[Virtual-pkg-base-maintainers] Bug#620297: base: vmstat and /proc/loadavg disagree

[Brandon Metcalf]

Package: base
Severity: important

After a system running squeeze has been under constant load, the load
averages in /proc/loadavg seem to be calculated incorrectly with
/proc/loadavg showing values that are much too low.  Below is vmstat
output along with the contents of /proc/loadavg.  The fourth field
of /proc/loadavg seems correct, but the 1-minute avg is definitely wrong:

vmstat:
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
11  0      0 12569200  22832 7468824    0    0     9   135   22   96 39  3 58  0
loadavg:
0.34 4.98 6.47 10/468 13072
vmstat:
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
15  0      0 12566340  22832 7469612    0    0     9   135   23   96 39  3 58  0
loadavg:
0.34 4.98 6.47 16/468 13075

Although this is just two samples, the behavior is consistent over a long
period of time.


-- System Information:
Debian Release: 6.0.1
  APT prefers stable
  APT policy: (500, 'stable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.32-5-amd64 (SMP w/16 CPU cores)
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash