prometheus

markdown

# Prometheus 监控部署指南

## 1 JMX Exporter

### 1.1 安装 jmx_agent

- 下载

  - jmx_prometheus_javaagent-0.9.jar

    ```bash
    wget -c https://repo1.maven.org/maven2/io/prometheus/jmx/jmx_prometheus_javaagent/0.9/jmx_prometheus_javaagent-0.9.jar
    mv jmx_prometheus_javaagent-0.9.jar kafka/libs
    ```

  - kafka-0-8-2.yml

    ```bash
    wget -c https://raw.githubusercontent.com/prometheus/jmx_exporter/master/example_configs/kafka-0-8-2.yml
    mv kafka-0-8-2.yml kafka/config
    ```

- 配置 kafka-server-start.sh

  ```patch
  + export JMX_PORT=${JMX_PORT:-9999}
  + export PROMETHEUS_PORT=${PROMETHEUS_PORT:-7072}
  COMMAND=$1
  ...
  - exec $base_dir/kafka-run-class.sh $EXTRA_ARGS kafka.Kafka "$@"
  + exec $base_dir/kafka-run-class.sh $EXTRA_ARGS -javaagent:$base_dir/../libs/jmx_prometheus_javaagent-0.9.jar=$PROMETHEUS_PORT:$base_dir/../config/kafka-0-8-2.yml kafka.Kafka "$@"
  ```

- 配置采集任务

  ```bash
  SERV_NAME=KAFKA
  SERV_ENDPOINT=localhost:7072
  cat >> prometheus/prometheus.yml << EOF
    - job_name: '$SERV_NAME'
      static_configs:
      - targets: ['$SERV_ENDPOINT']
  EOF
  ```

## 2 EMQ Exporter

### 2.1 安装 emq_exporter

```bash
VERSION=0.4.1
DIR=emq_exporter-$VERSION-linux
wget -c https://github.com/nuvo/emq_exporter/releases/download/v$VERSION/$DIR.tar.gz
tar zxvf $DIR.tar.gz
ln -s $DIR emq_exporter
mv $DIR.tar.gz $DIR
```

### 2.2 配置 emq_exporter

- 用户名和密码使用 emqtt 管理 dashboard 账号密码 (默认是 admin/public)

  ```bash
  EMQ_USERNAME=admin
  EMQ_PASSWORD=public
  cat > emq_exporter/auth.json << EOF
  {
    "username":$EMQ_USERNAME,
    "password":$EMQ_PASSWORD
  }
  ```

### 2.3 启动 emq_exporter

- 前台启动

  ```bash
  EMQ_USERNAME=admin EMQ_PASSWORD=public emq_exporter/emq_exporter
  ```

- 后台启动 

  ```bash
  BIN=emq_exporter
  nohup $BIN/$BIN > /tmp/$BIN.stdout 2>&1 & 
  ```

## 3 Node Exporter

### 3.1 安装 node_exporter

```bash
VERSION=0.17.0
DIR=node_exporter-$VERSION-linux-amd64
wget -c https://github.com/prometheus/node_exporter/releases/download/v$VERSION/$DIR.tar.gz
tar zxvf $DIR.tar.gz
ln -s $DIR node_exporter
mv $DIR.tar.gz $DIR
```

### 3.2 启动 node_exporter

- 前台启动

  ```bash
  node_exporter/node_exporter
  ```

- 后台启动 

  ```bash
  BIN=node_exporter
  nohup $BIN/$BIN > /tmp/$BIN.stdout 2>&1 & 
  ```

## 4 Prometheus 

### 4.1 安装 prometheus

```bash
VERSION=2.8.0 
DIR=prometheus-$VERSION.linux-amd64
wget -c https://github.com/prometheus/prometheus/releases/download/v2.8.0/prometheus-2.8.0.linux-amd64.tar.gz
tar zxvf $DIR.tar.gz 
ln -s $DIR prometheus
mv $DIR.tar.gz prometheus
```

### 4.2 配置采集任务

```bash
SERV_NAME=EMQ
SERV_ENDPOINT=localhost:9540
cat >> prometheus/prometheus.yml << EOF
  - job_name: '$SERV_NAME'
    static_configs:
    - targets: ['$SERV_ENDPOINT']
EOF
```

### 4.3 启动 prometheus

- 前台启动

  ```bash
  prometheus/prometheus
  ```

- 后台启动 

  ```bash
  BIN=prometheus
  nohup $BIN/$BIN > /tmp/$BIN.stdout 2>&1 & 
  ```

## 5 Grafana

### 5.1 安装 grafana

```bash
VERSION=5.4.2
DIR=grafana-$VERSION
wget -c https://dl.grafana.com/oss/release/grafana-$VERSION.linux-amd64.tar.gz
tar zxvf grafana-$VERSION.linux-amd64.tar.gz
ln -s $DIR grafana
mv grafana-$VERSION.linux-amd64.tar.gz $DIR
```

### 5.2 启动 grafana

- 前台启动

  ```bash
  grafana/bin/grafana-server -homepath grafana/
  ```

- 后台启动 

  ```bash
  BIN=grafana
  nohup $BIN/bin/$BIN-server > /tmp/$BIN.stdout 2>&1 & 
  ```

### 5.3 添加 Prometheus 数据源

- 新建数据源[Prometheus类型] -> 填写服务地址 -> 测试连接 -> 保存并返回

### 5.4 导入 Kafka Dashboard

- 公网

  - 导入 Dashboard -> 选择 Dashboard ID : 7589 -> 选择数据源

- 私网

  - 导入 Dashboard -> 选择 Json Model -> 选择数据源 [7589.json](https://gist.githubusercontent.com/zhangli2946/5138f537ea328610aa26b674c6b0c9ae/raw/4fde05979950ac2e64698c2d1ea871894547ea0e/7589.json)

### 5.5 导入 EMQ Dashboard

- 公网

  - 导入 Dashboard -> 选择 Dashboard ID : 9963 -> 选择数据源

- 私网

  - 导入 Dashboard -> 选择 Json Model -> 选择数据源 [9963.json](https://gist.githubusercontent.com/zhangli2946/b20e7f4921d77b3a4885707dfa506e0d/raw/244f9b6d865b48aea4a9065d0228e2f54d75fb41/9963.json)

### 5.6 导入 Node Dashboard

- 公网

  - 导入 Dashboard -> 选择 Dashboard ID : 8919 -> 选择数据源

- 私网

  - 导入 Dashboard -> 选择 Json Model -> 选择数据源 [8919.json](https://gist.githubusercontent.com/zhangli2946/b20e7f4921d77b3a4885707dfa506e0d/raw/244f9b6d865b48aea4a9065d0228e2f54d75fb41/9963.json)