Linux CPU实时监控命令mpstat介绍,linuxmpstat


1,简介
       mpstat是Multiprocessor  Statistics的缩写,是实时系统监控工具。其报告是CPU的一些统计信息,这些信息存放在/proc/stat文件中。在多CPUs系统里,其不但能查看所有CPU的平均状况信息,而且能够查看特定CPU的信息。mpstat最大的特点是:可以查看多核心cpu中每个计算核心的统计数据;而类似工具vmstat只能查看系统整体cpu情况。

2,安装
[root@ora10g ~]# mpstat
-bash: mpstat: command not found
[root@ora10g ~]# mount -o loop -t iso9660 /dev/cdrom /mnt/cdrom
[root@ora10g ~]# cd /mnt/cdrom/Server/
[root@ora10g Server]# rpm -ivh sysstat-7.0.2-3.el5.i386.rpm 
warning: sysstat-7.0.2-3.el5.i386.rpm: Header V3 DSA signature: NOKEY, key ID 37017186
Preparing...                ########################################### [100%]
   1:sysstat                ########################################### [100%]

3,实例
用法
 mpstat -V  显示mpstat版本信息
 mpstat -P ALL 显示所有CPU信息
 mpstat -P n 显示第n个cup信息,n为数字,计数从0开始 
 mpstat n m  每个n秒显示一次cpu信息,连续显示m次,最后显示一个平均值
 mpstat n    每个n秒显示一次cpu信息,连续显示下去
 
查看每个cpu核心的详细当前运行状况信息,输出如下:   
 [root@ora10g ~]# mpstat -P ALL
Linux 2.6.18-194.el5 (ora10g.up.com)    11/05/14
09:13:02     CPU   %user   %nice    %sys %iowait    %irq   %soft  %steal   %idle    intr/s
09:13:02     all    0.62    0.01    0.54    3.48    0.00    0.02    0.00   95.32   1039.58
09:13:02       0    0.92    0.01    1.18    8.77    0.01    0.05    0.00   89.06   1030.23
09:13:02       1    0.27    0.00    0.31    1.46    0.00    0.01    0.00   97.96      1.00
....
09:13:02      14    1.12    0.02    0.45    2.99    0.00    0.01    0.00   95.39      7.74
09:13:02      15    0.18    0.00    0.22    0.70    0.00    0.01    0.00   98.90      0.59

查看多核CPU核心的当前运行状况信息, 每2秒更新一次
[root@ora10g ~]# mpstat -P ALL 2

查看某个cpu的使用情况,数值在[0,cpu个数-1]中取值
[root@ora10g ~]# mpstat -P 2
Linux 2.6.18-194.el5 (ora10g.up.com)    11/05/14
10:19:28     CPU   %user   %nice    %sys %iowait    %irq   %soft  %steal   %idle    intr/s
10:19:28       2    0.08    0.00    0.04    0.22    0.00    0.01    0.00   99.64      0.55

查看多核CPU核心的当前运行状况信息, 每2秒更新一次,显示5次
[root@ora10g ~]# mpstat -P ALL 2 5

4,字段含义如下
英文解释:
CPU:Processor number. The keyword all indicates that statistics are calculated as averages among all processors.
%user:Show the percentage of CPU utilization that occurred while executing at the user level (application).
%nice:Show the percentage of CPU utilization that occurred while executing at the user level with nice priority.
%sys:Show  the  percentage  of CPU utilization that occurred while executing at the system level (kernel). Note that
              this does not include time spent servicing interrupts or softirqs.
%iowait:Show the percentage of time that the CPU or CPUs were idle during which the system had an outstanding disk  I/O request.
%irq:Show the percentage of time spent by the CPU or CPUs to service interrupts.
%soft:Show  the  percentage  of time spent by the CPU or CPUs to service softirqs.  A softirq (software interrupt) is
              one of up to 32 enumerated software interrupts which can run on multiple CPUs at once.
%steal:Show the percentage of time spent in involuntary wait by the virtual CPU or CPUs while the hypervisor was  ser-vicing another virtual processor.
%idle:Show  the percentage of time that the CPU or CPUs were idle and the system did not have an outstanding disk I/O request.
intr/s:Show the total number of interrupts received per second by the CPU or CPUs.

参数解释 从/proc/stat获得数据
CPU 处理器 ID
user 在internal时间段里,用户态的CPU时间(%),不包含 nice值为负 进程 (usr/total)*100  
nice 在internal时间段里,nice值为负进程的CPU时间(%)   (nice/total)*100  
system 在internal时间段里,核心时间(%)   (system/total)*100
iowait 在internal时间段里,硬盘IO等待时间(%) (iowait/total)*100
irq 在internal时间段里,硬中断时间(%)      (irq/total)*100
soft 在internal时间段里,软中断时间(%)    (softirq/total)*100
idle 在internal时间段里,CPU除去等待磁盘IO操作外的因为任何原因而空闲的时间闲置时间(%)(idle/total)*100
intr/s 在internal时间段里,每秒CPU接收的中断的次数intr/total)*100

CPU总的工作时间=total_cur=user+system+nice+idle+iowait+irq+softirq
total_pre=pre_user+ pre_system+ pre_nice+ pre_idle+ pre_iowait+ pre_irq+ pre_softirq
user=user_cur – user_pre
total=total_cur-total_pre
其中_cur 表示当前值,_pre表示interval时间前的值。上表中的所有值可取到两位小数点。

相关内容