vmstat and system utilization on 5.1A PK3

From: Eiler, James A. (James.Eiler@alcoa.com)
Date: Tue Oct 29 2002 - 15:45:11 EST

All,

We've added PK3 and have noticed that the system utilization numbers from
"vmstat 1" don't seem to make sense any more.

If I do a series of vmstat commands, the utilization looks as follows:

ager99>vmstat
Virtual Memory Statistics: (pagesize = 8192)
  procs memory pages intr cpu
  r w u act free wire fault cow zero react pin pout in sy cs us
sy id
  7 621 27 46K 5529 10K 37M 5M 12M 367 7M 0 459 381K 1K 9
34 57
ager99>vmstat
Virtual Memory Statistics: (pagesize = 8192)
  procs memory pages intr cpu
  r w u act free wire fault cow zero react pin pout in sy cs us
sy id
  9 619 27 46K 5529 10K 37M 5M 12M 367 7M 0 459 381K 1K 9
34 57
ager99>vmstat
Virtual Memory Statistics: (pagesize = 8192)
  procs memory pages intr cpu
  r w u act free wire fault cow zero react pin pout in sy cs us
sy id
  7 621 27 46K 5529 10K 37M 5M 12M 367 7M 0 459 381K 1K 9
34 57

But if I use the 1 second option, the first line printed looks like the
others, then the succeeding lines look strange (including "fault cow zero
react pin"):

ager99>vmstat 1
Virtual Memory Statistics: (pagesize = 8192)
  procs memory pages intr cpu
  r w u act free wire fault cow zero react pin pout in sy cs us
sy id
  8 620 27 46K 5529 10K 37M 5M 12M 367 7M 0 459 381K 1K 9
34 57
  5 623 27 46K 5527 10K 2 16 19 0 28 0 1K 882K 2K 22
78 0
  5 623 27 46K 5528 10K 0 0 0 0 0 0 998 892K 2K 22
78 0
  9 619 27 46K 5528 10K 0 0 0 0 0 0 978 897K 2K 22
78 0
  7 621 27 46K 5528 10K 0 0 0 0 0 0 1K 880K 2K 23
77 0
  8 620 27 46K 5528 10K 0 0 0 0 0 0 1K 885K 2K 21
79 0
  8 620 27 46K 5528 10K 0 0 0 0 0 0 1K 883K 2K 23
77 0

Did something get broken in PK3?

Thanks!

Jim

This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:58 EDT