Kubernetes集群部署Prometheus和Grafana
一、环境规划
K8S集群角色 Ip 主机名 控制节点 192.168.84.155 master1 工作节点 192.168.84.156 node1 工作节点 192.168.84.157 node2
二、node-exporter安装和配置
2.1、node-exporter介绍
node-exporter可以采集机器（物理机、虚拟机、云主机等）的监控指标数据，能够采集到的指标包括CPU, 内存，磁盘，网络，文件数等信息。
2.2、node-exporter安装
# 创建监控namespace [root@master ~# kubectl create ns monitoring namespace/monitoringcreated # 创建node-export.yaml [root@master ~]# cat node-export.yaml apiVersion: apps/v1 kind: DaemonSet # 可以保证k8s集群的每个节点都运行完全一样的pod metadata: name: node-exporter namespace: monitoring labels: name: node-exporter spec: selector: matchLabels: name: node-exporter template: metadata: labels: name: node-exporter spec: hostPID: true hostIPC: true hostNetwork: true containers: - name: node-exporter image: prom/node-exporter:v0.16.0 ports: - containerPort: 9100 resources: requests: cpu: 0.15 # 这个容器运行至少需要0.15核cpu securityContext: privileged: true # 开启特权模式 args: - --path.procfs - /host/proc - --path.sysfs - /host/sys - --collector.filesystem.ignored-mount-points - '"^/(sys|proc|dev|host|etc)($|/)"' volumeMounts: - name: dev mountPath: /host/dev - name: proc mountPath: /host/proc - name: sys mountPath: /host/sys - name: rootfs mountPath: /rootfs tolerations: - key: "node-role.kubernetes.io/master" operator: "Exists" effect: "NoSchedule" volumes: - name: proc hostPath: path: /proc - name: dev hostPath: path: /dev - name: sys hostPath: path: /sys - name: rootfs hostPath: path: / # hostNetwork、hostIPC、hostPID都为True时，表示这个Pod里的所有容器，会直接使用宿主机的网络，直接与宿主机进行IPC（进程间通信）通信，可以看到宿主机里正在运行的所有进程。加入了hostNetwork:true会直接将我们的宿主机的9100端口映射出来，从而不需要创建service 在我们的宿主机上就会有一个9100的端口 # 更新node-exporter.yaml文件 [root@master ~]# kubectl apply -f node-export.yaml # 查看node-exporter是否部署成功 [root@master ~]# kubectl get pods -n monitoring -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES node-exporter-nl5qz 1/1 Running 0 13s 192.168.40.181 k8s-node1 <none> <none> node-exporter-nxwkf 1/1 Running 0 13s 192.168.40.180 k8s-master1 <none> <none> node-exporter-x494t 1/1 Running 0 13s 192.168.40.182 k8s-node2 <none> <none> # 通过node-exporter采集数据 curl http://主机ip:9100/metrics # node-export默认的监听端口是9100，可以看到当前主机获取到的所有监控数据 [root@master ~]# curl http://192.168.84.155:9100/metrics | grep node_cpu_seconds # HELP node_cpu_seconds_total Seconds the cpus spent in each mode. # TYPE node_cpu_seconds_total counter node_cpu_seconds_total{cpu="0",mode="idle"} 9429.89 node_cpu_seconds_total{cpu="0",mode="iowait"} 3.96 node_cpu_seconds_total{cpu="0",mode="irq"} 0 node_cpu_seconds_total{cpu="0",mode="nice"} 2.81 node_cpu_seconds_total{cpu="0",mode="softirq"} 45.77 node_cpu_seconds_total{cpu="0",mode="steal"} 0 node_cpu_seconds_total{cpu="0",mode="system"} 527.92 node_cpu_seconds_total{cpu="0",mode="user"} 847.3 node_cpu_seconds_total{cpu="1",mode="idle"} 9432.26 node_cpu_seconds_total{cpu="1",mode="iowait"} 5.12 node_cpu_seconds_total{cpu="1",mode="irq"} 0 node_cpu_seconds_total{cpu="1",mode="nice"} 2.81 node_cpu_seconds_total{cpu="1",mode="softirq"} 58 node_cpu_seconds_total{cpu="1",mode="steal"} 0 node_cpu_seconds_total{cpu="1",mode="system"} 528.33 node_cpu_seconds_total{cpu="1",mode="user"} 814.66 [root@k8s-master1 prometheus]# curl http://192.168.84.155:9100/metrics | grep node_load # HELP node_load1 1m load average. # TYPE node_load1 gauge node_load1 0.44 # HELP node_load15 15m load average. # TYPE node_load15 gauge node_load15 0.89 # HELP node_load5 5m load average. # TYPE node_load5 gauge node_load5 0.74
三、Prometheus安装和配置
3.1、Prometheus安装
1）创建账号，做rbac授权
# 创建一个monitoring账号monitor [root@master ~]# kubectl create serviceaccount monitor -n monitoring # 把monitoring账号monitor通过clusterrolebing绑定到clusterrole上 [root@master ~]# kubectl create clusterrolebinding monitor-clusterrolebinding -n monitoring --clusterrole=cluster-admin --serviceaccount=monitoring:monitor
2）创建prometheus数据存储目录
# 将prometheus调度到node1节点 [root@node1 ~]# mkdir /data && chmod 777 /data
3）创建一个configmap存储卷，用来存放prometheus配置信息
[root@master ~]# cat prometheus-cfg.yaml --- kind: ConfigMap apiVersion: v1 metadata: labels: app: prometheus name: prometheus-config namespace: monitoring data: prometheus.yml: | global: scrape_interval: 15s scrape_timeout: 10s evaluation_interval: 1m scrape_configs: - job_name: 'kubernetes-node' kubernetes_sd_configs: - role: node relabel_configs: - source_labels: [__address__] regex: '(.*):10250' replacement: '${1}:9100' target_label: __address__ action: replace - action: labelmap regex: __meta_kubernetes_node_label_(.+) - job_name: 'kubernetes-node-cadvisor' kubernetes_sd_configs: - role: node scheme: https tls_config: ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token relabel_configs: - action: labelmap regex: __meta_kubernetes_node_label_(.+) - target_label: __address__ replacement: kubernetes.default.svc:443 - source_labels: [__meta_kubernetes_node_name] regex: (.+) target_label: __metrics_path__ replacement: /api/v1/nodes/${1}/proxy/metrics/cadvisor - job_name: 'kubernetes-apiserver' kubernetes_sd_configs: - role: endpoints scheme: https tls_config: ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token relabel_configs: - source_labels: [__meta_kubernetes_namespace, __meta_kubernetes_service_name, __meta_kubernetes_endpoint_port_name] action: keep regex: default;kubernetes;https - job_name: 'kubernetes-service-endpoints' kubernetes_sd_configs: - role: endpoints relabel_configs: - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scrape] action: keep regex: true - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scheme] action: replace target_label: __scheme__ regex: (https?) - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_path] action: replace target_label: __metrics_path__ regex: (.+) - source_labels: [__address__, __meta_kubernetes_service_annotation_prometheus_io_port] action: replace target_label: __address__ regex: ([^:]+)(?::\d+)?;(\d+) replacement: $1:$2 - action: labelmap regex: __meta_kubernetes_service_label_(.+) - source_labels: [__meta_kubernetes_namespace] action: replace target_label: kubernetes_namespace - source_labels: [__meta_kubernetes_service_name] action: replace target_label: kubernetes_name [root@master ~]# kubectl apply -f prometheus-cfg.yaml configmap/prometheus-config created
配置详解：
--- kind: ConfigMap apiVersion: v1 metadata: labels: app: prometheus name: prometheus-config namespace: monitoring data: prometheus.yml: | global: scrape_interval: 15s #采集目标主机监控据的时间间隔 scrape_timeout: 10s # 数据采集超时时间，默认10s evaluation_interval: 1m #触发告警检测的时间，默认是1m scrape_configs: # 配置数据源，称为target，每个target用job_name命名。又分为静态配置和服务发现 - job_name: 'kubernetes-node' kubernetes_sd_configs: # 使用的是k8s的服务发现 - role: node # 使用node角色，它使用默认的kubelet提供的http端口来发现集群中每个node节点 relabel_configs: # 重新标记 - source_labels: [__address__] # 配置的原始标签，匹配地址 regex: '(.*):10250' #匹配带有10250端口的url replacement: '${1}:9100' #把匹配到的ip:10250的ip保留 target_label: __address__ #新生成的url是${1}获取到的ip:9100 action: replace # 动作替换 - action: labelmap regex: __meta_kubernetes_node_label_(.+) #匹配到下面正则表达式的标签会被保留,如果不做regex正则的话，默认只是会显示instance标签 - job_name: 'kubernetes-node-cadvisor' # 抓取cAdvisor数据，是获取kubelet上/metrics/cadvisor接口数据来获取容器的资源使用情况 kubernetes_sd_configs: - role: node scheme: https tls_config: ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token relabel_configs: - action: labelmap # 把匹配到的标签保留 regex: __meta_kubernetes_node_label_(.+) #保留匹配到的具有__meta_kubernetes_node_label的标签 - target_label: __address__ # 获取到的地址：__address__="192.168.40.180:10250" replacement: kubernetes.default.svc:443 # 把获取到的地址替换成新的地址kubernetes.default.svc:443 - source_labels: [__meta_kubernetes_node_name] regex: (.+) # 把原始标签中__meta_kubernetes_node_name值匹配到 target_label: __metrics_path__ #获取__metrics_path__对应的值 replacement: /api/v1/nodes/${1}/proxy/metrics/cadvisor # 把metrics替换成新的值api/v1/nodes/k8s-master1/proxy/metrics/cadvisor # ${1}是__meta_kubernetes_node_name获取到的值 # 新的url就是https://kubernetes.default.svc:443/api/v1/nodes/k8s-master1/proxy/metrics/cadvisor - job_name: 'kubernetes-apiserver' kubernetes_sd_configs: - role: endpoints # 使用k8s中的endpoint服务发现，采集apiserver 6443端口获取到的数据 scheme: https tls_config: ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token relabel_configs: - source_labels: [__meta_kubernetes_namespace, __meta_kubernetes_service_name, __meta_kubernetes_endpoint_port_name] # endpoint这个对象的名称空间,endpoint对象的服务名,exnpoint的端口名称 action: keep # 采集满足条件的实例，其他实例不采集 regex: default;kubernetes;https #正则匹配到的默认空间下的service名字是kubernetes，协议是https的endpoint类型保留下来 - job_name: 'kubernetes-service-endpoints' kubernetes_sd_configs: - role: endpoints relabel_configs: - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scrape] action: keep regex: true # 重新打标仅抓取到的具有 "prometheus.io/scrape: true" 的annotation的端点，意思是说如果某个service具有prometheus.io/scrape = true annotation声明则抓取，annotation本身也是键值结构，所以这里的源标签设置为键，而regex设置值true，当值匹配到regex设定的内容时则执行keep动作也就是保留，其余则丢弃。 - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scheme] action: replace target_label: __scheme__ regex: (https?) # 重新设置scheme，匹配源标签__meta_kubernetes_service_annotation_prometheus_io_scheme也就是prometheus.io/scheme annotation，如果源标签的值匹配到regex，则把值替换为__scheme__对应的值。 - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_path] action: replace target_label: __metrics_path__ regex: (.+) # 应用中自定义暴露的指标，也许你暴露的API接口不是/metrics这个路径，那么你可以在这个POD对应的service中做一个"prometheus.io/path = /mymetrics" 声明，上面的意思就是把你声明的这个路径赋值给__metrics_path__，其实就是让prometheus来获取自定义应用暴露的metrices的具体路径，不过这里写的要和service中做好约定，如果service中这样写 prometheus.io/app-metrics-path: '/metrics' 那么你这里就要__meta_kubernetes_service_annotation_prometheus_io_app_metrics_path这样写。 - source_labels: [__address__, __meta_kubernetes_service_annotation_prometheus_io_port] action: replace target_label: __address__ regex: ([^:]+)(?::\d+)?;(\d+) replacement: $1:$2 # 暴露自定义的应用的端口，就是把地址和你在service中定义的 "prometheus.io/port = <port>" 声明做一个拼接，然后赋值给__address__，这样prometheus就能获取自定义应用的端口，然后通过这个端口再结合__metrics_path__来获取指标，如果__metrics_path__值不是默认的/metrics那么就要使用上面的标签替换来获取真正暴露的具体路径。 - action: labelmap #保留下面匹配到的标签 regex: __meta_kubernetes_service_label_(.+) - source_labels: [__meta_kubernetes_namespace] action: replace # 替换__meta_kubernetes_namespace变成kubernetes_namespace target_label: kubernetes_namespace - source_labels: [__meta_kubernetes_service_name] action: replace
4）通过deployment部署prometheus
[root@master ~]# cat prometheus-deploy.yaml --- apiVersion: apps/v1 kind: Deployment metadata: name: prometheus-server namespace: monitoring labels: app: prometheus spec: replicas: 1 selector: matchLabels: app: prometheus component: server #matchExpressions: #- {key: app, operator: In, values: [prometheus]} #- {key: component, operator: In, values: [server]} template: metadata: labels: app: prometheus component: server annotations: prometheus.io/scrape: 'false' spec: nodeName: k8s-node1 # 指定pod调度到哪个节点上 serviceAccountName: monitor containers: - name: prometheus image: prom/prometheus:v2.22.2 imagePullPolicy: IfNotPresent command: - prometheus - --config.file=/etc/prometheus/prometheus.yml - --storage.tsdb.path=/prometheus # 数据存储目录 - --storage.tsdb.retention=720h # 数据保存时长 - --web.enable-lifecycle # 开启热加载 ports: - containerPort: 9090 protocol: TCP volumeMounts: - mountPath: /etc/prometheus/prometheus.yml name: prometheus-config subPath: prometheus.yml - mountPath: /prometheus/ name: prometheus-storage-volume volumes: - name: prometheus-config configMap: name: prometheus-config items: - key: prometheus.yml path: prometheus.yml mode: 0644 - name: prometheus-storage-volume hostPath: path: /data type: Directory [root@master ~]# kubectl apply -f prometheus-deploy.yaml [root@master ~]# kubectl get pods -o wide -n monitoring NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES node-exporter-2xmmh 1/1 Running 2 (47m ago) 59m 192.168.84.155 master <none> <none> node-exporter-cslvg 1/1 Running 1 (55m ago) 59m 192.168.84.157 node2 <none> <none> node-exporter-q8lvs 1/1 Running 2 (47m ago) 59m 192.168.84.156 node1 <none> <none> prometheus-server-6f957c546b-m44wk 1/1 Running 0 14m 192.166.166.132 node1 <none> <none>
5）给prometheus pod创建一个service
[root@master# cat prometheus-svc.yaml
apiVersion: v1
kind: Service
metadata:
name: prometheus
namespace: monitoring
labels:
app: prometheus
spec:
type: NodePort
ports:
- port: 9090
targetPort: 9090
protocol: TCP
selector:
app: prometheus
component: server
[root@k8s-master]# kubectl apply -f prometheus-svc.yaml service/prometheus created
查看service在物理机映射的端口
[root@master]#kubectl get svc -n monitoring
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
prometheus NodePort 10.96.16.67 <none> 9090:31146/TCP 32m
3.2、Prometheus热加载
# 为了每次修改配置文件可以热加载prometheus，也就是不停止prometheus，就可以使配置生效，想要使配置生效可用如下热加载命令： [root@master]# kubectl get pods -n monitoring -o wide -l app=prometheus NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES prometheus-server-689fb8cdbc-kcsw2 1/1 Running 0 5m39s 192.166.166.131 node1 <none> <none> # 想要使配置生效可用如下命令热加载： [root@master]# curl -X POST http://192.166.166.131:9090/-/reload # 查看log [root@master]# kubectl logs -n monitoring prometheus-server-689fb8cdbc-kcsw2
# 热加载速度比较慢，可以暴力重启prometheus，如修改上面的prometheus-cfg.yaml文件之后，可执行如下强制删除： [root@master]# kubectl delete -f prometheus-cfg.yaml [root@master]# kubectl delete -f prometheus-deploy.yaml # 然后再通过apply更新： [root@master]# kubectl apply -f prometheus-cfg.yaml [root@master]# kubectl apply -f prometheus-deploy.yaml #注意：线上最好热加载，暴力删除可能造成监控数据的丢失
四、Grafana的安装和配置
4.1、Grafana介绍
Grafana是一个跨平台的开源的度量分析和可视化工具，可以将采集的数据可视化的展示，并及时通知给告警接收方。它主要有以下六大特点：
1）展示方式：快速灵活的客户端图表，面板插件有许多不同方式的可视化指标和日志，官方库中具有丰富的仪表盘插件，比如热图、折线图、图表等多种展示方式；
2）数据源：Graphite，InfluxDB，OpenTSDB，Prometheus，Elasticsearch，CloudWatch和KairosDB等；
3）通知提醒：以可视方式定义最重要指标的警报规则，Grafana将不断计算并发送通知，在数据达到阈值时通过Slack、PagerDuty等获得通知；
4）混合展示：在同一图表中混合使用不同的数据源，可以基于每个查询指定数据源，甚至自定义数据源；
5）注释：使用来自不同数据源的丰富事件注释图表，将鼠标悬停在事件上会显示完整的事件元数据和标记。
4.2、Grafana安装
# 准备yaml文件 [root@master]# cat grafana.yaml apiVersion: apps/v1 kind: Deployment metadata: name: monitoring-grafana namespace: kube-system spec: replicas: 1 selector: matchLabels: task: monitoring k8s-app: grafana template: metadata: labels: task: monitoring k8s-app: grafana spec: containers: - name: grafana image: grafana/grafana:latest ports: - containerPort: 3000 protocol: TCP volumeMounts: - mountPath: /etc/ssl/certs name: ca-certificates readOnly: true - mountPath: /var/lib/grafana name: grafana-storage env: - name: INFLUXDB_HOST value: monitoring-influxdb - name: GF_SERVER_HTTP_PORT value: "3000"
- name: GF_AUTH_BASIC_ENABLED value: "false" - name: GF_AUTH_ANONYMOUS_ENABLED value: "true" - name: GF_AUTH_ANONYMOUS_ORG_ROLE value: Admin - name: GF_SERVER_ROOT_URL value: / volumes: - name: ca-certificates hostPath: path: /etc/ssl/certs - name: grafana-storage emptyDir: {} --- apiVersion: v1 kind: Service metadata: labels: kubernetes.io/cluster-service: 'true' kubernetes.io/name: monitoring-grafana name: monitoring-grafana namespace: kube-system spec: ports: - port: 80 targetPort: 3000 selector: k8s-app: grafana type: NodePort # 更新yaml文件： [root@master]# kubectl apply -f grafana.yaml # 查看grafana是否创建成功： [root@master]# kubectl get pods -n kube-system -l task=monitoring -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES monitoring-grafana-675798bf47-z9dpx 1/1 Running 0 19s 192.166.104.12 node2 <none> <none> # 查看grafana前端的service [root@master]# kubectl get svc -n kube-system | grep grafana monitoring-grafana NodePort 10.96.52.234 <none> 80:30371/TCP 63s
4.3、配置Grafana
1）登陆grafana，在浏览器访问http://192.168.84.155:30371
2）开始配置grafana的web界面：选择Create your first data source
4.4、导入监控模板
官方链接搜索：https://grafana.com/dashboards?dataSource=prometheus&search=kubernetes
4.4.1、监控node状态
4.4.2、监控容器状态
Metrics Server部署
下载 Metrics Server 的部署清单： curl -LO https://github.com/kubernetes-sigs/metrics-server/releases/download/v0.5.0/components.yaml
找到 args 部分。在默认情况下，该部分应该如下所示
args: - --cert-dir=/tmp - --secure-port=4443 - --kubelet-preferred-address-types=InternalIP,Hostname,InternalDNS,ExternalDNS,ExternalIP
将 args 部分替换为以下内容
args: - --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname - --kubelet-insecure-tls
这些参数将允许 Metrics Server 在没有正确验证 SSL 证书的情况下与 kubelet 进行通信。
保存并关闭 components.yaml 文件，然后重新部署 Metrics Server：
kubectl apply -f components.yaml
等待一段时间，以确保 Metrics Server 正常运行。可以使用以下命令检查 Metrics Server 是否已正确安装：
kubectl get deployment metrics-server -n kube-system
参考：Kubernetes集群部署Prometheus和Grafana - 运维人在路上 - 博客园 (cnblogs.com)
(22条消息) k8s上部署Grafana_南方游牧的博客-CSDN博客
