/kubernetes-dashboard-latest 1 1 1 1 3d 21 22 NAMESPACE NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 23 default svc/kubernetes 10.254.0.1 <none> 443/TCP 17d 24 kube-system svc/kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 4d 25 kube-system svc/kubernetes-dashboard 10.254.12.102 <none> 80/TCP 3d 26 27 NAMESPACE NAME DESIRED CURRENT READY AGE 28 kube-system rs/kube-dns-778415672 1 1 1 4d 29 kube-system rs/kubernetes-dashboard-latest-3333846798 1 1 1 3d 30 31 NAMESPACE NAME READY STATUS RESTARTS AGE 32 default po/busybox2 1/1 Running 9 4d 33 kube-system po/kube-dns-778415672-q23st 4/4 Running 4 58m 34 kube-system po/kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 3d 35 [root@k8s-master ~]#
4、删除namespace，注意，特别危险，会删除namespace下所有的k8s资源。
1 [root@k8s-master ~]# kubectl delete namespace biehl 2 namespace "biehl" deleted 3 [root@k8s-master ~]# kubectl get namespace 4 NAME STATUS AGE 5 biehl Terminating 1m 6 default Active 17d 7 kube-system Active 17d 8 [root@k8s-master ~]# kubectl get namespace 9 NAME STATUS AGE 10 default Active 17d 11 kube-system Active 17d 12 [root@k8s-master ~]#
5、练习使用k8s的namespace进行测试。
1 [root@k8s-master ~]# cd k8s/ 2 [root@k8s-master k8s]# ls 3 book-master.war dashboard dashboard.zip deploy health pod rc skydns skydns.zip svc tomcat_demo tomcat_demo.zip 4 [root@k8s-master k8s]# mkdir namespace 5 [root@k8s-master k8s]# cd namespace/ 6 [root@k8s-master namespace]# ls 7 [root@k8s-master namespace]# cp ../rc/nginx_rc.yaml . 8 [root@k8s-master namespace]# cp ../svc/nginx_svc.yaml . 9 [root@k8s-master namespace]# ls 10 nginx_rc.yaml nginx_svc.yaml 11 [root@k8s-master namespace]#
修改nginx_rc.yaml配置文件，将namespace添加到metadata配置下面。
1 [root@k8s-master namespace]# vim nginx_rc.yaml
开始创建这个RC副本控制器，观察使用namespace的神奇之处，即使RC的名称相同，但是它们属于不同的namespace也是可以同时存在的。
1 [root@k8s-master namespace]# kubectl create namespace biehl 2 namespace "biehl" created 3 [root@k8s-master namespace]# kubectl create -f nginx_rc.yaml 4 replicationcontroller "myweb" created 5 [root@k8s-master namespace]# kubectl get all --namespace=biehl 6 NAME DESIRED CURRENT READY AGE 7 rc/myweb 2 2 2 23s 8 9 NAME READY STATUS RESTARTS AGE 10 po/myweb-5qwsn 1/1 Running 0 23s 11 po/myweb-6j0qs 1/1 Running 0 23s 12 [root@k8s-master namespace]# kubectl get all --all-namespaces 13 NAMESPACE NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE 14 kube-system deploy/kube-dns 1 1 1 1 4d 15 kube-system deploy/kubernetes-dashboard-latest 1 1 1 1 3d 16 17 NAMESPACE NAME DESIRED CURRENT READY AGE 18 biehl rc/myweb 2 2 2 31s 19 20 NAMESPACE NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 21 default svc/kubernetes 10.254.0.1 <none> 443/TCP 17d 22 kube-system svc/kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 4d 23 kube-system svc/kubernetes-dashboard 10.254.12.102 <none> 80/TCP 3d 24 25 NAMESPACE NAME DESIRED CURRENT READY AGE 26 kube-system rs/kube-dns-778415672 1 1 1 4d 27 kube-system rs/kubernetes-dashboard-latest-3333846798 1 1 1 3d 28 29 NAMESPACE NAME READY STATUS RESTARTS AGE 30 biehl po/myweb-5qwsn 1/1 Running 0 31s 31 biehl po/myweb-6j0qs 1/1 Running 0 31s 32 default po/busybox2 1/1 Running 9 4d 33 kube-system po/kube-dns-778415672-q23st 4/4 Running 4 1h 34 kube-system po/kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 3d 35 [root@k8s-master namespace]#
如果此时需要被外界访问，就需要创建SVC的，修改nginx_svc.yaml配置文件，将namespace添加到metadata配置下面。需要注意的是svc和rc必须处于同一个namespace下面才可以的
1 [root@k8s-master namespace]# vim nginx_svc.yaml
开始创建这个SVC，并进行查看，如下所示：
1 [root@k8s-master namespace]# kubectl create -f nginx_svc.yaml 2 service "myweb" created 3 [root@k8s-master namespace]# kubectl get all --namespace=biehl 4 NAME DESIRED CURRENT READY AGE 5 rc/myweb 2 2 2 4m 6 7 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 8 svc/myweb 10.254.154.202 <nodes> 80:30000/TCP 6s 9 10 NAME READY STATUS RESTARTS AGE 11 po/myweb-5qwsn 1/1 Running 0 4m 12 po/myweb-6j0qs 1/1 Running 0 4m 13 [root@k8s-master namespace]# kubectl get all --all-namespaces 14 NAMESPACE NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE 15 kube-system deploy/kube-dns 1 1 1 1 4d 16 kube-system deploy/kubernetes-dashboard-latest 1 1 1 1 3d 17 18 NAMESPACE NAME DESIRED CURRENT READY AGE 19 biehl rc/myweb 2 2 2 4m 20 21 NAMESPACE NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 22 biehl svc/myweb 10.254.154.202 <nodes> 80:30000/TCP 14s 23 default svc/kubernetes 10.254.0.1 <none> 443/TCP 17d 24 kube-system svc/kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 4d 25 kube-system svc/kubernetes-dashboard 10.254.12.102 <none> 80/TCP 3d 26 27 NAMESPACE NAME DESIRED CURRENT READY AGE 28 kube-system rs/kube-dns-778415672 1 1 1 4d 29 kube-system rs/kubernetes-dashboard-latest-3333846798 1 1 1 3d 30 31 NAMESPACE NAME READY STATUS RESTARTS AGE 32 biehl po/myweb-5qwsn 1/1 Running 0 4m 33 biehl po/myweb-6j0qs 1/1 Running 0 4m 34 default po/busybox2 1/1 Running 9 4d 35 kube-system po/kube-dns-778415672-q23st 4/4 Running 4 1h 36 kube-system po/kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 3d 37 [root@k8s-master namespace]#
可以访问一下，看是否已经可以被外界所访问了，如下所示：
1 [root@k8s-master namespace]# curl -I 192.168.110.133:30000 2 HTTP/1.1 200 OK 3 Server: nginx/1.13.12 4 Date: Mon, 22 Jun 2020 13:01:09 GMT 5 Content-Type: text/html 6 Content-Length: 612 7 Last-Modified: Mon, 09 Apr 2018 16:01:09 GMT 8 Connection: keep-alive 9 ETag: "5acb8e45-264" 10 Accept-Ranges: bytes 11 12 [root@k8s-master namespace]#
可以k8s的Dashboard的图形化界面进行观看。
6、Kubernetes（k8s）中反向代理访问k8s中的应用，使用proxy的方式访问k8s中的服务。
6.1、访问k8中应用的方式，需要创建一个server，其实service里面有不同的类型，比如说NodePort类型、ClusterIP类型。
1）、第一种，是通过NodePort的类型。
1 type: NodePort 2 ports: 3 - port: 80 4 targetPort: 80 5 nodePort: 30008
2）、第二种，是ClusterIP类型，比如k8s的databoard界面访问就是通过ClusterIP类型进行访问的。默认类型使用的是ClusterIP类型的。
1 type: ClusterIP 2 ports: 3 - port: 80 4 targetPort: 80
比如ClusterIP类型的访问，http://192.168.110.133:8080/api/v1/proxy/namespaces/kube-system/services/kubernetes-dashboard，可以替换namespace的名称，service的名称进行访问，如下所示：
1 [root@k8s-master namespace]# kubectl get all --namespace=biehl 2 NAME DESIRED CURRENT READY AGE 3 rc/myweb 2 2 2 22m 4 5 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 6 svc/myweb 10.254.154.202 <nodes> 80:30000/TCP 18m 7 8 NAME READY STATUS RESTARTS AGE 9 po/myweb-5qwsn 1/1 Running 0 22m 10 po/myweb-6j0qs 1/1 Running 0 22m 11 [root@k8s-master namespace]#
7、部署k8s的heapster监控。
7.1、Kubernetes扩展组件。
1）、kube-dns，负责为整个集群提供DNS服务。
2）、Ingress Controller，为服务提供外网入口。
3）、Heapster，提供资源监控。
4）、Dashboard，提供GUI。
5）、Federation，提供跨可用区的集群。
6）、Fluentd-elasticsearch，提供集群日志采集，存储与查询。
7.2、heapster监控的安装与部署。
1 [root@k8s-master ~]# cd k8s/ 2 [root@k8s-master k8s]# ls 3 book-master.war dashboard dashboard.zip deploy health namespace pod rc skydns skydns.zip svc tomcat_demo tomcat_demo.zip 4 [root@k8s-master k8s]# mkdir heapster 5 [root@k8s-master k8s]# cd heapster/ 6 [root@k8s-master heapster]# wget https://www.qstack.com.cn/heapster-influxdb.zip 7 --2020-06-25 15:05:44-- https://www.qstack.com.cn/heapster-influxdb.zip 8 Resolving www.qstack.com.cn (www.qstack.com.cn)... 111.202.85.37, 123.125.46.149 9 Connecting to www.qstack.com.cn (www.qstack.com.cn)|111.202.85.37|:443... connected. 10 HTTP request sent, awaiting response... 200 OK 11 Length: 2636 (2.6K) [application/zip] 12 Saving to: ‘heapster-influxdb.zip’ 13 14 100%[=======================================================================================================================================================================>] 2,636 --.-K/s in 0.001s 15 16 2020-06-25 15:05:45 (2.11 MB/s) - ‘heapster-influxdb.zip’ saved [2636/2636] 17 18 [root@k8s-master heapster]# unzip heapster-influxdb.zip 19 Archive: heapster-influxdb.zip 20 creating: heapster-influxdb/ 21 inflating: heapster-influxdb/grafana-service.yaml 22 inflating: heapster-influxdb/heapster-controller.yaml 23 inflating: heapster-influxdb/heapster-service.yaml 24 inflating: heapster-influxdb/influxdb-grafana-controller.yaml 25 inflating: heapster-influxdb/influxdb-service.yaml 26 [root@k8s-master heapster]# ls 27 heapster-influxdb heapster-influxdb.zip 28 [root@k8s-master heapster]# cd heapster-influxdb/ 29 [root@k8s-master heapster-influxdb]# ls 30 grafana-service.yaml heapster-controller.yaml heapster-service.yaml influxdb-grafana-controller.yaml influxdb-service.yaml 31 [root@k8s-master heapster-influxdb]#
修改配置文件，将地址指定为自己的apiServer的地址，如下所示：
1 [root@k8s-master heapster-influxdb]# vim heapster-controller.yaml
修改内容如下所示：
需要，注意的是每个配置文件里面都需要有它的一个镜像，这里的镜像都是从官方仓库下载的，也可以下载好上传到自己的私有仓库里面。
1 [root@k8s-master heapster-influxdb]# ls 2 grafana-service.yaml heapster-controller.yaml heapster-service.yaml influxdb-grafana-controller.yaml influxdb-service.yaml 3 [root@k8s-master heapster-influxdb]# kubectl create -f . 4 service "monitoring-grafana" created 5 replicationcontroller "heapster" created 6 service "heapster" created 7 replicationcontroller "influxdb-grafana" created 8 service "monitoring-influxdb" created 9 [root@k8s-master heapster-influxdb]#
上面的命令执行结束，五个都创建了。开始检查是否部署成功了，如下所示：
1 [root@k8s-master heapster-influxdb]# kubectl get svc --namespace=kube-system 2 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 3 heapster 10.254.196.163 <none> 80/TCP 2m 4 kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 6d 5 kubernetes-dashboard 10.254.12.102 <none> 80/TCP 6d 6 monitoring-grafana 10.254.196.53 <none> 80/TCP 2m 7 monitoring-influxdb 10.254.0.197 <none> 8083/TCP,8086/TCP 2m 8 [root@k8s-master heapster-influxdb]# kubectl get pod --namespace=kube-system 9 NAME READY STATUS RESTARTS AGE 10 heapster-26qj0 0/1 ImagePullBackOff 0 2m 11 influxdb-grafana-g1pg1 0/2 ErrImagePull 0 2m 12 kube-dns-778415672-q23st 4/4 Running 4 2d 13 kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 6d
发现svc是创建了，但是pod却创建失败了，这里我还是先下载下来，再上传到私有仓库进行创建吧。首先将heapster-amd64镜像拉取下来。
1 [root@k8s-master ~]# docker images 2 REPOSITORY TAG IMAGE ID CREATED SIZE 3 192.168.110.133:5000/tomcat latest 2eb5a120304e 2 weeks ago 647 MB 4 docker.io/tomcat latest 2eb5a120304e 2 weeks ago 647 MB 5 192.168.110.133:5000/mysql 5.7.30 9cfcce23593a 2 weeks ago 448 MB 6 docker.io/mysql 5.7.30 9cfcce23593a 2 weeks ago 448 MB 7 docker.io/busybox latest 1c35c4412082 3 weeks ago 1.22 MB 8 docker.io/registry latest 708bc6af7e5e 5 months ago 25.8 MB 9 192.168.110.133:5000/nginx 1.15 53f3fd8007f7 13 months ago 109 MB 10 docker.io/nginx 1.15 53f3fd8007f7 13 months ago 109 MB 11 192.168.110.133:5000/kubernetes-dashboard-amd64 v1.10.0 9e12bc435ba6 15 months ago 122 MB 12 registry.cn-qingdao.aliyuncs.com/wangxiaoke/kubernetes-dashboard-amd64 v1.10.0 9e12bc435ba6 15 months ago 122 MB 13 192.168.110.133:5000/nginx 1.13 ae513a47849c 2 years ago 109 MB 14 docker.io/nginx 1.13 ae513a47849c 2 years ago 109 MB 15 registry.access.redhat.com/rhel7/pod-infrastructure latest 99965fb98423 2 years ago 209 MB 16 192.168.110.133:5000/pod-infrastructure latest 34d3450d733b 3 years ago 205 MB 17 registry.cn-hangzhou.aliyuncs.com/google-containers/kubernetes-dashboard-amd64 v1.5.0 e5133bac8024 3 years ago 88.9 MB 18 192.168.110.133:5000/kubernetes-dashboard-amd64 v1.5.0 e5133bac8024 3 years ago 88.9 MB 19 myhub.fdccloud.com/library/kubedns-amd64 1.9 26cf1ed9b144 3 years ago 47 MB 20 myhub.fdccloud.com/library/dnsmasq-metrics-amd64 1.0 5271aabced07 3 years ago 14 MB 21 myhub.fdccloud.com/library/kube-dnsmasq-amd64 1.4 3ec65756a89b 3 years ago 5.13 MB 22 myhub.fdccloud.com/library/exechealthz-amd64 1.2 93a43bfb39bf 3 years ago 8.37 MB 23 [root@k8s-master ~]# docker ps 24 CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 25 9985f7777ad8 192.168.110.133:5000/pod-infrastructure:latest "/pod" 4 minutes ago Up 4 minutes k8s_POD.f88ae15c_influxdb-grafana-g1pg1_kube-system_18d4836b-b6b3-11ea-95a8-000c2919d52d_e8beeae9 26 48570ce53483 192.168.110.133:5000/pod-infrastructure:latest "/pod" 4 minutes ago Up 4 minutes k8s_POD.f88ae15c_heapster-26qj0_kube-system_1880f7ad-b6b3-11ea-95a8-000c2919d52d_f74ff6b4 27 9e809e332352 docker.io/busybox:latest "sleep 3600" 30 minutes ago Up 30 minutes k8s_busybox.f685e864_busybox2_default_d531ac21-b139-11ea-80b4-000c2919d52d_173f998d 28 9d9c246b8448 myhub.fdccloud.com/library/exechealthz-amd64:1.2 "/exechealthz '--c..." 2 days ago Up 2 days k8s_healthz.d09c0e9e_kube-dns-778415672-q23st_kube-system_38d64226-b47e-11ea-af83-000c2919d52d_5791ed6e 29 bf36e34ee5b5 myhub.fdccloud.com/library/dnsmasq-metrics-amd64:1.0 "/dnsmasq-metrics ..." 2 days ago Up 2 days k8s_dnsmasq-metrics.b0e0edc1_kube-dns-778415672-q23st_kube-system_38d64226-b47e-11ea-af83-000c2919d52d_25402444 30 d3f96f8c4a93 myhub.fdccloud.com/library/kube-dnsmasq-amd64:1.4 "/usr/sbin/dnsmasq..." 2 days ago Up 2 days k8s_dnsmasq.243653a3_kube-dns-778415672-q23st_kube-system_38d64226-b47e-11ea-af83-000c2919d52d_b6766883 31 617caefac177 myhub.fdccloud.com/library/kubedns-amd64:1.9 "/kube-dns --domai..." 2 days ago Up 2 days k8s_kubedns.80237e3f_kube-dns-778415672-q23st_kube-system_38d64226-b47e-11ea-af83-000c2919d52d_43df1c07 32 fff0e755ebbe 192.168.110.133:5000/pod-infrastructure:latest "/pod" 2 days ago Up 2 days k8s_POD.bec6e800_kube-dns-778415672-q23st_kube-system_38d64226-b47e-11ea-af83-000c2919d52d_e45259a0 33 66cd3f3b42e5 192.168.110.133:5000/pod-infrastructure:latest "/pod" 2 days ago Up 2 days k8s_POD.f88ae15c_busybox2_default_d531ac21-b139-11ea-80b4-000c2919d52d_fa900973 34 5e72b0961647 registry "/entrypoint.sh /e..." 7 days ago Up 2 days 0.0.0.0:5000->5000/tcp registry 35 [root@k8s-master ~]# docker search kubernetes/heapster:canary 36 INDEX NAME DESCRIPTION STARS OFFICIAL AUTOMATED 37 [root@k8s-master ~]# docker search kubernetes/heapster 38 INDEX NAME DESCRIPTION STARS OFFICIAL AUTOMATED 39 [root@k8s-master ~]# docker search heapster 40 INDEX NAME DESCRIPTION STARS OFFICIAL AUTOMATED 41 docker.io docker.io/fishchen/heapster-amd64 k8s.gcr.io/heapster-amd64 3 [OK] 42 docker.io docker.io/mirrorgooglecontainers/heapster-amd64 3 43 docker.io docker.io/ist0ne/heapster-grafana-amd64 gcr.io/google_containers/heapster-grafana-... 2 [OK] 44 docker.io docker.io/ist0ne/heapster-influxdb-amd64 gcr.io/google_containers/heapster-influxdb... 2 [OK] 45 docker.io docker.io/lvanneo/heapster-amd64 heapster-amd64 2 [OK] 46 docker.io docker.io/lvanneo/heapster-influxdb-amd64 heapster-influxdb-amd64 2 [OK] 47 docker.io docker.io/wavefronthq/heapster-amd64 2 48 docker.io docker.io/diamanti/heapster 1 49 docker.io docker.io/fishchen/heapster-influxdb-amd64 k8s.gcr.io/heapster-influxdb-amd64 1 [OK] 50 docker.io docker.io/ist0ne/heapster https://gcr.io/google_containers/heapster 1 [OK] 51 docker.io docker.io/lvanneo/heapster-grafana-amd64 heapster-grafana-amd64 1 [OK] 52 docker.io docker.io/vish/heapster 1 53 docker.io docker.io/alvintz/heapster_influxdb gcr.io/google_containers/heapster_influxdb 0 [OK] 54 docker.io docker.io/angelnu/heapster-grafana 0 55 docker.io docker.io/bitnami/heapster Bitnami Docker Image for Heapster 0 [OK] 56 docker.io docker.io/bonifaido/heapster Kubernetes Heapster that work on Rancher. 0 [OK] 57 docker.io docker.io/cedbossneo/heapster Heapster for Kubernetes 0.10.0 0 58 docker.io docker.io/ibmcom/heapster Docker Image for IBM Cloud private-CE (Com... 0 59 docker.io docker.io/ibmcom/heapster-ppc64le Docker Image for IBM Cloud Private-CE (Com... 0 60 docker.io docker.io/ist0ne/heapster-amd64 gcr.io/google_containers/heapster-amd64 0 [OK] 61 docker.io docker.io/pupudaye/heapster-influxdb-amd64 heapster-influxdb-amd64:v1.3.3 0 [OK] 62 docker.io docker.io/rancher/heapster-influxdb-amd64 0 63 docker.io docker.io/steady1211/heapster_grafana-v2.6.0-2 heapster_grafana-v2.6.0-2 0 [OK] 64 docker.io docker.io/vish/heapster-buddy-coreos 0 65 docker.io docker.io/zhaoqing/heapster-amd64 heapster-amd64:1.4.2 0 [OK] 66 [root@k8s-master ~]# docker pull docker.io/fishchen/heapster-amd64 67 Using default tag: latest 68 Trying to pull repository docker.io/fishchen/heapster-amd64 ... 69 sha256:694fedc23a10a39c8396dd0cec3625df11b809a0a4d7d215edc3becfc356835c: Pulling from docker.io/fishchen/heapster-amd64 70 ee522dc3e6e3: Pull complete 71 7f01af7be3bc: Pull complete 72 Digest: sha256:694fedc23a10a39c8396dd0cec3625df11b809a0a4d7d215edc3becfc356835c 73 Status: Downloaded newer image for docker.io/fishchen/heapster-amd64:latest 74 [root@k8s-master ~]#
然后将heapster_influxdb-amd64镜像拉取下来，如下所示：
1 [root@k8s-master ~]# docker search heapster_influxdb 2 INDEX NAME DESCRIPTION STARS OFFICIAL AUTOMATED 3 docker.io docker.io/ist0ne/heapster_influxdb-amd64 https://gcr.io/google_containers/heapster_... 1 [OK] 4 docker.io docker.io/aking666/heapster_influxdb Automated build heapster_influxdb image. 0 [OK] 5 docker.io docker.io/alvintz/heapster_influxdb gcr.io/google_containers/heapster_influxdb 0 [OK] 6 docker.io docker.io/arthas/heapster_influxdb heapster_influxdb 0 [OK] 7 docker.io docker.io/digbull/heapster_influxdb heapster_influxdb 0 [OK] 8 docker.io docker.io/dingzh/heapster_influxdb-v0.7 heapster_influxdb:v0.7 0 [OK] 9 docker.io docker.io/dockermonster/heapster_influxdb heapster_influxdb 0 [OK] 10 docker.io docker.io/eviloop/heapster_influxdb 0 11 docker.io docker.io/fociceo/heapster_influxdb heapster_influxdb 0 [OK] 12 docker.io docker.io/forestgun007/heapster_influxdb heapster_influxdb v0.7 0 [OK] 13 docker.io docker.io/haojianxun/heapster_influxdb 0 14 docker.io docker.io/hasura/heapster_influxdb Clone of gcr.io/google_containers/heapster... 0 15 docker.io docker.io/ist0ne/heapster_influxdb https://gcr.io/google_containers/heapster_... 0 [OK] 16 docker.io docker.io/locutus1/heapster_influxdb heapster_influxdb 0 [OK] 17 docker.io docker.io/maodouio/heapster_influxdb 0 18 docker.io docker.io/mirrorgooglecontainers/heapster_influxdb 0 19 docker.io docker.io/nelcy/heapster_influxdb 0 20 docker.io docker.io/sailsxu/heapster_influxdb heapster_influxdb:v0.6 0 [OK] 21 docker.io docker.io/shaloulcy/heapster_influxdb heapster_influxdb:v0.7 0 [OK] 22 docker.io docker.io/shenshouer/heapster_influxdb gcr.io/google_containers/heapster_influxdb... 0 23 docker.io docker.io/siriuszg/heapster_influxdb gcr.io/google_containers/heapster-influxdb... 0 [OK] 24 docker.io docker.io/storm2016/heapster_influxdb heapster_influxdb 0 [OK] 25 docker.io docker.io/typhoon1986/heapster_influxdb heapster_influxdb for kubernetes 0 26 docker.io docker.io/vish/heapster_influxdb 0 27 docker.io docker.io/zhaosijun/heapster_influxdb mirror from gcr.io/google_containers/heaps... 0 [OK] 28 [root@k8s-master ~]# docker pull docker.io/dingzh/heapster_influxdb-v0.7 29 Using default tag: latest 30 Trying to pull repository docker.io/dingzh/heapster_influxdb-v0.7 ... 31 sha256:2c9a3ac89f208147ab1d8731920cec681bf13596dcec381035623587a164cca1: Pulling from docker.io/dingzh/heapster_influxdb-v0.7 32 bbe5368a0432: Pull complete 33 98ff17f2ae39: Pull complete 34 b2c6d7e5c802: Pull complete 35 577d9278897e: Pull complete 36 a3ed95caeb02: Pull complete 37 e6bcb7eeeab6: Pull complete 38 ae45ea99a302: Pull complete 39 Digest: sha256:2c9a3ac89f208147ab1d8731920cec681bf13596dcec381035623587a164cca1 40 Status: Downloaded newer image for docker.io/dingzh/heapster_influxdb-v0.7:latest 41 [root@k8s-master ~]#
然后将heapster_grafana镜像拉取下来，如下所示：
1 [root@k8s-master ~]# docker search heapster_grafana 2 INDEX NAME DESCRIPTION STARS OFFICIAL AUTOMATED 3 docker.io docker.io/ist0ne/heapster_grafana https://gcr.io/google_containers/heapster_... 1 [OK] 4 docker.io docker.io/aking666/heapster_grafana Automated build heapster_grafana image. 0 [OK] 5 docker.io docker.io/alvintz/heapster_grafana gcr.io/google_containers/heapster_grafana 0 [OK] 6 docker.io docker.io/arthas/heapster_grafana heapster_grafana 0 [OK] 7 docker.io docker.io/caicloud/heapster_grafana 0 8 docker.io docker.io/cloudrti/heapster_grafana fork of the original to disable anonymous ... 0 [OK] 9 docker.io docker.io/digbull/heapster_grafana heapster_grafana 0 [OK] 10 docker.io docker.io/dingzh/heapster_grafana-v3.1.1 heapster_grafana:v3.1.1 0 [OK] 11 docker.io docker.io/dockermonster/heapster_grafana heapster_grafana 0 [OK] 12 docker.io docker.io/forestgun007/heapster_grafana heapster_grafana:v3.1.1 0 [OK] 13 docker.io docker.io/gcrxio/heapster_grafana 0 14 docker.io docker.io/ist0ne/heapster_grafana-amd64 https://gcr.io/google_containers/heapster_... 0 [OK] 15 docker.io docker.io/lupan/heapster_grafana heapster_grafana-v2.6.0-2 0 16 docker.io docker.io/mirrorgooglecontainers/heapster_grafana 0 17 docker.io docker.io/pkhanhao/heapster_grafana heapster_grafana 0 [OK] 18 docker.io docker.io/sailsxu/heapster_grafana heapster_grafana:v2.6.0 0 [OK] 19 docker.io docker.io/shaloulcy/heapster_grafana heapster_grafana 0 [OK] 20 docker.io docker.io/shenshouer/heapster_grafana gcr.io/google_containers/heapster_grafana:... 0 21 docker.io docker.io/siriuszg/heapster_grafana heapster_grafana 0 22 docker.io docker.io/steady1211/heapster_grafana-v2.6.0-2 heapster_grafana-v2.6.0-2 0 [OK] 23 docker.io docker.io/storm2016/heapster_grafana heapster_grafana 0 [OK] 24 docker.io docker.io/visenzek8s/heapster_grafana 0 25 docker.io docker.io/vish/heapster_grafana 0 26 docker.io docker.io/zhaosijun/heapster_grafana mirror from gcr.io/google_containers/heaps... 0 [OK] 27 docker.io docker.io/zzoujinn/heapster_grafana heapster_grafana 0 [OK] 28 [root@k8s-master ~]# docker pull docker.io/steady1211/heapster_grafana-v2.6.0-2 29 Using default tag: latest 30 Trying to pull repository docker.io/steady1211/heapster_grafana-v2.6.0-2 ... 31 sha256:5074ecd1033ca000ff642231b7a0f3af9a3e599ffd48ddd56c74670139eab63d: Pulling from docker.io/steady1211/heapster_grafana-v2.6.0-2 32 03e1855d4f31: Pull complete 33 a3ed95caeb02: Pull complete 34 7f1ce4d71e93: Pull complete 35 23d149931be4: Pull complete 36 2e86b9218e3a: Pull complete 37 db71c66d238d: Pull complete 38 de3678928269: Pull complete 39 Digest: sha256:5074ecd1033ca000ff642231b7a0f3af9a3e599ffd48ddd56c74670139eab63d 40 Status: Downloaded newer image for docker.io/steady1211/heapster_grafana-v2.6.0-2:latest 41 [root@k8s-master ~]#
将之前创建的删除掉svc、pod，需要根据namespace进行删除的哈，不然会找不到的。
1 [root@k8s-master heapster-influxdb]# kubectl get pod --namespace=kube-system 2 NAME READY STATUS RESTARTS AGE 3 heapster-26qj0 0/1 ImagePullBackOff 0 3m 4 influxdb-grafana-g1pg1 0/2 ErrImagePull 0 3m 5 kube-dns-778415672-q23st 4/4 Running 4 2d 6 kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 6d 7 [root@k8s-master heapster-influxdb]# kubectl delete svc --namespace=kube-system heapster monitoring-grafana monitoring-influxdb 8 service "heapster" deleted 9 service "monitoring-grafana" deleted 10 service "monitoring-influxdb" deleted 11 [root@k8s-master heapster-influxdb]# kubectl get svc --namespace=kube-system 12 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 13 kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 6d 14 kubernetes-dashboard 10.254.12.102 <none> 80/TCP 6d 15 [root@k8s-master heapster-influxdb]# kubectl get pod --namespace=kube-system 16 NAME READY STATUS RESTARTS AGE 17 heapster-26qj0 0/1 ImagePullBackOff 0 11m 18 influxdb-grafana-g1pg1 0/2 ImagePullBackOff 0 11m 19 kube-dns-778415672-q23st 4/4 Running 4 2d 20 kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 6d 21 [root@k8s-master heapster-influxdb]# kubectl delete pod --namespace=kube-system heapster-26qj0 influxdb-grafana-g1pg1 22 pod "heapster-26qj0" deleted 23 pod "influxdb-grafana-g1pg1" deleted 24 [root@k8s-master heapster-influxdb]# kubectl get pod --namespace=kube-system 25 NAME READY STATUS RESTARTS AGE 26 heapster-6fz3z 0/1 ContainerCreating 0 5s 27 influxdb-grafana-ms803 0/2 ContainerCreating 0 5s 28 kube-dns-778415672-q23st 4/4 Running 4 2d 29 kubernetes-dashboard-latest-3333846798-j8zjc 1/1 Running 1 6d 30 [root@k8s-master heapster-influxdb]#
此时，将下载下来的镜像上传到自己的私有仓库里面。
1 [root@k8s-master ~]# docker tag docker.io/fishchen/heapster-amd64:latest 192.168.110.133:5000/docker.io/fishchen/heapster-amd64:latest 2 [root@k8s-master ~]# docker push 192.168.110.133:5000/docker.io/fishchen/heapster-amd64:latest 3 The push refers to a repository [192.168.110.133:5000/docker.io/fishchen/heapster-amd64] 4 0c8eef97f390: Pushed 5 7034bc6e734f: Pushed 6 latest: digest: sha256:694fedc23a10a39c8396dd0cec3625df11b809a0a4d7d215edc3becfc356835c size: 739 7 [root@k8s-master ~]#
1 [root@k8s-master ~]# docker tag docker.io/dingzh/heapster_influxdb-v0.7:latest 192.168.110.133:5000/docker.io/dingzh/heapster_influxdb-v0.7:latest 2 [root@k8s-master ~]# docker push 192.168.110.133:5000/docker.io/dingzh/heapster_influxdb-v0.7:latest 3 The push refers to a repository [192.168.110.133:5000/docker.io/dingzh/heapster_influxdb-v0.7] 4 5f70bf18a086: Pushed 5 59b1df063c4d: Pushed 6 1a2eb7707e1e: Pushed 7 737f40e80b7f: Pushed 8 82b57dbc5385: Pushed 9 19429b698a22: Pushed 10 9436069b92a3: Pushed 11 latest: digest: sha256:3db1a13192c3ecb973c316b7dce485e61332011cfd60fa6a77f884583cdccce0 size: 3011 12 [root@k8s-master ~]#
1 [root@k8s-master ~]# docker push 192.168.110.133:5000/docker.io/steady1211/heapster_grafana-v2.6.0-2:latest 2 The push refers to a repository [192.168.110.133:5000/docker.io/steady1211/heapster_grafana-v2.6.0-2] 3 5f70bf18a086: Mounted from docker.io/dingzh/heapster_influxdb-v0.7 4 edec8b16494f: Pushed 5 ca627f7178ed: Pushed 6 a89b3190964f: Pushed 7 8683f0f614c9: Pushed 8 0828a6c7d921: Pushed 9 78dbfa5b7cbc: Pushed 10 latest: digest: sha256:40cf6e4b47c79092d299e658449ae687f42ffa9ec2de1dae3e4d7f550879b762 size: 3018 11 [root@k8s-master ~]#
此时，还需要进行修改配置文件，将镜像的地址填写为自己的私有仓库的地址。
Kubernetes Dashboard的界面默认是这样的，如下所示：
开始启动heapster，如下所示：
如果一个Pod里面有多个服务都需要被外界访问的时候，就需要为这两个服务分别建立service。
1 [root@k8s-master heapster-influxdb]# ls 2 grafana-service.yaml heapster-controller.yaml heapster-service.yaml influxdb-grafana-controller.yaml influxdb-service.yaml 3 [root@k8s-master heapster-influxdb]# kubectl create -f . 4 service "monitoring-grafana" created 5 replicationcontroller "heapster" created 6 service "heapster" created 7 replicationcontroller "influxdb-grafana" created 8 service "monitoring-influxdb" created 9 [root@k8s-master heapster-influxdb]# kubectl get pod -o wide --namespace=kube-system 10 NAME READY STATUS RESTARTS AGE IP NODE 11 heapster-vx4f2 1/1 Running 0 24s 172.16.47.5 k8s-master 12 influxdb-grafana-2q47p 2/2 Running 0 24s 172.16.47.6 k8s-master 13 kube-dns-778415672-q23st 4/4 Running 8 5d 172.16.47.4 k8s-master 14 kubernetes-dashboard-latest-3333846798-3mm68 1/1 Running 0 2m 172.16.16.3 k8s-node3 15 [root@k8s-master heapster-influxdb]#
检查heapster是否部署成功，如下所示：
1 [root@k8s-master heapster-influxdb]# kubectl get svc --namespace=kube-system 2 NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE 3 heapster 10.254.69.6 <none> 80/TCP 6m 4 kube-dns 10.254.230.254 <none> 53/UDP,53/TCP 9d 5 kubernetes-dashboard 10.254.127.32 <none> 80/TCP 8m 6 monitoring-grafana 10.254.65.105 <none> 80/TCP 6m 7 monitoring-influxdb 10.254.159.92 <none> 8083/TCP,8086/TCP 6m 8 [root@k8s-master heapster-influxdb]# kubectl get pod -o wide --namespace=kube-system 9 NAME READY STATUS RESTARTS AGE IP NODE 10 heapster-vx4f2 1/1 Running 0 6m 172.16.47.5 k8s-master 11 influxdb-grafana-2q47p 2/2 Running 0 6m 172.16.47.6 k8s-master 12 kube-dns-778415672-q23st 4/4 Running 8 5d 172.16.47.4 k8s-master 13 kubernetes-dashboard-latest-3333846798-3mm68 1/1 Running 0 8m 172.16.16.3 k8s-node3 14 [root@k8s-master heapster-influxdb]#
部署完毕heapster之后，刷新你的Kubernetes Dashboard的界面，就变成了这个样子了，如下所示：
8、Kubernetes架构图，如下所示：
每个Node节点上面都有kubelet服务和kube-proxy服务，在早期每个Node节点上还需要安装一个CAdvisor服务，但是新版中CAdvisor服务都集称到了kubelet服务中了，CAdvisor服务内部可以访问，如果需要外部访问的话，就需要配置一下kubelet的配置文件，将--cadvisor-port=8080添加到kubelet配置文件中，指定cadvisor对外暴漏的端口，不然无法进行访问。
1 [root@k8s-node2 ~]# vim /etc/kubernetes/kubelet
然后重启kubelet服务，如下所示：
1 [root@k8s-node2 ~]# systemctl restart kubelet.service 2 [root@k8s-node2 ~]#
重启之后，查看是否暴漏了8080端口号，如下所示：
然后访问这个服务器的8080端口号，如下所示：
heapster的各个cpu指标都是 从cAdvisor的指标里面取出来的。
