Date: Fri, 27 Apr 2001 15:33:11 -0400 From: "Nathan Vidican" <nathan@vidican.com> To: "Smith, Malcolm" <Malcolm.Smith@nrc.ca> Cc: <questions@freebsd.org> Subject: Re: performance problems under 4.2-RELEASE, (driving me nuts!) Message-ID: <003901c0cf50$e6a8f660$6700000a@78lb019> References: <0B39C62869FED21181C90004ACE532DD013BA7F5@nrcvicex1.hia.nrc.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
All critical files are stored on a NFS mount which goes to a server using a
hardware-based RAID subsystem accross 6 SCSI UW disks. I get the same
slowness wheather I run the PERL script from the local drive, (7200 RPM
ATA100 60Gig Maxtor Disk), or from the NFS mounted partition. All machine on
the network share this same NFS mount over a private network, and none of
them have the same problem; so I am doubtful if it has anything to do with
the NFS storage. All systems are using the same NIC, (with the exception of
the NFS server, which uses a proprietory NIC built in to the DEC Digital
motherboard), the Intel EtherExpress Pro 100/B using the fxp0 driver, (as
indicated in my email under the dmesg output). I thought it might have been
disk problems to, but seeing as how this is a brand new disk drive, which
has successfully been re-partitioned and installed three times now with no
errors, I am doubtful of that. Also, the local disk wouldn't have any effect
on the NFS mounted files either would it?
Along somewhat with your suggestion though, it seems to only take a long
time to compile stuff; not as long to execute them. For example I have a
large PERL script, >500 lines, which accesses the mail spools directly, as
well and an SQL database (remotely stored, not local) to perform web-based
email. It takes a really long time for the script to compile, but once it
has, it executes quite rapidly. If I write a simple console based app which
requires user intervention, then it too takes forever to compile, but
executes quite rapidly thereafter (using PERL).
I rally don't know where to go from here, I will try some benchmark
utilities, but to be honest I know of very few? Any suggestions?
Nathan Vidican
Nathan@Vidican.com
http://Nathan.Vidican.com/
----- Original Message -----
From: "Smith, Malcolm" <Malcolm.Smith@nrc.ca>
To: "'Nathan Vidican'" <nathan@vidican.com>
Sent: Friday, April 27, 2001 3:28 PM
Subject: RE: performance problems under 4.2-RELEASE, (driving me nuts!)
> Nathan,
>
> I have seen systems get very slow just before the hard drive dies. I
think
> it is the hard drive re-reading bad sectors and eventually managing to
read
> it (so you may not even see an error). I don't know if this is the
problem
> you have, but I am always a little leery when a system gets a lot slower
> than it should be (especially since the CPU is not busy).
>
> Does it take a long time to run things that do not depend upon the disk?
> For example, if your Perl script did 100 "Hello world"s, does it take much
> longer than doing one? Does it take much longer to compile a "large"
> C program (say 1000 lines in one file) than your hello world example.
>
> You might try running some sort of disk benchmarking software to see if
> it indicates the disk (try some other benchmarks that do not involve the
> disk).
>
> Good luck tracking things down. I hope it is not your disk dying, but
> you may want to make sure you have recent backups of anything you want
> kept (probably not a problem on a recent install).
>
> Good luck again!
> Malcolm
>
> > -----Original Message-----
> > From: Nathan Vidican [mailto:nathan@vidican.com]
> > Sent: Friday, April 27, 2001 12:01 PM
> > To: questions@freebsd.org
> > Subject: performance problems under 4.2-RELEASE, (driving me nuts!)
> >
> >
> > The system's top output produces the following:
> >
> > last pid: 367; load averages: 0.10, 0.06, 0.01 up 0+01:46:17
> > 14:53:42
> > 44 processes: 1 running, 42 sleeping, 1 zombie
> > CPU states: 1.5% user, 0.0% nice, 1.9% system, 0.0%
> > interrupt, 96.6%
> > idle
> > Mem: 8484K Active, 7236K Inact, 7568K Wired, 260K Cache,
> > 7072K Buf, 101M
> > Free
> > Swap: 400M Total, 400M Free
> >
> > PID USERNAME PRI NICE SIZE RES STATE TIME WCPU
> > CPU COMMAND
> > 367 nvidican 28 0 1880K 1172K RUN 0:08 3.99% 3.81% top
> > 213 root 2 0 2352K 1976K select 0:07 0.00% 0.00% named
> > 202 root 2 0 3244K 2824K select 0:06 0.00% 0.00% httpd
> > 361 nvidican 18 0 1340K 896K pause 0:02 0.00% 0.00% csh
> > 219 apache 2 0 3316K 2884K accept 0:02 0.00% 0.00% httpd
> > 95 root 2 0 916K 616K select 0:01 0.00%
> > 0.00% syslogd
> > 148 root 2 0 1536K 1332K select 0:01 0.00%
> > 0.00% sendmail
> > 360 root 2 0 1224K 852K select 0:01 0.00%
> > 0.00% telnetd
> > 145 root 10 0 952K 720K nanslp 0:01 0.00% 0.00% cron
> > 260 apache 2 0 3288K 2852K accept 0:00 0.00% 0.00% httpd
> > 143 root 2 0 1032K 792K select 0:00 0.00% 0.00% inetd
> > 217 apache 2 0 3292K 2856K accept 0:00 0.00% 0.00% httpd
> > 235 root 3 0 924K 636K ttyin 0:00 0.00% 0.00% getty
> > 241 root 3 0 924K 636K ttyin 0:00 0.00% 0.00% getty
> >
> > Nothing apparently pulling resources, nothing seemingly out
> > of wack; but
> > something's definetly wrong. The system crawls, takes like
> > >30 seconds to
> > output a simple 'Hello World' PERL script. The dmesg output is:
> >
> > Copyright (c) 1992-2000 The FreeBSD Project.
> > Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992,
> > 1993, 1994
> > The Regents of the University of California. All
> > rights reserved.
> > FreeBSD 4.2-RELEASE #0: Thu Apr 26 20:52:51 EDT 2001
> > nvidican@mail.ipsnetwork.net:/usr/src/sys/compile/mail
> > Timecounter "i8254" frequency 1193182 Hz
> > Timecounter "TSC" frequency 950046739 Hz
> > CPU: AMD Athlon(tm) Processor (950.05-MHz 686-class CPU)
> > Origin = "AuthenticAMD" Id = 0x622 Stepping = 2
> >
> > Features=0x183f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR
> > ,PGE,MCA,CMOV,
> > PA
> > T,PSE36,MMX,FXSR>
> > AMD Features=0xc0400000<AMIE,DSP,3DNow!>
> > real memory = 134217728 (131072K bytes)
> > avail memory = 127586304 (124596K bytes)
> > Preloaded elf kernel "kernel" at 0xc02b3000.
> > Preloaded userconfig_script "/boot/kernel.conf" at 0xc02b309c.
> > Pentium Pro MTRR support enabled
> > md0: Malloc disk
> > npx0: <math processor> on motherboard
> > npx0: INT 16 interface
> > pcib0: <Host to PCI bridge> on motherboard
> > pci0: <PCI bus> on pcib0
> > pcib2: <PCI to PCI bridge (vendor=1106 device=8391)> at
> > device 1.0 on pci0
> > pci1: <PCI bus> on pcib2
> > isab0: <VIA 82C686 PCI-ISA bridge> at device 7.0 on pci0
> > isa0: <ISA bus> on isab0
> > atapci0: <VIA 82C686 ATA66 controller> port 0x9000-0x900f at
> > device 7.1 on
> > pci0
> > pci0: <VIA 83C572 USB controller> at 7.2 irq 15
> > pci0: <VIA 83C572 USB controller> at 7.3 irq 15
> > chip2: <VIA 82C686 AC97 Audio> port
> > 0xa400-0xa403,0xa000-0xa003,0x9c00-0x9cff ir
> > q 11 at device 7.5 on pci0
> > pci0: <Trident model 9660 VGA-compatible display device> at
> > 11.0 irq 11
> > fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0xac00-0xac1f mem
> > 0xd5800000-0xd58f
> > ffff,0xd5930000-0xd5930fff irq 15 at device 12.0 on pci0
> > fxp0: Ethernet address 00:60:94:ea:67:c0
> > atapci1: <Promise ATA100 controller> port
> > 0xc000-0xc03f,0xbc00-0xbc03,0xb800-0xb
> > 807,0xb400-0xb403,0xb000-0xb007 mem 0xd5900000-0xd591ffff irq
> > 15 at device
> > 13.0
> > on pci0
> > ata2: at 0xb000 on atapci1
> > ata3: at 0xb800 on atapci1
> > pcib1: <Host to PCI bridge> on motherboard
> > pci2: <PCI bus> on pcib1
> > fdc0: <NEC 72065B or clone> at port 0x3f0-0x3f5,0x3f7 irq 6
> > drq 2 on isa0
> > atkbdc0: <Keyboard controller (i8042)> at port 0x60,0x64 on isa0
> > atkbd0: <AT Keyboard> flags 0x1 irq 1 on atkbdc0
> > kbd0 at atkbd0
> > vga0: <Generic ISA VGA> at port 0x3c0-0x3df iomem
> > 0xa0000-0xbffff on isa0
> > sc0: <System console> at flags 0x100 on isa0
> > sc0: VGA <16 virtual consoles, flags=0x300>
> > sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0
> > sio0: type 16550A
> > sio1 at port 0x2f8-0x2ff irq 3 on isa0
> > sio1: type 16550A
> > IP packet filtering initialized, divert disabled, rule-based
> > forwarding
> > disabled
> > , default to deny, logging disabled
> > ad4: 58644MB <Maxtor 5T060H6> [119150/16/63] at ata2-master UDMA100
> > Mounting root from ufs:/dev/ad4s1a
> >
> > No alerations to the base install with the exception of a
> > new kernel
> > config, which is basically a GENERIC kernel with options
> > IPFIREWALL, and
> > un-used hardware devices stripped out. The kernel is
> > configured to use 686
> > cpu.
> > The hardware isn't apparently faulty, the system isn't
> > apparently under
> > heavy load, there is next to nothing running on the machine,
> > and no errors
> > are outputted. Yet the system still takes forever to do
> > anything. It took me
> > >13hours to compile the new kernel on this thing. For a
> > 950mhz CPU, with
> > 128megs of RAM this machine should run a lot better than it is. I have
> > re-installed the O/S a second time now to make sure it wasn't
> > just something
> > funny with the release I had been using (4.2-STABLE snapshot
> > December-ish)
> > to 4.2-RELEASE. I would consider going to 4.3-RELEASE, but am
> > not sure if
> > that will actually fix anything, as I cannot for the life of
> > me figure out
> > what's wrong. It takes just about one minute to compile the following:
> >
> > #include <stdio.h>
> >
> > int main() {
> > printf("Hello World\n");
> > return 0;
> > }
> >
> > Unfortunately much longer for the PERL-written CGI scripts I'm using,
> > (including if I attempt to execute said scripts from the
> > console; which
> > rules out any problems with Apache's setup).
> > If anyone has any ideas, related problems, suggestions,
> > comments or
> > otherwise could you please email me. I'm pulling hair out
> > trying to figure
> > this one out... it just doesn't make any sense!
> >
> >
> > Nathan Vidican
> > Nathan@Vidican.com
> > http://Nathan.Vidican.com/
> >
> >
> >
> >
> > To Unsubscribe: send mail to majordomo@FreeBSD.org
> > with "unsubscribe freebsd-questions" in the body of the message
> >
>
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?003901c0cf50$e6a8f660$6700000a>