Date: Thu, 12 Aug 1999 17:19:08 -0700 From: Lamont Lucas <lamont@segasoft.com> To: freebsd-questions@freebsd.org Subject: 3.2-R performance questions on possibly broken hardware. Message-ID: <199908130019.RAA26068@adventure.segasoft.com>
next in thread | raw e-mail | index | archive | help
Hi. I just bought some new hardware, installed it, and have installed 3.2-R
on it. I'm noticing some amazingly poor performance on it, and wanted to get
some information as to how to narrow down what the problem may be. (make
world has, thusfar, taken 2 days, before I interrupted it, make search in
/usr/ports takes about 16 seconds, etc)
I'm trying to narrow down if it's the processor or the disks. This is a
dual-cpu board, running the GENERIC 3.2-R kernel, and thus using only one
processor, A 333 Celeron using some very cheap motherboard (who's name I don't
recall, but can find on the box at home). Disks are ultra wide scsi 2 7200
ibm drives, running on the TekRam(?) u2w controller, comes up as <ncr 53c875
fast20 wide scsi>) 128 megs of ram, 256 megs of swap.
When I get home tonight, I'm going to experiment with removing one of the
cpus, to see if that helps. I'll also look into the scsi bios setup to see if
I can spot any obvious errors (roughly akin to my method of fixing cars, open
the hood and look for a part waving a big flag with the word help on it)
I have a fleet of HP LPR dual cpu PIII/500's at work, and they're doing
fantastically, even before I recompiled SMP. (mmmm, web farm...)
Intrestingly, when the machine is almost totally idle, here's the output of
iostat -oI 10:
bash-2.03# iostat -oI 10
tty da0 da1 cd0 fd0
cpu
tin tout blk xfr msps blk xfr msps blk xfr msps blk xfr msps us ni sy in
id
0 26 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 77 0 13 2
8
2 2031384 85 118 0 0 0.0 0 0 0.0 0 0 0.0 10 0 8 3
79
0 8 16 110030 0 0 0.0 0 0 0.0 0 0 0.0 0 0 5 2
93
0 8 48 3 3343 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0 2
97
0 10 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 0 0 1 1
98
0 13 42 4 2508 0 0 0.0 0 0 0.0 0 0 0.0 1 0 7 2
90
0 169 80 5 2006 0 0 0.0 0 0 0.0 0 0 0.0 9 0 21 3
67
0 117 32 2 5015 0 0 0.0 0 0 0.0 0 0 0.0 2 0 1 2
95
0 8 80 5 2006 0 0 0.0 0 0 0.0 0 0 0.0 0 0 5 1
93
0 8 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0 2
97
0 8 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0 2
97
0 8 16 110030 0 0 0.0 0 0 0.0 0 0 0.0 0 0 5 1
93
0 8 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0 2
98
0 8 0 0 0.0 0 0 0.0 0 0 0.0 0 0 0.0 3 0 4 3
91
0 8 32 2 5015 0 0 0.0 0 0 0.0 0 0 0.0 0 0 5 3
92
whereas, when I do a large find or some other semi-intensive disk-oriented
activity, the numbers look more like:
0 8 74 9 1114 0 0 0.0 0 0 0.0 0 0 0.0 4 0 4 2
90
0 21 758 358 28.0 0 0 0.0 0 0 0.0 0 0 0.0 21 0 41 5
33
0 83546 423 23.7 0 0 0.0 0 0 0.0 0 0 0.0 39 0 29 6
27
0 8 784 368 27.3 0 0 0.0 0 0 0.0 0 0 0.0 26 0 36 6
32
0 84380 510 19.7 0 0 0.0 0 0 0.0 0 0 0.0 27 0 34 7
32
0 83452 509 19.7 0 0 0.0 0 0 0.0 0 0 0.0 23 0 36 7
34
0 81006 392 25.7 0 0 0.0 0 0 0.0 0 0 0.0 19 0 37 4
39
0 83486 494 20.3 0 0 0.0 0 0 0.0 0 0 0.0 14 0 42 7
37
0 81182 420 23.9 0 0 0.0 0 0 0.0 0 0 0.0 14 0 40 5
41
0 83772 555 18.1 0 0 0.0 0 0 0.0 0 0 0.0 11 0 41 8
40
0 82698 516 19.4 0 0 0.0 0 0 0.0 0 0 0.0 10 0 44 7
38
0 251236 429 23.4 0 0 0.0 0 0 0.0 0 0 0.0 11 0 46 5
38
0 83272 530 18.9 0 0 0.0 0 0 0.0 0 0 0.0 11 0 42 8
39
I'm a reasonably accomplished solaris and irix syadmin, but I'm still fairly
new with FreeBSD.
Can anyone suggest other things to check or look at? Any other information
that I can provide that may help?
--
Lamont Lucas, Sysadmin, lamont@segasoft.com
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199908130019.RAA26068>