Menu

文档

添加监控参数

编写指标数据获取接口

下面以io_read为例,在task目录的metric_task.py中编写如下io_read指标获取函数:

def io_read():
     child1 = subprocess.Popen(['pidstat', '-d'], stdout=subprocess.PIPE, shell=False)
     child2 = subprocess.Popen(['grep', 'gaussd[b]'], stdin=child1.stdout, stdout=subprocess.PIPE, shell=False)
     result = child2.communicate()
     if not result[0]:
         return 0.0
     else:
         return result[0].split()[3].decode('utf-8')

添加指标监控参数

在task目录下的metric_task.conf中添加io_read的section:

[io_read]
minimum = 30
maximum = 100
data_period = 1H
forecast_interval = 2H
forecast_period = 30M

重启服务

命令参考获取帮助

重启agent:
    本地服务器: 
        python main.py start --role agent
        python main.py stop --role agent
    远程服务器:
        python main.py start --user USER --host HOST --project-path PROJECT_PATH --role agent    
        python main.py stop --user USER --host HOST --project-path PROJECT_PATH --role agent    
重启server:
    本地服务器: 
        python main.py start --role server
        python main.py stop --role server
    远程服务器:
        python main.py start --user USER --host HOST --project-path PROJECT_PATH --role server
        python main.py stop --user USER --host HOST --project-path PROJECT_PATH --role server
重启monitor:
    本地服务器: 
        python main.py start --role monitor
        python main.py stop --role monitor
    远程服务器:
        python main.py start --user USER --host HOST --project-path PROJECT_PATH --role monitor
        python main.py stop --user USER --host HOST --project-path PROJECT_PATH --role monitor

本文档遵循知识共享许可协议CC 4.0 (http://creativecommons.org/Licenses/by/4.0/)。