Sponsored Content
Operating Systems AIX Problem with nmon, actual CPU usage per process Post 302779107 by zaxxon on Tuesday 12th of March 2013 06:19:54 AM
Old 03-12-2013
@DGPickett

Thanks, but the options I used are valid switches I get displayed when I issue an nmon -h or if I check the man page, as well in the documentation in the IBM Wiki.

@-=Xray=-
Thanks, but tprof is for monitoring a program that you start, like sleep. I have processes that are running already and I have to check them for let's say an interval of 10 seconds and have to get their actual CPU usage for exactly this interval.
I don't want it since the process is alive (that would be regular ps) nor do I have the option to start a program.
For the explanation of -f together with -c and -s for nmon it seems to be a pity that this option is valid as it looks to me, that the output file is being written and does not get any addition 5 or 10 seconds later.
But I got your point and I think I have tried that already. I will have another look into it, but iirc, the values I got back were not that what I was looking for.


The background of all this is that I am about to write a plugin for Nagios which shall just capture the the actual CPU usage of the specified interval for a process.
Those plugins I have found do usually issue a ps and work with values that are the average since the process came to life.
When this is several days and you currently have a workload peak, you get for example just 12% CPU usage back while it is actually 77%.
So this approach is in my eyes "useless".

Maybe I explained it better now. If you have any other ideas, let me know. Thanks a lot so far for your efforts both.

Last edited by zaxxon; 03-12-2013 at 07:25 AM..
 

10 More Discussions You Might Find Interesting

1. Solaris

Process CPU usage in Solaris 10

Hi All, Please let me know the command (expect top) to view the cpu usage of every process in Solaris 10. Thanks in Advance, Arun (1 Reply)
Discussion started by: arun.viswanath
1 Replies

2. AIX

CPU usage of a process

I'm trying to monitor the CPU usage of a process and output that value to a file or variable. I know topas or nmon can tell me this in interactive mode but what I need is topas-looking output that allows me to write to a file after a discrete interval. Unlike nmon data collection to a file on top... (5 Replies)
Discussion started by: robot23
5 Replies

3. UNIX for Advanced & Expert Users

Finding CPU usage by a Unix Process

Hi, I am designing a load balancer for an application. I am trying to find out the CPU usage by a specifc Unix process (PID is known). I guess I can use ps command to find that. can somebody help me in finding what exact command I should use to find? It is on AIX 5.3. Regards Asutosh (2 Replies)
Discussion started by: asutoshch
2 Replies

4. UNIX for Dummies Questions & Answers

process cpu usage

Trying to come up with a command that will show all processes sorted from highest cpu usage to lowest. Any ideas? (9 Replies)
Discussion started by: cwsmichigan
9 Replies

5. Shell Programming and Scripting

CPU Usage for a particular process

Hi, I have a shell script. But, upon execution of the same, the cpu usage is sometimes getting 100 % (checked executing top command). At that point of time, my process hangs, doesn't run anymore. I need to kill it manually. My concern is, is there any default method, by which I can check... (1 Reply)
Discussion started by: jitendriya.dash
1 Replies

6. AIX

How to trace cpu/memory usage for a process

I don't know when the process will start and end, I need write a script to trace it's cpu/memory usage when it is runing. How to write this script? (2 Replies)
Discussion started by: rainbow_bean
2 Replies

7. Shell Programming and Scripting

CPU Usage of a process

Hi guys, I am currently writing a JAVA script to monitor certain unix processes through JConsole. Upon having lots of trouble with runtime.exec, i decided to bypass the top/ps command call and just get the information straight from /proc/*pid*/whatever. Now i can pull back any... (0 Replies)
Discussion started by: matt9949
0 Replies

8. UNIX for Dummies Questions & Answers

Average CPU and RAM usage for a process

Hi, I will be creating a process myself and I want to know the average CPU and RAM used by the process over the lifetime of the process. I see that there are various tools available(pidstat) for doing , I was wondering if it possible to do it in a single command while creation. Thanks in... (3 Replies)
Discussion started by: koustubh
3 Replies

9. Shell Programming and Scripting

Tricky situation with process cpu usage - AIX

OS: AIX so we frequently receive a lot of cpu related alerts. all types of checks have been created to keep an eye on the cpu but a lot of these checks make too much noise as the CPU is always being seen as high. the system and application owners say there's no issue with the cpu. so now,... (6 Replies)
Discussion started by: SkySmart
6 Replies

10. UNIX for Beginners Questions & Answers

How can I check actual memory usage each Process?

I can't check actual memory usage in the Redhat 5.5... The used memory is 14214 Mb of Total 15919 by Free -m command. I sum the RSS value on PS aux command result and the value is 5428.66Mb. Yes It's quite different actual usage memory and RSS value. So I added Shared memory value... (5 Replies)
Discussion started by: tom8254
5 Replies
cpustat(1)						      General Commands Manual							cpustat(1)

Name
       cpustat - report CPU statistics

Syntax
       cpustat [-cfhsv] [interval [count]]

Description
       The  command  displays  statistics about the use and state of each CPU in the system.  The information about the CPU might change while the
       command is running, so the information it gives is a snapshot taken at a given time.

       Normally, the command displays the statistics and state only once. You can cause the command to repeat the display by specifying a  decimal
       number  in  the	interval argument. The command repeats the display until it is interrupted, for example, by a control character entered at
       the terminal.

       You can limit the number of times the command repeats the display by specifying an integer in the count argument.

       The command provides an interactive interface that you can invoke using the -f option.

       The following describes the fields in the display:

       Statistics: Information about how each CPU's time is being used

	    us%      Percent of time spent in user mode

	    ni%      Percent of time spent in nice mode

	    sy%      Percent of time spent in system mode

	    id%      Percent of time spent idle

	    csw      Number of context switches

	    sys      Number of system calls

	    trap     Number of traps

	    intr     Number of device interrupts

	    ipi      Number of interprocessor interrupts

	    ttyin    Number of characters input to the terminal

	    ttyout   Number of characters output to the terminal

       State: Information about the different states of each CPU

	    cpuid    Unique identifier of the CPU.

	    state    CPU state.
			B - boot CPU
			D - disable soft errors
			S - stopped
			R - running
			T - TB needs invalidation
			P - panicked

	    ipi-mask interprocessor interrupt mask
			P - panic
			R - console print
			S - schedule
			D - disable
			T - TB invalidation
			H - stop CPU

	    proc     Indication of whether the CPU has an associated process. A Y indicates that an associated process exists, while  an  N  indi-
		     cates that no associated process exists.

	    pid      Process ID of the running process.

	    If any statistic field value exceeds 9999, it is shown in a scaled representation with the suffix k, which indicates multiplication by
	    1000, or with the suffix m, which indicates multiplication by 1,000,000.  For example, the value 12,345 would appear as 12k.

Options
       -c   Displays only the statistics for each CPU.

       -f   Displays statistics and state information on a full screen.  If you use the -f option, you are prompted to enter one of the  following
	    commands:

	    c	 Displays only statistics about each CPU.

	    d	 Uses the default mode of display.

	    h	 Displays the help screen. To return to the original display screen, enter any character.

	    q	 Quits from the full screen display.

	    s	 Displays only state information about each CPU.

	    v	 Displays information about each vector processor and omits information about other processors.

       -s   Displays only state information about each CPU.

       -h   Provides help information about the usage of the command.

       -v   Displays  information about each vector processor and omits information about other processors. When displays information about vector
	    processors, it adds the following information to the display:

	    vp state  Indicates that the vector processor is operational (OK), not usable (DEAD), enabled (ENA), or disabled  (DIS).  The  command
		      display may contain a combination of these states.

	    vp owner  PID of the vector process whose context is stored in the vector processor.

	    chp cxsw  Number  of context switches in which the process whose context changed is the only process using a particular vector proces-
		      sor. Because no other process uses the vector processor, the register data associated with the processor	does  not  change.
		      Therefore,  the  system  does  not  need	to save and restore the register data; the data remains in the registers until the
		      process is ready to use it.

	    exp cxsw  Number of context switches in which a number of processes might be using the same vector	processor.   When  more  than  one
		      process uses the same processor, the register data associated with the processor might be changed by any process. Therefore,
		      the system saves and restores the register data during a context switch for any process.

	    succ req  Number of times the vector process has successfully allocated the vector processor.

	    failed req
		      Number of times the vector process has been refused vector processor allocation.

       If you omit all options, reports a summary of the statistics since the system has been booted and the state of each CPU.

Examples
       To print the system status every five seconds ten times, enter the following command:
       % cpustat 5 10

Files
       Virtual main memory image

       System namelist

See Also
       iostat(1), vmstat(1)

																	cpustat(1)
All times are GMT -4. The time now is 05:49 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy