导读 | psutil(process and system utilities)进程和系统实用程序,是一个跨平台的库,用于检索Python中正在运行的进程和系统利用率(CPU,内存,磁盘,网络,传感器)的信息。 |
psutil
(process and system utilities)进程和系统实用程序,是一个跨平台的库,用于检索Python中正在运行的进程和系统利用率(CPU,内存,磁盘,网络,传感器)的信息。 它主要用于系统监视,概要分析和限制进程资源以及管理正在运行的进程。 它实现了经典UNIX行工具提供的许多功能,例如ps,top,iotop,lsof,netstat,ifconfig,free等。 psutil当前支持以下平台:
-
Windows
-
macOS
-
FreeBSD, OpenBSD, NetBSD
-
Sun Solaris
-
AIX
【注】本文章在 python3.6,centos7.8,windows 10 下测试通过,windows下可能部分返回参数没有
pip install psutil
psutil.cpu_times()
返回系统CPU时间
-
user:在用户模式下执行普通进程所花费的时间
-
system: 在内核模式下执行的进程所花费的时间
-
idle:空闲时间
-
nice :在用户模式下执行完好的(优先级)进程所花费的时间; 在Linux上,这还包括guest_nice时间
-
iowait : 等待I / O完成所花费的时间。
-
irq :服务硬件中断所花费的时间
-
softirq : 服务软件中断所花费的时间
-
steal : 在虚拟环境中运行的其他操作系统所花费的时间
-
guest :在Linux内核控制下为来guest系统运行虚拟CPU所花费的时间
-
guest_nice:运行良好的guest所花费的时间
如需显示所有CPU核心信息,应显示的指出psutil.cpu_times(percpu=Ture)
,返回一个列表, 列表的第一个元素是指第一个CPU,第二个元素是指第二个CPU,依此类推,一般情况下只关注user,system,以及idle
# Centos7.8
>>> import psutil
>>> psutil.cpu_times()
scputimes(user=56276.04, nice=6.9, system=46298.69, idle=7267935.28, iowait=1748.04, irq=0.0, softirq=288.37, steal=0.0, guest=0.0, guest_nice=0.0)
# windows 10
>>> psutil.cpu_times(percpu=True)
[scputimes(user=120708.328125, system=113197.0625, idle=926771.3593749999, interrupt=6374.4375, dpc=3362.828125),
scputimes(user=104530.18749999999, system=87954.64062500012, idle=968191.6406249999, interrupt=1626.5625, dpc=281.8125),
scputimes(user=127384.45312499999, system=110084.546875, idle=923207.453125, interrupt=1679.453125, dpc=336.640625),
scputimes(user=110810.578125, system=82547.015625, idle=967318.875, interrupt=1359.25, dpc=262.21875)]
psutil.cpu_percent(interval=None,percpu=False)
当interval> 0.0时,将比较该间隔前后的系统CPU时间(阻塞)。 当interval为0.0或无时,比较自上次调用或模块导入以来经过的系统CPU时间,立即返回。 这意味着第一次调用它会返回一个无意义的0.0值,应该忽略它。 在这种情况下,为确保准确性,建议在两次调用之间至少调用0.1秒。
# 每隔两秒刷新CPU使用率,累计8次
>>> for x in range(8):
... print(psutil.cpu_percent(interval=2, percpu=True))
[30.5, 17.8, 30.2, 24.6]
[16.3, 20.2, 14.0, 14.0]
[12.1, 3.9, 4.7, 3.9]
[26.4, 16.3, 18.6, 16.3]
[13.1, 9.3, 7.8, 10.1]
[33.6, 40.3, 48.4, 34.4]
[17.6, 8.5, 14.7, 7.8]
[23.1, 7.8, 21.5, 8.5]
# 默认统计CPU逻辑核心数
>>> psutil.cpu_count() # CPU逻辑核心数
4
>>> psutil.cpu_count(logical=False) # cpu物理核心数
2
# 显示当前cpu的主频率
>>> psutil.cpu_freq()
scpufreq(current=1876.0, min=0.0, max=1896.0)
# 以元组的形式返回最近1、5和15分钟内的平均系统负载。 在Windows上,这是通过使用Windows API模拟的,该API产生一个线程,该线程保持在后台运行,并每5秒更新一次结果,从而模仿UNIX行为。 因此,在Windows上,第一次调用此方法,在接下来的5秒钟内,它将返回无意义的(0.0,0.0,0.0)元组。
>>> psutil.getloadavg()
(0.0, 0.01, 0.05)
# windows 10 返回内存使用情况
>>> memory = psutil.virtual_memory()
>>> memory
svmem(total=8463876096, available=2596061184, percent=69.3, used=5867814912, free=2596061184)
>>> memory.total
8463876096
# centos7.8
>>> psutil.virtual_memory()
svmem(total=1927192576, available=1060913152, percent=45.0, used=693989376, free=103620608, active=1383866368, inactive=288067584, buffers=143511552, cached=986071040, shared=589824, slab=93745152)
返回的是单位是字节Byte
重点关注的参数:
-
total=8463876096 总共8.46GB
-
available=2596061184 可用2.59GB
-
percent=69.3 使用率69.3%
-
used=5867814912 已使用5.86GB
# windows 10
>>> psutil.swap_memory()
sswap(total=13326721024, used=9382895616, free=3943825408, percent=70.4, sin=0, sout=0)
In [35]: psutil.disk_partitions()
Out[35]: [sdiskpart(device='/dev/vda1', mountpoint='/', fstype='ext4', opts='rw,relatime,data=ordered')]
# centos 7.8
In [36]: psutil.disk_usage('/')
Out[36]: sdiskusage(total=42140479488, used=12435881984, free=27755200512, percent=30.9)
# windows 10
In [32]: psutil.disk_usage('C:\\')
Out[32]: sdiskusage(total=125917933568, used=89807781888, free=36110151680, percent=71.3)
# windows 10
In [37]: psutil.disk_io_counters()
Out[37]: sdiskio(read_count=7703266, write_count=11199399, read_bytes=209586859008, write_bytes=217863440384, read_time=46821, write_time=5137)
# 获取单个分区的 io个数,读写信息
In [38]: psutil.disk_io_counters(perdisk=True)
Out[38]:
{'PhysicalDrive0': sdiskio(read_count=980771, write_count=399173, read_bytes=39607087616, write_bytes=8268857344, read_time=38514, write_time=1396),
'PhysicalDrive1': sdiskio(read_count=6723984, write_count=10805960, read_bytes=169995229696, write_bytes=209674449408, read_time=8309, write_time=3743)}
# windows 10
In [39]: psutil.net_io_counters()
Out[39]: snetio(bytes_sent=45955043, bytes_recv=736052740, packets_sent=383832, packets_recv=589165, errin=0, errout=0, dropin=0, dropout=0)
# windows 10 输出每个网络接口的IO信息
In [40]: psutil.net_io_counters(pernic=True)
Out[40]:
{'Ethernet': snetio(bytes_sent=0, bytes_recv=0, packets_sent=0, packets_recv=0, errin=0, errout=0, dropin=0, dropout=0),
'Local Area Connection* 11': snetio(bytes_sent=0, bytes_recv=0, packets_sent=0, packets_recv=0, errin=0, errout=0, dropin=0, dropout=0),
......}
# inet 代表 IPv4 and IPv6
In [42]: psutil.net_connections(kind='inet')
Out[42]:
[pconn(fd=115, family=<AddressFamily.AF_INET: 2>, type=<SocketType.SOCK_STREAM: 1>, laddr=addr(ip='10.0.0.1', port=48776), raddr=addr(ip='93.186.135.91', port=80), status='ESTABLISHED', pid=1254),
pconn(fd=117, family=<AddressFamily.AF_INET: 2>, type=<SocketType.SOCK_STREAM: 1>, laddr=addr(ip='10.0.0.1', port=43761), raddr=addr(ip='72.14.234.100', port=80), status='CLOSING', pid=2987),
pconn(fd=-1, family=<AddressFamily.AF_INET: 2>, type=<SocketType.SOCK_STREAM: 1>, laddr=addr(ip='10.0.0.1', port=60759), raddr=addr(ip='72.14.234.104', port=80), status='ESTABLISHED', pid=None),
pconn(fd=-1, family=<AddressFamily.AF_INET: 2>, type=<SocketType.SOCK_STREAM: 1>, laddr=addr(ip='10.0.0.1', port=51314), raddr=addr(ip='72.14.234.83', port=443), status='SYN_SENT', pid=None)
...]
>>> psutil.net_if_addrs()
{'lo': [snicaddr(family=<AddressFamily.AF_INET: 2>, address='127.0.0.1', netmask='255.0.0.0', broadcast='127.0.0.1', ptp=None),
snicaddr(family=<AddressFamily.AF_INET6: 10>, address='::1', netmask='ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff', broadcast=None, ptp=None),
snicaddr(family=<AddressFamily.AF_LINK: 17>, address='00:00:00:00:00:00', netmask=None, broadcast='00:00:00:00:00:00', ptp=None)],
'wlan0': [snicaddr(family=<AddressFamily.AF_INET: 2>, address='192.168.1.3', netmask='255.255.255.0', broadcast='192.168.1.255', ptp=None),
snicaddr(family=<AddressFamily.AF_INET6: 10>, address='fe80::c685:8ff:fe45:641%wlan0', netmask='ffff:ffff:ffff:ffff::', broadcast=None, ptp=None),
snicaddr(family=<AddressFamily.AF_LINK: 17>, address='c4:85:08:45:06:41', netmask=None, broadcast='ff:ff:ff:ff:ff:ff', ptp=None)]}
>>>
>>> psutil.net_if_stats()
{'eth0': snicstats(isup=True, duplex=<NicDuplex.NIC_DUPLEX_FULL: 2>, speed=100, mtu=1500),
'lo': snicstats(isup=True, duplex=<NicDuplex.NIC_DUPLEX_UNKNOWN: 0>, speed=0, mtu=65536)}
>>> psutil.users()
[suser(name='giampaolo', terminal='pts/2', host='localhost', started=1340737536.0, pid=1352),
suser(name='giampaolo', terminal='pts/3', host='localhost', started=1340737792.0, pid=1788)]
In [49]: import datetime
# 获取系统开机时间,以Linux时间戳格式返回
In [50]: psutil.boot_time()
Out[50]: 1595251273.0
# 转换为人可读的方式
In [51]: datetime.datetime.fromtimestamp(psutil.boot_time()).strftime("%Y-%m-%d %H:%M:%S")
Out[51]: '2020-07-20 21:21:13'
# 获取所有进程pid
In [52]: psutil.pids()
Out[52]:
[1,
2,
4,
...]
# 实例化一个Process对象,参数为进程PID
In [2]: p = psutil.Process(4313)
In [3]: p
Out[3]: psutil.Process(pid=4313, name='ipython', status='running', started='21:23:58')
# 进程bin路径
In [4]: p.exe()
Out[4]: '/usr/bin/python3.6'
# 进程工作目录绝对路径
In [5]: p.cwd()
Out[5]: '/root'
# 进程号
In [7]: p.pid
Out[7]: 4313
# 父进程号
In [8]: p.ppid()
Out[8]: 4139
# 进程状态
In [11]: p.status()
Out[11]: 'running'
# 进程 rss,vms信息
In [19]: p.memory_info()
Out[19]: pmem(rss=44388352, vms=347344896, shared=5902336, text=4096, lib=0, data=113049600, dirty=0)
# 进程内存利用率
In [20]: p.memory_percent()
Out[20]: 2.303264995557974
Linux | Windows | macOS |
---|---|---|
speedup: +2.6x | speedup: +1.8x / +6.5x |
本文原创地址://q13zd.cn/python-sysinfo-psutil.html编辑:yangbinbin,审核员:逄增宝