Is FreeBSD performance that bad ?

Is FreeBSD performance that bad ?

Post by FreeBSD use » Wed, 18 Aug 1999 04:00:00



Hello,

Does anyone have compared the performance of Apache web server on a FreeBSD
box compared to Linux or any other platform? Our results are pretty *,
and I guess something is wrong with our configuration, so I'm calling from
help, where to look for.

The case is the following: a friend of mine downloaded a tool from the net
to test the performance of one of our production web servers (apache 1.3.6 +
fastCGI + SSL on redHat Linux 6.0 kernel 2.2.5-22) This tool (don't know the
name) first creates a directory big tree (somethign like dir-hir) full of
10K html pages, then starts requesting the pages randomly simulating several
concurrent users. After running the test for 5 minutes simulating 50 users,
the results were the following:

apache 1.3.6 +SSL+FastCGI + redHat 6.0 : 398 req/sec
 Hardware:
    Pentium II 350 MHz
    128 MB RAM
    10.6 GB IDE HD
    100 MB/S DLINK Ethernet half duplex
Users Logged in:  0

Windows NT server 4.0 + IIS 4.0:  200 req/sec
Hardware:
    Pentium 266 MMX
    128 MB RAM
    ??? GB Hard Drive
Users Logged in: 0

!!!!! CHECK THIS OUT !!!!

apache 1.3.6 + FreeBSD 3.2 Release:   18 req/sec
Hardware
    Pentium II 450 MHz
    128 MB HD
    10.6 GB IDE HardDrive
    100 MB/S 3COM Ethernet half duplex
Users Logged in: 0 (just after the computer was rebooted)

All of the computers were connected to the same 100MB/S switch

The httpd.conf was identical on both linux and FreeBSD machines are were
taken from www.mindcraft.com (the one they used for the linux-NT contest.

I DON'T BELIEVE FreeBSD is that bad, so I think that something has to be
wrong with our configuration. we;'re not running NFS, DNS, PPP, telnet. The
only services running on FreeBSD were INND, SSHD anonymous ftp and WWW. WWW
is running in standalone mode as a daemon, not inetd. All those services
were inactive at the time of the test. No users were logged in.

Apache 1.3.6 was installed from the prots collection and configured
following the notes on   "The Complete FREEBSD 3.2" and the apache
foundation documents.

Can anybody tell me where to look for problems? How can I fine tune the
performance of the freeBSD box? Actually I started getting into the FreeBSD
world to replace our linux boxes.

DISCLAIMER: I'm not an experienced or fulltime sysadmin, if this message
seems too stupid to you ,
please send replies  > /dev/null. To all others who decide to help. Thank
you very much.


Sometimes my news server is not up-to-date. Thanks.

 
 
 

Is FreeBSD performance that bad ?

Post by FreeBSD use » Wed, 18 Aug 1999 04:00:00



also after posting the message i realized the results of the test were not
clear enough, here they go:

under a simulated load of 50 users during 5 minutes, the results were:

WebServer/OS                            Request per second
===============                    =====================
apache 1.3.6/Linux                        398 request/sec
IIS 4.0/ NT server 4                       200 requests/sec
apache 1.3.6/FreeBSD-3.2            18 req/sec

we ran the tests several time and the results were pretty consistents. The
topology follows:

All boxes in the same physical network, same subnet, all of them connected
to a 100 MB/s switch

If someone needs more information feel free to ask.

Thanks

Luis A. Burgos


> Hello,

> Does anyone have compared the performance of Apache web server on a
FreeBSD
> box compared to Linux or any other platform? Our results are pretty *,
> and I guess something is wrong with our configuration, so I'm calling from
> help, where to look for.

> The case is the following: a friend of mine downloaded a tool from the net
> to test the performance of one of our production web servers (apache 1.3.6
+
> fastCGI + SSL on redHat Linux 6.0 kernel 2.2.5-22) This tool (don't know
the
> name) first creates a directory big tree (somethign like dir-hir) full of
> 10K html pages, then starts requesting the pages randomly simulating
several
> concurrent users. After running the test for 5 minutes simulating 50
users,
> the results were the following:

> apache 1.3.6 +SSL+FastCGI + redHat 6.0 : 398 req/sec
>  Hardware:
>     Pentium II 350 MHz
>     128 MB RAM
>     10.6 GB IDE HD
>     100 MB/S DLINK Ethernet half duplex
> Users Logged in:  0

> Windows NT server 4.0 + IIS 4.0:  200 req/sec
> Hardware:
>     Pentium 266 MMX
>     128 MB RAM
>     ??? GB Hard Drive
> Users Logged in: 0

> !!!!! CHECK THIS OUT !!!!

> apache 1.3.6 + FreeBSD 3.2 Release:   18 req/sec
> Hardware
>     Pentium II 450 MHz
>     128 MB HD
>     10.6 GB IDE HardDrive
>     100 MB/S 3COM Ethernet half duplex
> Users Logged in: 0 (just after the computer was rebooted)

> All of the computers were connected to the same 100MB/S switch

> The httpd.conf was identical on both linux and FreeBSD machines are were
> taken from www.mindcraft.com (the one they used for the linux-NT contest.

> I DON'T BELIEVE FreeBSD is that bad, so I think that something has to be
> wrong with our configuration. we;'re not running NFS, DNS, PPP, telnet.
The
> only services running on FreeBSD were INND, SSHD anonymous ftp and WWW.
WWW
> is running in standalone mode as a daemon, not inetd. All those services
> were inactive at the time of the test. No users were logged in.

> Apache 1.3.6 was installed from the prots collection and configured
> following the notes on   "The Complete FREEBSD 3.2" and the apache
> foundation documents.

> Can anybody tell me where to look for problems? How can I fine tune the
> performance of the freeBSD box? Actually I started getting into the
FreeBSD
> world to replace our linux boxes.

> DISCLAIMER: I'm not an experienced or fulltime sysadmin, if this message
> seems too stupid to you ,
> please send replies  > /dev/null. To all others who decide to help. Thank
> you very much.


> Sometimes my news server is not up-to-date. Thanks.


 
 
 

Is FreeBSD performance that bad ?

Post by Yann Rami » Wed, 18 Aug 1999 04:00:00


Well, I can tell you my results (which are localhost, which is not a
real comparison, etc etc) showed static keep-alive HTML serves on my
FreeBSD3.2-R box (350 AMD k6-2, 64MB of RAM, running X, KDE and the
works) at about 720 requests a second.  I can't test this network wise
since an ISA card over BNC cable to a 486DX4 running W95 is not a good
test situation.
--

--------------------------------------------------------------------

Atrus Trivalie Productions      www.redshift.com/~yramin
                                irm.it.montereyhigh.com
Monterey High IT                www.montereyhigh.com
ICQ                             46805627
Marina, CA      
--------------------------------------------------------------------

 
 
 

Is FreeBSD performance that bad ?

Post by Stephen Wait » Wed, 18 Aug 1999 04:00:00



> WebServer/OS                            Request per second
> ===============                    =====================
> apache 1.3.6/Linux                        398 request/sec
> IIS 4.0/ NT server 4                       200 requests/sec
> apache 1.3.6/FreeBSD-3.2            18 req/sec

[SNIP]

I know that your problem is not "FreeBSD" in general.  CDRom.com, often
touted (rightfully so I believe) as "the busiest ftp server in the
world", is based on FreeBSD.  

I've been using FreeBSD since 1.0, stuck with it through the whole
lawsuit *(way to * up an OS US legal system -- bah!), and use it
to this day.  In ALL of my experience, I've NEVER seen FreeBSD with such
drastically poor performance as compared to Win32 or Linux, or Solaris
(slowlaris?).

I'm interested in helping you figure this out -- unless you are
trolling...

My $0.02..

Good Luck,
Steve

 
 
 

Is FreeBSD performance that bad ?

Post by David Schwart » Wed, 18 Aug 1999 04:00:00


Quote:

> What's your dmesg for the FBSD 3.2 box?  I'm willing to bet that it's an
> issue with the NIC and the NWay negotiation of the switch. I have yet to
> see a switch that functions only at 100BT-HD.  The cards should be
> 100BT-FD.  Did you also check the usual hardware resources?

        I got burned by this myself. I wonder exactly what NIC cards he was
using. And why are they half-duplex to a switch? I don't know of many
100Mbps switches that are half duplex.

        DS

 
 
 

Is FreeBSD performance that bad ?

Post by Donn Mille » Wed, 18 Aug 1999 04:00:00



> Hello,

> Does anyone have compared the performance of Apache web server on a FreeBSD
> box compared to Linux or any other platform? Our results are pretty *,
> and I guess something is wrong with our configuration, so I'm calling from
> help, where to look for.

-snip-

Quote:> apache 1.3.6 + FreeBSD 3.2 Release:   18 req/sec
> Hardware
>     Pentium II 450 MHz
>     128 MB HD
>     10.6 GB IDE HardDrive

        ^^^^^^^^^^^^^

Here's the problem:  is this an UDMA drive?

Quote:> Can anybody tell me where to look for problems? How can I fine tune the
> performance of the freeBSD box? Actually I started getting into the FreeBSD
> world to replace our linux boxes.

Linux enables optimized DMA drive access by default, whereas
FreeBSD mounts all filesystems sync by default.  What you want to
do is recompile your kernel, with the flags 0xa0ffa0ff to the IDE
controller, like so:

# IDE controller and disks
controller      wdc0    at isa? port IO_WD1 irq 14 flags 0xa0ffa0ff
disk            wd0     at wdc0 drive 0
disk            wd1     at wdc0 drive 1

controller      wdc1    at isa? port IO_WD2 irq 15 flags 0xa0ffa0ff
disk            wd2     at wdc1 drive 0
disk            wd3     at wdc1 drive 1

See how I added the flags at the end of the line?  Now, edit your
/etc/fstab so that your filesystems are mounted async.

Do this, and recompile your kernel.  Let us know how it goes.
You know how to configre and compile the kernel, right?

--
  Donn

 
 
 

Is FreeBSD performance that bad ?

Post by Brad Knowl » Thu, 19 Aug 1999 04:00:00




> under a simulated load of 50 users during 5 minutes, the results were:

> WebServer/OS                            Request per second
> ===============                    =====================
> apache 1.3.6/Linux                        398 request/sec
> IIS 4.0/ NT server 4                       200 requests/sec
> apache 1.3.6/FreeBSD-3.2            18 req/sec

    A previous response has noted that Linux optimizes IDE hard drive
performance by default, but FreeBSD does not.  I'd like to expand on this.

    In general, Linux makes certain optimistic assumptions about
hardware.  FreeBSD tends to be more conservative in these regards, and is
pessimistic.

    Now, if your hardware meets the optimistic assumptions that Linux
makes, and nothing ever goes wrong, then Linux will probably perform
better than FreeBSD, unless you take care to configure your system so that
your installation of FreeBSD makes the same optimistic assumptions.

    A better way to solve this problem is to find out which assumptions
you can relatively safely make for your hardware, and implement those
assumptions both under Linux and FreeBSD.  Then compare the performance.

    If you do this, I believe you will find that FreeBSD will perform
roughly equal to Linux, if not better in virtually all cases.  In
particular, the Linux network code right now is not very good, and the
FreeBSD stack is much better.  Linux himself recognizes this problem, and
it's one of the many, many things that are planned to be fixed for kernel
version 2.4, but Ghu-only-knows how many years away that might be.

    Now, let's take this a step further.  By default, Linux mounts
filesystems asynchronously.  This increases the chances that your system
will not be able to repair the filesystem, should there ever be an
unexpected system crash.  The busier your filesystem that's mounted async,
the greater the chance that it will be permanently screwed up in a crash
and that you'll have to wipe it clean and restore your latest version from
tape.

    FreeBSD pessimistically assumes that you're willing to put up with
lower system performance in return for greater system reliability in the
face of a crash.

    However, if you look at the LINT kernel, you'll note at the bottom is
a feature called "SOFTUPDATES".  If you follow the instructions and enable
this feature on your busy filesystems, they should perform every bit as
fast as Linux does when mounting the filesystem async, but there should be
no more danger in them being permanently screwed up by a crash than if you
mounted them sync.  This way, you really can have your cake and eat it
too.

    Of course, you don't want to run something unless you actually need
it, so while you'll want to build your kernel and enable softupdates,
you'll actually only want to turn on softupdates for those busy
filesystems where it will be of use, primarily ones where lots of
temporary files are being created and then deleted very soon thereafter.

    So, softupdates would be good for a separate /var/tmp filesystem (you
might want to run /tmp as an mfs filesystem, in the same way that Solaris
allows), or for a sendmail /var/spool/mqueue, or for an INN
/var/spool/news.  Using softupdates may or may not buy you anything on a
busy /usr/local/share/apache/htdocs filesystem, depending on the nature of
the web pages you're serving up and the other filesystem activity.

    I'm not sure how long you've been in the computer industry, but you
may remember many years ago that a company called Borland had a compiler
called Turbo Pascal.  The reason it sold so many copies was because it
appeared to be blindingly fast, and was relatively inexpensive.  However,
it was "blindingly fast" because they turned off all checking by default.

    If you took that same kind of behaviour and implemented it with the
other major Pascal compilers at the time, they would run just as fast (or
faster) than Turbo Pascal.  The difference was that they had virtually all
checking turned on by default, so that if there was a problem with your
program, you'd find out about it sooner rather than later.

    These days, Linux is kind of like Turbo Pascal.  They run fast, but
they run with most safety checks turned off by default.  If you never fall
off the high wire, then you didn't need any of those safety checks to
begin with.  However, if you do fall, then not having them means it's
*really* going to hurt when you actually fall and go *SPLAT*.

    FreeBSD turns on all of the safety devices by default, and although
you'll run slower out-of-the-box, you'll really appreciate having all
those safety devices if/when you ever do fall off the high wire.

    Of course, people really only care about this problem once they've
actually fallen.  And if you've grown up in a more conservative OS
environment (e.g., FreeBSD), then you may not really appreciate all those
safety checks until such time as you're operating somewhere without them,
and you have a *really* rude surprise when there's nothing there to catch
you.

--

    <http://wwwkeys.pgp.net:11371/pks/lookup?op=get&search=0xE38CCEF1>

 
 
 

Is FreeBSD performance that bad ?

Post by Brad Knowl » Thu, 19 Aug 1999 04:00:00




>     If you do this, I believe you will find that FreeBSD will perform
> roughly equal to Linux, if not better in virtually all cases.  In
> particular, the Linux network code right now is not very good, and the
> FreeBSD stack is much better.  Linux himself recognizes this problem, and

    s/Linux himself/Linus himself/

    Sigh...

--

    <http://wwwkeys.pgp.net:11371/pks/lookup?op=get&search=0xE38CCEF1>

 
 
 

Is FreeBSD performance that bad ?

Post by Tom Jobbin » Thu, 19 Aug 1999 04:00:00


Quote:> See how I added the flags at the end of the line?  Now, edit your
> /etc/fstab so that your filesystems are mounted async.

This advice is a bit dangerous I would have thought.   The reason FreeBSD
doesnt mount async by default is that, should the filesystem crash, you
might find yourself needing to recreate your entire filesystem.  That extra
speed comes at a cost.

A safe solution would be to use the new softupdates kernel option, then
enable softupdates on all partitions except / .  This will give most of the
speed benefits of mounting async, without the associated risk.

 
 
 

Is FreeBSD performance that bad ?

Post by Nick Hilliar » Thu, 19 Aug 1999 04:00:00



> apache 1.3.6 + FreeBSD 3.2 Release:   18 req/sec
> Hardware
>     Pentium II 450 MHz
>     128 MB HD
>     10.6 GB IDE HardDrive
>     100 MB/S 3COM Ethernet half duplex

Are both the switch and the ethernet card hard-wired to using half-duplex?   You
can check the ethernet card by typing "ifconfig -a" and checking the output.  If
there's a duplex mismatch, you'll get lousy results.

Nick

 
 
 

Is FreeBSD performance that bad ?

Post by Luis A. Burgo » Thu, 19 Aug 1999 04:00:00


Okay, it's me again, (Luis A. Burgos, who post the original message)

I've read all the messages, I appreciate all of them.  Thaks to all of you
you decided to help.

I've seen many people requesting more information. Here it goes:

- 100 MB/s switch is hardwired (by our "sysadmin") to half duplex. All ports
on that switch are set to 10MBs except the one that goes to his computer
(surprise?!)

A lot of people have told me about these SOFTUPDATES. What's that? The
complete FreeBSD Book doesn't talk about it. Can anybody tell me how do I
configure my system to add these softupdates? Or at least, where can I learn
more about it?

This is the dmesg from the FreeBSD 3.2-RELEASE that many people had
requetsed:

--------------------------  start of dmesg ---------------------
Copyright (c) 1992-1999 FreeBSD Inc.
Copyright (c) 1982, 1986, 1989, 1991, 1993
 The Regents of the University of California. All rights reserved.
FreeBSD 3.2-RELEASE #3: Mon Aug 16 16:01:34 EDT 1999
    r...@mydomain.com:/usr/src/sys/compile/SIMBA
Timecounter "i8254"  frequency 1193182 Hz
Timecounter "TSC"  frequency 447235148 Hz
CPU: Pentium II/Xeon/Celeron (447.24-MHz 686-class CPU)
  Origin = "GenuineIntel"  Id = 0x652  Stepping=2

Features=0x183f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,
PAT,PSE36,MMX,<b24>>
real memory  = 134217728 (131072K bytes)
avail memory = 128217088 (125212K bytes)
Preloaded elf kernel "kernel" at 0xc0240000.
Probing for devices on PCI bus 0:
chip0: <Intel 82443BX host to PCI bridge> rev 0x02 on pci0.0.0
chip1: <Intel 82443BX host to AGP bridge> rev 0x02 on pci0.1.0
chip2: <Intel 82371AB PCI to ISA bridge> rev 0x02 on pci0.7.0
ide_pci0: <Intel PIIX4 Bus-master IDE controller> rev 0x01 on pci0.7.1
chip3: <Intel 82371AB Power management controller> rev 0x02 on pci0.7.3
xl0: <3Com 3c905-TX Fast Etherlink XL> rev 0x00 int a irq 10 on pci0.9.0
xl0: Ethernet address: 00:60:08:a3:39:69
xl0: autoneg complete, link status good (half-duplex, 100Mbps)
Probing for devices on PCI bus 1:
vga0: <Number Nine model 493d graphics accelerator> rev 0x00 int a irq 10 on
pci1.0.0
Probing for PnP devices:
Probing for devices on the ISA bus:
sc0 on isa
sc0: VGA color <16 virtual consoles, flags=0x0>
atkbdc0 at 0x60-0x6f on motherboard
atkbd0 irq 1 on isa
psm0 irq 12 on isa
psm0: model MouseMan+, device ID 0
sio0 at 0x3f8-0x3ff irq 4 flags 0x10 on isa
sio0: type 16550A
sio1: configured irq 3 not in bitmap of probed irqs 0
sio1 not found at 0x2f8
fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa
fdc0: FIFO enabled, 8 bytes threshold
fd0: 1.44MB 3.5in
wdc0 at 0x1f0-0x1f7 irq 14 on isa
wdc0: unit 0 (wd0): <WDC AC36400L>
wd0: 6149MB (12594960 sectors), 13328 cyls, 15 heads, 63 S/T, 512 B/S
wdc1 at 0x170-0x177 irq 15 on isa
wdc1: unit 0 (atapi): <  36X CD-ROM/VER 1.C0>, removable, accel, dma, iordy
acd0: drive speed 343 - 4125KB/sec, 128KB cache
acd0: supported read types: CD-R, CD-RW, CD-DA, packet track
acd0: Audio: play, 255 volume levels
acd0: Mechanism: ejectable tray
acd0: Medium: no/blank disc inside, unlocked
ppc0 at 0x378 irq 7 flags 0x40 on isa
ppc0: Generic chipset (NIBBLE-only) in COMPATIBLE mode
lpt0: <generic printer> on ppbus 0
lpt0: Interrupt-driven port
ppi0: <generic parallel i/o> on ppbus 0
plip0: <PLIP network interface> on ppbus 0
lpt0: <generic printer> on ppbus 0
lpt0: Interrupt-driven port
vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa
npx0 on motherboard
npx0: INT 16 interface
changing root device to wd0s1a

-------------------------- end of dmesg -----------------------

The kernel config file follows
--------------------------start of kernel config ------------------

#
# GENERIC -- Generic machine with WD/AHx/NCR/BTx family disks
#
# For more information read the handbook part System Administration ->
# Configuring the FreeBSD Kernel -> The Configuration File.
# The handbook is available in /usr/share/doc/handbook or online as
# latest version from the FreeBSD World Wide Web server
# <URL:http://www.FreeBSD.ORG/>
#
# An exhaustive list of options and more detailed explanations of the
# device lines is present in the ./LINT configuration file. If you are
# in doubt as to the purpose or necessity of a line, check first in LINT.
#
# $Id: GENERIC,v 1.143.2.12 1999/05/14 15:12:26 jkh Exp $

machine  "i386"
cpu  "I686_CPU"
ident  COMPUTERNAME
maxusers 32

options  INET   #InterNETworking
options  FFS   #Berkeley Fast Filesystem
options  FFS_ROOT  #FFS usable as root device [keep this!]
#options MFS   #Memory Filesystem
#options MFS_ROOT  #MFS usable as rootdevice, "MFS" req'ed
#options  NFS   #Network Filesystem
#options  NFS_ROOT  #NFS usable as root device, "NFS" req'ed
#options MSDOSFS   #MSDOS Filesystem
#options "CD9660"  #ISO 9660 Filesystem
#options "CD9660_ROOT" #CD-ROM usable as root."CD9660"req'ed
options  PROCFS   #Process filesystem
options  "COMPAT_43"  #Compatible with BSD 4.3 [KEEP THIS!]
#options  SCSI_DELAY=15000 #Be pessimistic about Joe SCSI device
#options  UCONSOLE  #Allow users to grab the console
options  FAILSAFE  #Be conservative
options  USERCONFIG  #boot -c editor
options  VISUAL_USERCONFIG #visual boot -c editor

config  kernel root on wd0

# To make an SMP kernel, the next two are needed
#options SMP   # Symmetric MultiProcessor Kernel
#options APIC_IO   # Symmetric (APIC) I/O
# Optionally these may need tweaked, (defaults shown):
#options NCPU=2   # number of CPUs
#options NBUS=4   # number of busses
#options NAPIC=1   # number of IO APICs
#options NINTR=24  # number of INTs

controller isa0
controller pnp0
controller eisa0
controller pci0

controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2
disk  fd0 at fdc0 drive 0
disk  fd1 at fdc0 drive 1

#options "CMD640" # work around CMD640 chip deficiency
controller wdc0 at isa? port "IO_WD1" bio irq 14
disk  wd0 at wdc0 drive 0
#disk  wd1 at wdc0 drive 1

controller wdc1 at isa? port "IO_WD2" bio irq 15
disk  wd2 at wdc1 drive 0
#disk  wd3 at wdc1 drive 1

options  ATAPI  #Enable ATAPI support for IDE bus
options  ATAPI_STATIC #Don't do it as an LKM
device  acd0  #IDE CD-ROM
#device  wfd0  #IDE Floppy (e.g. LS-120)

# A single entry for any of these controllers (ncr, ahb, ahc) is
# sufficient for any number of installed devices.
#controller ncr0
#controller ahb0
#controller ahc0
#controller isp0

# This controller offers a number of configuration options, too many to
# document here  - see the LINT file in this directory and look up the
# dpt0 entry there for much fuller documentation on this.
#controller      dpt0

#controller adv0 at isa? port ? cam irq ?
#controller adw0
#controller bt0 at isa? port ? cam irq ?
#controller aha0 at isa? port ? cam irq ?

#controller scbus0

#device  da0

#device  sa0

#device  pass0

#device  cd0 #Only need one of these, the code dynamically grows

#device  wt0 at isa? port 0x300 bio irq 5 drq 1
#device  mcd0 at isa? port 0x300 bio irq 10

#controller matcd0 at isa? port 0x230 bio

#device  scd0 at isa? port 0x230 bio

# atkbdc0 controlls both the keyboard and the PS/2 mouse
controller atkbdc0 at isa? port IO_KBD tty
device  atkbd0 at isa? tty irq 1
device  psm0 at isa? tty irq 12

device  vga0 at isa? port ? conflicts

# splash screen/screen saver
pseudo-device splash

# syscons is the default console driver, resembling an SCO console
device  sc0 at isa? tty
# Enable this and PCVT_FREEBSD for pcvt vt220 compatible console driver
#device  vt0 at isa? tty
#options  XSERVER   # support for X server
#options  FAT_CURSOR  # start with block cursor
# If you have a ThinkPAD, uncomment this along with the rest of the PCVT
lines
#options  PCVT_SCANSET=2  # IBM keyboards are non-std

device  npx0 at isa? port IO_NPX irq 13

#
# Laptop support (see LINT for more options)
#
#device  apm0    at isa? disable flags 0x31 # Advanced PowerManagement

# PCCARD (PCMCIA) support
#controller card0
#device  pcic0 at card?
#device  pcic1 at card?

device  sio0 at isa? port "IO_COM1" flags 0x10 tty irq 4
device  sio1 at isa? port "IO_COM2" tty irq 3
device  sio2 at isa? disable port "IO_COM3" tty irq 5
device  sio3 at isa? disable port "IO_COM4" tty irq 9

# Parallel port
device  ppc0 at isa? port? flags 0x40 net irq 7
controller ppbus0
device  lpt0 at ppbus?
device  plip0 at ppbus?
device  ppi0 at ppbus?
#controller vpo0 at ppbus?

#
# The following Ethernet NICs are all PCI devices.
#
#device ax0  # ASIX AX88140A
#device de0  # DEC/Intel DC21x4x (``Tulip'')
#device fxp0  # Intel EtherExpress PRO/100B (82557, 82558)
#device mx0  # Macronix 98713/98715/98725 (``PMAC'')
#device pn0  # Lite-On 82c168/82c169 (``PNIC'')
#device rl0  # RealTek 8129/8139
#device tl0  # Texas Instruments ThunderLAN
#device tx0  # SMC 9432TX (83c170 ``EPIC'')
#device vr0  # VIA Rhine, Rhine II
#device vx0  # 3Com 3c590, 3c595 (``Vortex'')
#device wb0  # Winbond W89C840F
device xl0  # 3Com 3c90x (``Boomerang'', ``Cyclone'')

# Order is important here due to intrusive probes, do *not* alphabetize
# this list of network interfaces until the probes have been fixed.
# Right now it appears that the ie0 must be probed before ep0. See
# revision 1.20 of this file.

#device ed0 at isa? port 0x280 net irq 10 iomem 0xd8000
#device ie0 at isa? port 0x300 net irq 10 iomem 0xd0000
#device ep0 at isa? port 0x300 net irq 10
#device ex0 at isa? port? net irq?
#device fe0 at isa? port 0x300 net irq ?
#device le0 at isa? port 0x300 net irq 5 iomem 0xd0000
#device lnc0 at isa? port 0x280 net irq 10 drq 0
#device ze0 at isa? port 0x300 net irq 10 iomem 0xd8000
#device zp0 at isa? port 0x300 net irq 10 iomem 0xd8000
#device cs0 at isa? port 0x300 net irq ?

pseudo-device loop
pseudo-device ether
pseudo-device sl 1
pseudo-device ppp 1
pseudo-device tun 1
pseudo-device pty 16
pseudo-device gzip  # Exec gzipped a.out's

# KTRACE enables the system-call tracing facility ktrace(2).
# This adds 4 KB bloat to your kernel, and slightly increases
# the costs of each syscall.
#options  KTRACE  #kernel tracing

# This provides support for System V shared memory and message ...

read more »

 
 
 

Is FreeBSD performance that bad ?

Post by Gardner Buchan » Thu, 19 Aug 1999 04:00:00




Quote:

> Does anyone have compared the performance of Apache web server on a FreeBSD

[snip]

Perhaps you have apache's hostnames in the logs option switched
on, combined with a bogus resolv.conf setup.

I shouldn't think that async mounting or softupdates would make a
big difference when the main work the FreeBSD box is reading the
disk.

Be sure to verify that the network bandwidth is sound by doing FTPs
back and forth to your test client.  If this doesn't check out,
then suspect your network connection - card, switch or cable.

Stick with it.

============================================================

Ottawa, ON             FreeBSD: Where you want to go. Today.