docker-compose.yml clickhouse metrics prometheus

docker-compose.yml clickhouse metrics prometheus

这里写目录标题

• 1. docker-compose.yml clickhouse metrics prometheus
• • 1.1. prometheus 与 pushgateway 的区别
  • 1.2. tree
  • 1.3. config.xml
  • 1.4. docker-compose.yml
  • 1.5. prometheus.yml
  • 1.6. restart.sh

1. docker-compose.yml clickhouse metrics prometheus 1.1. prometheus 与 pushgateway 的区别

• prometheus 是主动抓取方，一般用于需要主动采集的软件，例如 clickhouse。在配置文件 prometheus.yml 中填入 clickhouse 后会主动抓取相应 /metrics 的 metrics。
• pushgateway 是被动接收方，一般用于某些软件主动汇报 metrics 的情况，例如 Flink。在 Flink 配置文件中引入 jar 包并填入 pushgateway 的地址和端口即可。

搞清楚这两个的区别非常有助于了解下面的配置。

1.2. tree

|-- clickhouse_dir
|   |-- clickhouse-server
|   |   |-- config.xml
|   `-- clickhouse-server.bak
|       |-- ...
|-- docker-compose.yml
|-- prometheus_dir
|   |-- conf
|   |   `-- prometheus.yml
|   `-- conf.bak
|       |-- ...
`-- restart.sh

1.3. config.xml

0.0.0.0



    /metrics
    8001

    true
    true
    true
    true

1.4. docker-compose.yml

# https://onlineyamltools.com/prettify-yaml
# docker network create clickhouse-network
version: '2'
services:
    clickhouse-server:
        container_name: clickhouse-server
        image: 'clickhouse/clickhouse-server:latest'
        ports:
          - 8123:8123
          - 8001:8001
        ulimits:
            nofile:
                soft: 262144
                hard: 262144
        volumes:
          - '/mnt/hgfs/share/docker-composes/clickhouse/clickhouse_dir:/tmp/tmp'
          - '/mnt/hgfs/share/docker-composes/clickhouse/clickhouse_dir/clickhouse-server:/etc/clickhouse-server'
        depends_on:
          - prometheus

    prometheus:
        container_name: prometheus_clickhouse
        image: bitnami/prometheus:latest
        ports:
          - 9090:9090
        volumes:
          - '/mnt/hgfs/share/docker-composes/clickhouse/prometheus_dir:/tmp/tmp'
          - '/mnt/hgfs/share/docker-composes/clickhouse/prometheus_dir/conf:/opt/bitnami/prometheus/conf'
networks:
  default:
    external:
      name: clickhouse-network

1.5. prometheus.yml

# A scrape configuration containing exactly one endpoint to scrape:
# Here it's Prometheus itself.
scrape_configs:
  # The job name is added as a label `job=` to any timeseries scraped from this config.
  - job_name: "prometheus"

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
      - targets: ["localhost:9090"]

  - job_name: "clickhouse"

    # metrics_path defaults to '/metrics'
    # scheme defaults to 'http'.

    static_configs:
      - targets: ["clickhouse-server:8001"]

1.6. restart.sh

#/bin/sh

docker stop clickhouse-server
docker stop prometheus_clickhouse
docker rm clickhouse-server
docker rm prometheus_clickhouse
docker-compose up -d
docker ps -a