使用prometheus监控flink的时候，重新启动flink的推送的消息会缓存，比如总槽数会累加

配置Prometheus

修改Prometheus根目录prometheus.yml文件的scrape_config，如下图所示： scrape_configs: # The job name is added as a label job=<job_name> to any timeseries scraped from this config. - job_name: 'pushgateway'

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

static_configs:
- targets: ['10.180.210.172:9091']
  labels:
    instance: 'pushgateway'

• job_name: 'linux' static_configs:

  • targets: ['10.180.210.172:9100'] labels: instance: 'linux'

• job_name: prometheus static_configs:

  • targets: ['10.180.210.172:9090'] labels: instance: prometheus

启动prometheus, node_exporter和pushgateway

在命令行启动prometheus, node_exporter和pushgateway。

通过访问组件对应的URL确定是否成功启动。 端口号：

prometheus: 9090
node_exporter: 9100
pushgateway: 9091

来源：简书 著作权归作者所有。商业转载请联系作者获得授权，非商业转载请注明出处。