vmstat命令是最常见的Linux/Unix监控工具,可以展现给定时间间隔的服务器的状态值,包括服务器的CPU使用率,内存使用,虚拟内存交换情况,IO读写情况。这个命令是我查看Linux/Unix最喜爱的命令,一个是Linux/Unix都支持,二是相比top,我可以看到整个机器的CPU,内存,IO的使用情况,而不是单单看到各个进程的CPU使用率和内存使用率(使用场景不一样)。
一般vmstat工具的使用是通过两个数字参数来完成的,第一个参数是采样的时间间隔数,单位是秒,第二个参数是采样的次数,如:
root@ubuntu:~# vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo cs us sy id wa
2表示每个两秒采集一次服务器状态,1表示只采集一次。
实际上,在应用过程中,我们会在一段时间内一直监控,不想监控直接结束vmstat就行了,例如:
root@ubuntu:~# vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo cs us sy id wa
这表示vmstat每2秒采集数据,一直采集,直到我结束程序,这里采集了5次数据我就结束了程序。
好了,命令介绍完毕,现在开始实战讲解每个参数的意思。
r 表示运行队列(就是说多少个进程真的分配到CPU),我测试的服务器目前CPU比较空闲,没什么程序在跑,当这个值超过了CPU数目,就会出现CPU瓶颈了。这个也和top的负载有关系,一般负载超过了3就比较高,超过了5就高,超过了10就不正常了,服务器的状态很危险。top的负载类似每秒的运行队列。如果运行队列过大,表示你的CPU很繁忙,一般会造成CPU使用率很高。
b 表示阻塞的进程,这个不多说,进程阻塞,大家懂的。
swpd 虚拟内存已使用的大小,如果大于0,表示你的机器物理内存不足了,如果不是程序内存泄露的原因,那么你该升级内存了或者把耗内存的任务迁移到其他机器。
free 空闲的物理内存的大小,我的机器内存总共8G,剩余3415M。
buff Linux/Unix系统是用来存储,目录里面有什么内容,权限等的缓存,我本机大概占用300多M
cache cache直接用来记忆我们打开的文件,给文件做缓冲,我本机大概占用300多M(这里是Linux/Unix的聪明之处,把空闲的物理内存的一部分拿来做文件和目录的缓存,是为了提高 程序执行的性能,当程序使用内存时,buffer/cached会很快地被使用。)
si 每秒从磁盘读入虚拟内存的大小,如果这个值大于0,表示物理内存不够用或者内存泄露了,要查找耗内存进程解决掉。我的机器内存充裕,一切正常。
so 每秒虚拟内存写入磁盘的大小,如果这个值大于0,同上。
bi 块设备每秒接收的块数量,这里的块设备是指系统上所有的磁盘和其他块设备,默认块大小是1024byte,我本机上没什么IO操作,所以一直是0,但是我曾在处理拷贝大量数据(2-3T)的机器上看过可以达到140000/s,磁盘写入速度差不多140M每秒
bo 块设备每秒发送的块数量,例如我们读取文件,bo就要大于0。bi和bo一般都要接近0,不然就是IO过于频繁,需要调整。
in 每秒CPU的中断次数,包括时间中断
cs 每秒上下文切换次数,例如我们调用系统函数,就要进行上下文切换,线程的切换,也要进程上下文切换,这个值要越小越好,太大了,要考虑调低线程或者进程的数目,例如在apache和nginx这种web服务器中,我们一般做性能测试时会进行几千并发甚至几万并发的测试,选择web服务器的进程可以由进程或者线程的峰值一直下调,压测,直到cs到一个比较小的值,这个进程和线程数就是比较合适的值了。系统调用也是,每次调用系统函数,我们的代码就会进入内核空间,导致上下文切换,这个是很耗资源,也要尽量避免频繁调用系统函数。上下文切换次数过多表示你的CPU大部分浪费在上下文切换,导致CPU干正经事的时间少了,CPU没有充分利用,是不可取的。
us 用户CPU时间,我曾经在一个做加密解密很频繁的服务器上,可以看到us接近100,r运行队列达到80(机器在做压力测试,性能表现不佳)。
sy 系统CPU时间,如果太高,表示系统调用时间长,例如是IO操作频繁。
id 空闲 CPU时间,一般来说,id + us + sy = 100,一般我认为id是空闲CPU使用率,us是用户CPU使用率,sy是系统CPU使用率。
wt 等待IO CPU时间。
转自:http://www.cnblogs.com/ggjucheng/archive/2012/01/05/2312625.html
VMSTAT(8) Linux Administrator’s Manual VMSTAT(8)
vmstat - Report virtual memory statistics
vmstat [-a] [-n] [delay [ count]] vmstat [-f] [-s] [-m] vmstat [-S unit] vmstat [-d] vmstat [-p disk partition] vmstat [-V]
vmstat reports information about processes, memory, paging, block IO, traps, and cpu activity. The first report produced gives averages since the last reboot. Addi- tional reports give information on a sampling period of length delay. The process and memory reports are instantaneous in either case. Options The -a switch displays active/inactive memory, given a 2.5.41 kernel or better. The -f switch displays the number of forks since boot. This includes the fork, vfork, and clone system calls, and is equivalent to the total number of tasks created. Each process is represented by one or more tasks, depending on thread usage. This display does not repeat. The -m displays slabinfo. The -n switch causes the header to be displayed only once rather than periodically. The -s switch displays a table of various event counters and memory statistics. This display does not repeat. delay is the delay between updates in seconds. If no delay is speci- fied, only one report is printed with the average values since boot. count is the number of updates. If no count is specified and delay is defined, count defaults to infinity. The -d reports disk statistics (2.5.70 or above required) The -p followed by some partition name for detailed statistics (2.5.70 or above required) The -S followed by k or K or m or M switches outputs between 1000, 1024, 1000000, or 1048576 bytes The -V switch results in displaying version information.
Procs r: The number of processes waiting for run time. b: The number of processes in uninterruptible sleep. Memory swpd: the amount of virtual memory used. free: the amount of idle memory. buff: the amount of memory used as buffers. cache: the amount of memory used as cache. inact: the amount of inactive memory. (-a option) active: the amount of active memory. (-a option) Swap si: Amount of memory swapped in from disk (/s). so: Amount of memory swapped to disk (/s). IO bi: Blocks received from a block device (blocks/s). bo: Blocks sent to a block device (blocks/s). System in: The number of interrupts per second, including the clock. cs: The number of context switches per second. CPU These are percentages of total CPU time. us: Time spent running non-kernel code. (user time, including nice time) sy: Time spent running kernel code. (system time) id: Time spent idle. Prior to Linux 2.5.41, this includes IO-wait time. wa: Time spent waiting for IO. Prior to Linux 2.5.41, shown as zero.
Reads total: Total reads completed successfully merged: grouped reads (resulting in one I/O) sectors: Sectors read successfully ms: milliseconds spent reading Writes total: Total writes completed successfully merged: grouped writes (resulting in one I/O) sectors: Sectors written successfully ms: milliseconds spent writing IO cur: I/O in progress s: seconds spent for I/O
reads: Total number of reads issued to this partition read sectors: Total read sectors for partition writes : Total number of writes issued to this partition requested writes: Total number of write requests made for partition
cache: Cache name num: Number of currently active objects total: Total number of available objects size: Size of each object pages: Number of pages with at least one active object totpages: Total number of allocated pages pslab: Number of pages per slab
vmstat does not require special permissions. These reports are intended to help identify system bottlenecks. Linux vmstat does not count itself as a running process. All linux blocks are currently 1024 bytes. Old kernels may report blocks as 512 bytes, 2048 bytes, or 4096 bytes. Since procps 3.1.9, vmstat lets you choose units (k, K, m, M) default is K (1024 bytes) in the default mode vmstat uses slabinfo 1.1 FIXME
/proc/meminfo /proc/stat /proc/*/stat
iostat(1), sar(1), mpstat(1), ps(1), top(1), free(1)
Does not tabulate the block io per device or count the number of system calls.
Written by Henry Ware <[email protected]>. Fabian Frédérick <[email protected]> (diskstat, slab, partitions...)