Re: Security camera management software

[Jeff Noxon]

Have a look at Zoneminder (http://www.zoneminder.com/), a very nice
surveillance app for Linux.  Linux Media Labs makes 4 & 8-port analog
capture cards that work with this app.

On Mon, Aug 02, 2004 at 11:01:40AM -0700, Brian Nelson wrote:
> Paul Johnson <[EMAIL PROTECTED]> writes:
> 
> > Brian Nelson <[EMAIL PROTECTED]> writes:
> >
> >> A friend of mine recently asked me if any decent software exists for
> >> Linux for managing security cameras.  He intends to setup up 6 or 7
> >> cameras, and wants to be able to manage them from a MacOS X box.
> >
> > Seriously?  Or is this a toy project?  
> 
> It's for a home business of sorts (a 24-hour car wash located on their
> property but not visible from the house).  So yeah, it's serious, but
> not something he wants to dump a ton of money into.
> 
> > Having used security cameras as part of my job, I can safely say that
> > every digital camera system blows balls.  If you want usable footage,
> > go analog for this project.  Whoever has to look at the footage for
> > some minute detail will thank you for it.
> 
> What about using analog cameras through a capture card?  Would that be
> good enough quality?
> 
> Also, what kind of minute detail are you talking about?  I'm sure he'd
> like to be able to recognize faces, but I'm not sure he'd need more
> detail than that.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: A way of setting up a computer for routing *just* port 113?

[Jeff Noxon]

It sounds like you just need to install one of the fake identd packages,
like nullidentd, and make your router forward port 113 requests to that.
nullidentd will always answer "foobar" to any request.  I run it on my
router (a Debian machine) to fool silly IRC servers which require an
ident service before you can connect.

There isn't any way to "share" the port like you suggest.  Perhaps if you
explain why you need ident to work (it is almost never needed at all),
someone can help more.

Regards,

Jeff

On Sun, Mar 28, 2004 at 04:06:24AM +0100, Joseph Jones wrote:
> My router, which is not a computer, but rather an unbranded little 
> plastic box, will only allow ports to be "forwarded" to one IP on the 
> LAN's Class C subnet. This causes a problem when trying to use ident, 
> which uses port 113, as it means that only one computer in the house may 
> use ident without resetting the router.
> 
> Would it be possible to setup, say, my desktop machine, or any other 
> Debian machine, to be a router for *just* port 113? So I could forward 
> port 113 on the WAN to that machine, and then that machine could 
> automatically share port 113 with any machine on the home LAN? This 
> would include the Windows boxes that form the unfortunate majority on 
> the LAN. If so, what would be the requirements?
> 
> Please understand I'm no expert with networking *or* Debian, but I know 
> enough to setup small LANs and am generally capable of following 
> instructions :)
> 
> Many thanks in advance :D
> 
> Joe


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Cheap dual-port NIC, anyone?

[Jeff Noxon]

Does anyone know of a cheap ($100 or less) dual-port NIC card?  Or quadport?
The best deal I am aware of is $150 for an Intel Dual 10/100 Server adapter,
and that card requires two IRQs for some reason...

Thanks

Jeff

Re: Cheap printer?

[Jeff Noxon]

There is no such thing as a cheap inkjet printer IMHO.  They all cost a
fortune to operate.  They work on the razor blade principle -- they give
away the handle and expect to make money on the blades.

I would personally suggest a used laser printer.  I see HP laserjets at
garage sales all the time (although I am sure the situation may be very
different where you are.)  Toner costs a LOT less than ink.

Regards

Jeff

On 29 Jun 2000, Vicente Torres wrote:
> 
> I have recently bought an EPSON Stylus color 480
> for my home use.
> 
> It was a surprise for me to note that this printer has 
> no button to push and everything must be controlled from
> Windows (including changing or installing the ink cartridge!).
> 
> So, I must return it and buy another printer.
> Wich one would you recommend to me?
> It must be cheap; I do not want to use color on my printings
> and it must work fine with linux.

Re: Oracle 8i and Debian?

[Jeff Noxon]

I'm running Oracle 8 w/ Potato on a production server.  It was a PITA
to get running but it runs very well.  You'll find the discussion forums
at Oracle.com an invaluable resource.

Regards

Jeff

Re: VMWare, Samba, and Slink

[Jeff Noxon]

Instead of asking the Debian mailing list for help with VMware's Samba,
you should probably try using Debian's Samba -- which works just fine.  :)

Is VMWare 2.0 significantly faster than 1.x?  I used it for a while and
gave up.

Good luck,

Jeff

On Thu, Jun 15, 2000 at 04:24:17PM -0600, Robert Kerr wrote:
> Hi all,
> I have VMWare 2.0 installed on a slink machine, and I'm trying to set up
> samba to let my Win98 installation talk to the rest of the box.  Anyway,
> it's set up host-only networking, and telnet, ftp, netscape all work
> fine.  But, samba doesn't work.  I looked in the /var/log/ logs and found
> this line:
> VMWare[init]: /usr/bin/vmware-smbd: error in loading shared libraries
> VMWare[init]: : undefined symbol: setresuid
> 
> and sure enough, there's no symbol setresuid in and of the libs in
> /usr/lib or /lib
> Under RedHat 6.0, setresuid is defined in libc.a, but it's not on Debian
> 2.1.  Has anyone gotten vmware's samba running under slink?  Or does
> anyone know how I can get around this problem?
> Thanks

Re: Wheel mouse not working...oy.

[Jeff Noxon]

FWIW, I tried using icewm and was unable to get the mouse wheel working with
Netscape.  But when I switched to sawfish (from Helix Gnome), the mouse wheel
works in most programs including Netscape and Mozilla.

Regards,

Jeff

--
Help Microsoft end piracy!  Give Linux to a friend today!

On Tue, Jun 13, 2000 at 03:32:47PM +0800, Corey Popelier wrote:
> Just a thought, but all this mouse wheel talk is centered around
> Netscape. Have you tested anything else that utilises the mouse wheel? For
> instance either Mozilla, or even the XMMS playlist should react to the
> mouse wheel.
> 
> Cheers,
>  Corey Popelier
>  http://members.dingoblue.net.au/~pancreas
>  Work Email: [EMAIL PROTECTED]
> 
> On Tue, 13 Jun 2000, Tim Jump wrote:
> 
> > First off, thanks to both of you for replying...and making me feel
> > dumb.  Even after reading all the documentation and all the config files
> > and SEEING that the mouse wheel was represented as buttons 4 & 5, I
> > never thought about changing the number of buttons in XF86Config to 5. 
> > Duh.
> > 
> > Unfortunately, it didn't work.  Here's the MODIFIED section of my
> > XF86Config:
> > 
> > Section "Pointer"
> >Protocol"IMPS/2"
> >Device  "/dev/psaux"
> >Resolution  200
> >Buttons 5
> >ZAxisMapping4 5
> > EndSection
> > 
> > I *think* I've checked everything else.  I copied & pasted the Netscape
> > entry from the aforementioned wheel mouse page into my .Xresources again
> > just to be sure...no dice.  I then tried copying THAT over to
> > /etc/X11/Xresources/netscape to see if it would do anything...but it
> > didn't.  I even tried running imwheel again with it's default
> > configuration to no avail.  Suddenly I'm REALLY missing the days when
> > I'd use console mode exclusively.
> > 
> > I am again, in a word, stumped.  Any further help would be greatly
> > appreciated.
> > -- 
> > [EMAIL PROTECTED]   []   Bababooey Dragon   []  <-==UDIC==->
> > "The stupider it looks, the more important it probably is."
> >  -- J. R. "Bob" Dobbs --
> > Babylon Five Addict [] DEVOlved [] Dirty old man in training
> > 
> > 
> > -- 
> > Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
> > 
> 
> 
> -- 
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
> 

Re: PDF Writer for Linux ?

[Jeff Noxon]

On Sun, Jun 11, 2000 at 06:43:12PM +0200, Oliver Schoenknecht wrote:
> I am currently searching a program similar to Adobe Acrobat under Linux
> that makes PDF-files out of files of all kinds...

gs contains a program called ps2pdf that will convert any Postscript file to
PDF.  PDF is mostly postscript to begin with.

Regards,

Jeff

Re: Promise Ultra66 Controller...

[Jeff Noxon]

On my system, I see messages like this:

hda: Maxtor 54098U8, ATA DISK drive
hdc: Maxtor 91728D8, ATA DISK drive
hde: Maxtor 91152D8, ATA DISK drive
ide0 at 0xac00-0xac07,0xb002 on irq 5
ide1 at 0xb400-0xb407,0xb802 on irq 5
ide2 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: Maxtor 54098U8, 39082MB w/2048kB Cache, CHS=79406/16/63, UDMA(66)
hdc: Maxtor 91728D8, 16479MB w/512kB Cache, CHS=33483/16/63, UDMA(33)
hde: Maxtor 91152D8, 10991MB w/256kB Cache, CHS=22332/16/63, (U)DMA

So I presume you are saying that the drive models have been identified, but
it has not displayed the capacity, cache size, or other details.

My first guess would be a potential IRQ conflict.  If you boot the system
with no drives attached to the Promise controller, does /proc/interrupts
show the Promise controller sharing an IRQ?  If so, try the card in
another slot.  (It is OK for both channels on the Promise card to use
the same IRQ.  That is normal.  Mine are both on IRQ 5.)

Beyond that, I'm not sure what to suggest.  You could try asking on
Linux-kernel, or asking the author of the Promise driver.

I'm wondering why your ide1 is on IRQ 15.  That sounds like the second
IDE port on the motherboard, not the second channel on the Promise card.
I wonder what happened to the second channel on the Promise?

Regards,

Jeff

On Mon, Jun 05, 2000 at 05:11:44PM -0500, Kelly Corbin wrote:
> That did it, thanks.  I am now having problems with it hanging during
> the boot. It finds the card and drives, but it hangs after the lines:
> 
> ide0 at 0xblah, 0xblah on irq 10
> ide1 at 0xblah, 0xblah on irq 15
> 
> Any ideas why this might be hanging?

Re: Promise Ultra66 Controller...

[Jeff Noxon]

Basically, you do the following:

tar xfz linux-2.2.15.tar.gz  (This you have probably done already, but make
  sure you start with a clean source tree)

cd linux

gunzip -dc /wherever/ide.2.2.15.2509.patch.gz | patch -p1

make menuconfig  (make sure to enable Promise support.  If you are booting
  the Promise controller and not your onboard chipset, there
  is also an option for that...  Read the help.)

make-kpkg kernel_image (or whatever you normally do to make kernels)

I hope this helps.

Regards,

Jeff

On Mon, Jun 05, 2000 at 03:26:53PM -0500, Kelly Corbin wrote:
> OK, I got the file ide.2.2.15.2509.patch.gz, but I have never
> installed a patch before, I have always installed the complete source. 
> I have tried patching as suggested on kernel.org's readme, but to no
> avail.  How did you do it?  Thanks again.
> 
> Kelly Corbin
> [EMAIL PROTECTED]
> 
> Jeff Noxon wrote:
> > 
> > I have one, and it works great.  Download the IDE patches relative to
> > 2.2.15 from www.linux-ide.org.

Re: Promise Ultra66 Controller...

[Jeff Noxon]

I have one, and it works great.  Download the IDE patches relative to
2.2.15 from www.linux-ide.org.

Regards,

Jeff

On Mon, Jun 05, 2000 at 12:12:37PM -0500, Kelly Corbin wrote:
> Anyone get the Ultra66 controller to work in the kernel?  I don't want
> to use it as a module, and I don't see it in the kernel anywhere.  They
> say it has been native in the kernel since 2.2.10, but I can't find it
> in 2.2.13 or 2.2.15.  Any help would be appreciated.  Thanks all...

Re: UPS wars: APC vs Tripplite?

[Jeff Noxon]

Have you looked at Best Power?  The Fortress line is very competitive with
APC, and comes with software and source code!  Onvia.com carries them,
and with free shipping and the various $25 coupon codes floating around...
You can't beat the price!  I bought one on back-order, and it arrived in
two days.  I've had several Fortress units in use at the office for the
last year and a half, and they have been flawless so far.

APC's recent low-end/midrange products have had failure rates over 60%
(of a dozen units) for me in less than two years.  APC's older products
were much more reliable.  As always, your mileage may vary.

Good luck

Jeff

On Thu, May 18, 2000 at 04:20:52PM -0700, Ron Farrer wrote:
> 
> Hello,
> 
> I've been thinking about getting a new UPS. Previously I purchased APC
> products, but I want to hear about other experiences. I've pretty much
> narrowed it down to a product from APC or Tripplite. APC is more
> expensive, less Linux/UNIX friendly, but makes good products (IMHO).
> Tripplite is less expensive, more Linux/UNIX friendly, but I'm not sure
> how the quality of their product compares to APC. 
> 
> Does anyone know of some sort of comparison between these two companies
> products? I'm interested in features/quality more then price. 
> 
> 
> TIA,
> 
> Ron
> -- 
> Email: 
> Home:  
> 
> Alpha Linux Organization: 
> Bellingham Linux Users Group: 



Regards,

Jeff

--
Diagnostics for your VW/Audi - http://www.planetfall.com/~jeff/obdii/

Trick serial port question!

[Jeff Noxon]

I need to be able to open a dumb-terminal session to a serial port, while
forcing RTS low.  Strange, yes, but an absolute requirement.  Can this be
done with any existing utilities?  kermit?

Thanks

Jeff

Re: Frozen, Potato or Woody?

[Jeff Noxon]

Frozen and Potato are the same thing (for now).   You probably don't want
to mess with Woody, particularly without doing a Potato upgrade first.

In other words, upgrade to Potato.

Apt-get is your best bet for a successful upgrade.  You will need more
disk space -- enough to hold new versions of everything.  And be prepared
to wait, because a 486 is not a rocket.  :)

If you can borrow another hard drive and add it to your machine
temporarily, that might help get things done.  If not, I suggest you
stick with Slink, or wait for Potato CD-ROMs to arrive.

Enjoy,

Jeff

Re: Asus K7V motherboard

[Jeff Noxon]

If it didn't work, you'd be seeing posts about it.  It's very unusual
for an off-the-shelf motherboard to have problems with Linux these days.

I'm about to buy a K7V myself.  Linux works just fine on my K7M.

Regards,

Jeff

On Fri, Apr 14, 2000 at 11:01:29AM +0200, Alfredo Amiel P. Leonardia wrote:
> Anybody out there using an Asus K7V motherboard?  It's very new and I
> haven't found anything on the web concerning compatibility with linux or
> lack thereof.  Any info?

Re: postgresql 6.5.3 vs. debian

[Jeff Noxon]

Why, pray tell, are you trying to install RedHat postgresql RPMs on a Debian
system?  Debian has postgresql 6.5.3 already, in frozen.  What you are doing
with RPMs is just asking for trouble.

The easiest thing to do is just upgrade your whole system to frozen.  If you
can't, then set up apt (/etc/apt/sources.list) to access your nearest mirror
and run "apt-get install postgresql" ... this should upgrade just postgresql
and the parts of your system it depends on.

Regards,

Jeff

On Tue, Apr 11, 2000 at 03:10:35PM -0500, w trillich wrote:
> aaugh!
> 
> i'm having my fourth day of troubles trying to upgrade
> my 6.3 pgsql (came on the debian cd) to 6.5.3, and i'm hoping
> someone might have the insight i need to get over this hurdle...
> 
> these are the postgres files i'd love to install:
> postgresql-6.5.3-3.i386.rpm
> or
> postgresql-devel-6.5.3-3.i386.rpm
> postgresql-perl-6.5.3-3.i386.rpm
> postgresql-server-6.5.3-3.i386.rpm
> postgresql-test-6.5.3-3.i386.rpm
> onto
> Linux 2.0.36 #2 Sun Feb 21 15:55:27 EST 1999 i586 unknown
> 
> here's the group of error messages i run into when merely
> CONVERTING (via 'alien') the *.rpm's to *.deb's:
> 
> dpkg-shlibdeps: warning: unknown output from ldd on
> `debian/tmp/usr/bin/pg_dump': ` libreadline.so.3 => not found'
> dpkg-shlibdeps: warning: unknown output from ldd on
> `debian/tmp/usr/bin/pg_dump': ` libhistory.so.3 => not found'
> dpkg-shlibdeps: warning: unknown output from ldd on 
> `debian/tmp/usr/bin/pg_id':
> `   libreadline.so.3 => not found'
> dpkg-shlibdeps: warning: unknown output from ldd on 
> `debian/tmp/usr/bin/pg_id':
> `   libhistory.so.3 => not found'
> dpkg-shlibdeps: warning: unknown output from ldd on
> `debian/tmp/usr/bin/psql': `libreadline.so.3 => not found'
> dpkg-shlibdeps: warning: unknown output from ldd on
> `debian/tmp/usr/bin/psql': `libhistory.so.3 => not found'
> dpkg: /lib/libtermcap.so.2 not found.
> 
> so i hopped on the web and searched for *.rpm's that would take care
> of filling in the missing holes:
> 
> libtermcap-2.0.8-20.i386.rpm
> -- contains
> /lib/libtermcap.so.2.0.8
> (to which i linked the sought-after libtermcap.2.0 via
> 'ln -s libtermcap.so.2.0.8 libtermcap.so.2.0' thinking that 
> internal tweaks wouln't break the interface)
> readline-2.2.1-6.i386.rpm
> --  contains 
> /usr/lib/libhistory.so.3.0
> /usr/lib/libreadline.so.3.0
> (which i linked to the sought-after filenames in similar fashion
> to the method i used for libtermcap)
> 
> but doing so seems to not helped one iota, even after going into
> 'dselect' and choosing to update package info from the mounted directory
> where the *deb packages reside.
> 
> nearly giving up, i tried using the RPM installer instead:
> 
> # rpm -i libtermcap-2.0.8-20.i386.rpm
> failed dependencies:
> /etc/termcap is needed by libtermcap-2.0.8-20
> /sbin/ldconfig   is needed by libtermcap-2.0.8-20
> /bin/sh   is needed by libtermcap-2.0.8-20
> ld-linux.so.2 is needed by libtermcap-2.0.8-20
> libc.so.6 is needed by libtermcap-2.0.8-20
> libc.so.6(GLIBC_2.0) is needed by libtermcap-2.0.8-20
> libc.so.6(GLIBC_2.1) is needed by libtermcap-2.0.8-20
> # rpm -i libtermcap-2.0.8-20.i386.rpm 
> failed dependencies:
> /etc/termcap is needed by libtermcap-2.0.8-20
> /sbin/ldconfig   is needed by libtermcap-2.0.8-20
> /bin/sh   is needed by libtermcap-2.0.8-20
> ld-linux.so.2 is needed by libtermcap-2.0.8-20
> libc.so.6 is needed by libtermcap-2.0.8-20
> libc.so.6(GLIBC_2.0) is needed by libtermcap-2.0.8-20
> libc.so.6(GLIBC_2.1) is needed by libtermcap-2.0.8-20
> 
> i'd appreciate any direction... thanks (i've still got some hair
> left, and would like to keep it).
> 
> [accidentally posted this to debian-devel this morning. whoops.]

Re: VMware, and acceleratedX

[Jeff Noxon]

On Sat, Mar 04, 2000 at 11:47:10PM -0500, Aaron Solochek wrote:
> Has anyone used this combination before?  I would like to be able to use
> full screen, or atleast be able to make the window larger than it
> currently seems to allow.  Also, I'm using vmware to access a windows98
> partition on a dual boot system.  Right now windows finds all this new
> hardware, is there a way to make windows see the same hardware it
> usually does so this doesn't happen?

I asked the AccelX folks about it, and they have no intention of
supporting it.  Apparently it's a huge hack in the first place and
it could comprimise system stability.  (Actually, on my system, it
tremendously affects system stability.)

Regards,

Jeff

Re: Transparent network bridge+filter?

[Jeff Noxon]

You have an interesting idea, but it won't work in my case.  I have to
put this between a pair of Cisco routers running EIGRP.  They won't see
each other if the router discovery packets (etc.) aren't forwarded by
a bridge.  I also can't guarantee that the address of the router on one
side won't change -- it is not under my control.

Thanks!!

Jeff

On Wed, Jan 19, 2000 at 08:42:00AM +0100, Onno Ebbinge wrote:
> At 02:08 PM 1/18/00 -0600, Jeff Noxon wrote:
> >Can anything that runs on Linux do reliable network bridging & filtering?
> >I need a transparent filter that I can drop into an existing network.
> >
> >Ipfilter will do the job with Open/NetBSD.  It may work on Linux, but
> >requires kernel 2.0.35 and isn't compatible with glibc.
> 
> Another guy ask -something like that- before,
> I replied with an answer that worked ;-)
> 
> Here is my reply and maybe you can use 
> parts of it:
> (You don't want to use this route config ;-)
> 
> > This has been a while but here it goes:
> > 
> > Please test if the next settings will do the trick.
> > The debian box cannot be reached from the inet or lan,
> > We can do something about the lan connection though...
> > 
> > Note: Filtering firewall is WIDE open!
> > Note: There is a route for all IP's because they are 
> >   on the same subnet (netmask) but NOT on the 
> >   same network device!
> > Note: Youre gateway is 63.225.131.78
> > 
> > root# ifconfig lo 127.0.0.1
> > root# ifconfig eth0 0.0.0.0 promisc
> > root# ifconfig eth1 0.0.0.0 promisc
> > 
> > root# route add 63.225.131.73 eth0
> > root# route add 63.225.131.74 eth0
> > root# route add 63.225.131.75 eth0
> > root# route add 63.225.131.76 eth0
> > root# route add 63.225.131.77 eth0
> > root# route add 63.225.131.78 eth1
> > 
> > root# ipchains -P input ACCEPT
> > root# ipchains -P forward ACCEPT
> > root# ipchains -P ouput ACCEPT
> > root# ipchains -F
> > root# ipchains -X
> > 
> > Please send me your results

Transparent network bridge+filter?

[Jeff Noxon]

Can anything that runs on Linux do reliable network bridging & filtering?
I need a transparent filter that I can drop into an existing network.

Ipfilter will do the job with Open/NetBSD.  It may work on Linux, but
requires kernel 2.0.35 and isn't compatible with glibc.

Any suggestions would be great.  Please copy me on replies.

Thanks!

Jeff

Re: Sound Blaster Live!

[Jeff Noxon]

On Wed, Oct 06, 1999 at 07:57:16PM -, Pollywog wrote:
> I am using SB 16 (Vibra I believe it is called) and I use the OSS-Linux
> drivers, which cost me $20.  They only allow half-duplex.  Are there other
> drivers that allow this card to be used in full-duplex mode?

I believe those cards only allow pseudo-full-duplex operation (ie there is
some kind of catch to it).

The ALSA drivers probably do what you want:

http://www.alsa-project.org/

ALSA is still in development, so it changes often, but it works very well.

Regards

Jeff

Re: Flat-panel monitors?

[Jeff Noxon]

On Wed, Sep 29, 1999 at 09:59:42AM -0400, Noah L. Meyerhans wrote:
> > I'm using an SGI 1600SW (17.4" wide) with the #9 card it's bundled with.
> 
> A co-worker of mine owns this same monitor (boy is it sweet!) and uses the
> Acclerated X server.  It's much faster and less buggy.  You may want to
> check this out.

It is much faster and seems bug-free, but unfortunately it doesn't support
DGA or XKEYBOARD, and those help out VMware quite a bit.  And there are
political issues within the company where I work -- getting software
"certified" is a pain.  I can use anything in Debian because Debian is
certified, but another product is not.

> Of course, the ideal solution would use an entirely different video card,
> as the #9 card is not a great card.  But there are very few 100% digital
> cards out there.

I looked into that, and even talked to SGI about it.  I forget which
DFP standard the 1600SW uses, but it's different from the two standards
offered by everyone else (Matrox, ATI, etc).  They said they are working
on a product that will adapt the monitor to work with other types of
DFP interfaces.  They realize they made a mistake when choosing the
interface.  Who knows, that product may even be shipping now... :)

Regards,

Jeff

Re: Flat-panel monitors?

[Jeff Noxon]

On Tue, Sep 28, 1999 at 05:57:27PM -0400, B. Szyszka wrote:
> I was wondering what luck, if any, you guys have had with flat panel
> monitors in Debian. Has everything worked as it should?

I'm using an SGI 1600SW (17.4" wide) with the #9 card it's bundled with.
The X-Server for this card really stinks and is full of bugs, but it's a
sacrifice I'm willing to make for this beautiful screen.  :)  I'd change
cards, but there aren't any other cards on the market that support this
monitor.

Regards,

Jeff

Re: large hard disks (again)

[Jeff Noxon]

On Wed, Sep 22, 1999 at 10:15:37AM -0700, Kenneth Scharf wrote:
> Exactly what I suspected.  Only how to figure out the
> geometry in the first place?  Anyway I now have that
> information for the disk in question.  As soon as
> potato is release it will be water under the bridge.

Possible sources:
- Label on the drive itself
- Web site of drive manufacturer
- BIOS (Award BIOS in particular has an auto-detect feature, which is not
the same as setting the drive type as "AUTO")
- The partition table already on the drive, if the drive has already been
  partitioned by another OS.

And others, I'm sure.

Regards,

Jeff

Re: large hard disks (again)

[Jeff Noxon]

The 2.2 kernels will properly detect >1024 cylinders.  For other kernels,
you can put fdisk in "x"pert mode and override its detection, or you
can also specify the disk geometry on the kernel command line.

Regards,

Jeff

On Wed, Sep 22, 1999 at 05:14:25AM -0700, Kenneth Scharf wrote:
> I have posted this before, how to fdisk a > 8.4gb
> drive.  Problem was that with slink fdisk saw only
> 1024 cy, 255h, 63s for a total of 8.4gb.  If I booted
> a RedHat 6.0 or Mandrake 6.0 CD (also stormix) their
> partition utilities (fdisk, disk druid, etc) would
> report the correct number of cyl's (2100) to give a
> 17.2gb total disk size. 
> 
> After partitioning with Stormix I stopped the install,
> and booted my slink cd and then installed slink
> (skipping the partition step, which was already done).
>  This worked.  Now if I run fdisk, it still reports
> the wrong(?) number of cyl's but does show the correct
> partition sizes (p command).  It also complains about
> different logical and physical starting and/or ending
> cyl numbers for the last two partitions on the disk. 
> However if I go to expert mode (x) and change the
> number of cylinders (c) to 2100, the p command now
> gives the exact same printout as fdisk under Mandrake
> or Redhat fdisk.  So...why does fdisk get the correct
> size info from the kernel under Mandrake or Redhat? 
> Can it be a difference between the 2.0.36 kernel and
> the 2.2 series?  Or is it a later version of fdisk
> itself (BTW cfdisk and sfdisk behave the same way...)
> I now think the difference is in the 2.2 kernel.
> Any ideas?  I now know I can partition under debian by
> giving fdisk the correct final cyl number.  But I
> still had to use Mandrake fdisk to find this number!

Re: Riva TNT

[Jeff Noxon]

On Fri, Aug 27, 1999 at 10:27:33AM -0400, Samantha Summers wrote:
> I have a Velocity 4400 which uses the Riva TNT.
> 
> I thought this was built in the SVGA X server, but if it is, XF86Setup
> doesn't see it.
> 
> I think I have the most current XF86Setup and SVGA server offered in stable.
> They are both 3.3.2.3a-11.

You need a newer one.  You can download a binary of the SVGA server from
nvidia's web site or from one of the Xfree86 mirrors.

Regards

Jeff

Re: Mylex DAC960

[Jeff Noxon]

Installing Debian on a system with a DAC960 is not easy.  I've done it
several times, *by hand*, using the Debian rescue disk only to get a shell
prompt.  Let me repeat that it is very difficult, and I wouldn't be able
to walk you through it.  It works wonderfully once installed, however.

I would suggest that the "easy" way to do this is to put in an IDE
drive temporarily, install Debian onto it, get it so it sees the DAC960,
and then move everything over to the DAC960 disk and remove the IDE drive.

Good luck,

Jeff


On Wed, Aug 11, 1999 at 10:42:45AM +0200, Fabrizio Roccato wrote:
> I will install debian on a computer with only this Mylex DAC960 Raid
> Controller (no other disks...). The official resc. disk does not work, so
> i have recompiled a
> new kernel with patch from www.dandelion.com, created the devices and
> substitute the kernel in the rescue-disk... lilo and fdisk seems to be of
> the correct version, so i dont substitute them.  
> Now at boot time the kernel recognize the RAID controller and the volume
> but the installation procedure doesnt.
> Any idea on how to install?
> Will this raid controller supported in the next debian release? 

Re: 8MB PCI 1600x1200@76Hz video card suggestion?

[Jeff Noxon]

On Thu, Jun 10, 1999 at 12:59:31PM -0400, Peter S Galbraith wrote:
> I found:  
>  STB Velocity 4400
>  Creative Labs Graphics Blaster Riva TNT
>  ASUS V3400 TNT (AGP only)
>  Diamond Viper V550
>  Hercules Dynamite TNT
> 
> What version of XFree is required, and what xserver do we use?

Just about any version of XFree will do.  You may need to download a newer
X server from xfree86.org, or upgrade to potato.  If I recall correctly, the
SVGA server contains the TNT support.  You might want to double-check on
the xfree86.org website.

Regards

Jeff

Re: 8MB PCI 1600x1200@76Hz video card suggestion?

[Jeff Noxon]

On Wed, Jun 09, 1999 at 03:14:40PM -0400, Sean wrote:
> I've often wanted to go with a TNT card, but I'm concerned about the card's 
> heat
> output.  Every query I've ever made about "just how hot does this card run" 
> has been
> left largely unanswered, which has just increased my trepidation about the 
> card in
> general.

If you're looking for actual temperatures, check out http://sysdoc.pair.com.

In general, don't worry about it.  If you don't overclock the card, heat is
not a problem.  Mine uses a passive heatsink and it's just fine.  The Diamond
and Hercules and Asus use fans.  The Asus and Hercules cards run cold.

Regards,

Jeff

Re: 8MB PCI 1600x1200@76Hz video card suggestion?

[Jeff Noxon]

I'm not sure who started this thread, but I recommend the nVidia RIVA
TNT chip.  You can buy a 16MB TNT card (PCI or AGP) for $99 at Best Buy.
You can find them cheaper on the 'net.  Creative Labs makes one, and so
do several other companies.  They work very well under both Windoze and X.
And the driver is now open source.

Regards

Jeff

Re: Mission critical Debian

[Jeff Noxon]

I'm using it for two mission critical point-of-sale support applications
for a Fortune 10 company.  In fact we have used Linux in one of these
roles since mid-95 (starting w/ Slackware and migrating to Debian later.)
Plans for others are in the works.  Slackware was a mess, but Debian is
very elegant and easy to manage/upgrade.

If I say which company, I will likely be fired.  I work here as a
contractor.  Suffice it to say that over 8,000 stores are involved, and
the role of Linux (Debian) is increasing.  We are replacing AIX systems
with it.

The management in this corner of the corporation has been convinced that
Linux is every bit as good, and in many ways better, than AIX in our
highly customized environment.  Believe it or not, support is one major
reason for the switch.  Others are ease-of-use and cost, both up-front
and ongoing.

Regards,

Jeff

Re: Corel : GNOME vs KDE

[Jeff Noxon]

Oh, no.  Let's not start another GNOME-vs-KDE thread.  Please!

On Thu, Apr 22, 1999 at 08:47:47PM +0200, Sami Dalouche wrote:
> I don't understand why Corel has chosen KDE instead of Gnome.
[ Rest deleted ]

Re: Potato

[Jeff Noxon]

At least 99% of Slink works with 2.2.x kernels.  I have several machines
running kernels 2.2.5/2.2.6 on Slink.

I'm running Potato on one machine, and at the moment it is stable, but
that will change from time to time during the course of the development
cycle.  Unless you want to deal with the occasional hassles presented
by glibc2, I'd suggest sticking with Slink.

Regards

Jeff

On Wed, Apr 21, 1999 at 11:01:25AM -0400, Madel, Kurt wrote:
> For those of you using Potato, I was wondering how unstable it is.  I have a
> Zip Plus drive and know that it is naturally supported by the 2.2.X  kernel,
> so would like to move to a 2.2.X distribution as soon as possible and would
> like to use Debian because I believe in freedom.

Re: new ATI graphics boards

[Jeff Noxon]

You might be better off avoiding ATI graphics boards under Linux.  ATI is
not cooperative with the developer community and drivers tend to take a
while to get written & stabilize.  I've suffered through problems with
several generations of Mach64 cards.  Maybe the new boards work great
already, but I doubt it.

Regards,

Jeff

Re: Citrix ICA and Debian Linux 2.0

[Jeff Noxon]

Try running 'ldd' on the executable to see what is missing.  I.e.,

ldd winframe

Or whatever the name of the executable is.  Follow up here.

Regards,

Jeff

On Thu, Apr 15, 1999 at 04:59:41PM +0200, Peter Niessink wrote:
> I'm a novice linux user, coming the spoiled-brat world of M$ ;-)
> 
> Has anyone gotten Citrix ICA (as avaliable from http://www.citrix.xom) for
> Linux running succesfully on Debian 2.0 ? So far i'm getting problems
> installing it because it is missing obscure libraries (i've already gott
> libc5 installed for backward compatability) but it is requiring components
> that i cannot relate to any package.
> 
> Who can help me out 
> 
> If you want to try a server to test it on I'll be glad to give you a test
> account (on the Internet at hydra.aimcons.nl).

Re: K6 error

[Jeff Noxon]

On Tue, Apr 13, 1999 at 02:40:17PM -0500, ktb wrote:
> AMD K6 stepping B detected - probably OK (after B9730)

Check the serial number on your CPU.  If it's > this number, you're fine.
If you have less than 32 MB, you're fine, unless you plan to add more.
If you have a buggy chip, you can call AMD and get it replaced with a
shiny new K6-2-300 for free!  (Seriously, they will do it.)

Some K6's had bugs that really only affected use under Linux with >32MB.
I recently had one replaced, but I had to get a new motherboard to
support the K6-2.

Regards,

Jeff

Re: cd-to-cd burning

[Jeff Noxon]

On Tue, Apr 13, 1999 at 02:51:16PM -0400, Dan Brosemer wrote:
> Haven't tried this one, but:
> 
> dd if=/dev/cdrom1 of=/dev/stdout bs=1048576|cdrecord dev= speed=?? -v
> -xa2 /dev/stdin
> 
> (try it with the -dummy flag first as I've never done this!)

It won't work because cdrecord has to know the size of the ISO image
before it can write it.

If you can't do CD-CD copying successfully in the way I described earlier,
you can do something like:

dd if=/dev/scd0 of=cdimage.iso bs=64k

To copy the image to the hard drive first.  Then burn it:

cdrecord dev=5,0 speed=4 -v -dummy cdimage.iso

Jeff

Re: cd-to-cd burning

[Jeff Noxon]

On Tue, Apr 13, 1999 at 01:34:50PM -0500, Matt Garman wrote:
> 
> What is the best way to do CD-to-CD copying under Linux?
> Specifically, I have a copy of the "official" debian 2.1 (slink) CD
> of which I want to burn an identical copy.
> 
> Is it necessary to make an ISO9660 image from the current CD and then
> burn?  Surely there is an easier way to get around this.

I used something like:

cdrecord -dev=5,0 -isosize -speed=4 /dev/scd0

You might also want to add -dummy and -v (verbose) when you do your
first test, and of course, use the correct SCSI ID for your recorder.

Note that this technique will only copy a single-session CD with one
ISO9660 data track.

Regards,

Jeff

Document processing? (TeX/SGML?)

[Jeff Noxon]

I'm looking for the best tool to create a professional-looking document
once, and then render it in the following formats:

HTML
Postscript
PDF
ASCII

I'm not concerned so much about a learning curve as I am about flexibility
and results.  Microsoft Word and other GUI-based products drive me nuts
because they don't give me the control (or stability) I need.

What tools can tackle this kind of chore and what are their relative
merits?  Should I be looking at TeX or SGML tools?  Or should I just stick
with plain HTML?

Naturally, the tools I use should be free and available as part of Debian.

Thanks in advance for your suggestions,

Jeff

Re: purge will not remove autofs

[Jeff Noxon]

Add an "exit 0" to the top of the prerm script and try again.

i.e.,

#!/bin/bash
exit 0

Regards,

Jeff

On Fri, Apr 09, 1999 at 03:31:18PM -0400, Paul Kirschner wrote:
> In the process of installing and removing autofs and amd, I got autofs
> locked up so that I cannot remove, purge, install or unpack it. All I
> get is...
> 
> dpkg --purge autofs
> (Reading database ... 28494 files and directories currently installed.)
> Removing autofs ...
> Stopping automounter.
> dpkg: error processing autofs (--purge):
>  subprocess pre-removal script returned error exit status 1
> Starting automounter:.
> Errors were encountered while processing:
>  autofs
> 
> 
> The pre and post-rm script just stop autofs and run OK by hand.
> 
> How can I completely remove or install or fix this?
> 
> TIA.
> 

Re: CDDB updates

[Jeff Noxon]

On Fri, Feb 12, 1999 at 06:21:03PM +, M.C. Vernon wrote:
> Is there any way to upload a record to a CDDB server (either the cddb one,
> or in a local record)? I have a cd in front of me that gcd refuses to play
> because cddb doesn't produce a match for that CD id

xmcd?  That's the program that started the CDDB.

Regards,

Jeff

Re: SAMBA and Peer Networks?

[Jeff Noxon]

On Thu, Feb 04, 1999 at 08:21:24PM +, tracheotomy bob wrote:
>   I have installed SAMBA on my machine at work but what I really want
> to do is to be able to read other peoples shared Windows9x folder et al. It 
> this
> possible with SAMBA or is SAMBA just a server-side application?

SAMBA will do what you want.

It includes a program called smbclient.

Regards,

Jeff

Re: Search and Replace

[Jeff Noxon]

On Fri, Jan 08, 1999 at 08:49:10AM -0700, John Greer wrote:
> I know that this is not Debian specific but I thought I would give it a 
> shot anyway.  I need to search a series of files for a text string 
> (grep I know) and then I need to replace that string with another.  Is 
> there a command or string of commands that I can do this in?  If 
> this is possible it will make my life much easier!!  Thanks 

You can do what you want with sed or with a simple perl script.

Example:

sed s/oldstring/newstring/g newfile

If you have a lot of files to do,

for x in file1 file2 file3 
do mv $x $x.old ; sed s/old/new/g <$x.old >$x ; done

You can use wildcards instead of file1...file3

Regards,

Jeff

IDE RAID 1 Hardware - Experiences?

[Jeff Noxon]

I want to use a PCI RAID-1 IDE controller on a mission-critical Linux box.
Are there any problems I'm likely to face?  How will I know if a disk
has gone bad?

I want to avoid the software RAID driver because I need the system to
be bootable even after a disk failure, without operator intervention.

Any comments would be appreciated!

Thanks,

Jeff

Re: SIIG SCSI AP-10 PCI Adapter Debian Help?

[Jeff Noxon]

There is an Initio driver available on the Initio website.  The driver
is also in the latest kernel snapshots from Alan Cox.  If this information
does not help you, let me know and I'll get more specific.

SIIG does not make its own cards.  They just buy cheap ones and sell
them with their name pasted on.  ;)

Regards,

Jeff

On Mon, Dec 21, 1998 at 11:21:39PM +, Art Lemasters wrote:
> I installed a Micropolis SCSI drive with a SIIG AP-10
> adapter ("Fast" SCSI-2), and the kernel did not find it.
> Does anyone have a kernel, module or boot disk to support it?
> A year and a half, and I'm still a newbie here.  :-) 
> 
> /proc/scsi/scsi showed "Attached devices: none"  Here's the
> extract from /proc/pci.
> 
> ---
> PCI devices found:
>   
> [...]
> 
>   Bus  0, device  17, function  0:
> SCSI storage controller: Initio Corp Unknown device (rev 1).
>   Vendor id=1101. Device id=9400.
>   Medium devsel.  IRQ 15.  Master Capable.  Latency=32.  
>   I/O at 0x6100.
>   Non-prefetchable 32 bit memory at 0xe041.
> -

Re: 64megs to 192megs of ram

[Jeff Noxon]

On Sun, Dec 20, 1998 at 02:33:18PM -0500, [EMAIL PROTECTED] wrote:
> On 20 Dec, Nuno Carvalho wrote:
> >  If you're using lilo you need to add to your lilo.conf the following
> > line:
> > 
> >  append="mem=128M"
> > 
> >  I not sure if 2.1.x kernels need that !

> Yes, all kernels need it.  The 64meg barrier is a lilo thing, not
> kernel afaik.

2.1.x kernels do not need it.

2.0.36 does not need it.

It has nothing to do with LILO.  LILO does not determine memory size for
the Linux kernel.

The BIOS call to determine memory size could originally return a maximum
of 64MB.  Recent (i.e. >= Pentium) machines support an alternate BIOS call
that can report >64MB.  Newer kernels use this new BIOS call if available.

Some really old machines with >64MB may still require something like
"mem=128M", but I doubt many old machines support that much RAM anyway.

Regards,

Jeff

Re: Compiling Netscape with LessTif?

[Jeff Noxon]

On Fri, Dec 18, 1998 at 11:28:38PM +0100, Daniel Elenius wrote:
> I vaguely recall some-one talking about compiling Netscape with
> lesstif, saving a lot of memory. Is this possible, and how would it be 
> done. I searched in the debian-user archives, but all I found were
> some generic advice on compiling with lesstif.

Are you thinking of Mozilla?  Source code is not available for Netscape.
Mozilla sort-of works with Lesstif, but you're better off with a real
Netscape.  Mozilla is switching to GNOME these days...  No more Motif.

Regards,

Jeff

Re: 8G ide harddrive limit?

[Jeff Noxon]

On Fri, Dec 18, 1998 at 09:47:39AM -0800, Kenneth Scharf wrote:
> I just brought 17.2G ide harddrive, and when I try to use cfdisk to
> partition
> it, it only sees 8G. FWIW, I updated cfdisk to ones in frozen. Bios and
> Linux
> bootup sees it as 17.2G, so I don't know why cfdisk will not.:(
> 
> Running 2.0.36(my own roll) and mostly Debian 2.0 with few updates.

I installed a 17.2GB Maxtor in my system.  2.1.130, and I had some strange
problems.  The BIOS and Linux both report correct geometry if it's on
hdc, but on hda, Linux thinks it's a 528MB drive.  The BIOS setup shows
correct geometry no matter where I put it.  I used fdisk, not cfdisk.

I ended up overriding the Linux values in lilo.conf to get lilo working:

... snip ...
append="hda=33483,16,63"
... snip ...

Use whatever values your BIOS reports.  Note that if you do this, your
kernel files must be below cylinder 1,024.  The easiest way to do this
is to make a small partition for /boot, and put it at the beginning of
the drive.

I had to pass the same values on the command line using the Debian rescue
disk in order to get it partitioned correctly and get lilo installed
once I messed everything up.  :-)

Maybe someone else has a better solution...

Regards,

Jeff

Re: Corel® WordPerfect® 8 for Linux® is here! (fwd)

[Jeff Noxon]

On Thu, Dec 17, 1998 at 12:33:00PM -0500, Person, Roderick wrote:
> Here's the link. According to the stats I was 3rd to download, so I figured
> it was hard to find.

The stats said that for me too, so something must be broken.

(And I was so impressed by my own timing...)

;)

Regards,

Jeff

Re: how to put apache under hosts.{allow,deny}'s control?

[Jeff Noxon]

On Wed, Dec 16, 1998 at 02:04:04PM -0500, Shaleh wrote:
> Subject says it all.  How can I put non-inetd services under
> hosts.{allow,deny}'s control?

If you really want that, the best thing to do is run Apache from inetd.
According to the manpage it can be run that way.

Regards,

Jeff

Re: Segmentation faults

[Jeff Noxon]

On Thu, Dec 10, 1998 at 06:22:54PM -0500, Tom wrote:
> Could anyone please answer these questions about segmentation faults?
> What is a Segmentation Fault? How do I trace its cause? How are they
> normally fixed. Why don't I get a core dump when they happen?  Thank you

It's a memory access violation.  It could happen (for example) by writing
to read-only memory or by writing outside of the process memory space.
It could be caused by a null-pointer dereference.

If your executable that's crashing has debug symbols in it, you can run
it from the gdb debugger.  Reproduce the problem and gdb will show you
the line of offending code.  You may also learn something by using strace
or ltrace on the program.

You don't get a core dump because you have core dumps disabled.  The
command "ulimit -c" controls the creation and maximum size of core files.
A core file will probably not be useful to you unless the binary that's
crashing has debug info in it.

Regards,

Jeff

Re: Creating 1.743MB Debian Rescue Disk

[Jeff Noxon]

I can't even get superformat/dd to create resc1743.  I think we really
need a 1.44MB rescue disk for systems that can't handle the oddball
disk sizes.  In my case, superformat formats/verifies just fine, but dd
fails with a bunch of floppy driver errors.

I ended up installing slink on a machine the brute-force way using the
tomsrtbt single-disk Linux distribution.  What a pain.  :)

Regards,

Jeff

Re: quicky postgreSQL question

[Jeff Noxon]

On Wed, Nov 25, 1998 at 02:47:53PM -0500, Shaleh wrote:
> How can I read in a comma delimited file and have the contents added to a SQL
> table?  I seem to recall there being a program that did this for postgres.

I've always done that kind of thing using Perl.  You can use the postgres
libraries, or you can just convert the file into SQL insert statements
and run it from psql.  Heck, you could even use sed.  I don't know of
an easier way.

Regards,

Jeff

Re: SQL Database performance

[Jeff Noxon]

On Thu, Nov 19, 1998 at 04:58:00PM +, Martin Oldfield wrote:
> Does anyone have a feel for the relative performance of the various
> SQL databases in Debian ? I'd also be interested to know how these
> compare to something like Oracle (under either NT or Linux).
> 
> I'm looking to manage two databases: one of roughly ten thousand
> records a few k in size, the other 100,000 rather smaller records. It
> might well be feasible to use something like gdbm for the latter one.

Postgresql runs like greased lightning.  In some tests involving
0.5 million records of a few K in size, it blew away MS SQL Server.
I think it was version 6.  Postgresql runs well, although it does have
some limitations.  ODBC was the main weakness I noticed.  We're currently
deploying Oracle on a Debian server...

Oracle is bloatware even by Microsoft standards.  I can't comment on
speed yet.

Good luck,

Jeff

Re: HELP slink upgrade: netscape: locale `C' not supported

[Jeff Noxon]

Re-install xlib6g and it should work.  Now if only I could get Acrobat
working again...

Regards,

Jeff

Re: source code for /bin/login?

[Jeff Noxon]

On Thu, Nov 12, 1998 at 02:09:33PM -0500, John B. Fink wrote:
> 
> Hi folks,
> 
> Due to an extreme amount of idiocy at the place where I work, I have need
> to modify the /bin/login program so that instead of displaying this --
> 
> adler login:
> 
> it will display this --
> 
> login:
> 
> [snipped]

If you make the changes, how about adding the option to login.defs, and
forwarding the changes upstream?  That would be most cool...

You can find out where to find source code for a given package like this:

$ dpkg --status login

Package: login
Essential: yes
Status: install ok installed
Priority: required
Section: base
Installed-Size: 153
Maintainer: Guy Maor <[EMAIL PROTECTED]>
Source: shadow
 this is what you want
Version: 980403-0.3
Replaces: shadow-login, shadow-passwd
Pre-Depends: libc6
Conflicts: shadow-login
Conffiles:
 /etc/login.defs 8f524e3b2937d29b4cc54fcfae9f4155
 /etc/login.access d30c542d565436bd5333a80466d5ccf8
 /etc/securetty 2561f31ce2a1b00b28cacf3e9ad54734
 /etc/porttime 5b19db99247a9275bd94a6c8e2140b52
 /etc/limits 9d253c5f92441741739064c5bf92ef6f
 /etc/init.d/logoutd 48a23db21d84dc9ea7f4f3f9adf97da2
Description: Sign on to the system.
 login and newgrp change the user and group.


Regards,

Jeff

Re: How much RAM do I need?

[Jeff Noxon]

On Wed, Nov 11, 1998 at 01:58:16PM -0800, Eric House wrote:
> I need to decide whether to upgrade my Debian laptop to 48 or 80
> meg (from the current 16).
> 
> Is there any way to log how much swap is currently getting used during
> the activities I run all the time?

Try vmstat & free

> Any other advice on how much is enough?  This machine is for personal
> use (e.g. Pilot software development) and will never be a server.

If you run X, I'd say 80, if you just run in text mode 48 should be fine.
You may get better battery life with 80 since you'll probably hit the
disk less.  Keep in mind that any "free" memory you have is used to buffer
executables and cache the filesystem...  The more the better.

On the other hand, if your chipset can't cache more than 64MB, going to
80MB might actually make some things slower.  I'm not sure if telling the
kernel "mem=64M" on an 80MB machine would be enough to fix that problem
or not.

Regards,

Jeff

Re: Multiple IP addresses on one Machine

[Jeff Noxon]

On Tue, Nov 10, 1998 at 02:58:21PM -0500, Collin Rose wrote:
> I am trying to setup Multiple IP addresses on one Machine. CAn anyone tell
> me how to do this? I.e. Apache will bind to one IP and the server itself
> uses another.

You can use IP aliasing to do that.  You need to have it enabled in the
kernel, and then you can use ifconfig to create a new "alias."  If your
network is eth0, you can create aliases like eth0:0, eth0:1, etc.

Then you need to change the Apache configuration to bind to the address
you want.

Regards,

Jeff

Re: [SMP] P2 Database Srvr: 1x450 or 2x350 CPUs?

[Jeff Noxon]

On Tue, Nov 10, 1998 at 11:47:18AM -0800, George Bonser wrote:
> On Tue, 10 Nov 1998, Jeff Noxon wrote:
> 
> > I'm setting up a Debian box for my client as an Oracle server.  I'm torn
> > between Dual 350's or a single 450 MHz chip.  The price is the same.
> > The machine will have 256MB of RAM and a 16MB RAID-5 controller (either
> > Mylex or AMI), and four Cheetah 9GB drives.
> 
> 
> If you are going to run a 2.0.x kernel, do *>NOT<* use an SMP solution. I
> would suggest you go to 2.1.x now ... find a kernel that is stable on your
> box (2.1.125 and .127 are good candidates) and use the SMP box. Everything
> will then migrate happilly over to 2.2 once it is released. If you MUST
> have the server now and MUST run 2.0.x kernels, use the single processor
> solution though you will be able to upgrade later.
> 
> Compromise: Get the SMP system and run a non-SMP kernel until you switch
> to 2.2.x

What is the basis for this advice?

Is 2.0.x SMP just a crappy performer because it doesn't have granular
locking?

I'm using 2.1.x on my Alpha box and it seems stable.  I'm a little
hesitant to run it on a production machine, but I can give it a shot.

With the kernel issue out of the way, I'm still not sure whether SMP
is the better solution.

Thanks,

Jeff

[SMP] P2 Database Srvr: 1x450 or 2x350 CPUs?

[Jeff Noxon]

I'm setting up a Debian box for my client as an Oracle server.  I'm torn
between Dual 350's or a single 450 MHz chip.  The price is the same.
The machine will have 256MB of RAM and a 16MB RAID-5 controller (either
Mylex or AMI), and four Cheetah 9GB drives.

Linux kernel will probably be 2.0.36pre, at least until 2.2.x is solid
enough for production use.  Our other Linux server had a 450-day uptime
and I'd like this one to be just as stable, if not better.  :-)

Single 450: No need for SMP kernel.

Dual 350: Extra overhead of SMP kernel.  More "raw" power.

I'm guessing they'd be about equal in performance.  With a single 450 I
can always add another CPU later if necessary.  Oracle couldn't offer any
advice.

Is anyone using AMI's MegaRaid card on Linux?  It looks better than the
Mylex, but I trust Leonard's ability to write rock-solid drivers.

Any tips or comments would be great!

Regards,

Jeff

Re: xf86setup_3.3.2.3a-6

[Jeff Noxon]

On Tue, Oct 27, 1998 at 07:33:06PM +, Ruud de Bruin wrote:
> I am having problems in using xf86setup from slink. I am starting from
> scratch and have choosen a "complete development environment" during the
> initial setup. I am trying to get all the programs from slink via apt.
> 
> I am having problems with xf86setup. This programs refuses to start
> because many programs are not yet installed on my HD. However, it is very
> difficult to find the programs that are needed on the slink distribution.
> (E.g. on one occasion it needed a README file).
> 
> Can someone tell me which .deb files need to be installed before xf86setup
> can succesfully be launched?

I tried setting up a machine for a friend last night and ran into the
same problem.  I'm guessing that new X packages will solve these problems,
since we've been warned to put X on hold.  ;)  One thing I noticed is that
there don't appear to be many (or any?) of the necessary dependencies
between the new X packages.  Also, the old X font packages are still on
the FTP site.

I managed to get things running, but not with x86setup.  I had to tweak
a lot of stuff by hand and I had to keep going back to find yet another
X package...

J

Re: SOLVED: 2 XServers on 1 terminal?

[Jeff Noxon]

On Tue, Oct 20, 1998 at 06:16:40PM -0500, Alexander Kushnirenko wrote:
> Hi, 
> 
> Disregard my previous message.  I already learned how to do it.
> 
> > Could I run 2 X servers on one comupter?  So that Alt-F7 correspond to 
> > first 
> > session, and Alt-F8 for the second session?  There are 2 people working on 
> > one 
> > computer, and we don't want to mess each other X sessions.
> 
> Sasha.

How?

Thanks for the enlightenment,

Jeff

Re: DRIVERS FOR DFE-530TX (D-LINK)

[Jeff Noxon]

On Fri, Oct 16, 1998 at 01:48:27PM +0100, jrivas wrote:
> Anybody knows where i can get DFE-530TX (D-Link) drivers?

Isn't that one using a DEC chipset?  If so, try the tulip driver.

Regards,

Jeff

--
It's time to close windows and open source.
Linux is a trademark of Linus Torvalds.

Re: Oracle

[Jeff Noxon]

On Fri, Oct 09, 1998 at 10:12:33PM -0400, Bedrock LAN Administrator wrote:
> alien -i oracle.rpm
> (or whatever the filename turns out to be)

Oracle is a ~145mb .tar.gz file.  It has its own installer.

Regards,

Jeff

--
It's time to close windows and open source.
Linux is a trademark of Linus Torvalds.

Re: Oracle

[Jeff Noxon]

On Fri, Oct 09, 1998 at 11:25:56AM -0400, Alex Yukhimets wrote:
> Debian is not a company to "sign up with". You can use Oracle on Debian with
> no problems whatsoever. It is even distributed in plain tarballs and not
> rpms. You can download it right now, if you want.

I have a problem with that.  I'm using Debian in a corporate setting.
We started with Slackware in 1995 for a low-budget project.  We were
forced to stop using it in 1997 and migrated that machine to AIX,
which is our standard platform.  Now, thanks to the good Linux press,
we're using Debian on a few machines with plans to deploy more.  One of
those will need to run Oracle.  If companies are only going to support
RedHat or one of the other commercial distros, Management is going to
tell us that we need to run RedHat, not Debian.

We may both realize that there isn't any problem technically with running
Oracle (or any other app) on Debian.  But that's not the issue.

I wish decisions were made on technical merit alone, but this is the
real world.  For that reason, it would be nice if Debian had a marketing
interface to corporations like Sybase, Oracle, etc.  And we need to know
key things like how big our user base is.  The problem is that this is
a full-time job...  :-(

Regards,

Jeff

--
It's time to close windows and open source.
Linux is a trademark of Linus Torvalds.

Re: Do I have a win-sound card?

[Jeff Noxon]

Try using isapnptools to configure the sound chip.  It's probably an ISA
PnP device.

Jeff

Re: Off Topic: Samba

[Jeff Noxon]

On Fri, Oct 02, 1998 at 04:28:09PM -0400, dpk wrote:
> I have searched the Debian and Samba website for answers, as well as
> /usr/doc, and newsgroups for a resolution.
> 
> I would like to use encrypted passwords, but not have to maintain a
> samba password file as well as /etc/passwd.  Is it possible to do this
> with Debian/samba?  If so, could someone point me in the right
> direction of sites and solutions I should research?

I do not believe this is currently possible.  Unix passwords are one type
of one-way hash, and Samba passwords are another.  It might be possible
someday when Debian supports PAM.

I guess a workaround would be a front-end passwd program that changes both
passwords.  I don't know if one exists.

Someone please correct me if I'm wrong.

Regards,

Jeff

Linux tools for Atmel AVR?

[Jeff Noxon]

Does anyone know of Linux tools for the AVR?

Anyone interested in helping me write some?  I'd like to see a macro
assembler, simulator, and programming software.  None of that stuff
should be terribly complicated.

Thanks

Jeff

Re: Apache mod_ssl

[Jeff Noxon]

On Wed, Sep 16, 1998 at 05:32:49PM +0100, Oliver Thuns wrote:
> >> No but there is apache-ssl, wich is also a debian package (in hamm), find
> >> it at www.debian.org -> packages -> search  I think it's in the non-US
> >> section
> >
> >Does anyone know why it's such an old version of Apache?
> 
> There should be 1.3.1/SSL in unstable (slink).

It's 1.2.5.

Jeff

Re: Apache mod_ssl

[Jeff Noxon]

On Wed, Sep 16, 1998 at 04:11:28PM +0200, Ulisses Alonso Camaro wrote:
> On Wed, 16 Sep 1998, Oliver Thuns wrote:
> > Is there a Apache mod_ssl package under development?
> 
> No but there is apache-ssl, wich is also a debian package (in hamm), find
> it at www.debian.org -> packages -> search  I think it's in the non-US
> section

Does anyone know why it's such an old version of Apache?

Thanks

Jeff

Re: GTKICQ

[Jeff Noxon]

On Wed, Sep 16, 1998 at 06:30:27AM +, Phillip Neumann wrote:
> Is was a dep problem. I found a directory inside another directory called
> .deps. I remove it and the compilation went ok. Why is this problem? The
> conculson of this is that the maker of this program used redhat
> .?

My conclusion is that their dependency mechanism is broken.  Normally
(IMHO) dependencies should be based on program-specific header files,
and exclude the system-wide ones.  They aren't necessarily in the same
place on every system, as you found out...

Jeff

Re: Softmodems and Debian

[Jeff Noxon]

On Tue, Sep 15, 1998 at 12:11:40PM -0400, Person, Rod wrote:
> I FINALLY got Debian up and running to my liking with the help of a few
> if not more of you guys thanks. 
> 
> But here is something that hit me on the way to work today. I have a
> Digicom Softmodem. I have to load an algorithm before I can use it. In
> DOS it was quite simple.
> 
>   path=c:\smodem;c\
>   dl data144b.xtd
> 
> That's all my autoexec.bat needed. I just realized that is why Seyon is
> hanging for me. Does anyone out there use a softmodem? Are the loader
> and algorithm ported to Linux (I know the answer is no!!). I think
> I'm going to have to load the DOS algorithm from Debian. How do I do
> this? After all I'm still a newbie.

Two possibilities come to mind:

1) Boot DOS first, and load linux from loadlin instead of LILO.  You can
use FreeDOS if you don't want to use DOS.

2) Use dosemu to load the code after Debian has booted.

I used #2 once to get a Digicom modem working.  I ended up returning it
to the store anyway.

Jeff

Unable to create a boot disk for hamm install

[Jeff Noxon]

I'm forwarding this for a friend.  You can reply directly to me or to
the list.  I already told him how to boot from the rescue disk and the
CD-ROM, but I've never seen this problem before and we would appreciate
some insight.

Thanks!

Jeff

---

I got stuck on my installation last night.  I can't get my floppy
to create a boot floopy.  As such, I'm afraid to reboot my system to
continue the installation since I have not way of rebooting it.  *smile*
When it tries to create my boot floppy, it gives me:

  "Problem -- Creation of boot floppy failed.  Please make sure that the
  floppy was not write-protected, and that you put it in the first drive. 
  Try another floppy if the problem persists." 

I've tried several disks and they all give me the same error.  I've tried
"mke2fs /dev/fd0 1440" and while it is writing the inode table I get:

  "Warning: could not write 8 blocks in inode table starting at 5:
  Attempt to write block from filesystem resulted in short write".  

This error persists starting at "13", "21", "29", "37", and "45" and ends
with "ext2fs_mkdir: Attempt to write block from filesystem resutled in
short write while creating root dir".  Thus when I try a "mount -t ext2
/dev/fd0 /floppy" I get 

  "mount: wrong fs type, bad option, bad superblock on /dev/fd0, or too 
   many mounted file systems"

These errors are consistant across many disks.  I can however mount the
disk I booted the system off of orginally (fat32).  All other disks don't
seem to work.

Any suggestions?

Re: Where's my mouse?

[Jeff Noxon]

On Fri, Sep 04, 1998 at 08:37:04AM -0700, Kenneth Scharf wrote:
> 
> I change out my serial mouse for a ps2 mouse, and now my mouse
> can't be found.
> What do I do now?
> 
> -
> In 'X': re-run XF86Setup and select /dev/psaux for your mouse port.
> 
> also set up a sym link between /dev/mouse->/dev/psaux

Look at /proc/misc.  Do you have a line like this?

  1 ps2aux

If not you need a kernel with PS/2 mouse support.

Jeff

Re: Problem with superformat

[Jeff Noxon]

On Thu, Sep 03, 1998 at 10:33:49PM +0800, zjdwdz wrote:
> When I used superformat I got an error message:
> 
> ...
> mformat error.
> mformat: command not found
> 
> Why did this happen? How to solve it? I've installed the fdutils package. TIA!

You need the mtools package.  fdutils recommends it.  You can ignore
the error if you don't need a DOS (FAT) filesystem on the floppy.  It
seems to me that this error could be cleaned up a bit.  :)

Jeff

Re: Linux BOOTP server: questions

[Jeff Noxon]

On Wed, Sep 02, 1998 at 01:38:18PM -0500, Nathan E Norman wrote:
> I wonder what "when it discovers that the configuration file has
> changed" means.  Does it "keep an eye" on the file?

Yes.

Jeff

Re: autonice or default priority of program

[Jeff Noxon]

On Sat, Aug 29, 1998 at 03:37:09PM -0500, Torry Akins wrote:
>Is there a way to autonice a program when it is run?  We use matlab
> running on debian linux.  When a couple of these memory hogs run, the X
> display gets really choppy?  It would be nice if matlab was set to a
> lower priority when it is loaded.

rename it:

$ mv matlab matlab-real

And wrap a script round it:

#!/bin/sh
nice -whatever /usr/bin/matlab

Jeff

hamm: hwclock y2k problem

[Jeff Noxon]

FYI, to those whom may be concerned:

There appears to be a bug with the "hwclock" program and y2k wraparound,
on at least one system.

I have several systems using the AMI model 721 motherboard.  You'd think
that AMI, being a BIOS manufacturer, would get this right...

In any case, this board wraps to 1980 in y2k.  Any attempt to display
the date with hwclock fails in mktime() after the rollover.  The system
time does correctly wrap to 2000.  /proc/rtc shows 2000.  Rebooting the
system puts it in 1980, however.

It seems to me that /proc/rtc and hwclock would show the 1980 date if
they were functioning correctly.  The only workaround for this bug I
can see is to put a crontab entry in the system to have it rewrite the
CMOS time after the wraparound occurs.  Other Y2K testing has shown that
other Y2K dates do not present a problem for this system.  Even y2k leap
years work.  Odd.

Thanks,

Jeff

--
Please Cc: me on replies sent to mailing lists.

PHP 3.0.3 available on my webpage

[Jeff Noxon]

I couldn't wait any more for a PHP release that works with Apache 1.3.1,
so I did it myself.  You can download it from my homepage:

http://www.planetfall.com/~jeff/debian/

This is PHP 3.0.3 for Apache 1.3.1 (Slink).  I was going to build it for
hamm, but I noticed that Apache 1.3.1 is no longer in stable-updates.

I have built binaries for Alpha *only* ... It's tested and working,
and should be trivial to compile on x86.  All the necessary changes have
been done.  If anyone builds x86 binaries, please send them to me.

Note that I'm not a developer, and there are certain to be problems with
this release -- particularly with dependencies.  I've done the best I
could do in a limited amount of time, and it works for me.

Enjoy,

Jeff

Upgrade bo->hamm w/SOCKS proxy, how?

[Jeff Noxon]

I have a friend running bo, which he installed from CD.

How can he upgrade to Hamm through his NT proxy server, running SOCKS?
I can think of several difficult ways, but is there a beginners-approved
way?

Thanks!

Jeff


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

Re: high end SCSI card recommendation

[Jeff Noxon]

On Thu, Jul 09, 1998 at 08:10:08PM +0200, Jens Ritter wrote:
> Allen Ahoffman <[EMAIL PROTECTED]> writes:
> 
> > Anyway, if this is a controller issue, whats the best controller for busy
> > web servers running Debian LInux 1.3, kernel 2.0.33 and apache?
> > I can just get one and be done with it, recompile and that particular one,
> > and ...
> 
> I heard a rumor that buslogic cards are the better supported cards in
> linux as buslogic itself writes the drivers. 
> 
> Can this be confirmed?

BusLogic drivers are not written by BusLogic, although BusLogic has
provided equipment, test facilities, engineering assitance, and so on.
The BusLogic driver is rock solid.

For a strictly budget-minded user, Advansys is not a bad choice.
They write their own drivers.  If you can spring a few more bucks and
get BusLogic, though, I highly recommend their products.

Jeff


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

Re: PHP3 and PGSQL

[Jeff Noxon]

You need to edit /etc/php3.ini.  RTFM in the /usr/doc/php3-pgsql
directory.  (Look at README.Debian.)

It works great.

On Wed, Jul 08, 1998 at 10:43:18PM +0200, Adam Heczko wrote:
> Hi everybody !
> 
> I've got problem running my *.php3 files, in which I use PostgreSQL as a
> database server. Generally pg_connect("localhost", "", "", "",
> "dbname"); function generates following error message :
> Fatal error: Call to unsupported or undefined function pg_connect() in
> /var/www/my_file.php3 on line 15
> 
> My system configuration is : 
> 
> 
> // from dpkg -l :
> 
> ii  apache  1.3.0-2Versatile, high-performance HTTP
> server
> ii  php33.0-2  A server-side, HTML-embedded
> scripting langu
> ii  php3-doc3.0-2  Documentation for PHP3
> ii  php3-pgsql  3.0-2  PostgreSQL module for PHP3
> ii  postgresql  6.3.2-8Object-relational SQL database,
> descended fr
> 
> 
> // from /etc/apache/httpd.conf :
> 
> LoadModule php3_module /usr/lib/apache/1.3/libphp3.so
> 
> //from postmaster.init :
> 
> PGALLOWTCPIP=yes
> PGPORT=5432
> 
> All files resides in their default (Debian) locations. Could anybody
> help me ?
> 
> Thanks,
> 
> Adam Heczko.
> 
> 
> --  
> Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null
> 


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

Re: Mirrored Drives?

[Jeff Noxon]

On Fri, Jul 03, 1998 at 06:13:35PM -0400, Chris Brown wrote:
> I'm looking to increase the reliability (fault tolerance) of a system 
> that's using IDE drives.
> 
> Is there a way to add a second identical drive and have the the OS 
> mirror the data on the partitions? such that if one drive actually 
> crashed, the OS would keep running or at least be restarted with 
> little data loss?

Take a look at the mdutils package.  You can do mirroring (RAID 1), but
you'll have to re-partition your disks.  I doubt it can continue to run
during a hardware failure, though, and it's a pain in the butt to boot
from a Linux md (RAID) device.

Newer kernels offer RAID 5, which may be a better solution.  I think
you need three or more disks.  Patches are probably available for 2.0.x
kernels.  For this, look at the raidtools package.

Jeff


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED] < /dev/null

Re: postgres intallation failed

[Jeff Noxon]

On Mon, Jun 29, 1998 at 06:29:17AM +0200, Oliver Elphick wrote:
> The bug is not in base-passwd but in postgresql.  I used adduser to create
> the account if it did't exist, but I had not noticed that `adduser --system'
> assigns /bin/false as the shell.  I have changed postgresql-6.3.2-11
> to use useradd instead.  This is bug#24036.

If you don't use --system, will the uid still be <1000?  And shouldn't this
uid be consistent across multiple Debian installations?

Has someone already closed the base-passwd bug, or should I?

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: postgres intallation failed

[Jeff Noxon]

On Sat, Jun 27, 1998 at 11:14:19AM +0200, Networking Wizard wrote:
> After Shared PostgreSQL library intsallation (libpgsql 6.3.2-8)
> i've been trying to inttall postgresql (6.3.2-8) severl times,
> but i did not succeed; dpkg terminates because of unspecified errors.
> Here is the log:
> 
> > Unpacking replacement postgresql ...
> > Setting up postgresql (6.3.2-8) ...
> > Now installing the PostgreSQL database files in /var/postgres/data
> > su - postgres -c 
> > PATH=/usr/uxs:/root/uxs:/root/bin:/sbin:/usr/sbin:/usr/local/sbin:/bin:/usr/bin:/usr/local/bin:/usr/X11R6/bin:/usr/lib/postgresql/bin:/usr/lib/postgresql/bin;
> >  initdb -l /usr/lib/postgresql/lib -r /var/postgres/data -u postgres
> > dpkg: error processing postgresql (--install):

[snip]

I think the bug is in base-passwd.

You need to edit /etc/passwd so that the shell is /bin/sh or similar.  I
filed a bug report on base-passwd about this.

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Postgres, php question

[Jeff Noxon]

On Thu, Jun 25, 1998 at 12:27:56PM +1000, Hamish Moffatt wrote:
> On Wed, Jun 24, 1998 at 07:51:19AM -0400, Paul McDermott wrote:
> > hello my debian user friends,  I've installed Postgres.deb, php3.deb,
> > apache.deb and www-pgsql from the slink area. I've done the postgres
> > tutorial and it was great.  Now i want to start on makeing a web interface
> > for a database i put together.  I read through the documentation, but i am
> > confused.  Is there a tutorial for the php3 program?  If so where can i
> 
> That's why I got in to www-sql in the first place -- it is easier to use/learn
> than php but is less flexible as a result. I have done some reasonably
> advanced stuff with www-sql and have created some big messes to do it.

FWIW, I read the beginning of this thread this morning, then installed
php3 and php3-pgsql, and now have some wickedly nice code.  I don't know
how I ever lived without php3 before!  It is so sweet!  I've already
replaced most of my shtml and CGI with it.  The learning curve is very
short if you know C.

If anyone needs any specific help, please let me know.

To make php3-pgsql work, you have to enable it in /etc/php3.ini.  Then
you just do simple stuff like:



and so on.

I still haven't done a form with it yet, but that's next on my agenda.

One of the niftier things you can do is write functions and include them
from any web page...  I'm impressed.  It's fast, too.

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Dvorak keyboard in X?

[Jeff Noxon]

On Fri, Jun 19, 1998 at 02:00:33PM -0700, Keith Beattie wrote:
> Ian Keith Setford wrote:
> > 
> > I was wondering if the current xbase has support for a "dvorak" type
> > keyboard. If so, is it an option within xf86config?
> > 
> 
[snip]
> Perhaps there is another way to compensate for a "Dvoraked" keyboard
> (one where the keys have been re-arranged) but my guess is that would
> have to happen at the device-driver level.
[snip]

You can do this with the kbd package.  AFAIK there is no need to even
mess with X.  kbd comes with dvorak-l and dvorak-r key translation tables.
Try "man loadkeys"

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: staroffice

[Jeff Noxon]

On Fri, Jun 19, 1998 at 03:56:00PM -0400, Patrick Ouellette wrote:
> Ok, I messed that up.  Version 5 is scheduled for the second quarter.
> (It has only been 6 or 7 years since I had to *use* my poor excuse for
> German)

Second quarter of 97 or 98?  (Or if they were truly emulating Microsoft
it would be 2Q96.)

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: staroffice

[Jeff Noxon]

On Fri, Jun 19, 1998 at 11:59:53AM -0700, Brian Weiss wrote:
> > StarOffice 4 is commercial -- i.e. not free.  IMHO it's well worth the $100
> > since it's virtually a clone of MS Office.  It's a bit sluggish though.
> > 
> > Jeff
> 
> That's not correct. I downloaded the full software package from their site
> for absolutely nothing. It's not an evaluation copy and doesn't require
> you to register or spend money in any way. Try downloading it from their
> site and if you still have trouble getting it drop me an E-mail and I'll
> see what I can do.

If it's an evaluation copy, that means it's not free.  It's for sale
all over the place, and I remember reading a statement from the company
that the Linux version is no longer free -- although it may be possible
to download (and apparently is).

StarOffice 3 was just an experiment to test the viability of a Linux
port.  It was distributed under different terms.

Someone please correct me if I'm wrong.  I don't remember if this info
came from C.O.L.A. or Linux Journal.

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: staroffice

[Jeff Noxon]

On Fri, Jun 19, 1998 at 01:13:49PM -0400, Brian Morgan wrote:
> Where is that Staroffice4 that you're talking about?  I can only find 3.1-8 in
> hamm/contrib/binary-i386/editors.

StarOffice 4 is commercial -- i.e. not free.  IMHO it's well worth the $100
since it's virtually a clone of MS Office.  It's a bit sluggish though.

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Apple II Emulator

[Jeff Noxon]

On Wed, Jun 17, 1998 at 09:13:44AM -0400, [EMAIL PROTECTED] wrote:
> Bad newsI can't find xgs anywhere!
> I just spent some time digging around the web, I found its
> offcial homepage is gone...I was able to find the authors homepage
> but it contained no mention of xgs, all I could fine was xgs-dos, and
> running a GS emulator, through dosemu, thats just too twisted even for me.
> -Steve

ftp.apple.asimov.net.

Jeff


--  
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

NTSC Composite TV Output from X or svgatextmode?

[Jeff Noxon]

I'm looking for a way to get NTSC composite video from my computer.
I know that a number of external converter boxes are available, but
I'd rather have it come directly from the video card, if possible.
Do any cards support this under Linux?

I tried an ATI Xpression PC2TV card, and it only worked in 80x25 text
mode -- and looked like crap on my TV.  I'd rather use 40x25 for this,
or possibly 320x200 under X.  I'm developing a custom home theater
application under Linux...

Thanks

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Still can't sort out MBR -- Linux and NT

[Jeff Noxon]

On Sat, May 09, 1998 at 05:54:08PM +0100, Tristan Day wrote:
> When I load from a boot disk (created by "format /s" at dos prompt in
> Win95), fdisk doesn't work because it doesn't exist, thus
> 
> fdisk /MBR
> 
> doesn't work. If I try this in linux, it says MBR not found  case too>

Of course.  You'll have to copy fdisk.exe onto the boot disk...
Actually, the best thing to do is to use Win95's boot disk utility to
make a boot disk.  Use the help facility, search for boot disk, and
you're on your way.  It will copy a whole bunch of utilities to the
floppy.  Everyone who runs Windoze should have one or two handy.

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Books or Debian wrong? and other stuff

[Jeff Noxon]

On Fri, May 08, 1998 at 02:05:21PM -0500, Pete Harlan wrote:
> > Kermit is full of bugs, and hamm does not have a current version.
> 
> Kermit is easy to download, compile and install, and works well.
> Ctrl-\ is the escape character, but you have to follow it with another
> character to cause anything to happen.  For example, Ctrl-\ C will get
> you back to the kermit prompt, and Ctrl-\ ? will give you a list of
> other escaped commands.

I think some of the issues I've had with kermit are libc6-related,
because libc5 versions always worked well for me.  Among other problems,
I've found it impossible to exit from kermit on occasion.  I have to
put it in the background and kill it.  :)

Thanks

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Books or Debian wrong? and other stuff

[Jeff Noxon]

On Fri, May 08, 1998 at 01:07:44PM -0400, Thomas J. Malloy wrote:
> When I, an linux and unix novice, find that commands I am entering are not
> yielding the results I expect how do I know if this failure is caused by a
> program bug, an error in the book or man page, my error or something else?
> For example on page 104 of "Learning the Bash Shell" O'reilly there is the
> following command
>  vi $(grep -l 'command substitution' ch*)
> 
>  According to the text should load into the vi editor a file that is a list
> of the files in the PWD that begin with "ch" which contain the string  
> "command substition".  The man page for grep would seem to confirm this
> However when I typed
>  vi $(grep -l 'linux' *.txt) 
> it loaded all the documents into vi not a list of documents.  Is  the book 
> wrong?

The book is wrong.  Debian's behavior is correct.  You'd have to save the
list of files _to_ a file before vi would be able to read it as a list:

grep -l linux *.txt > tmpfile
vi tmpfile

Instead, what you're doing is feeding the output of the grep command to
the command line.  Try this for clarification:

echo $(grep -l linux *.txt)

Or this:

echo `grep -l linux *.txt`

> And as long as I am here, I have noticed that the escape charactor in 
> kermit does not work ^\.  Neither does there seem to be anyway to exit 
> dosemu other than killing the process.

Kermit is full of bugs, and hamm does not have a current version.

You can exit dosemu by running the "exitemu" program.

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Diskless Debian, Shared /usr, etc...

[Jeff Noxon]

On Thu, May 07, 1998 at 03:10:57PM -0600, Marcelo E. Magallon wrote:
> On Sun, May 03, 1998 at 09:13:58PM -0500, Jeff Noxon wrote:
> 
> > Thanks -- that was an excellent idea.  I had to massage nfsroot into
> > working with libc6, but it was a great starting point.  My client is
> > now taking up 2100K -- not bad.
> 
> Could you plubish the patched nfsroot somewhere? It would be useful for
> quite a few ppl.

That's opening a can of worms.  :-)

I haven't fixed it, but I did use it to figure out what I needed to make
everything work.  In my opinion, the package is far more complicated
than it needs to be -- at least for most people.

If you want to use it (or fix it), the main thing to look at are the
different libraries.  Hamm uses libc6, while nfsroot still tries to
use libc5.

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Debian 1.31

[Jeff Noxon]

One other thing to keep in mind is the tendency Microsoft software has
to rewrite the boot record (i.e. remove LILO).  So always keep a boot
disk handy -- you never know when you'll need it.  (Like after that
Win98 upgrade...)

:)

Jeff

On Thu, May 07, 1998 at 04:55:41PM -0400, Will Lowe wrote:
> On Thu, 7 May 1998, Jeff Noxon wrote:
> 
> > off the secondary drive?  Is there a graceful and free way to do this
> > without installing LILO in the primary drive's boot sector?
> 
> I don't think so.  But you do
> boot=/dev/hda
> root=/dev/hdb1
> 
>   for your linux kernel stanza and 
> 
> boot=/dev/hda1 
> 
>   for your win95 stanza and (I think) you should be ok.  I
> dual-booted linux off the second hard drive for a while with Win95 on the
> first disk using lilo and didn't have any problems.  I've since deleted
> that lilo.conf,  so I don't remember exactly how I did it,  other than
> that it was pretty straightforward.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Debian 1.31

[Jeff Noxon]

On Thu, May 07, 1998 at 04:42:23PM -0400, Will Lowe wrote:
> > I want to have the ability to dual boot with Debian and Windows 95 ? (I
> > know with Windows NT and Windows 95 this is possible). Will such thing
> > be possible Debian & Windows 95?

> Absolutely.  Lots of us do this all the time.  Very easy to configure and
> use.

It might be helpful to explain how to do it.  I know the default
installation will only let you boot off the primary drive, as LILO is
installed into the partition and not the boot sector.  It's always easy
to boot from the boot disk made during installation, but booting Linux
off the secondary drive?  Is there a graceful and free way to do this
without installing LILO in the primary drive's boot sector?

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Radius Server Authentication

[Jeff Noxon]

On Tue, May 05, 1998 at 02:02:26PM -0500, Matthew D. Myers wrote:
> Ok... what is PAM ?

Pluggable Authentication Modules.

Install the pam-doc package and go from there.

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Radius Server Authentication

[Jeff Noxon]

On Tue, May 05, 1998 at 01:36:54PM -0500, Matthew D. Myers wrote:
> I would like to know if there is some way to make linux authenticate telnet
> and ftp sessions from a radius servers' user list?

The answer is probably, if you recompile login, ftp, etc. to use PAM,
and then configure PAM appropriately.  But it sounds like a big pain,
and I'm not sure how you'd get UID's and GID's from a RADIUS server.
You'd still need /etc/passwd and /etc/group entries.  And you'd need to
provide a way for users to change their passwords... and so on.

Sounds like a better idea might be to hack RADIUS to use NIS for
authentication.  But then again I can only guess about what you're
trying to accomplish.

Good luck,

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Diskless Debian, Shared /usr, etc...

[Jeff Noxon]

On Sat, May 02, 1998 at 10:57:39AM -0700, Jim Pick wrote:
> I use Joost's nfsroot package, which sets most of the stuff up.  It's
> a good start - but You still need to do a fair amount of hacking to
> get it to work.
> 
> I think the package got wiped out by the latest freeze, so you need to
> fetch the source out of /project/orphaned.

Thanks -- that was an excellent idea.  I had to massage nfsroot into
working with libc6, but it was a great starting point.  My client is
now taking up 2100K -- not bad.

I also found a WaveLAN packet driver and burned a boot ROM using the
netboot package.  It works!  How exciting...  :)

Jeff


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]