Login or Register to Ask a Question and Join Our Community

Sponsored Content
Full Discussion: vmstat
Top Forums UNIX for Dummies Questions & Answers vmstat Post 11011 by Perderabo on Tuesday 27th of November 2001 09:22:55 AM
Old 11-27-2001
No, a new process will cause real page faults that require stuff to be read into core. For a minor fault to happen, the stuff must already be in core.

Imagine a process running in a cpu. It does i/o and must wait for the i/o to complete. So the cpu finds another process and runs it. When the i/o completes the process will run again. But if the system has several cpus, it may wind up in a different one. The TLB is in the cpu. This cpu is seeing this process for the first time. The cpu will generate lots of page faults and all the kernel needs to do is shove some ptes in this cpu's tlb. I thought this would be the most common case.

But last night I looked this up in a book I have at home called Solaris Internals. The authors say that the most common source of minor faults is from processes that attach shared segments to themselves, particularly shared libraries. This makes sense, shared libraries are very commonly used.

If you do "vmstat -s" you will see that Solaris also has a notion of "micro faults". These are page faults due to illegal addresses and cause the kernel to simply deliver a sigsegv to the offending process. I am not sure if these are counted in the "mf" field of the regular vmstat output. That book ignores them.
 

10 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Vmstat

I have MATLAB INSTALLED IN MY SUN MACHINE >> WHENEVER I USE IT THE CPU USAGE SHOWS ABT 90% Seeing the vmstat shows that system calls and context switch counters reach a very high value . What are these counters ( Man pages do not give much info on that) .... The only thing i can make out that... (1 Reply)
Discussion started by: DPAI
1 Replies

2. UNIX for Dummies Questions & Answers

vmstat

When I exeute vmstat (e.g. vmstat 30 2), in some machines I get some wierd result as the first line. like: -117% or 208% for CPU idle percentage. But the second line is alright. Could someone explain this please. Thanks ! Chaadana (4 Replies)
Discussion started by: chaandana
4 Replies

3. UNIX for Advanced & Expert Users

vmstat

Hi, what does mean the free colomne in out put of vmstat ? is it free espace of physical memory or of swap space on hard disk ? Thank you (4 Replies)
Discussion started by: big123456
4 Replies

4. UNIX for Dummies Questions & Answers

vmstat

Hi I wanted to collect data by using vmstat -I 60 >xxxx.txt & using my own account It was stopped by it self after 2 hours try again same result We want to collect day date by succession how to collect data using vmstat for day Thank you (2 Replies)
Discussion started by: Syed_45
2 Replies

5. Solaris

Huge PI in vmstat

This is something nowbody around me can explain: vmstat (-S 5) shows a huge number of PI but when I try to monitor it in parallel with iostat - there is no IO activity to be seen that would correspond to this. I have 16G RAM and 32G swap file. I'll really appreciate if somebody can explain it.... (9 Replies)
Discussion started by: dkvent
9 Replies

6. Linux

vmstat help

Hi everyone, I need to see some VM manager performance/behavior information on some Linux boxes regarding pages scanned/activation of the paging algorithm in order to get an idea if a given server needs more memory and is actually paging. In Aix servers, by using the vmstat cmd you... (1 Reply)
Discussion started by: jcpetela
1 Replies

7. Shell Programming and Scripting

vmstat

Hi I need to write a script to display VMSTAT every 5 seconds and I just need the memory columns - swap free re and just the numbers and the headers arent required. For example bash-3.00$ vmstat 5| awk '{print $4" "$5" "$6}' disk faults cpu ------ This header isnt required swap... (3 Replies)
Discussion started by: kapilk
3 Replies

8. AIX

Need guidance on VMStat

I need some guidance on the differences in observations, not sure how significantly different are they. Also, It would be nice to hear on the values and what the obvious tuning for performance missing. Observation 1 ending vmstat -v 3948544 memory pages ending vmstat -v ... (1 Reply)
Discussion started by: Snipper
1 Replies

9. Solaris

Vmstat | nawk

I'm trying to parse vmstat output with this: vmstat | nawk '/0/{printf "%s\ \n", $5}' but output is different on two sparc Solaris 10 servers, one is missing line with 'free'. why ? (3 Replies)
Discussion started by: orange47
3 Replies

10. Linux

Vmstat

I m checking idle time using vmstat, below are the results var=$(ssh wmtmgr@$hostname vmstat | tail -1 | awk '{print $15}') 89 and now im subtracting 89 with 100 & im getting expected results expr 100 - $var 11 Now How can I get the result 11 in one line code? (4 Replies)
Discussion started by: sam@sam
4 Replies

LEARN ABOUT PLAN9

cpu

CPU(1)							      General Commands Manual							    CPU(1)

NAME

       cpu - connection to cpu server

SYNOPSIS

       cpu [ -h server ] [ -c cmd args ...  ]

DESCRIPTION

       Cpu  starts  an	rc(1) running on the server machine, or the machine named in the $cpu environment variable if there is no -h option.  Rc's
       standard input, output, and error files will be /dev/cons in the name space where the cpu command was invoked.  Normally, cpu is run in	an
       81/2(1)	window	on a terminal, so rc output goes to that window, and input comes from the keyboard when that window is current.  Rc's cur-
       rent directory is the working directory of the cpu command itself.

       The name space for the new rc is an analogue of the name space where the cpu command was invoked: it is the same except	for  architecture-
       dependent bindings such as /bin and the use of fast paths to file servers, if available.

       If a -c argument is present, the remainder of the command line is executed by rc on the server, and then cpu exits.

       The  name  space is built by running /usr/$user/lib/profile with the root of the invoking name space bound to /mnt/term.  The service envi-
       ronment variable is set to cpu; the cputype and objtype environment variables reflect the server's architecture.

FILES

       The name space of the terminal side of the cpu command is mounted on the CPU side on directory /mnt/term.

SOURCE

       /sys/src/cmd/cpu.c

SEE ALSO

       rc(1), 81/2(1)

BUGS

       Binds and mounts done after the terminal lib/profile is run are not reflected in the new name space.

																	    CPU(1)
All times are GMT -4. The time now is 11:02 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy