Home > Article > System Tutorial > Introduction to commands for Linux performance monitoring
Introduction to commands for Linux performance monitoring
- WBOYforward
- 2023-12-29 19:43:271108browse
| Introduction | This is part of our ongoing series on Linux commands and performance monitoring. Both vmstat and iostat commands are available on all major unix-like systems |
If the vmstat and iostat commands are not available on your system, please install the sysstat package. The vmstat, sar and iostat commands are all included in the sysstat (system monitoring tool) package. The iostat command generates statistics for the CPU and all devices. You can download the source code package from (http://sebastien.godard.pagesperso-orange.fr/download.html) to compile and install sysstat, but we recommend installing it through the YUM command.
#yum -y install sysstat
vmstat - Brief information about memory, processes, paging, etc.
iostat - CPU statistics, input/output statistics for devices and partitions.
In the following example, 6 columns are output. The meaning of each column parsed in the man page of vmstat. The most important ones are the free attribute in the memory and the si and so attributes in the swap partition.
[root@tecmint ~]# vmstat -a procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu----- r b swpd free inact active si so bi bo in cs us sy id wa st 1 0 0 810420 97380 70628 0 0 115 4 89 79 1 6 90 3 0
Free – free memory space
•si – The amount of data swapped from disk into memory per second (in KB).
•so – The amount of data (in KB) swapped from memory to disk per second.
Note: If you execute the vmstat command without parameters, it will output a summary report since the system started.
2. Execute vmstat every X seconds, N times in totalThe following command will execute vmstat every 2 seconds and will automatically stop executing after 6 times.
[root@tecmint ~]# vmstat 2 6 procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 0 0 0 810420 22064 101368 0 0 56 3 50 57 0 3 95 2 0 0 0 0 810412 22064 101368 0 0 0 0 16 35 0 0 100 0 0 0 0 0 810412 22064 101368 0 0 0 0 14 35 0 0 100 0 0 0 0 0 810412 22064 101368 0 0 0 0 17 38 0 0 100 0 0 0 0 0 810412 22064 101368 0 0 0 0 17 35 0 0 100 0 0 0 0 0 810412 22064 101368 0 0 0 0 18 36 0 1 100 0 03. vmstat command with timestamp
Execute the vmstat command with the -t parameter. This command will put a timestamp after each line of output, as shown below.
[tecmint@tecmint ~]$ vmstat -t 1 5 procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ ---timestamp--- r b swpd free buff cache si so bi bo in cs us sy id wa st 0 0 0 632028 24992 192244 0 0 70 5 55 78 1 3 95 1 0 2012-09-02 14:57:18 IST 1 0 0 632028 24992 192244 0 0 0 0 171 514 1 5 94 0 0 2012-09-02 14:57:19 IST 1 0 0 631904 24992 192244 0 0 0 0 195 600 0 5 95 0 0 2012-09-02 14:57:20 IST 0 0 0 631780 24992 192244 0 0 0 0 156 524 0 5 95 0 0 2012-09-02 14:57:21 IST 1 0 0 631656 24992 192244 0 0 0 0 189 592 0 5 95 0 0 2012-09-02 14:57:22 IST4. Statistics of various counters
The -s parameter of the vmstat command will output various event counters and memory statistics.
vmstat命令的-s参数，将输出各种事件计数器和内存的统计信息。
[tecmint@tecmint ~]$ vmstat -s
1030800 total memory
524656 used memory
277784 active memory
185920 inactive memory
506144 free memory
26864 buffer memory
310104 swap cache
2064376 total swap
0 used swap
2064376 free swap
4539 non-nice user cpu ticks
0 nice user cpu ticks
11569 system cpu ticks
329608 idle cpu ticks
5012 IO-wait cpu ticks
79 IRQ cpu ticks
74 softirq cpu ticks
0 stolen cpu ticks
336038 pages paged in
67945 pages paged out
0 pages swapped in
0 pages swapped out
258526 interrupts
392439 CPU context switches
1346574857 boot time
2309 forks
5. Disk statistics
The -d parameter of vmstat will output statistical information of all disks.
[tecmint@tecmint ~]$ vmstat -d
disk- ------------reads------------ ------------writes----------- -----IO------
total merged sectors ms total merged sectors ms cur sec
ram0 0 0 0 0 0 0 0 0 0 0
ram1 0 0 0 0 0 0 0 0 0 0
ram2 0 0 0 0 0 0 0 0 0 0
ram3 0 0 0 0 0 0 0 0 0 0
ram4 0 0 0 0 0 0 0 0 0 0
ram5 0 0 0 0 0 0 0 0 0 0
ram6 0 0 0 0 0 0 0 0 0 0
ram7 0 0 0 0 0 0 0 0 0 0
ram8 0 0 0 0 0 0 0 0 0 0
ram9 0 0 0 0 0 0 0 0 0 0
ram10 0 0 0 0 0 0 0 0 0 0
ram11 0 0 0 0 0 0 0 0 0 0
ram12 0 0 0 0 0 0 0 0 0 0
ram13 0 0 0 0 0 0 0 0 0 0
ram14 0 0 0 0 0 0 0 0 0 0
ram15 0 0 0 0 0 0 0 0 0 0
loop0 0 0 0 0 0 0 0 0 0 0
loop1 0 0 0 0 0 0 0 0 0 0
loop2 0 0 0 0 0 0 0 0 0 0
loop3 0 0 0 0 0 0 0 0 0 0
loop4 0 0 0 0 0 0 0 0 0 0
loop5 0 0 0 0 0 0 0 0 0 0
loop6 0 0 0 0 0 0 0 0 0 0
loop7 0 0 0 0 0 0 0 0 0 0
sr0 0 0 0 0 0 0 0 0 0 0
sda 7712 5145 668732 409619 3282 28884 257402 644566 0 126
dm-0 11578 0 659242 1113017 32163 0 257384 8460026 0 126
dm-1 324 0 2592 3845 0 0 0 0 0 2
6. Output statistical information in MB units
The -S and -M parameters of vmstat (uppercase and MB) will be output in MB units. vmstat outputs statistical information in KB by default.
[root@tecmint ~]# vmstat -S M 1 5 procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu----- r b swpd free buff cache si so bi bo in cs us sy id wa st 0 0 0 346 53 476 0 0 95 8 42 55 0 2 96 2 0 0 0 0 346 53 476 0 0 0 0 12 15 0 0 100 0 0 0 0 0 346 53 476 0 0 0 0 32 62 0 0 100 0 0 0 0 0 346 53 476 0 0 0 0 15 13 0 0 100 0 0 0 0 0 346 53 476 0 0 0 0 34 61 0 1 99 0 06 examples of Iostat commands under Linux 1. Output CPU and input/output (I/O) statistics
The iostat command without parameters will output the CPU and output/output statistics of each partition, as shown below.
[root@tecmint ~]# iostat
Linux 2.6.32-279.el6.i686 (tecmint.com) 09/03/2012 _i686_ (1 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
0.12 0.01 1.54 2.08 0.00 96.24
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
sda 3.59 161.02 13.48 1086002 90882
dm-0 5.76 159.71 13.47 1077154 90864
dm-1 0.05 0.38 0.00 2576 0
2. Only output CPU statistics
The -c parameter of the iostat command only outputs CPU statistical information, as shown below.
[root@tecmint ~]# iostat -c
Linux 2.6.32-279.el6.i686 (tecmint.com) 09/03/2012 _i686_ (1 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
0.12 0.01 1.47 1.98 0.00 96.42
3. Output only the input/output statistics of the disk
The -d parameter of the iostat command only outputs the input/output statistics of all partitions of the disk, as shown below.
[root@tecmint ~]# iostat -d Linux 2.6.32-279.el6.i686 (tecmint.com) 09/03/2012 _i686_ (1 CPU) Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn sda 3.35 149.81 12.66 1086002 91746 dm-0 5.37 148.59 12.65 1077154 91728 dm-1 0.04 0.36 0.00 2576 04. Output only the input/output statistics of a certain disk
By default, the iostat command will output the statistical information of all partitions, but if you add the -p parameter and the disk device name after the iostat command, the command will only output the input/output statistical information of the listed disks. As follows.
[root@tecmint ~]# iostat -p sda
Linux 2.6.32-279.el6.i686 (tecmint.com) 09/03/2012 _i686_ (1 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
0.11 0.01 1.44 1.92 0.00 96.52
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
sda 3.32 148.52 12.55 1086002 91770
sda1 0.07 0.56 0.00 4120 18
sda2 3.22 147.79 12.55 1080650 91752
5. 输出逻辑卷管理（LVM）的统计信息
iostat命令的-N(大写)参数将会输出LVM（LCTT译注：LVM是linux环境下对磁盘分区进行管理的一种机制，是磁盘分区和文件系统间的一个逻辑层）的统计信息，如下所示。
[root@tecmint ~]# iostat -N
Linux 2.6.32-279.el6.i686 (tecmint.com) 09/03/2012 _i686_ (1 CPU)
avg-cpu: %user %nice %system %iowait %steal %idle
0.11 0.01 1.39 1.85 0.00 96.64
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
sda 3.20 142.84 12.16 1086002 92466
vg_tecmint-lv_root 5.13 141.68 12.16 1077154 92448
vg_tecmint-lv_swap 0.04 0.34 0.00 2576 0
6. iostat版本信息
iostat的-V(大写)参数将会输出iostat的版本信息，如下所示。
[root@tecmint ~]# iostat -V sysstat version 9.0.4 (C) Sebastien Godard (sysstat orange.fr)
注意：vmstat和iostat的输出中包含多列的数据和标志，限于篇幅无法在本文中进行详细的解析。如果你想知道更多的信息，请查看vmstat和iostat的man帮助手册。如果你觉得本文对你有价值，请在下面的评论框中与你的朋友分享。
The above is the detailed content of Introduction to commands for Linux performance monitoring. For more information, please follow other related articles on the PHP Chinese website!