Description of problem:  
vmstat gives invalid data for idle cpu.  
  
  
Version-Release number of selected component (if applicable):  
procps-2.0.17-13.7  
  
How reproducible:  
just run it.  
  
Additional info:  
I have a notion that this might be related to the fact that the server has  
been running for 559 days now so the values in /proc/stat that vmstat uses  
have become larger than 32 bits.  
This is a problem for us, as nagios starting giving alerts because of the 0 in 
idle cpu. The alerts started about 1 week ago. 
An upgrade to the latest version of procps did not help. 
We also have munin running on the same server, and it works OK with cpu loads. 
 
Here is the output of /proc/stat:  
#cat /proc/stat  
cpu  361122015 9493 117779454 4345053198 7367033 2792545 4092073  
cpu0 361122015 9493 117779454 4345053198 7367033 2792545 4092073  
intr 4726461882 1137517882 1085 0 0 0 0 0 0 0 0 563659663 3025282925 326 0 1 0  
ctxt 13140683776  
btime 1101370711  
processes 156740468  
procs_running 1  
procs_blocked 0  
  
and of vmstat:  
# vmstat  
procs                      memory      swap          io     system         cpu  
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa  
 5  0      0 156564 125272 1631688    0    0     2     0    0     0  4  7  0   
2