k8s下kubeprometheus监控ingressnginx

首先需要已经配置好了一个ingress-nginx亦或者使用ACK上的ingress-nginx鉴于对ingress-nginx的状态，或者流量的监控是有一定的必要性，配置监控的指标有助于了解更多细节

通过使用kube-prometheus的项目来监控ingress-nginx，首先需要在nginx-ingress-controller的yaml中配置10254的端口，并且配置一个service，最后加入到ServiceMonitor即可。

start

如果是helm，则需要如下修改

• helm

  ..
  controller:
  metrics:
    enabled: true
    service:
      annotations:
        prometheus.io/port: "10254"
        prometheus.io/scrape: "true"
  ..

  如果不是 helm，则必须像这样编辑清单：服务清单：

    - name: prometheus
      port: 10254
      targetPort: prometheus

prometheus将会在service中被调用

apiVersion: v1
kind: Service
metadata:
 annotations:
   prometheus.io/scrape: "true"
   prometheus.io/port: "10254"
..
spec:
  ports:
    - name: prometheus
      port: 10254
      targetPort: prometheus
      ..

deployment

apiVersion: v1
kind: Deployment
metadata:
 annotations:
   prometheus.io/scrape: "true"
   prometheus.io/port: "10254"
..
spec:
  ports:
    - name: prometheus
      containerPort: 10254
      ..

测试10254的/metrics的url能够被访问到

bash-5.1$ curl 127.0.0.1:10254/metrics
# HELP go_gc_duration_seconds A summary of the pause duration of garbage collection cycles.
# TYPE go_gc_duration_seconds summary
go_gc_duration_seconds{quantile="0"} 1.9802e-05
go_gc_duration_seconds{quantile="0.25"} 3.015e-05
go_gc_duration_seconds{quantile="0.5"} 4.2054e-05
go_gc_duration_seconds{quantile="0.75"} 9.636e-05
go_gc_duration_seconds{quantile="1"} 0.000383868
go_gc_duration_seconds_sum 0.000972498
go_gc_duration_seconds_count 11
# HELP go_goroutines Number of goroutines that currently exist.
# TYPE go_goroutines gauge
go_goroutines 92
# HELP go_info Information about the Go environment.

Service And ServiceMonitor

另外需要配置一个ServiceMonitor, 这取决于kube-promentheus的发行版

spec部分字段如下

spec:
  endpoints:
  - interval: 15s   # 15s频率
    port: metrics   # port的名称
    path: /metrics  # url路径
  namespaceSelector:
    matchNames:
    - kube-system   # ingress-nginx所在的名称空间
  selector:
    matchLabels:
      app: ingress-nginx # ingress-nginx的标签

最终配置如下：

service在ingress-nginx的名称空间下配置，而ServiceMonitor在kube-prometheus的monitoring名称空间下，使用endpoints定义port名称，使用namespaceSelector.matchNames指定了ingress pod的名称空间，selector.matchLabels和标签

apiVersion: v1
kind: Service
metadata:
  name: ingress-nginx-metrics
  namespace: kube-system
  labels:
    app: ingress-nginx
  annotations:
    prometheus.io/port: "10254"
    prometheus.io/scrape: "true"
spec:
  type: ClusterIP
  ports:
  - name: metrics
    port: 10254
    targetPort: 10254
    protocol: TCP
  selector:
    app: ingress-nginx 
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
    name: ingress-nginx-metrics
    namespace: monitoring
spec:
  endpoints:
  - interval: 15s
    port: metrics
    path: /prometheus
  namespaceSelector:
    matchNames:
    - kube-system
  selector:
    matchLabels:
      app: ingress-nginx

• grafana

在grafana的dashboards中搜索ingress-nginx，得到的和github的官网的模板一样

https://grafana.com/grafana/dashboards/9614?pg=dashboards&plcmt=featured-dashboard-4

或者下面这个模板

这些在prometheus的targets中被发现

参考

ingress-nginx monitoringprometheus and grafana install