网站创建人/b站推广形式

本节重点介绍 :

• 打镜像，导出镜像，传输到各个节点并导入
• 运行该项目
• 配置prometheus和grafana

打镜像

本地build

docker build -t ink8s-pod-metrics:v1 .

build过程

导出镜像

docker save  ink8s-pod-metrics > ink8s-pod-metrics.tar 

传输到各个node节点上

scp ink8s-pod-metrics.tar k8s-node01:~

各个node节点上导入镜像

使用docker

docker load < ink8s-pod-metrics.tar

使用containerd

ctr --namespace k8s.io images import ink8s-pod-metrics.tar

运行该项目

 kubectl apply -f rbac.yamlkubectl apply -f deployment.yaml

检查

[root@k8s-master01 ink8s-pod-metrics]# kubectl get pod -o wide 
NAME                                           READY   STATUS    RESTARTS   AGE    IP              NODE         NOMINATED NODE   READINESS GATES
grafana-d5d85bcd6-f74ch                        1/1     Running   0          3d9h   10.100.85.199   k8s-node01   <none>           <none>
grafana-d5d85bcd6-l44mx                        1/1     Running   0          3d9h   10.100.85.198   k8s-node01   <none>           <none>
ink8s-pod-metrics-deployment-85d9795d6-95lsp   1/1     Running   0          13m    10.100.85.207   k8s-node01   <none>           <none>

• 日志

[root@k8s-master01 ink8s-pod-metrics]# kubectl logs -l app=ink8s-pod-metrics  -f  
2021-08-23 20:34:35.377256 INFO app/get_k8s_objs.go:128 [pod.label:map[component:etcd tier:control-plane]]
2021-08-23 20:34:35.377266 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-apiserver tier:control-plane]]
2021-08-23 20:34:35.377274 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-controller-manager tier:control-plane]]
2021-08-23 20:34:35.377292 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:85c698c6d4 k8s-app:kube-proxy pod-template-generation:1]]
2021-08-23 20:34:35.377299 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:85c698c6d4 k8s-app:kube-proxy pod-template-generation:1]]
2021-08-23 20:34:35.377317 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-scheduler tier:control-plane]]
2021-08-23 20:34:35.377324 INFO app/get_k8s_objs.go:128 [pod.label:map[app.kubernetes.io/name:kube-state-metrics app.kubernetes.io/version:v1.9.7 pod-template-hash:564668c858]]
2021-08-23 20:34:35.377331 INFO app/get_k8s_objs.go:128 [pod.label:map[k8s-app:metrics-server pod-template-hash:7dbf6c4558]]
2021-08-23 20:34:35.377336 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:prometheus-5b9cdcfd6c k8s-app:prometheus statefulset.kubernetes.io/pod-name:prometheus-0]]
2021-08-23 20:34:35.377358 INFO app/get_k8s_objs.go:143 server_pod_ips_result][num_pod:11][time_took_seconds:6.189551999]
2021-08-23 20:34:39.197614 INFO app/get_k8s_objs.go:107 server_node_ips_result][num_node:2][time_took_seconds:0.009575987]
2021-08-23 20:34:39.200824 INFO app/get_k8s_objs.go:128 [pod.label:map[k8s-app:kube-dns pod-template-hash:68b9d7b887]]
2021-08-23 20:34:39.200857 INFO app/get_k8s_objs.go:128 [pod.label:map[k8s-app:kube-dns pod-template-hash:68b9d7b887]]
2021-08-23 20:34:39.200871 INFO app/get_k8s_objs.go:128 [pod.label:map[component:etcd tier:control-plane]]
2021-08-23 20:34:39.200889 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-apiserver tier:control-plane]]
2021-08-23 20:34:39.200903 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-controller-manager tier:control-plane]]
2021-08-23 20:34:39.200920 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:85c698c6d4 k8s-app:kube-proxy pod-template-generation:1]]
2021-08-23 20:34:39.200934 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:85c698c6d4 k8s-app:kube-proxy pod-template-generation:1]]
2021-08-23 20:34:39.200947 INFO app/get_k8s_objs.go:128 [pod.label:map[component:kube-scheduler tier:control-plane]]
2021-08-23 20:34:39.200961 INFO app/get_k8s_objs.go:128 [pod.label:map[app.kubernetes.io/name:kube-state-metrics app.kubernetes.io/version:v1.9.7 pod-template-hash:564668c858]]
2021-08-23 20:34:39.200981 INFO app/get_k8s_objs.go:128 [pod.label:map[k8s-app:metrics-server pod-template-hash:7dbf6c4558]]
2021-08-23 20:34:39.200992 INFO app/get_k8s_objs.go:128 [pod.label:map[controller-revision-hash:prometheus-5b9cdcfd6c k8s-app:prometheus statefulset.kubernetes.io/pod-name:prometheus-0]]
2021-08-23 20:34:39.201022 INFO app/get_k8s_objs.go:143 server_pod_ips_result][num_pod:11][time_took_seconds:0.013052527]

node上请求 pod 的metrics

• curl pod的ip:8080/metrics

[root@k8s-master01 ink8s-pod-metrics]# curl -s 10.100.85.207:8080/metrics |grep ink8s
# HELP ink8s_pod_metrics_get_node_detail k8s node detail each
# TYPE ink8s_pod_metrics_get_node_detail gauge
ink8s_pod_metrics_get_node_detail{containerRuntimeVersion="containerd://1.4.4",hostname="k8s-master01",ip="172.20.70.205",kubeletVersion="v1.20.1"} 1
ink8s_pod_metrics_get_node_detail{containerRuntimeVersion="containerd://1.4.4",hostname="k8s-node01",ip="172.20.70.215",kubeletVersion="v1.20.1"} 1
# HELP ink8s_pod_metrics_get_node_last_duration_seconds get node last duration seconds
# TYPE ink8s_pod_metrics_get_node_last_duration_seconds gauge
ink8s_pod_metrics_get_node_last_duration_seconds 0.008066143
# HELP ink8s_pod_metrics_get_pod_control_plane_pod_detail k8s pod detail of control plane
# TYPE ink8s_pod_metrics_get_pod_control_plane_pod_detail gauge
ink8s_pod_metrics_get_pod_control_plane_pod_detail{component="etcd",ip="172.20.70.205",pod_name="etcd-k8s-master01"} 1
ink8s_pod_metrics_get_pod_control_plane_pod_detail{component="kube-apiserver",ip="172.20.70.205",pod_name="kube-apiserver-k8s-master01"} 1
ink8s_pod_metrics_get_pod_control_plane_pod_detail{component="kube-controller-manager",ip="172.20.70.205",pod_name="kube-controller-manager-k8s-master01"} 1
ink8s_pod_metrics_get_pod_control_plane_pod_detail{component="kube-scheduler",ip="172.20.70.205",pod_name="kube-scheduler-k8s-master01"} 1
# HELP ink8s_pod_metrics_get_pod_last_duration_seconds get pod last duration seconds
# TYPE ink8s_pod_metrics_get_pod_last_duration_seconds gauge
ink8s_pod_metrics_get_pod_last_duration_seconds 0.01159838

prometheus target页面检查pod

• 发现pod已经发现到了，但是 报错：向http的server发送https的请求

prometheus和grafana配置

检查 kubernetes-pods的job

• 如果之前配置的https，需要改为http的

- job_name: kubernetes-podshonor_timestamps: truescrape_interval: 30sscrape_timeout: 10smetrics_path: /metricsscheme: httpfollow_redirects: truerelabel_configs:- source_labels: [__meta_kubernetes_pod_annotation_prometheus_io_scrape]separator: ;regex: "true"replacement: $1action: keep- source_labels: [__meta_kubernetes_pod_annotation_prometheus_io_path]separator: ;regex: (.+)target_label: __metrics_path__replacement: $1action: replace- source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_port]separator: ;regex: ([^:]+)(?::\d+)?;(\d+)target_label: __address__replacement: $1:$2action: replace- separator: ;regex: __meta_kubernetes_pod_label_(.+)replacement: $1action: labelmap- source_labels: [__meta_kubernetes_namespace]separator: ;regex: (.*)target_label: kubernetes_namespacereplacement: $1action: replace- source_labels: [__meta_kubernetes_pod_name]separator: ;regex: (.*)target_label: kubernetes_pod_namereplacement: $1action: replacekubernetes_sd_configs:- role: podkubeconfig_file: ""follow_redirects: true

在prometheus中检查指标

• 查询 ink8s_pod_metrics_get_node_detail
• 

ink8s_pod_metrics_get_node_detail{app="ink8s-pod-metrics", containerRuntimeVersion="containerd://1.4.4", hostname="k8s-master01", instance="10.100.85.207:8080", ip="172.20.70.205", job="kubernetes-pods", kubeletVersion="v1.20.1", kubernetes_namespace="default", kubernetes_pod_name="ink8s-pod-metrics-deployment-85d9795d6-95lsp", pod_template_hash="85d9795d6"}
1
ink8s_pod_metrics_get_node_detail{app="ink8s-pod-metrics", containerRuntimeVersion="containerd://1.4.4", hostname="k8s-node01", instance="10.100.85.207:8080", ip="172.20.70.215", job="kubernetes-pods", kubeletVersion="v1.20.1", kubernetes_namespace="default", kubernetes_pod_name="ink8s-pod-metrics-deployment-85d9795d6-95lsp", pod_template_hash="85d9795d6"}

配置grafana

• 举例图片
• 
• 
• 
• 

本节重点总结 :

• 打镜像，导出镜像，传输到各个节点并导入
• 运行该项目
• 配置prometheus和grafana