【修复】centos定时任务python top不能输出

问题

用python监控进程占用硬件资源时,手动执行正常运行,定时任务执行出错

在centos的定时任务中,用python的os模块执行top命令时终端没有输出

python 复制代码
cmd = 'top -d 1 -n 1 -p {} | grep -E "PID|{}|%Cpu|buff/cache" '.format(process_id, process_id, )
sys_out = os.popen(cmd)
sys_out = sys_out.read()

修复

  1. top -n 1不能在 crontab中使用,要加一个参数top -b -n 1
  2. 用 /bin/top 代替 top 命令
python 复制代码
cmd = '/bin/top -b -d 1 -n 1 -p {} | grep -E "PID|{}|%Cpu|buff/cache" '.format(process_id, process_id, )

sys_out = os.popen(cmd)
sys_out = sys_out.read()  

引用

https://blog.51cto.com/zdz8207/3552214

源代码

python 复制代码
def record_process_occupy(process_id=0, process_log_file="", ):
    """
    记录一条这个进程的资源占用情况
    :param process_id: 进程的ID
    :param log_file: 保存top输出的文件
    :return: None
    """

    cmd = '/bin/top -b -d 1 -n 1 -p {} | grep -E "PID|{}|%Cpu|buff/cache" '.format(process_id, process_id, )

    sys_out = os.popen(cmd)
    print(cmd)
    sys_out = sys_out.read()
    sys_out = sys_out.replace("\x1b[7m", "")
    sys_out = sys_out.replace("\x1b(B\x1b[m\x1b[39;49m\x1b(B\x1b[m\x1b[39;49m\x1b[K", "")
    sys_out = sys_out.replace("\x1b(B\x1b[m\x1b[39;49m\x1b[K", "")
    sys_out = sys_out.replace("\x1b(B\x1b[m\x1b[39;49m", "")
    sys_out = sys_out.replace("\x1b(B\x1b[m", '')
    sys_out = sys_out.replace("\x1b[1m", "")
    sys_out = sys_out.split('\n')

    assert len(sys_out) >= 4
    # print(sys_out)
    first_line = "time,"
    first_line += "us,sy,ni,id,wa,hi,si,st,"
    first_line += "total, free, used, buff/cache,"
    first_line += "PID,USER,PR,NI,VIRT,RES,SHR,S,%CPU,%MEM,TIME+,COMMAND"
    line = ",".join([
        datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
        ",".join([x for x in sys_out[0].replace(",", " ").split(" ") if x.replace(".", "").isdigit()]),  # cpu
        ",".join([x for x in sys_out[1].replace(",", " ").split(" ") if x.replace(".", "").isdigit()]),  # mem
        ",".join([x for x in sys_out[3].split(" ") if x]),  # 进程
    ])
    # print(first_line)
    # print(line)
    print("process_log_file", process_log_file)
    if not process_log_file:
        process_log_file = os.path.join(BACKUP_PATH, "process_{}_resource_occupy.csv".format(process_id))
        print()
    print("process_log_file:", process_log_file)
    is_new_file = not os.path.exists(process_log_file)
    with open(process_log_file, 'a') as f:
        if is_new_file:
            f.write(first_line + "\n")
        f.write(line + "\n")

    return True
相关推荐
爱笑的k116 分钟前
nano-vllm 源码剖析:千行 Python 手写离线推理引擎的调度、显存与并行时序
开发语言·python·vllm
用户8356290780511 小时前
使用 Python 控制 Excel 工作表的分页位置
后端·python
用户8356290780511 小时前
Python 如何在 Word 中添加和管理文档变量
后端·python
AR-26710-3 小时前
Linux Day33——定时任务+lvm
linux·运维
用户019027581614 小时前
Dify / Coze 低代码 Agent 怎么接 A 股行情?
python
墨痕诉清风4 小时前
SSRF漏洞实战记录+OSS桶命令执行+MCP漏洞
linux·服务器·数据库
byte轻骑兵4 小时前
实战 4:BlueZ 源码中添加自定义设备过滤规则
linux·物联网·bluez·电脑蓝牙·嵌入式蓝牙
Dumbness4 小时前
不传 day 参数,座位图显示全空:用 Python 标准库手写 Web 框架,请求到业务代码前的 5 个细节
后端·python
瓦瓦gogo5 小时前
从一个 TypeScript 函数开始理解 MCP 协议
linux·ubuntu·typescript
databook5 小时前
收上来一堆 Excel,怎么保证没改错
python·数据分析