I have:
2 Ultra 2 170's 2 CPUs w/ 512MB RAM, both are Sybase dataservers, both have 3
SCSI controllers with various Boxhill drives. Swap is 512M, both have in
/etc/system:
*
* Increase shared memory capability for Sybase dataserver
*
set shmsys:shminfo_shmmin=1
set shmsys:shminfo_shmmax=536870912
set shmsys:shminfo_shmmni=256
set shmsys:shminfo_shmseg=100
The dataservers are configured to use 300MB of memory. Liveback is also in
use.
On the good system, which runs 2 dataservers, the CPUs are relatively busy,
paging is light, there is virtually no I/O waiting, and iostat -x 30 is all
fine.
On the not-so-good one, `vmstat 4` looks like:
r b w swap free re mf pi po fr de sr s1 s3 s1 s1 in sy cs us sy
id
0 0 0 726040 12208 0 1111 0 0 0 0 0 0 0 13 0 208 4160 213 8 13 79
0 0 0 724904 11000 0 317 6 14 14 0 0 0 0 2 0 169 1591 147 10 34 55
1 0 0 725152 11296 0 393 0 0 0 0 0 0 0 13 0 265 4826 196 28 46 26
0 0 0 726424 12504 0 567 0 0 0 0 0 0 0 3 0 140 1488 136 3 7 90
0 0 0 725672 11928 0 625 0 0 0 0 0 0 0 9 0 194 3029 165 6 10 84
0 0 0 725096 11104 0 644 0 0 0 0 0 0 0 5 0 166 3445 167 24 30 46
0 0 0 725680 11944 0 705 2 0 0 0 0 0 4 11 0 240 5474 306 7 10 83
0 0 0 725344 11320 0 175 0 0 0 0 0 0 0 5 0 175 2181 114 21 30 50
0 0 0 726480 12544 0 10 0 0 0 0 0 0 0 0 0 122 98 64 0 0 99
0 0 0 726480 12544 0 131 0 0 0 0 0 0 0 1 0 135 344 87 1 1 98
1 0 0 725040 11240 0 1098 0 0 0 0 0 0 0 7 0 208 5217 234 17 59 24
0 0 0 725696 11896 0 414 0 0 0 0 0 0 0 14 0 244 3081 154 6 18 76
0 0 0 725704 11784 0 163 0 0 0 0 0 0 0 4 0 153 1998 113 20 23 57
0 0 0 726136 12296 0 939 0 0 0 0 0 0 0 9 0 182 3111 195 7 9 84
0 0 0 725504 11680 0 278 0 0 0 0 0 0 0 4 0 166 1822 123 4 19 77
0 0 0 725296 11352 0 1020 0 0 0 0 0 0 0 12 0 209 4367 213 27 22 51
0 0 0 725584 11776 0 334 0 0 0 0 0 0 0 5 0 176 2089 142 4 16 79
0 0 0 725624 11608 0 284 0 0 0 0 0 0 0 4 0 161 2122 123 21 18 61
0 0 0 725664 11800 0 252 0 0 0 0 0 0 0 0 0 147 1404 130 8 29 63
and performance is sluggish.
My question is why are there so many minor faults considering there is little
or no paging?
We have swapped out the motherboard, and the disks will soon be upgraded.
Any and all ideas, comments, and questions are appreciated, and I will
summarize.
TIA,
Jon Anderson
System Admin
Goldman Sachs & Co.