python练手脚本-定时检测无响应进程并重启

背景

总有一些程序在windows平台表现不稳定，动不动一段时间就无响应，但又不得不用，每次都是发现问题了手动重启，现在写个脚本定时检测进程是否正常，自动重启。

涉及知识点

1. schedule定时任务调度

2. os.popen运行程序并读取解析运行结果

代码分解

脚本主入口

if __name__ == '__main__':
    #每5秒执行检查任务
    schedule.every(5).seconds.do(check_job)
    #此处固定写法，意思是每秒钟schedule看下是否有pending的任务，有就执行
    while True:
        schedule.run_pending()
        time.sleep(1)

schedule的其它示例

import schedule
import time

def job(message='stuff'):
    print("I'm working on:", message)

#每10分钟
schedule.every(10).minutes.do(job)
#每小时
schedule.every().hour.do(job, message='things')
#每天10点30分
schedule.every().day.at("10:30").do(job)

while True:
    schedule.run_pending()
    time.sleep(1)

检查无响应进程并重启

def check_job():
    process_name = "xx.exe"
    not_respond_list = list_not_response(process_name)
    if len(not_respond_list) <= 0:
        return
    pid_params = " ".join(["/PID " + pid for pid in not_respond_list])
    os.popen("taskkill /F " + pid_params)
    if len(list_process(process_name)) <= 0:
        start_program(r'E:\xx\xx.exe')
｝

查找符合条件的进程列表

def list_process(process_name, not_respond=False):
    cmd = 'tasklist /FI "IMAGENAME eq %s"'
    if not_respond:
        cmd = cmd + ' /FI "STATUS eq Not Responding"'
    output = os.popen(cmd % process_name)
    return parse_output(output.read())

def list_not_response(process_name):
    return list_process(process_name, True)

解析命令执行结果

def parse_output(output):
    print(output)
    pid_list = []
    lines = output.strip().split("\n")
    if len(lines) > 2:
        for line in lines[2:]:
            pid_list.append(line.split()[1])
    return pid_list

tasklist示例输出

映像名称                       PID 会话名              会话#       内存使用
========================= ======== ================ =========== ============
WizChromeProcess.exe          1620 Console                    1     32,572 K

完整代码

import os
import time

import schedule


def parse_output(output):
    print(output)
    pid_list = []
    lines = output.strip().split("\n")
    if len(lines) > 2:
        for line in lines[2:]:
            pid_list.append(line.split()[1])
    return pid_list


def list_not_response(process_name):
    return list_process(process_name, True)


def list_process(process_name, not_respond=False):
    cmd = 'tasklist /FI "IMAGENAME eq %s"'
    if not_respond:
        cmd = cmd + ' /FI "STATUS eq Not Responding"'
    output = os.popen(cmd % process_name)
    return parse_output(output.read())


def start_program(program):
    os.popen(program)


def check_job():
    process_name = "xx.exe"
    not_respond_list = list_not_response(process_name)
    if len(not_respond_list) <= 0:
        return
    pid_params = " ".join(["/PID " + pid for pid in not_respond_list])
    os.popen("taskkill /F " + pid_params)
    if len(list_process(process_name)) <= 0:
        start_program(r'E:\xxx\xx.exe')


if __name__ == '__main__':
    schedule.every(5).seconds.do(check_job)
    while True:
        schedule.run_pending()
        time.sleep(1)