一、为什么要服务发现（sd：service discovery ）

希望服务发现能做什么

1. 自动添加 target ，海量的服务添加实在不便；
2. 自动配置 lable ，lable 的设计和添加比 target 添加还麻烦；
3. 无需重启 prometheus ；

二、prometheus 当前支持的 sd_config 配置

自带配置	适配	说明
azure_sd_config	微软 Azure 云	Azure SD configurations allow retrieving scrape targets from Azure VMs.
consul_sd_config
digitalocean_sd_config
docker_sd_config
dockerswarm_sd_config
dns_sd_config
ec2_sd_config
openstack_sd_config
ovhcloud_sd_config
puppetdb_sd_config
file_sd_config
gce_sd_config
hetzner_sd_config
http_sd_config
ionos_sd_config
kubernetes_sd_config
kuma_sd_config
lightsail_sd_config
linode_sd_config
marathon_sd_config
nerve_sd_config
nomad_sd_config
serverset_sd_config
triton_sd_config
eureka_sd_config
scaleway_sd_config
uyuni_sd_config
vultr_sd_config

三、 http_sd_config 和 file_sd_config 对比

Item	File SD	HTTP SD
Event Based	Yes, via inotify	No
Update frequency	Instant, thanks to inotify	Following refresh_interval
Format	Yaml or JSON	JSON
Transport	Local file	HTTP/HTTPS
Security	File-Based security	TLS, Basic auth, Authorization header, OAuth2

四、http_sd_config

需要开发部署一个HTTP SD endpoint，prometheus周期性的http访问该endpoint获取targets。如下是返回格式范例：
Example response body:

[
  {
    "targets": [ "<host>", ... ],
    "labels": {
      "<labelname>": "<labelvalue>", ...
    }
  },
  ...
]

[
    {
        "targets": ["10.0.10.2:9100", "10.0.10.3:9100", "10.0.10.4:9100", "10.0.10.5:9100"],
        "labels": {
            "__meta_datacenter": "london",
            "__meta_prometheus_job": "node"
        }
    },
    {
        "targets": ["10.0.40.2:9100", "10.0.40.3:9100"],
        "labels": {
            "__meta_datacenter": "london",
            "__meta_prometheus_job": "alertmanager"
        }
    },
    {
        "targets": ["10.0.40.2:9093", "10.0.40.3:9093"],
        "labels": {
            "__meta_datacenter": "newyork",
            "__meta_prometheus_job": "alertmanager"
        }
    }
]

五、file_sd_config

prometheus.yml 配置文件增加如下

scrape_configs:
- job_name: 'node'
  file_sd_configs:
  - files:
    - 'targets.json'

编辑 targets.json

[
  {
    "targets": [
      "localhost:9100"
    ],
    "labels": {
      "job": "node"
    }
  },
  {
    "targets": [
      "localhost:9200"
    ],
    "labels": {
      "job": "node"
    }
  }
]

[
  {
    "labels": {
      "job": "logstash1",
      "env": "dev"
    },
    "targets": [
      "localhost:9501",
      "localhost:9502",
      "localhost:9503"
   ]
  }
]

[
  {
    "targets": ["localhost:9090", "example.org:443"]
  }
]

yaml 格式

- targets:
  [ - '<host>' ]
  labels:
    [ <labelname>: <labelvalue> ... ]

- targets: ['localhost:9090', 'example.org:443']

- targets: ['10.98.29.9:9100', '10.98.29.12:9100', '10.98.29.15:9100']
  labels:
    job: rtms-nacos
    env: prd

- targets:
  - 10.98.29.9:9100
  - 10.98.29.12:9100
  - 10.98.29.15:9100
  labels:
    job: rtms-nacos
    env: prd