Share your knowledge at the LQ Wiki.
Go Back > Forums > Linux Forums > Linux - Distributions > Red Hat
User Name
Red Hat This forum is for the discussion of Red Hat Linux.


  Search this Thread
Old 12-29-2006, 09:37 AM   #1
Registered: Jan 2006
Location: USA
Posts: 579

Rep: Reputation: 62
issue with VMSTAT

so what gives. different man pages show different things for IO bi and bo ???

shouldn't bo mean "blocks out" and bi "blocks in" from perspective of the cpu ??

does redhat compile their vmstat differently ??
bi: Blocks sent to a block device (blocks/s).
bo: Blocks received from a block device (blocks/s).
bi: Blocks received from a block device (blocks/s).
bo: Blocks sent to a block device (blocks/s).

[my system]
Red Hat Linux release 9 (Shrike) [and yeah, i know, very old OS]
vmstat -V procps version 2.0.11
man -V man, version 1.5k

man vmstat says:
bi: Blocks sent to a block device (blocks/s).
bo: Blocks received from a block device (blocks/s).

Last edited by Linux_Kidd; 12-29-2006 at 09:55 AM.
Old 12-30-2006, 06:15 AM   #2
Registered: Jun 2006
Location: Mariposa
Distribution: Slackware 9.1
Posts: 938

Rep: Reputation: 30
vmstat man page issue

My system is Slackware 9.1, kernel 2.4.22, but the experiment I performed should also work with your system.

Before we begin, my vmstat(8) man page says:

bi: blocks sent out to a block device (in blocks/s)
bo: blocks received from a block device (in blocks/s)

Do the following experiment on a fairly idle system. Do it only once. It relies on there being not too many directory blocks in main memory at the beginning.

If you need to repeat the experiment, then either reboot your system or perform a lot of other activity on it first.

The first step of the experiment is to do this:

vmstat 10

This causes the screen to show incremental results every 10 seconds. It should soon settle down to 0 and 0 for the bi and bo columns.

The second step of the experiment is done on a different screen, while the vmstat command is still running. It can be done in either of two ways.

The first way (preferable, because you can avoid logging in as root) is to choose a user account on your system which owns many, many files. Log in and do this:

du -s .

That's a period at the end of the command, with a space character just before it.
The second way is to log in as root, and do this:

du -s /

If this command takes a long time to finish, that's ok. Don't wait for it to finish. Just let it keep running and continue with the third step.

The third step is to return to the original screen where the vmstat command is running.

You might already see a line where the bi or bo quantity is far from 0. If you don't, wait for 10 seconds or so, and it will appear.

This far-from-zero quantity reflects the input of directory blocks.

On my system, that was the "bi" column.

On my system, the man page lies.

Hope this helps.


redhat, shrike

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off

Similar Threads
Thread Thread Starter Forum Replies Last Post
vmstat vs. sar danieldinu Solaris / OpenSolaris 10 10-27-2006 04:07 PM
vmstat UltraSoul Solaris / OpenSolaris 9 07-14-2006 10:11 PM
regarding vmstat output cnkhandar Linux - Newbie 1 07-07-2006 08:01 AM
vmstat shows swpd as 0 Skazi Linux - General 4 03-11-2005 03:39 AM
vmstat segmentation fault mbeason Debian 0 09-07-2004 05:35 PM > Forums > Linux Forums > Linux - Distributions > Red Hat

All times are GMT -5. The time now is 11:46 AM.

Main Menu
Write for LQ is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration