Linux-Setup Digest #684, Volume #20 Thu, 22 Feb 01 06:13:04 EST
Contents:
Re: LILO, boot problems ("Peter T. Breuer")
gal && glade installation (SC Patton)
user management....how to? (Robert Abbate)
Re: mounting udf cds as nonroot ("Eric")
Re: cd-rom trouble ("Eric")
Re: Linux newbie<HELP!>: dual boot problem(Linux/win98) freeze at "L"( as in LILO)!
("Eric")
Re: linux changes active partition ("Eric")
Re: LILO, boot problems ("Eric")
How can i install old-version kernel? (Kihong Lee)
Re: LILO problem ("Eric")
Re: rmmod (Michael Heiming)
Re: mount can't find hda1 ("Eric")
Re: Hard Disk won't boot, floppy will ("Eric")
Re: Redhat Boot Issue 6.2 ("Eric")
Re: linux booting problem ("X.P. Wei")
Re: linux booting problem (benoit mordelet)
Re: How can i install old-version kernel? (Eggert Ehmke)
Garbled output if there had been no paper in the printer (CUPS) (Mario Auer)
Re: poweroff does not power my box off, even with APM enabled (Eggert Ehmke)
----------------------------------------------------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: LILO, boot problems
Date: Thu, 22 Feb 2001 06:46:25 GMT
Vlar Schreidlocke <[EMAIL PROTECTED]> wrote:
> I made the floppy after recompiling the kernel. The original
> installation version was 2.2.17. The new kernel is 2.2.18. I made the
> floppy using mkbootdosk after compiling the kernel. When I boot from
> the floppy it says "2.2.18", not 2.2.17.
OK .. that's definitive, and good detail. The only problem is that I
really don't know anything about your "mkbootdisk" command. It could do
anything. I suspect that it makes a boot disk that depends on an
initial ram disk (initrd) for part of the boot process, thus allowing
you to boot a kernel that has not enough drivers in, then load the
drivers from the ram disk image, then go on to access and mount your
hard disk controller and partitions.
In all probability you don't want to do that. You want to build a
straightforward kernel with the hard disk drivers and fs drivers built
into it (but not the ethernet drivers).
>>You did, but who cares? First convince us that your newly compiled
>>kernel has support for your ide controller and boot file system BUILT
>>IN. Then you can worry about boot sequences.
>>
> I ran the Mandrake installation program and chose "Upgrade". It
I have no idea what that means or does.
> allowed me to "Install Boot Loader" after doing that and finishing the
> "Upgrade" the computer booted correctly with the new LILO. Then it
> wouldn't boot again without the floppy.
I somewhat suspect that you have grub, not lilo.
>>Well, on second thoughts .. scratch that. First convince me that you
>>have a boot loader on your hard disk! What makes you think you do?
>>
> I assume from the previous comment that there was a boot loader and it
> allowed me to boot one time.
I mean "what happens when you boot from the hard disk".? If the letetrs
LILO show - even partially - then you have a boot loader. Or are you
using grub instead?
>>That's a "geometry error", or possibly just "no kernel" at the point
>>where you tried to jump to. Either you changed the disk geometry or you
>>replaced the kernel.
>>
> I tried manually inputting the CHS info instead of selecting "AUTO" in
> the BIOS setup.
The bios is not really implicated at this stage, but if it is, then
just selecting lba mode in the bios shoudl help. What you need to do is
select the same mode that the kernel is told. What are the bios c/h/s
figures?
But it sounds like you just moved/replaced the kernel.
>>Anything.
>>
>>> and have the "ether=5,0x280,eth0 ether=4,0x300,eth1" parameters
>>> entered automatically?
>>
> For some reason I was never able to get two NE2000 cards to run from
> just the modules. I tried every configuration in the ethernet,
> network, ipmasq, etc. HOWTO's. I used the NE2000 support compiles into
> the kernel and it worked. I have had this same problem on three other
> computers and have resorted to the kernel support. It worked so I had
> no reason to keep hacking away at the module problem. Everyone told me
> they should have been able to load as modules, but after two weeks of
> trying everything I had to move on to what worked, I was just
> repeating that in this case.
There is no difference between running as a module and running as a
built-in, except that a built-in can do some probes that a module
cannot. One problem is that you look to be running the ne driver not
ne2k-pci since your params refer to isa irqs and io ports. I presume that
is correct. An ne2000 isa card is problematic in itself, but should
work fine .. I am using an ne2000 pcmcia card as I write this.
> How do I run lilo on the floppy? I know I have seen that somwhere, but
> I'm frazzled and can't remember where.
If your floppy has a lilo.conf in its /etc dir, just mount /floppy,
edit /floppy/etc/lilo.conf, and /sbin/lilo -r /floppy.
Peter
------------------------------
From: SC Patton <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: gal && glade installation
Date: Thu, 22 Feb 2001 00:40:00 -0600
Hi everyone!
I'm trying to upgrade some of my GNOME libraries,
compiling them from source code that came from the files
gal-0.4.1.tar.gz and glade-0.5.11.tar.gz.
QUESTION:
Is glade (from glade-0.5.11.tar.gz) supposed to create a
/usr/lib/libGladeConf.sh file? I seem to need this file so that GAL's
configure script (from gal-0.4.1.tar.gz) can detect the presence of
libGlade. I've tried and tried, and I'm stumped.
Can someone help me?
BACKGROUND:
GAL PROBLEMS (gal-0.4.1.tar.gz):
After unpacking the .tar.gz, I run "./configure" as
instructed. /.configure complains:
"checking for Glade libraries >= 0.13... configure: error:
Did not find libGlade installed"
I had already installed Glade (glade-0.5.11.tar.gz), so I poked around
configure script for GAL, to see if I could find out how it was
detecting Glade. On (or about) line 5194 of the configure script, it
says:
if gnome-config --libs libglade > /dev/null 2>&1; then
<snip>
else {echo "configure: error: Did not find libGlade installed 1>&2; \
exit 1; }
So when I tried to run "gnome-config --libs libglade" by hand, it
returned with: "Unknown library `libglade'". I then tried to find out
how gnome-config detects the presence of libraries by running the
command "gnome-config --help", it said that (after looking at some
known libraries), it looks for /usr/lib/<LIBRARY>Conf.sh for whatever
information is called for. After installing libGlade (from
glade-0.5.11.tar.gz) I DO NOT HAVE A /usr/lib/libGladeConf.sh !!
GLADE PROBLEMS (glade-0.5.11.tar.gz):
I cannot figure out how this version of libGlade creates a
/usr/lib/libGladeConf.sh file. (I even tried the command "grep -n
'Conf' *" from the glade-0.5.11 directory--nothing returned hinted at
the creation of this file).
Please help!
Thank you *very much* in advance!
Steven Patton
[EMAIL PROTECTED]
------------------------------
From: Robert Abbate <[EMAIL PROTECTED]>
Subject: user management....how to?
Date: Thu, 22 Feb 2001 01:52:37 -0500
I am really new to Redhat Linux 7, and I have set it up as a web hosting
server. I am also using the KDE X-window GUI interface.
I was wondering if there was a good admin program/RPM which allows you to
direct incoming I.P. requests (Domain names), and control user accounts?
Thanks!~
============================================================
Get your FREE web-based e-mail and newsgroup access at:
http://MailAndNews.com
Create a new mailbox, or access your existing IMAP4 or
POP3 mailbox from anywhere with just a web browser.
============================================================
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: Re: mounting udf cds as nonroot
Date: Thu, 22 Feb 2001 07:55:59 +0100
"Gregory Davis" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I made a post a few weeks ago (maybe) about not being able to mount a
> udf disk as a user. I thought the solution lay in upgrading my version
> of utils-linux to the version required by kernel 2.4.1, i upgraded that
> and still no dice. This is the output directly from console (no kde gui
> or anything):
>
> greg@linux:~ > mount /dev/cdrom /cdrom -t udf
> mount: only root can do that
> greg@linux:~ >
>
> If it helps, I will tack on my fstab and filesystems files, but I have
> pored over them, and I am almost positive that they are correct. I can
> mount the a udf cd as root successfully. What's up???
>
Like you were told last time:
`mount /dev/cdrom`
only root is allowed to give more than one parameter to the mount command
All other users must use the fstab entry.
PS. this info is in the manpages (man mount)
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: cd-rom trouble
Date: Thu, 22 Feb 2001 07:57:49 +0100
> I think I am having a similar if not the same problem you are having. I
> reinstalled Mandrake 7.2 and my floppy and my CD-RW show in file manager
> with a padlock as part of the icon. Whenever I try to access the CD-ROM I
> get 'You do not have Access Rights" type of error. There must be some way
> to unlock them. I also wonder if there is a way to get Linux to re-detect
> my CD-ROM to make sure the the hardware is properly installed with drivers
> and initialized and such. JH
What is in your /etc/fstab file?
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: Linux newbie<HELP!>: dual boot problem(Linux/win98) freeze at "L"( as in
LILO)!
Date: Thu, 22 Feb 2001 08:00:28 +0100
> p.s.
> I checked my /etc/lilo.conf and I saw that it said linear and not
lba32,
> should I change that? Everything is working as it should right now....
>
>
Then don't change it.
It's only important when your kernel image (/boot) is (partly) above cyl.
1024
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: linux changes active partition
Date: Thu, 22 Feb 2001 08:06:14 +0100
> when I boot to linux and then exit and boot to Win98 I get a filesystem
not
> found error ( I have a separate FAT-32 data partiton) I have to start
> Windows 98 with the boot disk and use Fdisk to set the partion to active
> again. When I restart Windows 98 I have no problem. Is lilo overwriting
my
> active partition information?
Not on it's own
You claim that each boot to linux, alters the partitiontable?
I cannot believe that.
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: LILO, boot problems
Date: Thu, 22 Feb 2001 08:09:54 +0100
`fdisk -l /dev/hd[a-d]`
I suspect the recompilation of your kernel placed it beyond cyl. 1024
Eric
------------------------------
From: Kihong Lee <[EMAIL PROTECTED]>
Subject: How can i install old-version kernel?
Date: Thu, 22 Feb 2001 16:12:32 +0900
Hi, all,
At first, i'm sorry for my poor english.
Recently, I started to study linux-kernel.
So i tried to install kernel version 0.01 or 0.1 or 1.0.
Because i think that old-version kernel is simple so that it makes to
easy
to study than latest version kernel.
I'm using RedHat Linux 6.1.
For down-grade my RedHat Linux 6.1, i downloaded old-version kernel,
and tried to compile so many times.
But i failed. It is not so easy.
The point of question is this.
I want to down-grade my linux kernel to old-verison.
(I don't care it is a down-grade or new-install to old-version kernel)
Another words, I want to compile successfully old-version linux kernel
and re-booting my computer using that.
That's all.
I don't need other function(networking.. X windows.. and so on).
All i need is just compile and re-boot using old-version kernel.
If you ever tried this, please let me know.
Thank you for reading.
Bye.
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: LILO problem
Date: Thu, 22 Feb 2001 08:11:52 +0100
> Now the only way I can get to boot windows is to change the boot device in
> the bios, which is not much good. Does anyone know how I can get either
the
> win2k bootloader to load LILO, or preferably LILO to boot the win2k
> bootloader?
>
Use the "map-drive= to=" lines in your lilo.conf
(it's in the manpages/lilo documentation)
Eric
------------------------------
Date: Thu, 22 Feb 2001 08:08:20 +0100
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: rmmod
Paul Kimoto wrote:
> In article <[EMAIL PROTECTED]>, Michael Heiming wrote:
> > jeff wrote:
> >> Is there any good reason to be running rmmod -as every ten minutes???
>
> > IMHO not, the kernel will perfectly take care of unloading modules
> > himself, if your memory gets to low,
>
> No, it doesn't. kerneld used to remove unused modules, but that was back
> in the days of 2.0.*.
Thx....I didn't knew this had changed
Michael Heiming
>
>
> > otherwise they will stay in memory with state (autoclean).
>
> --
> Paul Kimoto
> This message was originally posted on Usenet in plain text. Any images,
> hyperlinks, or the like shown here have been added without my consent,
> and may be a violation of international copyright law.
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: mount can't find hda1
Date: Thu, 22 Feb 2001 08:14:03 +0100
As Herman suggested, check the major/minor numbers, and look at the
partitiontable. (fdisk -l /dev/hda)
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: Hard Disk won't boot, floppy will
Date: Thu, 22 Feb 2001 08:15:46 +0100
Stop posting that same question.
It's the third thread I ran into.
I answered in the first one I encountered.
Eric
------------------------------
From: "Eric" <[EMAIL PROTECTED]>
Subject: Re: Redhat Boot Issue 6.2
Date: Thu, 22 Feb 2001 08:21:38 +0100
> LILO boot:
> Loading linux............................
> Uncompressing Linux...
>
> invalid compressed format (err=2)
used the wrong image?
> -- System halted-
>
>
> ---------------------------------------------------------------
>
> I have 3 valid kernel's to boot from
says who?
> and none seem to be taking.
Then they're not valid kernels.
> Any insight would be much appreciated
get a correct kernel
don't use the /usr/src/linux/vmlinux
The correct kernel image is in
/usr/src/linux/arch/i386
Eric
------------------------------
From: "X.P. Wei" <[EMAIL PROTECTED]>
Subject: Re: linux booting problem
Date: Wed, 21 Feb 2001 23:51:55 -0800
Reply-To: "X.P. Wei" <[EMAIL PROTECTED]>
That solved the problem. Thanks.
X. Wei
"David" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "X.P. Wei" wrote:
> >
> > Hi,
> >
> > I had a linux computer running RedHat 6.2. I tried to install a tape
driver
> > a couple of days ago, but did not succeed.
> > After that I had problem booting the compuer. It always failed at file
> > system check and the error message is as follows:
> > " Checking filesystems
> > fsck.ext2 could this be a zero-length partition?
> > :Attempt to read block from file system resulted in short read while
> > trying to open /dev/st0( might be the tape driver)
> > /dev//hda1: clean, 24/6024 files,3256/24066 blocks
> > [FAILED]
> > *** An error occured during the file system check."
> >
> > I was then put into a shell mode, but I could not startx X and network
> > connection was lost. The "/boot"
> > was empty the first time when the problem occured. I tried to restore
the
> > "/boot" directory by re-compiling and
> > installing the kernel. However, the same thing happened everytime I
boot
> > the computer.
> >
> > What's the problem and how would I be able to fix it?
>
>
> :Attempt to read block from file system resulted in short read while
> trying to open /dev/st0( might be the tape driver)
>
> It is trying to read the tape drive at boot time. You don't mount a tape
> drive. If you added any lines to /etc/fstab for the tape drive you need
> to remove them.
>
> --
> Confucius say: He who play in root, eventually kill tree.
> Registered with the Linux Counter. http://counter.li.org
> ID # 123538
> Completed more W/U's than 99.078% of seti users. +/- 0.01%
------------------------------
From: benoit mordelet <[EMAIL PROTECTED]>
Subject: Re: linux booting problem
Date: Thu, 22 Feb 2001 10:13:16 +0100
David wrote:
>
> "X.P. Wei" wrote:
> >
> > Hi,
> >
[...]
> >
> > What's the problem and how would I be able to fix it?
>
> :Attempt to read block from file system resulted in short read while
> trying to open /dev/st0( might be the tape driver)
>
> It is trying to read the tape drive at boot time. You don't mount a tape
> drive. If you added any lines to /etc/fstab for the tape drive you need
> to remove them.
>
or add the noauto flag in this line's options.
ben
------------------------------
From: Eggert Ehmke <[EMAIL PROTECTED]>
Subject: Re: How can i install old-version kernel?
Date: Thu, 22 Feb 2001 11:22:22 +0100
Reply-To: [EMAIL PROTECTED]
On Thu, 22 Feb 2001 16:12:32 +0900, Kihong Lee <[EMAIL PROTECTED]>
wrote:
>Recently, I started to study linux-kernel.
>So i tried to install kernel version 0.01 or 0.1 or 1.0.
I don't think that's a good idea. By the time you spend to make that old
kernel run, you could even start to study an up to date kernel. They had
different C-libraries at that time, different file system, different object
file format, different compiler versions ... lot of things have changed.
Most likely applications that rely on glibc will not run. Take a new kernel
like 2.2.18 or 2.4.1 and throw out all features you don't need. The kernel
sources are well documented, it should not be so hard. But maybe I'm wrong,
and it can be done with reasonable effort. Good luck.
--
Eggert Ehmke
Email: [EMAIL PROTECTED]
------------------------------
From: Mario Auer <[EMAIL PROTECTED]>
Subject: Garbled output if there had been no paper in the printer (CUPS)
Date: Thu, 22 Feb 2001 11:23:48 +0100
Hello,
I have some troubles using CUPS. If I send a job to the printer while it is
out of paper the job doesn't resume properly. After refilling the paper the
printer starts working but the output looks garbled. I have to reset the
printer several times and clear the queue. It looks like some data had been
droped so that some control sequences were not interpreted correctly.
I have a Lasejet 5L attached to /dev/lp0. I even tried the latest "drivers"
(ppd files), but the problem remains. The packages installed on my Mandrake
system running kernel 2.4.1 are:
libcups1-1.1.6-2mdk
cups-common-1.1.6-3.1mdk
cups-1.1.6-5mdk
cups-drivers-1.0-7mdk
Thanks for any input,
Mario
------------------------------
From: Eggert Ehmke <[EMAIL PROTECTED]>
Subject: Re: poweroff does not power my box off, even with APM enabled
Date: Thu, 22 Feb 2001 11:30:57 +0100
Reply-To: [EMAIL PROTECTED]
On Tue, 20 Feb 2001 23:01:26 +0000, Kevin Milburn
<[EMAIL PROTECTED]> wrote:
>I managed to get my system to do this, the solution was to
>exclude 'Use real mode APM BIOS call to power off'.
>After doing this my system halted and powered off correctly..
After I enabled the APM stuff (following your advice) things were even
worth. The system still did not switch off, and even the power switch at the
front panel was disabled ! I have to switch off using the power switch at
the back panel, beside the power plug.
The system powers down ok when Windows ME is told so.
--
Eggert Ehmke
Email: [EMAIL PROTECTED]
------------------------------
** 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.setup.
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
******************************
