Linux-Misc Digest #39, Volume #27 Tue, 6 Feb 01 14:13:01 EST
Contents:
Re: 1 screen, 1 keyboard, 2 copmutors (Robert Heller)
Re: How to display texinfo help pages? (Jean-Pierre Moreau)
Re: RedHat 7.0 and WABI ([EMAIL PROTECTED])
Re: Disk space not marked as free? ("Julius Longauer")
Resources available in csh vs. bash in cgi's under Apache (Bill Cripe)
Re: How to display texinfo help pages? (Bob Tennent)
Re: Won't compile anymore!!! ([EMAIL PROTECTED])
Re: setup issue: boot partition size too big ("Jeremy Paiz")
partitioning problems (Holo)
Re: Which Linux distribution is best? ([EMAIL PROTECTED])
Re: How to check for process existence ("Chris Coyle")
PCI: "the same IRQ is used by device" problem ([EMAIL PROTECTED])
Re: Linux for a Portable Dell (Per Weisteen)
Re: problems with kernel 2.4.1 - me too (Joe)
Re: setup issue: boot partition size too big (Mark Bratcher)
Re: Samba 2.0.7 and large files (>2 GB) ([EMAIL PROTECTED])
Re: Converting text files from Microsoft to Linux and Linux to Microsoft ("Jeff
Susanj")
Re: make menuconfig (Tim Limbert)
Re: Unpacking ISO-images without a CD burner (Richard Snow)
----------------------------------------------------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: 1 screen, 1 keyboard, 2 copmutors
Date: 6 Feb 2001 10:08:34 -0600
Natanael Copa <[EMAIL PROTECTED]>,
In a message on Tue, 6 Feb 2001 10:51:28 +0100, wrote :
NC> I wonder how I connect on terminal to another computer. Alt-F1 to Alt-F6
NC> gives the tty1-6 but I want to redirect tty6 to ttyS0. So when I press
NC> Alt-F6 I see the login screen on the second box.
NC>
NC> How? (Yes I have read mgetty docs)
This is *exactly* what I do with my print server, smaug:
In /etc/rc.d/rc.local:
/usr/sbin/rconsold &
/usr/sbin/rconsold contains:
#!/bin/tcsh -f
while (1)
minicom </dev/tty8 >&/dev/tty8 smaug
end
/etc/minirc.smaug contains (actually a symlink to minirc.ttyS16):
# Machine-generated file - use "minicom -s" to change parameters.
pr port /dev/ttyS16
pu baudrate 9600
pu rtscts No
(/dev/ttyS16 is a port on my Boca BB-1004 4-port serial card -- I have
ttyS0 busy as my serial mouse and ttyS1 busy with my USR Courier
V.Everything modem.)
/dev/ttyS16 is wired via a Null modem cable to /dev/ttyS0 on smaug -- an
old '486 Dell box with two parallel ports, an AHA-1540B SCSI controller,
a 4gig SCSI disk, and a WD EtherNet card. Smaug runs RH 6.2 and its
lilo.conf looks like:
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
default=linux
serial=0,9600n8
image=/boot/vmlinuz-2.2.12-20
label=linux
initrd=/boot/initrd-2.2.12-20.img
read-only
root=/dev/sda1
append="console=ttyS0,9600n8 console=tty0"
note the 'serial=0,9600n8' and 'console=ttyS0,9600n8' options. I have
also didled with stuff in /dev:
smaug.deepsoft.com% dir -l /dev/console /dev/ttyS0
lrwxrwxrwx 1 root root 5 May 4 2000 /dev/console -> ttyS0
crw------- 1 root root 4, 64 Feb 4 19:08 /dev/ttyS0
And /etc/inittab:
smaug.deepsoft.com% grep S0 /etc/inittab
con:2345:respawn:/sbin/mingetty ttyS0
While sitting at my main box I can hit Ctrl-Alt-F8 and presto, smaug's
console. I only actually need to do this if I need single user mode on
smaug or such like.
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
Posted Via Nuthinbutnews Premium Usenet Newsgroup Services
==========================================================
** SPEED ** RETENTION ** COMPLETION **
==========================================================
http://www.nuthinbutnews.com
------------------------------
Subject: Re: How to display texinfo help pages?
From: Jean-Pierre Moreau <[EMAIL PROTECTED]>
Date: Tue, 06 Feb 2001 16:12:09 GMT
Jean-Pierre Moreau <[EMAIL PROTECTED]> writes:
> David <[EMAIL PROTECTED]> writes:
>
> > Bill S wrote:
> > > . . .
> > > . . . [ All about info ]
>
> If path of YourApp.info file in INFOPATH, *and* YourApp.info cited in some
> INFOPATH file named 'dir' (like '/usr/local/info/dir'):
> $ info YourApp
> will work.
>
> If YourApp.info file not installed in INFOPATH, do
> $ info -f path/filename
>
> To see your INFOPATH, the one defined at 'info' command compile time,
> $ strings `which info` | less # / INFOPATH
> 'INFOPATH' is a list of `:' separated paths, and can be an environment var,
> I think.
>
> I do not know how else get the default INFOPATH of info command.
>
> To know how to navigate with info command: $ info info
>
> And if you know Emacs, 'C-h i' will bring you in some 'dir' (or all '.../dir')
> in your INFOPATH, or something like that.
And to add to it, I did not show you how to fish your fishes, i.e. read some
manuals too for the commands cited above:
$ man info
$ man strings
$ man which
$ man less # or more, or ... your preferred viewer.
--
Jean-Pierre Moreau,
[EMAIL PROTECTED]
(substitute/notuser/jpmoreau/ ; substitute/noexist/jfactory)
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: RedHat 7.0 and WABI
Date: 6 Feb 2001 16:23:24 GMT
That is all well and good, however, my use for wine is to run just two
apps at the office. WINE may have matured a lot since I last looked at
it (about a year ago), however, WABI has served me very well for four
years. The error I am getting is telling me that the binary (or more
likely some shared library) is missing. The binary naturally is not
missing. Furthermore, I receive the same error when trying to run an
old utility (unix2dos).
I certainly can investigate whether WINE will now run the two apps that
I need (lotus notes and an off-the-shelf trouble-ticketing tool), but I
would still like to know that I can fall back on running WABI too.
E J <[EMAIL PROTECTED]> wrote:
: WABI is pretty old and was tested to run the top 10 Windows programs at
: the time.
: You should switch to WINE. It is much improved and can run some of the
: current windows based game (suprise!!!)
: [EMAIL PROTECTED] wrote:
:> I have recently rebuilt my long-standing system which has run
:> several releases of slackware and then redhat 6.x. Since
:> the rebuild to RH7.0, I have been unable to run WABI (Sun's old
:> Windoze 3.x-under-linux environment). Has anyone else tried
:> to run WABI under RH7, and, if so, did it still work?
--
Chris Cox, N0UK/G4JEC NIC Handle: CC345
UnitedHealthGroup, Inc., MN10-W116, UNIX Services & Consulting
6300 Olson Memorial Highway, Golden Valley, MN 55427
email: [EMAIL PROTECTED] (work) [EMAIL PROTECTED] (home)
A computer without MS Windows is like a dog without bricks
tied to its head.
LINUX - aka Microsoft Windows/NT - The Final Service Pack
------------------------------
From: "Julius Longauer" <[EMAIL PROTECTED]>
Subject: Re: Disk space not marked as free?
Date: Tue, 6 Feb 2001 17:24:19 +0100
Fritz <[EMAIL PROTECTED]> writes
> Julius Longauer wrote:
>
> > Fritz <[EMAIL PROTECTED]> writes:
> > > Hi,
> > >
> > > I got the following problem:
> > > Having two Linux partitions, one at moutpoint '/' (hda9) and one at
> > > '/usr' (hda13), it appears that my root partition was filled up to
87%,
> > > so I decided to move the '/opt' directory (580mb) to a newly created
> > > partition. I did it like this:
> > > 1. move /opt to a spare place
> > > 2. delete /opt
> > > 3. create new /opt (md)
> > > 4. add new partition to mountpoint /opt
> > > 5. copy back old /opt /... to new
> > > I expected to have my disk usage for hda9 thus lowered by 580M, about
> > > 50% of the partition, however df still shows 87% of usage. On the
other
> > > hand,
> > I
> > > see 34% usage on hda12, the new partition.
> > > I verified that /opt now resides on hda12 (it does) and counted up
> > manually
> > > the values given by du for hda9 (which now are really about 300M), so
I
> > am
> > > suspecting that the disk usage for hda9 is not correct.
> > > Can anybody advise me ? I am not an experienced Linux user and might
> > > have
> > > done something wrong. How can I get rid of this? Are there any
> > > utilities to map directories and free space to partitions?
> > >
> > Let's guess: The whole contents of your former /opt directory is still
> > there on hda9 and consumes disk space but you (and du) can't see it
while
> > hda12 is mounted. So umount hda12 and look if there are any files in it.
> >
> > Julius
> >
> Thank you,
>
> I did as suggested, but no new perceptions...
> df -> hda9 87%, hda12 37%
> dir /opt -> all stuff
> umount hda12
> dir /opt -> only directory (1K)
> df -> hda9 87%...
> Besides this, I set up a little routine utilizing the 'stat' system call,
> which tells that /opt is on 3, 12 (hda12) when mounted, and on 3,9 (hda9)
> if not...
> Still hoping for a solution. Fritz
>
That's weird. Are you sure that you don't have the contents of
/opt floating around somwhere on hda9? Did you remove your
temp directory? What says 'du -sx / '? Try 'find' or 'locate' on
files you know exactly that there are only in /opt.
Julius
------------------------------
From: [EMAIL PROTECTED] (Bill Cripe)
Subject: Resources available in csh vs. bash in cgi's under Apache
Date: Tue, 06 Feb 2001 16:27:09 GMT
Recently I was moving some csh scripts used for cgi from an old SCO
machine to a RedHat 7.0 machine running an uptodate Apache server. One
of the scripts fires off a number crunching program that is
fairly resource intensive, at least as cgi's go. (By default it grabs
two megabytes of core at startup.) This program was core dumping when
called from within a cgi, but would run fine otherwise.
On this programs command line you can set the amount of RAM you want
it to use at startup, and I knew from past experience that setting this
to a low number, 200k for example, could induce a core dump. It seemed
plausible that Apache would limit the resources available to cgi
scripts, but I discovered that Apache's settings for RlimitMEM and
RlimitCPU default to the system's max. I tried setting these explicitly
nevertheless, but still got the core dump.
For some desperate reason I took a shot at converting the script
to bash, and this worked just fine. Does anyone have an explaination
of why in a cgi bash would aparently receive the resources it needed,
but csh would not?
Bill Cripe
--
------------------------------
From: [EMAIL PROTECTED] (Bob Tennent)
Subject: Re: How to display texinfo help pages?
Date: 6 Feb 2001 16:19:21 GMT
Reply-To: [EMAIL PROTECTED]
On Tue, 06 Feb 2001 14:59:44 GMT, Bill S wrote:
>I am running rh5.2. When I look for help in the man pages, I read the
>message "...the Texinfo documentation is now the authoritive source".
>My question is, what is the syntax for displaying texinfo pages?" How
>am I sure they are installed?
info command
will work but isn't very user-friendly. I suggest installing tkinfo,
which has a usable interface.
Bob T.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Won't compile anymore!!!
Date: Tue, 06 Feb 2001 16:23:27 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Juan Meneses) wrote:
> [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote in article
> <95mk64$je$[EMAIL PROTECTED]>:
> >I have a RedHat 7.0 machine
> >make Xconfig works
> >make clean/dep/bzImage won't work.
> >I get the "can't find rule" every time.
> >Should I use the kernel from kernel.org ?
>
> Yes. (Or preferably from your nearest mirror.)
>
> --
> Juan Meneses \ "Ford, you're turning into a penguin.
> �Ay ay ay, la polic�a! \ Stop it." -- Arthur Dent
> [EMAIL PROTECTED] \ (Douglas Adams, "Hitchhiker's Guide
> http://www.iki.fi/juan/ \ to the Galaxy")
>
Well, turns out there is no rule because the whole /arch/i386/boot
directory is empty. How did that happen? you might ask. I don't know.
I'll just get the kernel from kernel.org and while I'm at it, upgrade to
2.2.18 instead of 2.2.16.
Thanks anyway.
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Jeremy Paiz" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions
Subject: Re: setup issue: boot partition size too big
Date: Tue, 6 Feb 2001 11:33:52 -0500
okay, so here's my plan: i'm going to configure lilo to boot either windows
or linux. in redhat linux 6.2 installation, lilo installation immediately
follows the partitioning part. i tried partitioning as normal "/",
"<swap>", "/usr", "/home" and "/var", along with "/boot" (which appeared in
red indicating it failed). when i clicked to move ahead to the lilo
installation, the partitioning dialog froze and i was forced to restart.
this happened twice more. my question now is: do i skip partitioning
altogether and proceed with lilo installation? do i leave out the "/boot"
partition? is there something else i should do?
--
________________________________________________________________________
JEREMY M PAIZ
Software Engineer
Research & Development Division
Welding Technology Corporation
24775 Crestview Court
Farmington Hills MI 48335-1507
Email: [EMAIL PROTECTED]
Web: http://www.weldtechcorp.com
Phone: (248) 477-3900 x3362
Fax: (248) 477-8897
Mobile: (248) 568-1592
------------------------------
From: Holo <[EMAIL PROTECTED]>
Subject: partitioning problems
Date: Wed, 07 Feb 2001 01:24:44 +0800
I've a harddisk with 2 windows partitions and two linux partitions. I
wanted to remove the linux partitions so that I can install windows on
the harddisk for my friend. After I used dos fdisk to remove all the
partitions, the dos fdisk shows that there is no partition defined, so I
rebooted the harddisk and tried to create a single dos partition but
wasn't able to format the partition created. What should I do so that I
can create a single dos partition on the harddisk.
Any suggestion is appreciated. Thanks
Regards
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Which Linux distribution is best?
Date: Tue, 06 Feb 2001 17:10:12 GMT
In article <95l5cj$s24$[EMAIL PROTECTED]>,
bhogak <[EMAIL PROTECTED]> wrote:
> I have a P133 with 32MB RAM and 10G HD, but my System Bios has this
8.4
> Gig barrier. I am not getting BIOS upgrade for the motherboard. When I
> try to install Win 98, at the time of installation, the scandisk
> freezes the machine.
> Would I have the same problem with Linux too? If not, which
> distribution is good?
>
> Thanks.
> bhogak
>
> Sent via Deja.com
> http://www.deja.com/
>
I first got mandrake, which was slow as shit, but it worked fine with my
Celeron 500, w/ 32MB ram, except StarOffice took upwards of 10 minutes
to load, so I don't use it that often anymore, except now, when I need
it to write papers since my printer doesn't work anymore(I installed
RH7.0, and my printer went to hell, never to return, until I finish my
College stuff....and the saga continues...). Everyone says not to run
KDE, but that is bullshit. I ran X, KDE, XMMS, and a shitload of other
apps at the same time, and very little slowdown.
Justin Hibbits
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Chris Coyle" <[EMAIL PROTECTED]>
Subject: Re: How to check for process existence
Date: Tue, 6 Feb 2001 12:35:09 -0500
"Jean-David Beyer" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Thomas Hahn wrote:
> >
> > OK, so I want to find out within my C program if process "foo" exists, and I
> > want to
> > do it efficiently. Doing system("killall -0 foo") or popen("/bin/ps -e") takes
> > too long
> > (like 30ms). I don't want to pay for execing a shell or starting a new process.
> >
> > I could probably grovel around and find some old ps source, but I was hoping
> > someone
> > could just go "I've done this, here's the code". Oh, and it needs to work on
> > Irix too.
> >
> Do you happen to know the pid of process foo? If so, you could kill(2)
> it and look at the return code. kill(pid, 0), so you do not annoy pid. I
> worked on a system with a very large (> 200) number of cooperating
> sequential processes, and we did this to see if particular ones were
> still alive.
>
Some processes write their pid into a file at startup.
If you own `foo' why don't you make it do this.
Then you can read the pid from the file.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.hardware
Subject: PCI: "the same IRQ is used by device" problem
Date: Tue, 06 Feb 2001 17:28:56 GMT
Hello,
My PCI subsystem (kernel 2.4.1) issues "the same IRQ is used
by device..." message when I try to use, for example, my network
card running the adsl-start script. The IRQ in question is 9 and
it's shared by my network card (3Com 3c905B "Cyclone"), the sound
card (SB 256 Live!) and the USB controller (VIA chipset); my
motherboard is Asus A7V.
Any suggestions how to get the PCI working? Thanks.
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Per Weisteen <[EMAIL PROTECTED]>
Subject: Re: Linux for a Portable Dell
Date: Tue, 06 Feb 2001 18:48:41 +0100
Angel Belda wrote:
> Do you know if I am going to have any problem to install RedHat 6.2
> with XWindow support.
>
> Thanks
I'm running RedHat 7.0 on a Dell Latitude CPi without any problems. X works ok
at 800x600.
--
Per Weisteen !Internet support, Norsk Hydro !Email: [EMAIL PROTECTED]
N-0246 Oslo, Norway !Phone: (+47) 2253 8227/ !Web: http://www.hydro.com/
------------------------------
From: Joe <[EMAIL PROTECTED]>
Subject: Re: problems with kernel 2.4.1 - me too
Date: Tue, 06 Feb 2001 13:14:00 -0500
Thanks,
That was it. I thought I had everything I needed, but I found out otherwise.
Got it compiled and running this morning.
-Joe Morton
------------------------------
From: [EMAIL PROTECTED] (Mark Bratcher)
Crossposted-To: comp.os.linux.setup,comp.os.linux.questions
Subject: Re: setup issue: boot partition size too big
Date: Tue, 06 Feb 2001 18:14:49 GMT
In article <95p91j$a7k$[EMAIL PROTECTED]>, Jeremy Paiz wrote:
>okay, so here's my plan: i'm going to configure lilo to boot either windows
>or linux. in redhat linux 6.2 installation, lilo installation immediately
>follows the partitioning part. i tried partitioning as normal "/",
>"<swap>", "/usr", "/home" and "/var", along with "/boot" (which appeared in
>red indicating it failed). when i clicked to move ahead to the lilo
>installation, the partitioning dialog froze and i was forced to restart.
>this happened twice more. my question now is: do i skip partitioning
>altogether and proceed with lilo installation? do i leave out the "/boot"
>partition? is there something else i should do?
>
>
Linux cannot run without /boot. You're going to have to get /boot in
there somewhere.
You must be choosing the "custom" or "expert" installation. Why don't
you pick one of the automatic ones ("Workstation" for example)?
--
Mark Bratcher
To reply, remove _UNSPAM from my email address
=========================================================
Escape from Microsoft's proprietary tentacles: use Linux!
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Samba 2.0.7 and large files (>2 GB)
Date: Tue, 06 Feb 2001 18:39:11 GMT
In article <MORf6.7075$[EMAIL PROTECTED]>,
"Nils O. Sel�sdal" <[EMAIL PROTECTED]> wrote:
> > P�r Lindahl
> if you do a ls -lh it is also 0sized?
> Could be you need some new file/utils ..
Thanks for your answer!
Yes, it comes out as 0 bytes even if I do a 'ls -lh'. I should mention
that according to my man-file for ls, there is no -h option, but it
seems to work anyway when I use it.
I checked the GNU fileutils homepage, and there is no newer version than
the one I am using (4.0), so I guess the error must lay elsewhere. :(
/P�r Lindahl
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Jeff Susanj" <[EMAIL PROTECTED]>
Subject: Re: Converting text files from Microsoft to Linux and Linux to Microsoft
Date: Tue, 6 Feb 2001 18:47:07 GMT
"Robert A. Knop Jr." <[EMAIL PROTECTED]> wrote in message
news:LYVe6.48573$[EMAIL PROTECTED]...
>
> Are you just talking plain text files? If so, there are utitilies
> "dos2unix" and "unix2dos" (you can doubtless find RPMs on
> www.rpmfind.net, if you like RPMs) which will convert the line terminator
> conventions and so forth.
>
In the DOS world I have seen similar utilities called U2D.EXE and D2U.EXE.
Jeff S.
------------------------------
From: [EMAIL PROTECTED] (Tim Limbert)
Subject: Re: make menuconfig
Date: 6 Feb 2001 18:47:15 GMT
[EMAIL PROTECTED] wrote in <[EMAIL PROTECTED]>:
>I installed the kernel in Easy Linux in SuSE Linux
>and in GoLinux but >>make menuconfig<< does not work
>in /usr/src/linux.GoLinux and Easy Linux said >>command
>not found<< and SuSE Linux said >> no rule to make menuconfig<<
>Do I have to install further
>packages?
>
I am having the exact same problem! Upgrading kernel in Turbolinux 6, I
get "command not found" when I 'make config'. Would sure like a solution
to this problem.
Tim Limbert
------------------------------
From: Richard Snow <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.admin,comp.os.linux.setup,comp.os.linux.questions
Subject: Re: Unpacking ISO-images without a CD burner
Date: Tue, 06 Feb 2001 19:01:48 GMT
[EMAIL PROTECTED] wrote:
>
> I try to download and install Linux on my computer, but I haven't got a CD
> burner.
> The only download methods I can find on the net is iso-images or all the
> individual files from FTP, which would take me days to download.
> So I'm wondering if there's any way to unpack or convert the ISO-images to
> individual files without having to burn a CD.
> Or if anyone know of a site to download SuSE Linux 7 as a ZIP or similar
> archive file.
> All help appreciated, and please no answers like "buy a CD burner".
>
> Audun
try www.cheapbytes.com and order a cd shipped to you. ONly a few
dollars plus shipping.
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
