获取帮助

启动调优程序之前，可以通过如下命令获取帮助信息：

  1. 源码方式：python main.py --help

输出帮助信息结果如下：

  1. usage:
  2.         python main.py start [--role {{agent,server,monitor}}]
  3.         # start local service.
  4.         python main.py stop [--role {{agent,server,monitor}}]
  5.         # stop local service.
  6.         python main.py start [--user USER] [--host HOST] [--project-path PROJECT_PATH] [--role {{agent,server,monitor}}]
  7.         # start the remote service.
  8.         python main.py stop [--user USER] [--host HOST] [--project-path PROJECT_PATH] [--role {{agent,server,monitor}}]
  9.         # stop the remote service.
  10.         python main.py deploy [--user USER] [--host HOST] [--project-path PROJECT_PATH]
  11.         # deploy project in remote host.
  12.         python main.py show_metrics
  13.         # display all monitored metrics(can only be executed on 'detector' machine).
  14.         python main.py forecast [--metric-name METRIC_NAME] [--forecast-periods FORECAST_PERIODS] [--forecast-method {{auto_arima, fbprophet}}] [--save-path SAVE_PATH]
  15.         # forecast future trend of metric(can only be executed on 'detector' machine).
  18. abnomaly_detection: a timeseries forecast and anomaly detection tool.
  20. positional arguments:
  21.   {start,stop,deploy,show_metrics,forecast}
  23. optional arguments:
  24.   -h, --help            show this help message and exit
  25.   --user USER           User of remote server.
  26.   --host HOST           IP of remote server.
  27.   --project-path PROJECT_PATH
  28.                         Project location in remote server.
  29.   --role {agent,server,monitor}
  30.                         Run as 'agent', 'server', 'monitor'. notes: ensure the
  31.                         normal operation of the openGauss in agent.
  32.   --metric-name METRIC_NAME
  33.                         Metric name to be predicted, if this parameter is not
  34.                         provided, all metric in database will be predicted.
  35.   --forecast-periods FORECAST_PERIODS
  36.                         Forecast periods of metric, it should be integer plus
  37.                         time unit format such as '1S', '2H', '10D'; now we
  38.                         support time unit: S: Second, M: Minute, H: Hour, D:
  39.                         Day, W: Week. notes: the specific value should be
  40.                         determined to the trainnig data.if this parameter is
  41.                         not provided, the default value '100S' will be used.
  42.   --forecast-method FORECAST_METHOD
  43.                         Forecast method, default method is 'auto_arima',if
  44.                         want to use 'fbprophet', you should install fbprophet
  45.                         first.
  46.   --save-path SAVE_PATH
  47.                         Save the results to this path using csv format, if
  48.                         this parameter is not provided,, the result wil not be
  49.                         saved.
  50.   -v, --version         show program's version number and exit
  52. epilog:
  53.      the 'a-detection.conf' and 'metric_task.conf' will be read when the program is running,
  54.      the location of them is:
  55.      a-detection.conf: xxx/a-detection.conf.
  56.      metric_config: xxx/metric_task.conf.