vmstat issue

Hi,

I'm seeing weird things with vmstat:
frodo [z70bf32] [/home/z70bf32]# vmstat 5
kthr memory page disk faults cpu
r b w swap free re mf pi po fr de sr m1 m1 m1 m2 in sy cs us sy id
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
1 0 0 1557664 654384 179 1261 0 0 0 0 0 0 0 0 0 410 1973 156 68 16 17
0 0 0 1559720 656976 0 0 0 0 0 0 0 0 0 0 0 422 28 112 0 2 98
0 0 0 1560080 657312 34 624 0 3 3 0 0 0 0 0 0 421 826 146 3 8 88
0 0 0 1559960 656592 285 2233 0 35 34 0 0 0 0 0 0 438 4059 189 28 29 43
1 0 0 1559744 656440 349 5138 53 3 3 0 0 0 0 0 0 451 8064 420 33 67 0
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1559648 656560 34 368 0 0 0 0 0 0 0 0 0 432 742 125 2 9 89
0 0 0 1559016 655624 285 2111 0 89 89 0 0 0 0 0 0 554 4078 184 31 27 42
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1615760 747104 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1559464 656784 79 1054 0 5 5 0 0 0 0 0 0 504 1481 194 7 14 79
1 0 0 1557496 654248 101 912 0 26 26 0 0 0 0 0 0 457 1613 131 76 14 10
0 0 0 1559648 657112 30 551 0 0 0 0 0 0 0 0 0 420 738 145 4 6 90
^Cfrodo [z70bf32] [/home/z70bf32]#

This is Solaris 9 on a Netra T1:
SunOS frodo 5.9 Generic_118558-10 sun4u sparc SUNW,UltraAX-i2

I know, the 1st line should be ignored, as it would be the aggregated values since last boot.
However, sometimes after the 1st line, I get the exact same values as in the 1st line. The longer the in between time, the higher the chance that vmstat will repost the same values (except sometimes for memory) as in the 1st line.
Effectively, a 'vmstat 290 2' (as issued by BB/larrd) will always report the same line twice (except for memory):
frodo [z70bf32] [/home/z70bf32]# vmstat 290 2
kthr memory page disk faults cpu
r b w swap free re mf pi po fr de sr m1 m1 m1 m2 in sy cs us sy id
0 0 0 1615752 747088 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
0 0 0 1615744 747080 69 577 2 1 1 0 0 0 0 0 0 440 365 144 5 8 88
frodo [z70bf32] [/home/z70bf32]#

What could cause this? Is this a bug in vmstat, or would something else in my system interfere?

--
Best regards,
Jeroen Besse
(to contact me: the nospam address actually exists)
.

Relevant Pages

  • Re: strange memory leak
    ... PHYSCAL memory is being used by the system at a given time ... prefer, i use vmstat. ... which ALWAYS makes it look like you have very little free memory, ... the "paging space page in's" are much LESS than the "paging ...
    (comp.unix.aix)
  • Re: Kernel memory leak in ATAPI/CAM or ATAng?
    ... and a tunable maximum allocation being too high for the ... We've noticed a number of tuned memory limits were set when ... >>tuning parameters for large memory systems. ... Because the field allocated in vmstat are not large enough, ...
    (freebsd-current)
  • Re: vmstat meaning
    ... >What does the first line in vmstat mean? ... Where I would worry is that very high %wio you see with sar -- the ... the memory consumption somehow. ...
    (comp.unix.admin)
  • Memory leak/hog on Sol9
    ... The free memory goes down, ... in vmstat. ... Now, after installing Sol9, on an *idle* ... I see that a kernel memory size ...
    (comp.unix.solaris)
  • Re: Memory leak/hog on solaris 9
    ... The free memory goes down, ... in vmstat. ... Now, after installing Sol9, on an *idle* ... I see that a kernel memory size ...
    (comp.unix.solaris)