首先看看nmon命令的帮助信息：
[java] view plaincopy[root@linux nmon]# ./nmon.sh -h
Hint: nmon.sh [-h] [-s ] [-c ] [-f -d -t -r ] [-x]
-h 查看完整的说明信息，有两种模式：a、命令行交互式模式 (h) b、对于数据收集模式 (-f) -f 电子表格的输出格式 [注意：默认 -s300 -c288] 可选 (300秒*288次=86400秒=60*60*24=1天) -s <seconds> 刷新屏幕频率的时间 [默认 2] -c <number> 刷新屏幕的次数 [默认 1000000] -d <disks> to increase the number of disks [default 256] -t spreadsheet includes top processes -x capacity planning (每15分钟1天 = -fdt -s 900 -c 96)
版本 - nmon 14g
对于命令行交互式模式
-s 刷新屏幕频率的时间 [默认 2]
-c 刷新屏幕的次数 [默认 1000000]
-g User Defined Disk Groups [hit g to show them]
- file = on each line: group_name space separated
- like: database sdb sdc sdd sde
- upto 64 disk groups, 512 disks per line
- disks can appear more than once and in many groups
-b 命令行交互模式的界面是黑色和白色 [默认的颜色]
例如: nmon.sh -s 1 -c 100 (说明：在命令行交互模式下，每秒钟刷新一次屏幕，总共采集100次)
对于数据收集模式 = 电子表格格式 （逗号分隔值）
Note: use only one of f,F,z,x or X and make it the first argument
-f 电子表格输出格式 [注意: default -s300 -c288]
输出文件是 _YYYYMMDD_HHMM.nmon
-F 等同于 -f 但是使用用户提供的文件名
-r 用于电子表格文件 [default hostname]
-t include top processes in the output
-T as -t plus saves command line arguments in UARG section
-s 采集数据的时间
-c 采集数据的次数
-d to increase the number of disks [default 256]
-l disks/line default 150 to avoid spreadsheet issues. EMC=64.
-g User Defined Disk Groups (see above) - see BBBG & DG lines
-N include NFS Network File System
-I Include process & disks busy threshold (default 0.1)
don’t save or show proc/disk using less than this percent
-m 生成的数据文件的路径
例如：在30秒的时间间隔收集的top procs，持续1小时
nmon.sh -f -t -r Test1 -s30 -c120
To load into a spreadsheet: sort -A *nmon >stats.csv transfer the stats.csv file to your PC Start spreadsheet & then Open type=comma-separated-value ASCII file The nmon analyser or consolidator does not need the file sorted.
Capacity planning mode - use cron to run each day
-x sensible spreadsheet output for CP = one day
每15分钟1天 ( i.e. -ft -s 900 -c 96)
-X sensible spreadsheet output for CP = busy hour
每30秒1小时 ( i.e. -ft -s 30 -c 120)
交互模式命令
key — Toggles to control what is displayed —
h = 联机帮助信息
r = 机器类型，机器名，缓存信息和OS版本+LPAR
c = CPU处理器统计条形图
l = 条形图长期CPU（超过75个快照）
m = 内存统计
L = 巨大的内存页面统计
V = 虚拟内存和交换统计
k = 内核内部统计
n = 网络统计和错误
N = NFS网络文件系统
d = 磁盘I/O图
D = 磁盘I/O统计
o = 磁盘I/O映射（每个磁盘上的一个字符显示它是多么繁忙）
j = 文件系统
t = 顶级进程统计使用1,3,4,5来选择数据及顺序
u = 顶级进程命令的详细信息
v = 详细简单的检查 - OK/Warn(警告)/Danger(危险)
b = 黑白模式（或使用- b选项）
. = 最小模式，即只显示繁忙的磁盘和进程
key --- Other Controls --- + = 双屏幕刷新时间 - = 一半的屏幕刷新时间 q = 退出 (also x, e or control-C) 0 = 零峰计数复位 (峰值 = ">") space = 立即刷新屏幕
Startup Control
If you find you always type the same toggles every time you start
then place them in the NMON shell variable. For example:
export NMON=cmdrvtan
Others:
a) To you want to stop nmon - kill -USR2
b) Use -p and nmon outputs the background process pid
c) To limit the processes nmon lists (online and to a file)
Either set NMONCMD0 to NMONCMD63 to the program names
or use -C cmd:cmd:cmd etc. example: -C ksh:vi:syncd
d) If you want to pipe nmon output to other commands use a FIFO:
mkfifo /tmp/mypipe
nmon -F /tmp/mypipe &
grep /tmp/mypipe
e) If nmon fails please report it with:
1) nmon version like: 14g
2) the output of cat /proc/cpuinfo
3) some clue of what you were doing
4) I may ask you to run the debug version
Developer Nigel Griffiths Feedback welcome - on the current release only and state exactly the problem No warranty given or implied.
在操作系统下面输入如下命令就会进到nmon的监控界面：
[java] view plaincopy[root@linux nmon]# ./nmon.sh +nmon-14g------[H for help]---Hostname=linux--------Refresh= 2secs ---04:22.50-----------------------------------------------------------------------------------------------------------------+ || | ------------------------------ For help type H or ... | | # # # # #### # # nmon -? - hint| | ## # ## ## # # ## # nmon -h - full| | # # # # ## # # # # # # | | # # # # # # # # # # To start the same way every time | | # ## # # # # # ## set the NMON ksh variable| | # # # # #### # # | | ------------------------------ | || | Use these keys to toggle statistics on/off:| |c = CPU l = CPU Long-term - = Faster screen updates| |m = Memoryj = Filesystems+ = Slower screen updates| |d = Disksn = Network V = Virtual Memory | |r = Resource N = NFS v = Verbose hints | |k = kernelt = Top-processes . = only busy disks/procs| |h = more options q = Quit | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
在这里就可以看见一些指令的介绍和一些信息，其中 Refresh= 2secs 就是表示监控界面2秒钟刷新一次，也可以在nmon命令后面跟 -s 参数来指定想要刷新的频率，输入 h 可以看见更详细指令的介绍：
[java] view plaincopy+nmon-14g------[H for help]---Hostname=linux--------Refresh= 2secs ---04:27.49-----------------------------------------------------------------------------------------------------------------+ | HELP ------------------------------------------------------------------------------------------------------------------------------------------------| |key --- statistics which toggle on/off --- | |h = This help information| |r = RS6000/pSeries CPU/cache/OS/kernel/hostname details + LPAR | |t = Top Process Stats 1=basic 3=CPU| | u = shows command arguments (hit twice to refresh) | |c = CPU by processor l = longer term CPU averages | |m = Memory & Swap stats L=Huge j = JFS Usage Stats | |n = Network statsN = NFS| |d = Disk I/O Graphs D=Statso = Disks %Busy Map | |k = Kernel stats & loadavg V = Virtual Memory | |g = User Defined Disk Groups [start nmon with -g <filename>]| |v = Verbose Simple Checks - OK/Warnings/Danger | |b = black & white mode | |--- controls --- | |+ and - = double or half the screen refresh time | |q = quit space = refresh screen now| |. = Minimum Mode =display only busy disks and processes| |0 = reset peak counts to zero (peak = ">") | |Developer Nigel Griffiths see | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
在这里可以看见所有输入的指令，以及会看到什么内容，这里要说一下在nmon中输入一次 h 会看见帮助信息，在敲一次 h 就会取消显示了，其它指令也同理，这里输入 r （机器类型，机器名，缓存信息和OS版本+LPAR）：
[cpp] view plaincopy| Linux and Processor Details -------------------------------------------------------------------------------------------------------------------------| | Linux: Linux version 2.6.18-164.el5 (mockbuild@x86-002.build.)| | Build: (gcc version 4.1.2 0704 (Red Hat 4.1.2-46)) | | Release : 2.6.18-164.el5 | | Version : #1 SMP Tue Aug 18 15:51:54 EDT | | cpuinfo: model name : Intel(R) Core(TM) i3-2310M CPU @ 2.10GHz | | cpuinfo: vendor_id : GenuineIntel | | cpuinfo: cpu MHz : 2093.260 | | cpuinfo: bogomips : 4186.52 | | # of CPUs: 1 --1颗cpu | | Machine : i686 | | Nodename : linux --hostname | | /etc/*ease[1]: Red Hat Enterprise Linux Server release 5.4 (Tikanga)--操作系统版本 | | /etc/*ease[2]: (null)| | /etc/*ease[3]: (null)| | /etc/*ease[4]: (null)| | lsb_release: Distributor ID: RedHatEnterpriseServer | | lsb_release: Description: Red Hat Enterprise Linux Server release 5.4 (Tikanga)| | lsb_release: Release: 5.4| | lsb_release: Codename:Tikanga | +---------Warning: Some Statistics may not shown------------------------------------------------------------------------------------------------------------------------+
在这里看见一些主机和操作系统的信息，再敲一次 r 就会取消显示了，然后输入 t （顶级进程统计使用1,3,4,5来选择数据及顺序），然后再按数字 5 ：
[csharp] view plaincopy| Top Processes Procs=85 mode=5 (1=Basic, 3=Perf 4=Size 5=I/O)---------------------------------------------------------------------------------------------------------------------------| | PID %CPU SizeRes ResResRes Shared Faults Command| | UsedKBSet Text DataLib KBMin Maj | |40500.5 12748 10548108 10896 0 832840 nmon.sh| | 10.0 207262432280 0 53200 init | | 20.0 0 0 0 0 0000 migration/0 | | 30.0 0 0 0 0 0000 ksoftirqd/0 | | 40.0 0 0 0 0 0000 watchdog/0 | | 50.0 0 0 0 0 0000 events/0 | | 60.0 0 0 0 0 0000 khelper| | 70.0 0 0 0 0 0000 kthread| | 100.0 0 0 0 0 0000 kblockd/0 | | 110.0 0 0 0 0 0000 kacpid| | 670.0 0 0 0 0 0000 cqueue/0 | | 700.0 0 0 0 0 0000 khubd | | 720.0 0 0 0 0 0000 kseriod| |1360.0 0 0 0 0 0000 pdflush| |1370.0 0 0 0 0 0000 pdflush| |1380.0 0 0 0 0 0000 kswapd0| |1390.0 0 0 0 0 0000 aio/0 | +---------Warning: Some Statistics may not shown-----------------------------------------------------------------------------------------------------------------------------------------------+
注意这个 mode=5 表示就是按I/O来排序了，这里还可以选择其它（1、3、4、5）方式排序，这里可以看见系统有85个进程（Procs=85），5就是按占用的cpu来排的降序，接着输入 u （顶级进程命令的详细信息）：
[csharp] view plaincopy| Top Processes Procs=85 mode=5 (1=Basic, 3=Perf 4=Size 5=I/O)---------------------------------------------------------------------------------------------------------------------------------| | PID %CPU ResSize CommandCommand| |UsedKB | | 4050 1.0 10660 ./nmon.sh | | 1 0.0624 init [3] | | 2 0.0 0 [migration/0] | | 3 0.0 0 [ksoftirqd/0] | | 4 0.0 0 [watchdog/0]| | 5 0.0 0 [events/0] | | 6 0.0 0 [khelper] | | 7 0.0 0 [kthread] | |10 0.0 0 [kblockd/0]| |11 0.0 0 [kacpid] | |67 0.0 0 [cqueue/0] | |70 0.0 0 [khubd]| |72 0.0 0 [kseriod] | |136 0.0 0 [pdflush] | |137 0.0 0 [pdflush] | |138 0.0 0 [kswapd0] | |139 0.0 0 [aio/0]| +---------Warning: Some Statistics may not shown-----------------------------------------------------------------------------------------------------------------------------------------------+
以上信息都很直观，我就不在多说了，接着看 c （CPU处理器统计条形图）：
[cpp] view plaincopy| CPU Utilisation -------------------------------------------------------------------------------------------------------------------------------------| |---------------------------+-------------------------------------------------+ | |CPU User% Sys% Wait% Idle|0|25 |50|75 100| | | 1 0.0 0.0 0.0 100.0| > | | |---------------------------+-------------------------------------------------+ | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
从上面可以看见系统非常闲（Idle=100%），其中“>”代表的是系统最高cpu的使用峰值，如果按数字0就会重置峰值为0了，接着看 l （条形图长期CPU）：
[
sql] view plaincopy| CPU +-------------------------------------------------------------------------+| |100%-|| | | 95%-|| | | 90%-|| | | 85%-|| | | 80%-|| | | 75%-|| | | 70%-|| | | 65%-|| | | 60%-|| | | 55%-|| | | 50%-|| | | 45%-|| | | 40%-|| | | 35%-|| | | 30%-|| | | 25%-|| | | 20%-|| | | 15%-|| | | 10%-|| | +---------Warning: Some Statistics may not shown-----------------------------------------------------------------------------------------------------------------------------------------------+
也是cpu使用情况的另一种显示，其中“|”和上面的“>”原理一样，接着看 m（内存统计）：
[cpp] view plaincopy| Memory Stats ----------------------------------------------------------------------------------------------------------------------------------------| |RAMHighLowSwap Page Size=4 KB| | Total MB 503.30.0 503.3 1027.6 | | Free MB 192.30.0 192.3 1027.6 | | Free Percent 38.2%0.0% 38.2% 100.0% | | MB MB MB| | Cached= 200.6Active= 107.4 | | Buffers= 46.4 Swapcached=0.0 Inactive = 169.7 | | Dirty =0.2 Writeback =0.0 Mapped =9.5 | | Slab = 26.8 Commit_AS = 130.0 PageTables=1.3 | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
从上面可以看见swap还100%没有使用，物理内存ram空闲38.2%，接着看 L （巨大的内存页面统计）：
[cpp] view plaincopyLarge (Huge) Page Stats -----------------------------------------------------------------------------------------------------------------------------| | There are no Huge Pages| | - see /proc/meminfo | || |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
这个东西我也没看太明白，接着看 j （文件系统）：
[sql] view plaincopy| Filesystems -----------------------------------------------------------------------------------------------------------------------------------------| |Filesystem SizeMB FreeMB Use% TypeMountPoint | |/dev/sda3 48502 36378 21% ext3/ | |/proc - -- procnot a real filesystem | |/sys - -- sysfs not a real filesystem | |/dev/pts - -- devpts not a real filesystem | |/dev/sda1 9982 12% ext3/boot | |/dev/shm - -- tmpfs not a real filesystem | |/proc/sys/fs/binfmt_misc - -- binfmt_m not a real filesystem | |/var/lib/nfs/rpc_pipefsrpc_pipe size=zero blocks! | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
这里可以看见磁盘的使用情况，非常直观，接着看 n （网络统计和错误）：
[cpp] view plaincopy| Network I/O -----------------------------------------------------------------------------------------------------------------------------------------| |I/F Name Recv=KB/s Trans=KB/s packin packout insize outsize Peak->Recv Trans | |lo0.00.0 0.00.00.0 0.0 0.20.2 | | eth00.00.1 0.50.5 60.0 218.0 52.4 99.0 | | sit00.00.0 0.00.00.0 0.0 0.00.0 | | Network Error Counters ------------------------------------------------------------------------------------------------------------------------------| |I/F Name iErrors iDrop iOverrun iFrame oErrors oDrop oOverrun oCarrier oColls | |lo 0 0 0 0 0 0 0 0 0| | eth0 0 0 0 0 0 0 0 0 0| | sit0 0 0 0 0 0 0 0 0 0| |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 N （NFS网络文件系统）：
[cpp] view plaincopy| Network Filesystem (NFS) I/O Operations per second --------------------------------------------------------------------------------------------------| | Version 2 Client Server Version 3Client Server | | null0.00.0null0.00.0 | | getattr0.00.0 getattr0.00.0 | | setattr0.00.0 setattr0.00.0 | | root0.00.0 lookup0.00.0 | | lookup0.00.0 access0.00.0 | | readlink0.00.0 readlink0.00.0 | | read0.00.0read0.00.0 | | wrcache0.00.0write0.00.0 | |write0.00.0 create0.00.0 | | create0.00.0mkdir0.00.0 | | remove0.00.0 symlink0.00.0 | | rename0.00.0mknod0.00.0 | | link0.00.0 remove0.00.0 | | symlink0.00.0rmdir0.00.0 | |mkdir0.00.0 rename0.00.0 | |rmdir0.00.0link0.00.0 | | readdir0.00.0 readdir0.00.0 | | fsstat0.00.0 readdirplus0.00.0 | +---------Warning: Some Statistics may not shown-----------------------------------------------------------------------------------------------------------------------------------------------+
接着看 d （磁盘I/O图）：
[cpp] view plaincopy| Disk I/O --/proc/diskstats----mostly in KB/s-----Warning:contains duplicates-------------------------------------------------------------------------| |DiskName Busy Read WriteKB|0|25 |50|75 100| | |sda 0% 0.0 0.0|> | | |sda1 0% 0.0 0.0|> | | |sda2 0% 0.0 0.0|> | | |sda3 0% 0.0 0.0|> | | |Totals Read-MB/s=0.0Writes-MB/s=0.0Transfers/sec=0.0| |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 D （磁盘I/O统计）：
[cpp] view plaincopyDisk I/O --/proc/diskstats----mostly in KB/s-----Warning:contains duplicates-------------------------------------------------------------------------| |DiskName Busy Read Write Xfers Size Peak% Peak-RW InFlight | |sda 0%0.00.0KB/s 0.0 0.0KB 0% 8.0KB/s 0 | | |sda1 0%0.00.0KB/s 0.0 0.0KB 0% 0.0KB/s 0 | | |sda2 0%0.00.0KB/s 0.0 0.0KB 0% 0.0KB/s 0 | | |sda3 0%0.00.0KB/s 0.0 0.0KB 0% 8.0KB/s 0 | | |Totals Read-MB/s=0.0Writes-MB/s=0.0Transfers/sec=0.0| |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 o （磁盘I/O映射）：
[cpp] view plaincopy| Disk %Busy Map --Key: @=90 #=80 X=70 8=60 O=50 0=40 o=30 +=20 -=10 .=5 _=0%--------------------------------------------------------------------------| | Disk No. 1 2 3 4 5 6| |Disks=40123456789012345678901234567890123456789012345678901234567890123 | |disk 0 to 63 ____ | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 k （内核内部统计）：
[cpp] view plaincopy| Kernel Stats ----------------------------------------------------------------------------------------------------------------------------------------| | RunQueue 1 Load Average CPU use since boot time | | ContextSwitch 30228.5 1 mins 0.00 Uptime Days= 0 Hours= 2 Mins=29 | | Forks 25.9 5 mins 0.00 Idle Days= 0 Hours= 2 Mins=28 | | Interrupts738415.7 15 mins 0.00 Average CPU use= 0.98% | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 V （虚拟内存和交换统计）：
[cpp] view plaincopy| Virtual-Memory --------------------------------------------------------------------------------------------------------------------------------------| |nr_dirty = 0 pgpgin= 0High Normal DMA | |nr_writeback= 0 pgpgout= 0 alloc 000 | |nr_unstable = 0 pgpswpin = 0 refill 000 | |nr_table_pgs=330 pgpswpout = 0 steal 000 | |nr_mapped =2438 pgfree= 0 scan_kswapd000 | |nr_slab=6852 pgactivate = 0 scan_direct000 | | pgdeactivate= 0| |allocstall = 0 pgfault= 7 kswapd_steal=0 | |pageoutrun = 0 pgmajfault = 0 kswapd_inodesteal=0 | |slabs_scanned= 0 pgrotated = 0 pginodesteal=0 | |----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
接着看 v （详细简单的检查）：
[cpp] view plaincopy| Verbose Mode ----------------------------------------------------------------------------------------------------------------------------------------| | Code Resource Stats Now Warn Danger | |OK -> CPU%busy 0.0% >80% >90%| |OK -> Top Disk%busy 0.0% >40% >60%| || || | HELP ------------------------------------------------------------------------------------------------------------------------------------------------|
这里显示了cpu和disk的一个诊断信息，Warn(警告)/Danger(危险)，可以看见cpu大于80%就是警告，大于90%就是危险，disk同理如下。nmon工具可以收集非常详细的系统信息，不过通常用的更多的是采样信息，来生成报表！！！
