本节重点介绍 :
- 基于文件的服务发现提供了一种配置静态目标的更通用的方法
- 可以摆脱对特定服务发现源的依赖
- 通常的做法是调用内部CMDB的接口获取target数据，打上标签，生成json文件发给prometheus采集
基于文件的服务发现模式
解决的问题
- 之前手动配置了很多个traget
- redis
- mysql
- blackbox
- pushgateway
- 手动配置维护成本高，还容易出错
基于文件的服务发现配置
特点
- 基于文件的服务发现提供了一种配置静态目标的更通用的方法
- 并充当了插入自定义服务发现机制的接口。
- 摆脱对特定服务发现源的依赖
- 只要能正确给出 json/yaml文件即可
- 和服务树的最好匹配方案
yaml文件类型
YAML yaml - targets: [ - '<host>' ] labels: [ <labelname>: <labelvalue> ... ]
- targets: - 172.20.70.205:9100- 172.20.70.215:9100labels:account: "aliyun-01" region: "ap-south-1"
json文件类型
json [ { "targets": [ "<host>", ... ], "labels": { "<labelname>": "<labelvalue>", ... } }, ... ]
[{"targets": ["172.20.70.205:9100"],"labels": {"account": "aliyun-01","region": "ap-south-1","env": "prod","group": "inf","project": "monitor","stree_gpa": "inf.monitor.prometheus"}},{"targets": ["172.20.70.215:9100"],"labels": {"account": "aliyun-02","region": "ap-south-2","env": "prod","group": "inf","project": "middleware","stree_gpa": "inf.middleware.kafka"}}
]
下面来解读一下
- targets 是一组实例地址的列表
- labels 是这组实例的标签，应用到列表中所有实例
- 如果想每个实例不同的标签，可以将targets列表保留一个实例即可
- 标签可以自定义，下面举几个例子
- account 代表公有云账户，多账户情况
- region 代表区域
- env 代表所属环境 prod代表生产，pre代表预发，test代表测试
- group代表业务大组
- project 代表项目
- stree_gpa 代表服务树三级标签
- 那么prometheus在采集对应target时就会将对应标签打入其metrics中
- 为后续我们按照标签过滤提供方便
配置举例
- files 代表 文件路径 支持通配符
- refresh_interval 代表 文件刷新间隔
- job_name: 'node_exporter'scrape_interval: 30sscrape_timeout: 10smetrics_path: /metricsscheme: httphonor_timestamps: falsefile_sd_configs:- files:- /opt/app/prometheus/sd/node_exporter.jsonrefresh_interval: 5m
改造为服务发现类型
- 将 blackbox-http 和 node_exporter 改为文件发现
- prometheus 配置
- job_name: 'blackbox-http'metrics_path: /probeparams:module: [http_2xx] scrape_interval: 15sscrape_timeout: 10sscheme: httphonor_timestamps: falsefile_sd_configs:- files:- /opt/app/prometheus/sd/blackbox_http.jsonrefresh_interval: 2mrelabel_configs:- source_labels: [__address__]target_label: __param_target- source_labels: [__param_target]target_label: instance- target_label: __address__replacement: 172.20.70.205:9115 - job_name: 'node_exporter'scrape_interval: 30sscrape_timeout: 10sscheme: httphonor_timestamps: falsefile_sd_configs:- files:- /opt/app/prometheus/sd/node_exporter.jsonrefresh_interval: 2m
mkdir -pv /opt/app/prometheus/sd/
cat <<EOF > /opt/app/prometheus/sd/node_exporter.json
[{"targets": ["172.20.70.205:9100"],"labels": {"name": "prome-master01","account": "aliyun-01","region": "ap-south-1","env": "prod","group": "inf","project": "monitor","stree_gpa": "inf.monitor.prometheus"}},{"targets": ["172.20.70.215:9100"],"labels": {"name": "prome-node01","account": "aliyun-02","region": "ap-south-2","env": "prod","group": "inf","project": "middleware","stree_gpa": "inf.middleware.kafka"}}
]
EOFcat <<EOF > /opt/app/prometheus/sd/blackbox_http.json
[{"targets": ["172.20.70.205:9115","http://prometheus.io","http://www.baidu.com","https://www.baidu.com"]}
]
EOF
- reload prometheus
- 观察target页面 和 sd结果页面
- 修改文件 blackbox_http.json 新增 https://github.com/ 的探测
cat <<EOF > /opt/app/prometheus/sd/blackbox_http.json
[{"targets": ["172.20.70.205:9115","http://prometheus.io","http://www.baidu.com","https://www.baidu.com","https://github.com/"]}
]
EOFcurl -X POST http://localhost:9090/-/reload
本节重点总结 :
- 基于文件的服务发现提供了一种配置静态目标的更通用的方法
- 可以摆脱对特定服务发现源的依赖
- 通常的做法是调用内部CMDB的接口获取target数据，打上标签，生成json文件发给prometheus采集
