Kubernetes + Prometheus + Grafana 集群监控

为什么80%的码农都做不了架构师？>>>   

地址: http://www.mknight.cn/post/631/

简介: Welcome to Prometheus! Prometheus is a monitoring platform that collects metrics from monitored targets by scraping metrics HTTP endpoints on these targets.

Prometheus

下载

1. wget https://github.com/prometheus/prometheus/releases/download/v1.5.2/prometheus-1.5.2.linux-amd64.tar.gz
2.  
3. tar -zxvf prometheus-1.5.2.linux-amd64.tar.gz -C /opt/prometheus --strip-components=1
4.  
5. cd /opt/prometheus
6. # 备份一下配置文件
7. mv prometheus.yml prometheus.yml-bak

修改配置文件,替换为自己的IP。

1. static_configs:
2. - targets: ['10.10.30.102:9090']

启动

1. /opt/prometheus/prometheus --config.file=prometheus.yml

检查9090端口是否监听，则正常。访问9090端口

Grafana

下载

1. wget https://s3-us-west-2.amazonaws.com/grafana-releases/release/grafana-4.0.1-1480694114.x86_64.rpm
2. yum localinstall grafana-4.0.1-1480694114.x86_64.rpm
3.  
4. service grafana-server start

配置数据源

保存后测试一下，如果不通过说明刚才的9090端口有问题。

增加dashboards模板

1. 下载dashboards
2. (https://github.com/percona/grafana-dashboards)
3.  
4. git clone https://github.com/percona/grafana-dashboards.git
5. cp -r grafana-dashboards/dashboards /var/lib/grafana/
6.  
7. 编辑 Grafana config
8.  
9. vi /etc/grafana/grafana.ini
10.  
11. [dashboards.json]
12. enabled = true
13. path = /var/lib/grafana/dashboards
14.  
15.  
16. systemctl restart grafana-server

Kubernetes监控

导入json模板

下载地址json模板。
然后Import

配置prometheus

最终配置文件如下所示：

1. # my global config
2. global:
3. scrape_interval: 15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
4. evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
5. # scrape_timeout is set to the global default (10s).
6.  
7. # Attach these labels to any time series or alerts when communicating with
8. # external systems (federation, remote storage, Alertmanager).
9. external_labels:
10. monitor: 'codelab-monitor'
11.  
12. # Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
13. rule_files:
14. # - "first.rules"
15. # - "second.rules"
16.  
17. # A scrape configuration containing exactly one endpoint to scrape:
18. # Here it's Prometheus itself.
19. scrape_configs:
20. # The job name is added as a label `job=` to any timeseries scraped from this config.
21. - job_name: 'prometheus'
22.  
23. # metrics_path defaults to '/metrics'
24. # scheme defaults to 'http'.
25.  
26. static_configs:
27. - targets: ['10.10.30.102:9090']
28.  
29. #以下部分为配置kubernetes,注意修改地址
30. - job_name: 'kubernetes-nodes-cadvisor'
31. kubernetes_sd_configs:
32. - api_server: 'http://10.10.30.102:8080'
33. role: node
34. relabel_configs:
35. - action: labelmap
36. regex: __meta_kubernetes_node_label_(.+)
37. - source_labels: [__meta_kubernetes_role]
38. action: replace
39. target_label: kubernetes_role
40. - source_labels: [__address__]
41. regex: '(.*):10250'
42. replacement: '${1}:10255'
43. target_label: __address__
44. - job_name: 'kubernetes_node'
45. kubernetes_sd_configs:
46. - role: node
47. api_server: 'http://10.10.30.102:8080'
48. relabel_configs:
49. - source_labels: [__address__]
50. regex: '(.*):10250'
51. replacement: '${1}:9100'
52. target_label: __address__

重启服务，选择kubernetes的Dashboard

转载于:https://my.oschina.net/binges/blog/1617425