Prometheus監(jiān)控

1陋桂、Prometheus概述

prometheus本身就是一監(jiān)控系統(tǒng)，也分為server端和agent端，server端從被監(jiān)控主機(jī)獲取數(shù)據(jù)杨箭，而agent端需要部署一個(gè)node_exporter，主要用于數(shù)據(jù)采集和暴露節(jié)點(diǎn)的數(shù)據(jù)储狭，那么 在獲取Pod級(jí)別或者是mysql等多種應(yīng)用的數(shù)據(jù)互婿，也是需要部署相關(guān)的exporter。我們可以通過PromQL的方式對(duì)數(shù)據(jù)進(jìn)行查詢辽狈，但是由于本身prometheus屬于第三方的 解決方案慈参，原生的k8s系統(tǒng)并不能對(duì)Prometheus的自定義指標(biāo)進(jìn)行解析，就需要借助于k8s-prometheus-adapter將這些指標(biāo)數(shù)據(jù)查詢接口轉(zhuǎn)換為標(biāo)準(zhǔn)的Kubernetes自定義指標(biāo)刮萌。

如上圖猜绣，每個(gè)被監(jiān)控的主機(jī)都可以通過專用的exporter程序提供輸出監(jiān)控?cái)?shù)據(jù)的接口灰殴，并等待Prometheus服務(wù)器周期性的進(jìn)行數(shù)據(jù)抓取。如果存在告警規(guī)則掰邢，則抓取到數(shù)據(jù)之后會(huì)根據(jù)規(guī)則進(jìn)行計(jì)算牺陶，滿足告警條件則會(huì)生成告警，并發(fā)送到Alertmanager完成告警的匯總和分發(fā)辣之。當(dāng)被監(jiān)控的目標(biāo)有主動(dòng)推送數(shù)據(jù)的需求時(shí)掰伸，可以以Pushgateway組件進(jìn)行接收并臨時(shí)存儲(chǔ)數(shù)據(jù)，然后等待Prometheus服務(wù)器完成數(shù)據(jù)的采集召烂。

任何被監(jiān)控的目標(biāo)都需要事先納入到監(jiān)控系統(tǒng)中才能進(jìn)行時(shí)序數(shù)據(jù)采集碱工、存儲(chǔ)、告警和展示奏夫，監(jiān)控目標(biāo)可以通過配置信息以靜態(tài)形式指定怕篷，也可以讓Prometheus通過服務(wù)發(fā)現(xiàn)的機(jī)制進(jìn)行動(dòng)態(tài)管理。下面是組件的一些解析：

監(jiān)控代理程序：如node_exporter：收集主機(jī)的指標(biāo)數(shù)據(jù)酗昼，如平均負(fù)載廊谓、CPU、內(nèi)存麻削、磁盤蒸痹、網(wǎng)絡(luò)等等多個(gè)維度的指標(biāo)數(shù)據(jù)。
kubelet（cAdvisor）：收集容器指標(biāo)數(shù)據(jù)呛哟，也是K8S的核心指標(biāo)收集叠荠，每個(gè)容器的相關(guān)指標(biāo)數(shù)據(jù)包括：CPU使用率、限額扫责、文件系統(tǒng)讀寫限額榛鼎、內(nèi)存使用率和限額、網(wǎng)絡(luò)報(bào)文發(fā)送鳖孤、接收者娱、丟棄速率等等。
API Server：收集API Server的性能指標(biāo)數(shù)據(jù)苏揣，包括控制隊(duì)列的性能黄鳍、請(qǐng)求速率和延遲時(shí)長(zhǎng)等等
etcd：收集etcd存儲(chǔ)集群的相關(guān)指標(biāo)數(shù)據(jù)
kube-state-metrics：該組件可以派生出k8s相關(guān)的多個(gè)指標(biāo)數(shù)據(jù)，主要是資源類型相關(guān)的計(jì)數(shù)器和元數(shù)據(jù)信息平匈，包括制定類型的對(duì)象總數(shù)框沟、資源限額藏古、容器狀態(tài)以及Pod資源標(biāo)簽系列等。
Prometheus 能夠 直接 把 Kubernetes API Server 作為 服務(wù) 發(fā)現(xiàn) 系統(tǒng) 使用 進(jìn)而 動(dòng)態(tài) 發(fā)現(xiàn) 和 監(jiān)控 集群 中的 所有 可被 監(jiān)控 的 對(duì)象忍燥。 這里 需要 特別 說(shuō)明 的 是校翔， Pod 資源 需要 添加 下列 注解 信息 才 能被 Prometheus 系統(tǒng) 自動(dòng) 發(fā)現(xiàn) 并 抓取 其 內(nèi)建 的 指標(biāo) 數(shù)據(jù)。

1） prometheus. io/ scrape： 用于 標(biāo)識(shí) 是否 需要 被 采集 指標(biāo) 數(shù)據(jù)灾前， 布爾 型 值， true 或 false孟辑。
2） prometheus. io/ path： 抓取 指標(biāo) 數(shù)據(jù) 時(shí) 使用 的 URL 路徑哎甲， 一般 為/ metrics。
3） prometheus. io/ port： 抓取 指標(biāo) 數(shù)據(jù) 時(shí) 使 用的 套 接 字 端口饲嗽， 如 8080炭玫。
另外， 僅 期望 Prometheus 為 后端 生成 自定義 指標(biāo) 時(shí) 僅 部署 Prometheus 服務(wù)器 即可貌虾， 它 甚至 也不 需要 數(shù)據(jù) 持久 功能吞加。 但 若要 配置 完整 功能 的 監(jiān)控 系統(tǒng)， 管理員 還需 要在 每個(gè) 主機(jī) 上 部署 node_ exporter尽狠、 按 需 部署 其他 特有 類型 的 exporter 以及 Alertmanager衔憨。

2、Prometheus監(jiān)控架構(gòu)

3袄膏、Prometheus監(jiān)控配置

3.1 首先在Kubernetes中搭建Prometheus,yaml文件如下

deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: prometheus-dev-app
  namespace: prometheus-exporter-dev-app
  labels:
    app: prometheus-dev-app
spec:
  selector:
    matchLabels:
      app: prometheus-dev-app
  template:
    metadata:
      labels:
        app: prometheus-dev-app
    spec:
      serviceAccountName: prometheus-dev-app-sa
      volumes:
      - name: prometheus-dev-pvc
        persistentVolumeClaim:
          claimName: prometheus-dev-pvc
      - name: prometheus-dev-cm
        configMap:
          name: prometheus-dev-cm
      initContainers:
      - name: fix-permissions
        image: busybox
        command: [chown, -R, "nobody:nobody", /prometheus]
        volumeMounts:
        - name: prometheus-dev-pvc
          mountPath: /prometheus
      containers:
      - name: prometheus-dev
        image: prom/prometheus:v2.24.1
        args:
        - "--config.file=/etc/prometheus/prometheus.yml"
        - "--storage.tsdb.path=/prometheus"
        - "--storage.tsdb.retention.time=120h"
        - "--web.enable-admin-api"
        - "--web.enable-lifecycle"
        ports:
        - name: http
          containerPort: 9090
        volumeMounts:
        - name: prometheus-dev-cm
          mountPath: "/etc/prometheus"
        - name: prometheus-dev-pvc
          mountPath: "/prometheus"
        resources:
          requests:
            cpu: 100m
            memory: 512Mi
          limits:
            cpu: 100m
            memory: 2048Mi

pv pvc

apiVersion: v1
kind: PersistentVolume
metadata:
  name: prometheus-dev-pv
  labels:
    app: prometheus-dev-pv
spec:
  accessModes:
  - ReadWriteOnce
  capacity:
    storage: 10Gi
  storageClassName: local-storage
  local:
    path: /data/k8s/prometheus
  nodeAffinity:
    required:
      nodeSelectorTerms:
      - matchExpressions:
        - key: kubernetes.io/hostname
          operator: In
          values: ["es-k8s-app-dev-nd03"]
  persistentVolumeReclaimPolicy: Retain
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: prometheus-dev-pvc
  namespace: prometheus-exporter-dev-app
spec:
  selector:
    matchLabels:
      app: prometheus-dev-pv
  accessModes:
  - ReadWriteOnce
  resources:
    requests:
      storage: 10Gi
  storageClassName: local-storage

Cluster role

kind: ClusterRole
metadata:
 name: prometheus-dev-app-cr
rules:
- apiGroups:
 - ""
 resources:
 - configmaps
 - secrets
 - nodes
 - pods
 - services
 - resourcequotas
 - replicationcontrollers
 - limitranges
 - persistentvolumeclaims
 - persistentvolumes
 - namespaces
 - endpoints
 verbs:
 - get
 - list
 - watch
- apiGroups:
 - ""
 resources:
 - configmaps
 - nodes/metrics
 verbs:
 - get
- nonResourceURLs:
 - /metrics
 verbs:
 - get
- apiGroups:
 - apps
 resources:
 - statefulsets
 - daemonsets
 - deployments
 - replicasets
 verbs:
 - list
 - watch
- apiGroups:
 - batch
 resources:
 - cronjobs
 - jobs
 verbs:
 - list
 - watch
- apiGroups:
 - autoscaling
 resources:
 - horizontalpodautoscalers
 verbs:
 - list
 - watch
- apiGroups:
 - authentication.k8s.io
 resources:
 - tokenreviews
 verbs:
 - create
- apiGroups:
 - authorization.k8s.io
 resources:
 - subjectaccessreviews
 verbs:
 - create
- apiGroups:
 - policy
 resources:
 - poddisruptionbudgets
 verbs:
 - list
 - watch
- apiGroups:
 - certificates.k8s.io
 resources:
 - certificatesigningrequests
 verbs:
 - list
 - watch
- apiGroups:
 - storage.k8s.io
 resources:
 - storageclasses
 - volumeattachments
 verbs:
 - list
 - watch
- apiGroups:
 - admissionregistration.k8s.io
 resources:
 - mutatingwebhookconfigurations
 - validatingwebhookconfigurations
 verbs:
 - list
 - watch
- apiGroups:
 - networking.k8s.io
 resources:
 - networkpolicies
 - ingresses
 verbs:
 - list
 - watch
- apiGroups:
 - coordination.k8s.io
 resources:
 - leases
 verbs:
 - list
 - watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
 name: prometheus-dev-app-crb
roleRef:
 apiGroup: rbac.authorization.k8s.io
 kind: ClusterRole
 name: prometheus-dev-app-cr
subjects:
- kind: ServiceAccount
 name: prometheus-dev-app-sa
 namespace: prometheus-exporter-dev-app

Service

apiVersion: v1
kind: Service
metadata:
 name: prometheus-dev-app
 namespace: prometheus-exporter-dev-app
 labels:
   app: prometheus-dev-app
spec:
 type: NodePort
 ports:
 - name: prometheus-dev
   port: 9090
   targetPort: http
   nodePort: 30097
 selector:
   app: prometheus-dev-app

Configmap

apiVersion: v1
kind: ConfigMap
metadata:
  name: prometheus-dev-cm
  namespace: prometheus-exporter-dev-app
data:
  prometheus.yml: |
    global:
      scrape_interval: 15s
      scrape_timeout: 15s
    scrape_configs:
    - job_name: 'prometheus'
      static_configs:
      - targets: ['localhost:9090']

    - job_name: 'nodes'
      kubernetes_sd_configs:
      - role: node
     # static_configs:
     # - targets: ['10.20.20.100:10250','10.20.20.101:10250','10.20.20.102:10250']
      relabel_configs:
      - action: replace 
        source_labels: ['__address__']
        regex: '(.*):10250'
        replacement: '${1}:9100'
        target_label: __address__
      - action: labelmap
        regex: __meta_kubernetes_node_label_(.+)
    
    - job_name: 'kubelet'
      kubernetes_sd_configs:
      - role: node
      scheme: https
      tls_config:
        ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
        insecure_skip_verify: true
      bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
      relabel_configs:
      - action: labelmap
        regex: __meta_kubernetes_node_label_(.+)
    
    - job_name: 'cadvisor'
      kubernetes_sd_configs:
      - role: node
      scheme: https
      tls_config:
        ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
        insecure_skip_verify: true
      bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
      relabel_configs:
      - action: replace
        source_labels: ['__metrics_path__']
        regex: '(.*)'
        replacement: '${1}/cadvisor'
        target_label: __metrics_path__
      - action: labelmap
        regex: __meta_kubernetes_node_label_(.+)

    - job_name: 'apiserver'
      kubernetes_sd_configs:
      - role: endpoints 
      scheme: https
      tls_config:
        ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
        insecure_skip_verify: true
      bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token
      relabel_configs:
      - action: keep
        source_labels: ['__address__']
        regex: (.*):6443
      - action: labelmap
        regex: __meta_kubernetes_service_label_(.+)
    - job_name: 'pod'
      kubernetes_sd_configs:
      - role: endpoints
      relabel_configs:
      - source_labels: [__meta_kubernetes_service_annotation_prometheus_io_scrape]
        action: keep
        regex: true 
      - action: labelmap
        regex: __meta_kubernetes_service_label_(.+)
      - source_labels: [__address__, __meta_kubernetes_service_annotation_prometheus_io_port]
        action: replace
        target_label: __address__
        regex: ([^:]+)(?::\d+)?;(\d+)
        replacement: $1:$2
      - source_labels: [__meta_kubernetes_namespace]
        action: replace
        target_label: kubernetes_namespace
      - source_labels: [__meta_kubernetes_service_name]
        action: replace
        target_label: kubernetes_name
      - source_labels: [__meta_kubernetes_pod_name]
        action: replace
        target_label: kubernetes_pod_name

3.2 在Kubernetes部署Kube-metrics-state和node-exporter

node-exporter

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: dev-app-node-exporter
  namespace: prometheus-exporter-dev-app
  labels:
    app: dev-app-node-exporter
spec:
  selector:
    matchLabels:
      app: dev-app-node-exporter
  template:
    metadata:
      labels:
        app: dev-app-node-exporter
    spec:
      hostPID: true
      hostIPC: true
      hostNetwork: true
      nodeSelector:
        kubernetes.io/os: linux
      containers:
      - name: node-exporter
        image: prom/node-exporter:v1.1.1
        args:
        - --web.listen-address=$(HOSTIP):9100
        - --path.procfs=/host/proc
        - --path.sysfs=/host/sys
        - --path.rootfs=/host/root
        - --collector.filesystem.ignored-mount-points=^/(dev|proc|sys|var/lib/docker/.+)($|/)
        - --collector.filesystem.ignored-fs-types=^(autofs|binfmt_misc|cgroup|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|mqueue|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|sysfs|tracefs)$
        ports:
        - containerPort: 9100
        env:
        - name: HOSTIP
          valueFrom:
            fieldRef:
              fieldPath: status.hostIP
        resources:
          requests:
            cpu: 150m
            memory: 180Mi
          limits:
            cpu: 150m
            memory: 180Mi
        securityContext:
          runAsNonRoot: true
          runAsUser: 65534
        volumeMounts:
        - name: proc
          mountPath: /host/proc
        - name: sys
          mountPath: /host/sys
        - name: root
          mountPath: /host/root
          mountPropagation: HostToContainer
          readOnly: true
      tolerations:
      - operator: "Exists"
      volumes:
      - name: proc
        hostPath:
          path: /proc
      - name: dev
        hostPath:
          path: /dev
      - name: sys
        hostPath:
          path: /sys
      - name: root
        hostPath:
          path: /

kube-state-metrics

參考鏈接：https://github.com/kubernetes/kube-state-metrics
部署:

git clone https://github.com/kubernetes/kube-state-metrics.git
kubectl apply -f kube-state-metrics/examples/standard
根據(jù)需求修改對(duì)應(yīng)yaml配置

3.2践图、搭建Prometheus監(jiān)控

我這里通過docker運(yùn)行

docker run -d -p 9090:9090 --name=prometheus -v /opt/prometheus/new-prometheus.yml:/etc/prometheus/prometheus.yml -v /opt/prometheus/rules/:/etc/prometheus/rules -v /opt/prometheus/prom_job_conf/:/etc/prometheus/prom_job_conf/ prom/prometheus

Prometheus配置目錄結(jié)構(gòu)

.
├── new-prometheus.yml
├── new-prometheus.yml.bak
├── prom_job_conf
│   ├── kubernetes
│   │   ├── demo-cn.json
│   │   ├── dev.json
│   │   ├── pre.json
│   │   └── qa.json
│   ├── mongodb
│   │   ├── demo_cn_mongodb.json
│   │   ├── dev_mongodb.json
│   │   ├── pre_mongodb.json
│   │   └── qa_mongodb.json
│   ├── mysql
│   │   ├── demo_cn_mysql.json
│   │   ├── dev_mysql.json
│   │   ├── pre_mysql.json
│   │   └── qa_mysql.json
│   └── node
│       ├── demo_cn_db.json
│       ├── demo_cn.json
│       ├── dev_db.json
│       ├── dev.json
│       ├── pre_db.json
│       ├── pre.json
│       ├── qa_db.json
│       └── qa.json
└── rules
    ├── kubernetes.yml
    ├── mongo.yml
    ├── mysql.yml
    └── node.yml

Prometheus配置文件

global:
  scrape_interval:    60s
  scrape_timeout:     60s
  evaluation_interval:   60s
alerting:
  alertmanagers:
  - static_configs:
    - targets: ["10.168.101.80:9093"]

rule_files:
- "/etc/prometheus/rules/kubernetes.yml"
- "/etc/prometheus/rules/mysql.yml"
- "/etc/prometheus/rules/mongo.yml"

scrape_configs:
  - job_name: 'kubernetes'
    scrape_interval: 1m
    honor_labels: true
    metrics_path: '/federate'
    params:
      'match[]':
         - '{job=~"apiserver"}'
         - '{job=~"cadvisor"}'
         - '{job=~"kubelet"}'
    #     - '{job=~"nodes"}'
         - '{job=~"pod"}'
    #     - '{job=~"prometheus"}'
    #     - '{job=~"dev-mysql"}'
    #     - '{job=~"dev-mongodb"}'
    #     - '{job=~"qa-mysql"}'
    #     - '{job=~"qa-mongodb"}'

    file_sd_configs:
      - files:
        - /etc/prometheus/prom_job_conf/kubernetes/*.json
        refresh_interval: 5m
  - job_name: 'mysql'
    scrape_interval: 30s
    file_sd_configs:
      - files:
        - /etc/prometheus/prom_job_conf/mysql/*.json
        refresh_interval: 5m
  - job_name: 'node'
    scrape_interval: 30s
    file_sd_configs:
      - files:
        - /etc/prometheus/prom_job_conf/node/*.json
        refresh_interval: 5m
  - job_name: 'mongodb'
    scrape_interval: 30s
    file_sd_configs:
      - files:
        - /etc/prometheus/prom_job_conf/mongodb/*.json
        refresh_interval: 5m
···