Bug 2229 - vmstat doesn't like Linux SMP
Summary: vmstat doesn't like Linux SMP
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Red Hat Linux
Classification: Retired
Component: psmisc
Version: 5.2
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Michael K. Johnson
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 1999-04-17 07:13 UTC by Louis Mandelstam
Modified: 2008-05-01 15:37 UTC (History)
2 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 1999-06-13 01:59:10 UTC
Embargoed:


Attachments (Terms of Use)

Description Louis Mandelstam 1999-04-17 07:13:50 UTC
1. vmstat CPU idle column does Strange Things<tm> when CPUs
get busy.

2. vmstat should perhaps show seperate idle columns per CPU.

Comment 1 Michael K. Johnson 1999-05-11 16:14:59 UTC
1) could you be explicit, perhaps paste in a line demonstrating what
   you are writing about?  I suggest that you download procps-1.2.12
   first, as it changes the output format of vmstat a bit and may
   answer this for you, since I don't know what the complaint is...

2) when /proc exports the information, perhaps vmstat will support
   it.  This is a kernel limitation, not a vmstat limitation.

Comment 2 Louis Mandelstam 1999-05-17 10:11:59 UTC
The problem does indeed appear to be fixed in later releases.
The only SMP system under my control that still exhibits the problem
(the rest having been upgraded) is a 5.0 box with procps-1.2.7-1 and
kernel 2.0.32.  Sample output:

 0 0 0     0  6772 24612 338432   0   0    0    0  146   53   0   2 98
 1 0 0     0  6756 24612 338432   0   0    0    0  147   71   0  54 46
 1 0 0     0  6732 24612 338432   0   0    0    0  120   20   0 102
41698710
 1 0 0     0  6724 24612 338432   0   0    0    0  121   15   0 102
41698710
 procs                  memory    swap        io    system         cpu
 r b w  swpd  free  buff cache  si  so   bi   bo   in   cs  us  sy  id
 2 0 0     0  6712 24612 338432   0   0    0   25  150   26   6  96
41698710
 1 0 0     0  6708 24612 338432   0   0    0    0  153   46   1 103
40904447
 1 0 0     0  6704 24612 338432   0   0    0    0  127   33   1 101
41698710
 0 0 0     0  6700 24612 338432   0   0    0    0  170  150   6  91 3
 0 0 0     0  6700 24612 338432   0   0    0    0  119   12   0   2 98

The problem appears to be historic/academic however now, so I'd close
it.  Sorry for the trouble.

Comment 3 Jeff Johnson 1999-06-13 01:59:59 UTC
This appears to be fixed. Please reopen if I'm wrong.


Note You need to log in before you can comment on or make changes to this bug.