Linux-Setup Digest #425, Volume #19 Fri, 18 Aug 00 15:13:15 EDT
Contents:
Re: Booting from a different kernel image (Jonathan)
Re: Help: Probme with Modem and Video Mode in new Kernel ("Peter T. Breuer")
Re: Linux on AMD (alex)
Re: installing lilo after setup (Craig Kelley)
Re: Can't get cgi to work? (easy question...) - why do I need "use CGI"? (Craig
Kelley)
Re: remote exexcution? (Craig Kelley)
Re: Where should be installed the LILO? (Marijan Peh)
Re: I've killed my RH 6.2!! Help!!! (Craig Kelley)
Re: Parallel ZIP drive problems (aflinsch)
Re: ProFTP config problem (Jeff Grossman)
Re: networking problem ("Sean")
ALR Quad 6 server and PPro Overdrives. (Kurt Savegnago)
Re: SCSI CD-ROM install (=?ISO-8859-1?Q?R=FCdiger?= Arp)
Re: KPPP connecting, hosts not found ("Gareth Cranny")
Is it a WINMODEM ?? ("KMX")
Re: can't set PATH (Colin Watson)
Re: Can't get cgi to work? (easy question...) (Colin Watson)
----------------------------------------------------------------------------
From: Jonathan <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux
Subject: Re: Booting from a different kernel image
Date: Fri, 18 Aug 2000 16:02:03 GMT
In article <[EMAIL PROTECTED]>,
jinp <[EMAIL PROTECTED]> wrote:
> I have a question on this one, I built a new kernel(on RedHat 6.0) and
> did all the lilo and link the System.map to the new file, and new
kernel
> booted ok, but the old one found a wrong System.map file. What should
> I do?
>
> Thanks
> James
>
> Akira Yamanita wrote:
>
> > [EMAIL PROTECTED] wrote:
> > >
> > > Hi,
> > > Can we specify which boot image to be used at LILO Prompt?
> > > Thanks in advance
> > >
> > > Sandy
> >
> > Sure. This is my /etc/lilo.conf file. Edit as necessary but you
> > should get the general idea. Basically you compile the new kernel,
> > move it to your boot (or root) partition, then add an entry for
> > the new kernel.
> >
> > boot=/dev/hda
> > map=/boot/map
> > install=/boot/boot.b
> > prompt
> > timeout=50
> > default=linux
> >
> > image=/boot/vmlinuz-2.2.12-20
> > label=old_kernel
> > initrd=/boot/initrd-2.2.12-20.img
> > read-only
> > root=/dev/hda8
> >
> > image=/boot/vmlinuz-2.2.16
> > label=linux
> > initrd=/boot/initrd-2.2.16.img
> > read-only
> > root=/dev/hda8
>
>
Just to shut the errors up I cp /usr/src/linux/arch/i386/boot/System.map
/boot
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux.slackware,aus.computers.linux,comp.os.linux,comp.os.linux.admin,comp.os.linux.misc
Subject: Re: Help: Probme with Modem and Video Mode in new Kernel
Date: 18 Aug 2000 16:01:22 GMT
In comp.os.linux.misc JP <[EMAIL PROTECTED]> wrote:
: 1. The internal modem on COM3 (/dev/ttyS2) fails to work. I have "resource
: or device busy" every time I try to access the modem port. I did not change
That's nice. And what are you doing to access it? Using isapnp to
set its params, then setserial to tell the kernel about them (after
disabling the same resources on the mobo), then using minicom to talk
to the line .... ?
: anything on the hardware. The modem has been working with the old kernel.
: 3. The KDE window fails to start by issuing "startx". The error message
: related to the framebuffer of 800x600 or something like that.
Why are you using it? Use vga mode.
: Can anyone please give me some suggestions or I will have to blow it out
Report what you are doing, the state of the machine, and the rsults
(verbatim) of your tests ...
Peter
------------------------------
From: [EMAIL PROTECTED] (alex)
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc
Subject: Re: Linux on AMD
Date: Fri, 18 Aug 2000 16:42:25 GMT
On 14 Aug 2000 17:06:16 -0400, [EMAIL PROTECTED] (David C.) wrote:
>blowfish <[EMAIL PROTECTED]> writes:
>> Open/Free-BSD. ;-)
>>
>> cd to the ports tree,
>> Just pick the app you want.
>> eg: cd /usr/ports/mysql123.tar.gz
>> make ; make install ; make clean
>>
>> That's it. Everything will be build from source. All dependency files
>> fetched, MD5 checked, patched, build, make, install all done
>> automatically, and custom tailord, configured to YOUR machine; WITHOUT
>> any baby sitting.
>
>This isn't much better than rebuilding Linux.
Sure it is. You are not recompiling the kernel(you *could* update the
system to the latest development version every night and recompile but
it wouldn't be very stable :-) Rather, you are updating the ports
collection -- the application index. Basically, the way it works is
that there is a directory /usr/ports which has subdirectories with
programs divided by categories. Each directory contains a makefile
skeleton and the patches needed for a smooth compile. To install
gnome, all you have to do is cd /usr/ports/x11/gnome and run make
install. It will fetch all dependencies from the internet, compile
all the libraries and programs and install it for you. Easy, clean,
quick. By updating the ports, blowfish means updating the skeletons
to reflect the newest apps.
>I don't understand why you consider it to be so evil for somebody else
>to pre-compile an operating system and distribute it in binary form.
IMHO, it is not evil(and, frankly speaking, I wish I tried to do a
binary install of mozilla on my old 166; the compile took forever but
apperently, the startup took even longer :-) However, with all things
equal, why not compile from source? You'll get a little extra
performance from custom switches. You can tweak the program if
needed. . .
>So it doesn't have every last optimization and tweak that could possibly
>be installed. I, for one, really don't care that much.
Fine. Install packages. I, personally, believe that freedom is
freedom to choose. Packages have their place and if you are not
interested in hacking around but rather need to get something done, I
can easily see how a package might be better than compiling source
code for you.
>> And you can CVS to updated the source everynight automatically.
>
>Oh yeah, that takes brains. So every morning, your computer is running
>a new version of the system. That's really great way to run a stable
>production system.
Nope. You'll be running with an updated ports collection(more/updated
apps you can install)
>> Or do a make world to update the whole system.
>>
>> No pain, no mess. Start it. Go out, go to bed. Find something else to do. ;-)
>>
>> .RPM sucks big time, anyway.
>
>Unless you want to do something with your life other than maintaining
>your OS.
As I see it, RPMs/packages are for people who need productivity and
who are not sysadmins/etc. Both packages and source compiles have
their place. It's up to you which one you want to use.
>-- David
------------------------------
Subject: Re: installing lilo after setup
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 18 Aug 2000 11:26:16 -0600
Andrew Chapman <[EMAIL PROTECTED]> writes:
> Hi there,
>
> I installed Mandrake recently, and skipped the lilo installation during
> setup (I've been using a boot floppy).
>
> Now, I'd like to get linux happening from the NT boot loader, so I need
> LILO installed on the linux partition (not MBR!), but I'm not sure how
> to do this after the main installation.
>
> One of those dead simple things you just can't find docs for...
Look for the NT boot MINI-HOWTO at your favorite LDP mirror.
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
Subject: Re: Can't get cgi to work? (easy question...) - why do I need "use CGI"?
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 18 Aug 2000 11:28:21 -0600
psantt writes:
> Major points goes to Greg! It works!
>
> Anyway, why am I required to use CGI?
You are not. In fact, it is a pretty ugly package who's only real use
is to make the retrieval of POST and GET data easier.
> My scripting before (since I wouldn't have to learn the perl CGI
> structure) was to print "Content-type: html" (the standard html
> header...) and then to print the rest using print commands in a loop.
You can do that just fine with print statements.
> Why on Linux, with Apache 1.3 (I think...) doesn't it work with the
> normal print commands?
This should work:
#!/usr/bin/perl
print "Content-type: text/plain\r\n\r\n";
print "Hello World.\n";
exit (0);
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
Subject: Re: remote exexcution?
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 18 Aug 2000 11:30:17 -0600
"Wilkie" <[EMAIL PROTECTED]> writes:
> is it possible to run a command or script on a remote machine and have the
> script run in a shell on the remote machine?
>
> ex... telnet to host
> run script from terminal with intentions of script executing in environment
> present on host
> NOT on telnet terminal
man rsh (although ssh is a MUCH better solution!)
> also is there any way to have 2 xservers running from the same machine?
> login one and login two both active with their own instance of gnome running
> simultaneously?
> i know i may be asking a lot
man Xnest
man X
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: [EMAIL PROTECTED] (Marijan Peh)
Crossposted-To: alt.os.linux,comp.os.linux
Subject: Re: Where should be installed the LILO?
Date: 18 Aug 2000 17:32:49 GMT
Reply-To: [EMAIL PROTECTED]
albert writes:
>Question:
>
>I have win98/ Red Hat 6.2 on one 8.4 Gb disk (splitted in several logical).
>I've configured LILO to reside in the MBR. I've tryed using Bootmagic, but i
>could'n make it work. now I'm reading that it is not advisable to leave it
>there? Why? If anything goes wrong is it not possible to recover ?
>I need some enlightnement over this issue, any comments are appreciated.
If u have 2 primary partitions it's best to install lilo on primary
partition on which is linux.Then with fdisk make this partition bootable.
If u have 1 primary partition with windows on it, it's best to install lilo
on the MBR or to use 'loadlin'.
--
linux: the choice of a GNU generation
------------------------------
Subject: Re: I've killed my RH 6.2!! Help!!!
From: Craig Kelley <[EMAIL PROTECTED]>
Date: 18 Aug 2000 11:31:49 -0600
[EMAIL PROTECTED] writes:
> I just installed the 2.2.16 updater over a stock 2.2.14 RedHat
> install. Now when LILO goes to the Uncompressing Linux...... line, it
> hangs.
Which "updater"?
> Is there a parameter or options I can give to Lilo to show me what is
> going on during boot? I need to see where the problem is and how to
> fix it.
You can always boot with the rescue floppy and re-run lilo manually to
fix it up.
--
The wheel is turning but the hamster is dead.
Craig Kelley -- [EMAIL PROTECTED]
http://www.isu.edu/~kellcrai finger [EMAIL PROTECTED] for PGP block
------------------------------
From: aflinsch <[EMAIL PROTECTED]>
Subject: Re: Parallel ZIP drive problems
Date: Fri, 18 Aug 2000 13:55:06 -0500
John Wilkinson wrote:
>
> Hi all,
> I have an Iomega parallel ZIP drive and I can rum modprobe ppa and then
> I cna mount the drive and all is OK.
> If I reboot then Linux does not recognise the drive until I rerun
> modprobe again.
> What do I need to do to stop having to rerun modprobe.
>
there are a number of different ways to do this
- kernel recompile & include the module in the kernel
- add to your startup scripts (rc.local comes to mind)
------------------------------
From: Jeff Grossman <[EMAIL PROTECTED]>
Subject: Re: ProFTP config problem
Date: Fri, 18 Aug 2000 11:03:58 -0700
Opps! <[EMAIL PROTECTED]> wrote:
: I am using RedHat 6.2 and removed wu-ftp from my rhl.
: Then I d/l the proftpd 1.2.0 rc2 tarball, complie and install it.
: I can get it running but none of my user can login,same for anonymous.
: I even login locally but still fail.
: I also tried inetd and standalone mode but still the same.
: Can anyone help me?
: If I use the RPM ver. from ftp.proftp.net ,then no problem..but all u
: guy know that the RPM package is quite old....
I would recommend subscribing to the mailing list at www.proftpd.net and
posting your question there.
Jeff
--
=====
Jeff Grossman ([EMAIL PROTECTED])
------------------------------
From: "Sean" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help,alt.os.linux.mandrake,comp.os.linux.networking
Subject: Re: networking problem
Date: Fri, 18 Aug 2000 18:17:34 GMT
This is a bit unusual. Are you sure your DSL modem is really a modem and not
a router?
If this is the case then it may be configured to firewall and reject outside
IP's. And if so, you would need to configure it. This usually can be done
through telnet but you will most likely need a username and password.
chris humphrey <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hi all,
>
> I am succesfully running on Mandrake 7.1, got my DSL running, and have a
> small problem.
>
> I need to be able to run servers from my computer and want to be able to
> connect to it from other omputers on the internet.
>
> The problem is when I ping my ip address, (or telnet) my DSL modem
> responds and does not forward the requests to the computer.
>
> Can someone tell me if this is a linux setup problem or a modem setup
> problem? I am not sure where to start to figure out what is hapening.
>
> my DSL modem has all the parameters to log into the internet and how to
> connect, there is no pppoe or anything like that here.
>
> Any help anyone could give would be very much appreciated.
>
> tia.
> -chris
>
>
------------------------------
From: Kurt Savegnago <[EMAIL PROTECTED]>
Crossposted-To:
intel.motherboards.pentium_pro,alt.os.linux.slackware,alt.sys.pc-clone.gateway2000
Subject: ALR Quad 6 server and PPro Overdrives.
Date: Fri, 18 Aug 2000 13:23:50 -0500
Hello,
I just wanted to report that I was able to get an ALR Quad 6
(4 cpu) server working with Slackware 7.1 and four Pentium Pro
Overdrive 333Mhz cpu's. I have yet to fully test it but I was
able to do a kernel compilation in an X-window under KDE and all
four of the cpu's were being utilized. I have an 18.2 Gig Fujitsu
drive, MAE3182LC, on a Tekram DC390U2W controller. The cpu's are
running at 333Mhz with no problem. It is still pretty early to
comment on system stability but it worked and nothing vaporized!
This is an ex-Gateway 2000 machine.
Intel states that the Overdrive is only good for 2 way SMP but
it so far is working under Linux. The Infomanager LCD screen is
functioning so far too! I DID not change the clock speed of the
motherboard as the PPro Overdrive was meant to be used in a system
that has been setup for 200Mhz. I am not going to experiment with
Overclocking cause I don't want to jepordize my $1200.00 investment
in Overdrive cpu's. I bought this machine as I had a bunch of
pricey ECC buffered 60ns DIMMS that I can't use on any other type of
machine. I picked up the bare Quad 6 at oemliquidators.com.
This could mean that the 6 X 6 motherboard might be able to
run with the OD cpu's although a search of the net did not find
anyone who has reported this. I think mine is the first in a
Quad server environment
Best regards,
Kurt Savegnago
[EMAIL PROTECTED]
------------------------------
From: =?ISO-8859-1?Q?R=FCdiger?= Arp <[EMAIL PROTECTED]>
Subject: Re: SCSI CD-ROM install
Date: Fri, 18 Aug 2000 18:23:09 GMT
>>>>>>>>>>>>>>>>>> Urspr=FCngliche Nachricht <<<<<<<<<<<<<<<<<<
Am 18.08.00, 17:41:45, schrieb [EMAIL PROTECTED] (Robert Delisle) zu=
m=20
Thema SCSI CD-ROM install:
> HI
> I use
> Red Hat 6.0
> The motherboard is tyan S1837 dual CPU (2X 500 P111), scsi controller=
> adaptec 7896 on bord.
> the hard disk is two IDE 20G, and the cd -rom is SCSI
> I start the instalation with floopy all work find but the system
> freeze with this display
> Scanning for adaptec 29...................
> Thanks
> Robert
hi robert,
I just read in the suse cdb the AIC-789x chipset (which I believe you=20
have) is supported...
The module to be loaded is aic7xxx (modprobe aic7xxx, don't replace x's =
with numbers)
good luck,
rude
------------------------------
From: "Gareth Cranny" <[EMAIL PROTECTED]>
Subject: Re: KPPP connecting, hosts not found
Date: Fri, 18 Aug 2000 14:38:38 -0400
Yes, I added both of them that my ISP has given me. This is annoying,
because i'd really like to be able to get outside from Linux :(
Gareth.
Joel Schultz wrote in message <8nib38$nqa$[EMAIL PROTECTED]>...
>Did you set up your Domain Name Servers IP address(es) in kppp?
>
>"Gareth Cranny" <[EMAIL PROTECTED]> wrote in message
>news:nkAm5.3432$[EMAIL PROTECTED]...
>> I have the same problem, if I found out I'll let you know!
>>
>> Gareth.
>>
>> Karen Heiby wrote in message <8ndktq$4sm$[EMAIL PROTECTED]>...
>> >Hello, this is probably a common problem. I recently moved to Germany
>> >where I have to use a dial-up connection. I used to have Roadrunner
>> >cable in the U.S. Anyway I disabled my ethernet card (eth0) from
>> >initializing at startup anymore. Sorry, I forget the exact name of the
>> >tool that I used; "nettool" or something like that. Anyway, I set KPPP
>> >to do its thing. It dials, connects, and looks good, but doesn't seem
>> >to be sending/receiving any more bytes after the connection is made.
>> >Netscape, GAIM, etc. cannot locate host servers.
>> >
>> >I read the Modem How-To and don't see anything. I did do "modemtool"
>> >and set the correct COM port (COM1).
>> >
>> >I have an external Elsa Microlink 56K (German specifications) modem.
>> >The manufacturer assures me that it works with Linux, although they do
>> >not technically support Linux.
>> >
>> >I use Sonnet.de for my ISP here in Germany. I also try Compuserve now
>> >and then (here in Germany Compuserve works more like a standard ISP;
>> >they don't have that cheeseball software interface and you can just
>> >dial it straight with KPPP or in Windows, Dial-Up Networking).
>> >
>> >If someone can point me in the right direction for what else I need to
>> >read, I'd appreciate it.
>> >
>> >Thanks,
>> >Karen
>> >
>> >
>> >Sent via Deja.com http://www.deja.com/
>> >Before you buy.
>>
>>
>
>
------------------------------
From: "KMX" <[EMAIL PROTECTED]>
Subject: Is it a WINMODEM ??
Date: Fri, 18 Aug 2000 19:41:57 +0100
Hi,
I have got a copy of SUSE 6.4 evaluation and will try and install on my pc.
there will be two operating systems hopefully, with a dual boot option or
something.
I have a WINMODEM, but have seen a internal modem for 20 in a local cheap
shop. there is no mention of WINMODEM on the box so ca I tell if it will
work with linux by looking at it for looking for some specification ?
Thanks
KMX
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Subject: Re: can't set PATH
Date: 18 Aug 2000 07:25:50 GMT
Don Belmore <[EMAIL PROTECTED]> wrote:
>Sam ([EMAIL PROTECTED]) wrote:
>: in the /etc/profile file. I add PATH=$PATH:/etc/jdk1.3/bin and EXPORT PATH
>
> Greetings all!
>
> I have somewhat a similar problem. I was going through xf86config
>and it says to make sure /usr/X11R6 is _before_ /usr/X386/bin. My problem
>I seem to be having is that when I type 'printenv' it lists PATH as :
>
>PATH=/sbin:/bin:/usr/sbin:/usr/bin:/usr/X11R6/bin:/root/bin:/etc/XR11
/etc/XR11? Mumble ... why do you need to start programs from there?
(What is that directory? :))
> yet in my .bash_profile is says:
>
>PATH=$PATH:$HOME/bin
>
> As you can see I tried to add the '/etc/XR11' to my PATH (a la
>export command) and it simply appended it to the end.
To clear up a bit of confusion here:
export PATH=/usr/bin:/bin
... say, is just an abbreviation in the Bourne shell language (used by
bash and some other shells) for the following:
PATH=/usr/bin:/bin
export PATH
The first line here sets the variable, and the second marks the variable
PATH as one of those that get passed on to the environment of any
command started by the current shell process.
Also, there isn't really any concept in the shell language of "adding"
an element to $PATH. It's just a character string, so all you can do is
textual substitutions. Adding elements before (PATH=$HOME/bin:$PATH) or
after (PATH=$PATH:$HOME/bin) is just about all that anyone bothers to
do. If you really have to then you can do tricks with command
substitution and sed to pull the string apart and put stuff into the
middle of it, but that's almost never necessary (and probably indicates
bad system administration if you need to).
>My question is how do I put '/etc/XR11' _before_ '/usr/X11R6/bin' when
>I can't even find out how my system gets all those other directories in
>the PATH? I have been hunting all over my system to find out where it
>defines '/sbin:/bin:...'
Probably in /etc/profile or thereabouts. bash has a default $PATH, I
believe, but I don't have the source code to hand to check exactly what
it is. At a guess, I would say that it's /bin:/usr/bin, and that your
/etc/profile replaces that with at least /sbin:/bin:/usr/sbin:/usr/bin
when you're running as root.
--
Colin Watson [[EMAIL PROTECTED]]
"There is perhaps in every thing of any consequence, secret history,
which it would be amusing to know, could we have it authentically
communicated." - James Boswell
------------------------------
From: [EMAIL PROTECTED] (Colin Watson)
Subject: Re: Can't get cgi to work? (easy question...)
Date: 18 Aug 2000 07:29:12 GMT
Craig Kelley <[EMAIL PROTECTED]> wrote:
>You forgot to import CGI and it's symbols:
>
>#!/usr/bin/perl
>
>use strict;
... which is almost indispensable in getting Perl programs right, but
also:
#! /usr/bin/perl -w
(Without -w, almost anything is legal Perl. :))
--
Colin Watson [[EMAIL PROTECTED]]
"Abandon the search for Truth; settle for a good fantasy."
------------------------------
** 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 (and comp.os.linux.setup) via:
Internet: [EMAIL PROTECTED]
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-Setup Digest
******************************
