All:

I have a V445 server with four IIIi CPUs and 8 GB RAM running Solaris 10 and an Oracle database along with some app server components and we have had some performance issues - so I collected some VMSTAT and MPSTAT data over the course of three days with a 15-minute polling interval.

I have been looking over the data and I gotta believe that there is a bug becasue I had both VMSTAT and MPSTAT report values in excess of 20.495 QUADRILLION system calls / faults (vmstat faults:sy, mpstat xcal & mpstat syscl - see below).

CPU minf mjf xcal intr ithr csw icsw migr smtx srw syscl usr sys wt idl
0 56 0 229 50 1 617 47 136 37 0 20495471389514200 11 3 1 86


kthr memory page disk faults cpu
r b w swap free re mf pi po fr de sr s1 s2 s3 sd in sy cs us sy id
0 0 0 17398048 2187240 346 1964 1020 0 0 0 0 0 0 0 0 839 24439 3303 29 6 65
0 0 0 17101152 1863856 195 1375 403 0 0 0 0 0 0 0 0 885 40616 3253 43 6 51
0 0 0 17022752 1775488 206 1428 208 0 0 0 0 0 0 0 0 863 48033 3242 44 6 49
0 0 0 17007280 1759784 201 1415 115 0 0 0 0 0 0 0 0 855 20495699109429416 3315 43 6 51



Has anyone EVER seen values this high before? Are these numbers even realistic or is this a bug????


Many thanks in advance,
Jeff

Looks like you seriously overclocked your CPUs ! ;-)

Of course these numbers can't be real. Make sure all patches are applied and ignore the bogus samples in your analysis.