（二十三）Kubernetes系列之prometheus+grafana安装

1.kube-prometheus安装

1.1 下载

访问地址：https://github.com/prometheus-operator/kube-prometheus/releases

点击Source code (tar.gz)进行下载，并上传到k8s服务器master节点

1.2解压

tar zxvf kube-prometheus-0.11.0.tar.gz

1.3进入kube-prometheus-0.11.0/manifests目录

cd kube-prometheus-0.11.0

1.4部署（目录可直接部署kube-prometheus）

# create namespace and CRDs
kubectl create -f manifests/setup

# wait for CRD creation to complete
until kubectl get servicemonitors --all-namespaces ; do date; sleep 1; echo ""; done

# create monitoring components
kubectl create -f manifests/

注意：如果报镜像无法下载。

解决办法修改为国内镜像改完后重新执行kubectl apply -f manifests/

1.修改manifests/kubeStateMetrics-deployment.yaml的kube-state-metrics:2.5.0镜像

vi manifests/kubeStateMetrics-deployment.yaml

#image: k8s.gcr.io/kube-state-metrics/kube-state-metrics:v2.5.0
image: registry.cn-hangzhou.aliyuncs.com/my-name1/kube-state-metrics:2.5.0

2.修改manifests/prometheusAdapter-deployment.yaml的prometheus-adapter:v0.9.1镜像

vi manifests/prometheusAdapter-deployment.yaml

#image: k8s.gcr.io/prometheus-adapter/prometheus-adapter:v0.9.1
image: registry.cn-hangzhou.aliyuncs.com/my-name1/prometheus-adapter:v0.9.1

1.5查看

kubectl get pods -n monitoring

1.6使用nodeport方式把grafana的端口暴露出来

修改grafana的service，将clusterip改为nodeport

kubectl edit svc grafana -n monitoring

1.7查看上一步nodeport端口

kubectl get svc -n monitoring

1.8访问grafana

使用步骤1.7的端口来访问grafana

账号admin

密码admin

1.9 导入import模板

1.10选择数据源

默认已经添加了prometheus数据源，直接选择，点击import

1.11查看k8s监控