Linux-Setup Digest #508, Volume #20 Fri, 26 Jan 01 14:13:12 EST
Contents:
Re: 2.4.0 & modules "not found" (Paul Lew)
A Linux disaster! (95459906)
Re: Installing RedHat 7.0 with Promise ATA/100 controller ("Keith Wheeler")
Re: A linux newbie ([EMAIL PROTECTED])
Shutdown hangs, /usr won't unmount (Hack Cabot)
Re: Question on LILO and recompiling your Kernel ("Keith Wheeler")
Re: Question on LILO and recompiling your Kernel ("Keith Wheeler")
Re: can't partition HD (benoit mordelet)
PCMCIA Modem Problem (rob)
Re: Trying to install dual boot WinNT + Linux ("Tyberghein Jorrit")
Re: Newbee modem set-up question (Uwe Malzahn)
Re: 7 CD changer (Norman Levin)
Re: Editing /etc/fStab during bad boot (Norman Levin)
Re: filesystem corruption (Norman Levin)
Re: 'force user' in linux ? (Norman Levin)
Re: general kernel compile question (Norman Levin)
Re: general kernel compile question (Norman Levin)
Re: Help, please, post install... (Norman Levin)
Re: Intel 810 chipset probs (Norman Levin)
Re: Line editor in bash (Norman Levin)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Paul Lew)
Subject: Re: 2.4.0 & modules "not found"
Date: Fri, 26 Jan 2001 17:10:34 GMT
On 26 Jan 2001 07:12:07 GMT, H.Bruijn <[EMAIL PROTECTED]> wrote:
>On Fri, 26 Jan 2001 07:02:26 GMT, Paul Lew allegedly wrote:
>>I was able to compile and get 2.4.0 "running" BUT it is unable to find
>>and load the modules like the tulip, vfat, fat, msdos and some other
>>modules. Now, I did notice that when I did "make modules", I did notice
>>some "nothing to do for modules" and the "make modules_install" did not
>>return errors.
>>
>>Looking at the /lib/modules/ I only see the pcima and I think the usb
>>directories and a "build" directory as a link to the /usr/src/linux-2.4.0.
>>Something went wrong somewhere....
>>
>>Any hints on how to get the modules made and installed? The kernel is
>>working but no etho, no network and no "dos, vfat filesystems"..no
>>nothing...
>
>You did do the required upgrades as mentioned in
>../src/linux-2.4.0/Changes file? Notably
>o binutils 2.9.1.0.25 # ld -v
>o util-linux 2.10o # fdformat --version
>o modutils 2.4.0 # insmod -V
OOPS....
>
>You did select stuff as modules, did you not? Simply do a
>"grep \=m ../src/linux-2.4.0/.config" to make sure. And you selected
>support for modules? CONFIG_MODULES=y?
>
Yup, I believe so.
In fact looking at it again, it looks like the modules were "installed",
but in /lib/modules/kernel/drivers and /lib/modules/kernel/fs....
Well now to check the ChangeLog...
thanks.
------------------------------
From: 95459906 <[EMAIL PROTECTED]>
Subject: A Linux disaster!
Date: Sat, 27 Jan 2001 01:00:06 +0800
Hi, all!
I've a disaster on linux! The super-block of one of my linux partitions
was found to be corrupted and don't know how to recover it! The partiion
here is /dev/hda6. I tried to install another linux partition on
/dev/hda7 and wanted to use e2fsck to fix that partition (e2fsck
/dev/hda6 or e2fsck -b (8193...16385...24577) /dev/hda6) but failed to
do this. Anyone can offer a help to me? Below is the message displayed
when I tried to fix the /dev/hda6 using fsck command on the linux
partition /dev/hda7!
[root@localhost /root]# fsck /dev/hda6
Parallelizing fsck version 1.15 (18-Jul-1999)
e2fsck 1.15, 18-Jul-1999 for EXT2 FS 0.5b, 95/08/09
Couldn't find ext2 superblock, trying backup blocks...
fsck.ext2: Bad magic number in super-block while trying to open
/dev/hda6
The superblock could not be read or does not describe a correct ext2
filesystem. If
the device is valid and it really contains an ext2 filesystem (and not
swap or ufs or something
else), then the superblock is corrupt, and you might try running e2fsck
with an alternate
superblock: e2fsck -b 8193 <device>
regards,
Laurence
------------------------------
From: "Keith Wheeler" <[EMAIL PROTECTED]>
Subject: Re: Installing RedHat 7.0 with Promise ATA/100 controller
Date: Fri, 26 Jan 2001 10:19:41 -0700
Steve,
I've been playing with RH 7 and the Asus A7V motherboard with an onboard
Promise ATA100 controller for about a month now. I played with the 2.4.0,
but it needs a bit more work for my uses. I ended up upgrading the RH 7
stock 2.2.16-22 kernel to 2.2.18.
I just went to http://www.kernel.org/ and downloaded the 2.2.18 kernel
source files, downloaded the latest Hedrick IDE patches for 2.2.18, applied
the patches, compiled the kernel, installed the kernel, reran LILO, and
viola!
RedHat has a HowTo on updating kernels on RH 7. Make sure you read that
first.
Good Luck,
KW
<[EMAIL PROTECTED]> wrote in message
news:94qi9a$i2b$[EMAIL PROTECTED]...
> I am looking for a way to install Red Hat 7.0 using
> a hard drive connected to a Promise ATA/100 controller.
> This is a PCI connected card.
>
> If anyone could point me in the direction as to where
> I could find this driver or other information that would
> help, I would greatly appreciate. If there is another
> compatible driver or some sort of workaround, that would
> also help.
>
> Thanks!
>
> Steve
>
> * Please reply CC: e-mail [EMAIL PROTECTED]
>
>
> Sent via Deja.com
> http://www.deja.com/
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: A linux newbie
Date: Fri, 26 Jan 2001 17:06:03 GMT
In article <LHmc6.12879$K6.101470@news>,
"gameboy" <[EMAIL PROTECTED]> wrote:
> Hi all:
>
> I have a caldera edesktop 2.4 box, and I want to set it up as a
client to a
> win2000 domain.
> Any idea/articles would be great.
>
> Thanks,
> gb
Start by looking into SAMBA (http://www.linux.org/docs/ldp/howto/SMB-
HOWTO.html).
You may also want to check out the various dual-boot HOWTOs if you
intend to use both on the same machine (same URL, up a directory).
Dave
Sent via Deja.com
http://www.deja.com/
------------------------------
From: Hack Cabot <[EMAIL PROTECTED]>
Subject: Shutdown hangs, /usr won't unmount
Date: Fri, 26 Jan 2001 17:33:47 GMT
Hi folks,
I recently created separate partitions for /tmp and /usr on my
Mandrake 7.2 box. I used partition magic, and had some problems with
/tmp, crashing my X, but I update installed, got it back and have X and
the /tmp and /usr partitions.
The partitions mount properly, but /usr won't unmount, I get a
Device Busy error when I have tried to umount it manually. I end up
pulling the plug, which isn't good for the system, and on boot fschck is
run, taking forever. It's only a matter of time before all my programs
get corrupted.
ps, lsof and fuser do not help. Killing processes individually does
not help. Could there be something in the old /tmp and /usr files that
somehow stayed in there when I emptied them? Is there some way to kill
every process in /tmp and /usr on shutdown so I do not get the device
busy error? (Have I inadvertently mounted the partitions on directories
that were not actually empty?). I don't know. I'll reinstall everything
if I have to, but It's been an awful lot of work just getting this far.
Thanks to all,
Hack
------------------------------
From: "Keith Wheeler" <[EMAIL PROTECTED]>
Subject: Re: Question on LILO and recompiling your Kernel
Date: Fri, 26 Jan 2001 10:35:49 -0700
<[EMAIL PROTECTED]> wrote in message news:94qqp8$p2o$[EMAIL PROTECTED]...
> I'm having the same problem, but a simpler situation: the machine is
> single OS (Red Hat Linux 6.2), and the only boot loader is LILO. Ergo,
> neither of those things is likely to be your problem.
>
> Interestingly, my original lilo.conf had the following lines for the
> image:
>
> image=/boot/vmlinuz-2.2.13.5.0
> label=linux
> initrd=/boot/initrd-2.2.14-5.0.img
> read-only
> root=/dev/hda1
>
> This raises several questions for me...
>
> (1) The Kernel-HOWTO says that it is an error to have "root=/dev/hda1"
> (it should be "root=/dev/hda"). Obviously, this worked for me though.
>
> (2) I don't know what the initrd line is and it isn't mentioned in the
> Kernel-HOWTO or the LILO mini-HOWTO. I tried reading the initrd man
> page and found myself in deep water...I guess I'll just read it a few
> more times and see if it becomes clearer by repetition. One fact that
> I thought I understood in the docs was that, if there was no initrd
> line in lilo.conf, the world would not end and loading would simply
> proceed. If I do in fact need the initrd line, how do I create the .img
> file?
The initrd is for modules. You can now compile several options in the kernel
to modular to reduce the size and complexity of the kernel itself. But some
of those modules may be required before the kernel has booted far enough to
have mounted your file system. Therefore you create a RAM disk, copy the
required modules to the RAM disk (your img file), and access them from there
during kernel load/bootup.
>
> (3) After coming to this tentative (and probably wrong) conclusion I
> changed the image lines to read as follows:
>
> image=/boot/vmlinuz-2.4.0
> label=linux-2.4.0
> read-only
> root=/dev/hda1
>
> image=/boot/vmlinuz-2.2.13.5.0
> label=linux-2.2.14
> initrd=/boot/initrd-2.2.14-5.0.img
> read-only
> root=/dev/hda1
>
> I made sure that /boot/vmlinuz-2.4.0 was actually there, and I ran
> lilo. The message I got back was:
>
> Added linux-2.4.0
> Added linux-2.2.14
> No images have been defined.
>
Looks like LILO encountered an error and didn't actually update the boot
tracks.
>
> It boots fine, but it boots straight into the old kernel, and when I
> hit TAB it doesn't show the new one. I tried changing it
> to "root=/dev/hda" for the 2.4.0 kernel, but that didn't help either.
>
> Anyone have any ideas?
>
> Dave
>
>
> Sent via Deja.com
> http://www.deja.com/
------------------------------
From: "Keith Wheeler" <[EMAIL PROTECTED]>
Subject: Re: Question on LILO and recompiling your Kernel
Date: Fri, 26 Jan 2001 10:37:34 -0700
Don't forget to copy the System.map file too. The new System.map file also
needs to be in /boot.
KW
Collector <[EMAIL PROTECTED]> wrote in message
news:94q7tg$qq4$[EMAIL PROTECTED]...
> Howdy,
>
> I had a couple questions. First of all, I was trying to recompile my
> kernel last night. I went through all of the steps, like 'make dep', 'make
> clean', etc. and got a new bzImage that I copied into my /boot directory
as
> zipkern. I then went into my lilo.conf file and added a new image section,
> and I added the following:
>
> image = /boot/zipkern
> root = /dev/hda3
> label = LinTest
>
> After this I typed "lilo" at the command prompt and it said it added the
new
> entry. Great. So I reboot, and when I get to the LILO prompt, it doesn't
> display my new image label, just the same old stuff as usual. I logged in
as
> normal and checked the lilo.conf file. It still had my changes in it. Did
I
> skip a step or something in adding the lilo information? My machine is
> Dual-Booting Windows 2k, and the Windows bootloader is the first one that
> pops up when I turn my machine on; from there I choose my linux OS and
THEN
> I get LILO. Does this have something to do with it? Because LILO is not
the
> primary booting utility? Why is it telling me it added the new information
> from the config file when its not reflecting it at bootup?
>
> My second question - when recompiling my kernel, it doesn't matter
WHERE
> I compile from does it? As long as I put the resulting image in the /boot
> directory and make sure I point to it in the boot config file right? After
I
> couldn't get LILO working last night, I backed up one of my working
kernels,
> and renamed the new one (zipkern) to the name of the default one
(vmlinuz).
> I rebooted, and as soon as it started loading linux it told me that it ran
> out of system resources and halted the system. OOPS! Now, does this mean
> that my new kernel is just bad, or is it because I cheated and tried
> renaming the default one? Am I not allowed to do that? ANY information on
> either of these problems would be appreciated. Thanks!
>
> For reference, my machine is a P2 450 w/ 128 MB of RAM. It dual-boots
> between Win2k and SuSE 6.4. The windows booting utility is the first one
to
> come up at bootup (if that makes any difference).
>
> Collector
>
>
------------------------------
From: benoit mordelet <[EMAIL PROTECTED]>
Subject: Re: can't partition HD
Date: Fri, 26 Jan 2001 18:36:53 +0100
Ron Sussman wrote:
>
> I'm trying to install Mandrake 7.2 onto a 30 gig IBM HD. The entire disk
> is one big "C" drive containing win 98SE. I want to resize the "C" partition
> & carve out a few gigs for linux. Both DiskDrake & Partition Magic fail. DD
> says "can't resize hard drive"; PM says that it can't partition the drive
> because it's full (PM lies; actually 27-28 gigs are free). Any ideas?
> Thanks
>
try using GNU parted. I don't guaranty that it will work, but it's
another piece of software aimed at repartitionning hard drives.
ftp://ftp.gnu.org/gnu/parted/
ben
------------------------------
From: rob <[EMAIL PROTECTED]>
Subject: PCMCIA Modem Problem
Date: Fri, 26 Jan 2001 17:40:10 GMT
Dear All:
I have a PCMCIA modem on my laptop. When I boot up Linux Mandrake 7.2
and go to dial up, I get a message saying the modem is busy. I f I pop
out the card (which is followed by a beep) and then pop it in again
(which is followed by two beeps), it works fine. Any ideas how I can
fix this?
Many thanks, rob
--
respectfully, rob
e: [EMAIL PROTECTED]
Sent via Deja.com
http://www.deja.com/
------------------------------
From: "Tyberghein Jorrit" <[EMAIL PROTECTED]>
Subject: Re: Trying to install dual boot WinNT + Linux
Date: Fri, 26 Jan 2001 18:16:33 GMT
> NT recognises FAT16 and NTFS partitions. All others might be
> problematic.
> But perhaps NT is smarter and doesn't look at the partition ID only,
> like win9x.
> show me your `fdisk -l /dev/hd[a-d]` listing. Then I might be able to
> say more.
Here is the output of 'fdisk -l /dev/hda':
Disk /dev/hda: 255 heads, 63 sectors, 730 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 * 1 261 2096451 7 HPFS/NTFS
/dev/hda2 262 730 3767242+ 85 Linux extended
/dev/hda5 262 452 1534176 6 FAT16
/dev/hda6 453 478 208813+ 82 Linux swap
/dev/hda7 479 730 2024158+ 83 Linux
Note that I tried Partition Manager 6. Strangely enough it is only able
to find the NTFS partition and doesn't recognize any of the other
partitions. In fact, it just sees one 2GIG NTFS partition and then
3.6GIG unknown space.
Greetings,
------------------------------
From: [EMAIL PROTECTED] (Uwe Malzahn)
Subject: Re: Newbee modem set-up question
Date: Fri, 26 Jan 2001 18:55:26 +0100
In article <[EMAIL PROTECTED]>,
Robert Petrillo <[EMAIL PROTECTED]> writes:
> I have a Hayes external modem attached to com1. setserial shows that
> ttyS0 is set to the correct interrupt and buffer address. minicom
> however returns "I/O Error" when run. If I cat /dev/modem I also get
> "I/O Error".
>
Do a ls -la /dev/modem, to check if it points to ttyS0. If not change the
link.
> Does this indicate that I do not have a modem driver? If so where can I
> get one and how do I install it? The disks that came with the modem only
> contain win 95/98/2000 and NT things.
If it's a proper modem (no winmodem), no driver is needed.
Cheers,
Uwe
------------------------------
Date: Thu, 25 Jan 2001 16:52:07 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: 7 CD changer
tboyd wrote:
>
> I have what seems like an unusual problem. I have two Nakamichi
> (Mountain) 7 CD changers. I had them on a win98 machine which worked,
> but I had to reboot so often (which took about 10 minutes), I've taken
> an old PC and set it up as a home server and DSL router using RH 6.2.
> When the machine boots up, it recognizes both SCSI ID's and all the
> LUN's for the CDs. After boot up, all seven of the 1st CD changer
> cdroms are recognized, but only the first CD on the 2nd changer is
> recognized? When I list /dev/cd*, I get the devices, but cdrom8 through
> cdrom13 are in red and flashing? I've never seen this before?
*** flashing red - a WARNING. You have a symbolic link naming a file that doesn't
exist. In this case scd8 thru 13.
> I can't
> mount them. I should mention that each mountable /dev/scd* is linked to
> /dev/cdrom*, but I don't see /dev/scd8-13.
>
> I looked at the SCSI-Howto, but it seems as though it's primarily for
> programmers. I'm surviving without all 14 cdroms, but with a digital
> camera, it sure would be nice to have them all recognized....
>
> Any help appreciated.
>
> Tom Boyd
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:05:53 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Editing /etc/fStab during bad boot
Jack Frosch wrote:
>
> How does one edit /etc/fStab?
>
> I opened /etc/fStab in an editor and inadvertently corrupted it when I
> closed the editor. Now of course, Linux won't boot. When I was dropped
> into bash I tried to fix the problem as root, but can't.
>
> I can see the errant text in the first line of fStab, but am not sure how to
> edit it. Vi won't let me save changes since it's a system file, and I can't
> chmod my way out of the problem.
** problem is, root filesystem is often mounted read/only. YOu need to use
mount with the option remount to get it rw. If though root can do anything, he can't
write to a r/o filesystem. Once you do that, you should be able to save fstab.
Norm (was there, did it, it was fun) Levin
I tried booting from the floppy disk,
> booting with "linux rescue" and "linux emergency LILO settings, but ran into
> the same problem.
>
> So how can I fix the problem I created??
>
> TIA.
>
> --
> Jack
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:13:38 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: filesystem corruption
Eric wrote:
>
> > > > Is this a gotcha I should be aware of (and defend against)
> > >
> > > Yes, partition tables are monsters
> > >
> > <chomp>
> > >
> > > You should post the *full* output of `fdisk -l /dev/hdc`
> > > I suspect that the problem is there, but as you stripped the important
> > > stuff, I can't tell yet.
> > >
> > > very likely, hda4 has the wrong partition ID
> > >
> > > Eric
> >
> > Thanks for your quick response, Eric. Here's the full output:
> >
> > Disk /dev/hdc: 16 heads, 63 sectors, 38792 cylinders
> > Units = cylinders of 1008 * 512 bytes
> >
> > Device Boot Start End Blocks Id System
> > /dev/hdc1 * 1 19 9544+ 83 Linux
> > /dev/hdc2 20 3894 1953000 83 Linux
> > /dev/hdc3 3895 7769 1953000 83 Linux
> > /dev/hdc4 7770 18449 5382720 5 Extended
> > /dev/hdc5 7770 8017 124960+ 82 Linux swap
> > /dev/hdc6 8018 8265 124960+ 82 Linux swap
> > /dev/hdc7 8266 12140 1952968+ 83 Linux
> > /dev/hdc8 12141 12636 249952+ 83 Linux
> > /dev/hdc9 12637 18449 2929720+ b Win95 FAT32
> >
> > Thanks for your help. G.
>
> As I expected. hda4 has the wrong ID.
> In your case it should be type 0x0F
** I'm not to sure about that. FDISK is certainly seeing this as an extended
partition.
I am surprised to see the ONLY fat partition as a logical partition. If this is
supposed
to be a bootable partition, MS system require primary. If you are just storing data
there, why make it FAT?
> (And most likely this is causing the problems you expierience)
>
> BTW, this disc has really large values for the cylinders. This is due to
> the low number
> of heads defined (16 instead of 255). Probably this is setup wrong in
> the BIOS.
> If you ever plan to reinstall everything on this disc, it's a good idea
> to change the
> detection in the BIOS.
>
> Eric
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:16:23 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: 'force user' in linux ?
David Efflandt wrote:
>
> On Thu, 18 Jan 2001, Todd Sargent <[EMAIL PROTECTED]> wrote:
> >This is a multi-part message in MIME format.
>
> Why? Use text for Unix/Linux/Perl newsgroups.
>
> >I don't know of a way to force new files in a directory to have the same owner,
> >but there is a way to have them have the same group owner. The setgid bit.
> >With the setgid bit set, no matter who creates the file in the given directory,
> >each file in that directory has the same group owner. This is a great way to
> >set up a group shared directory. Once setgid is set, simply make everyone who
> >needs access to files in a directory a member of the group and SHAZAM!! access
> >is granted.
> >
> >Here's how you make it happen. Once the target directory is created, simply
> >change the permissions.
> >
> > # chmod g+x, g+s dir_name
> >
> >There are a couple of caveats. First, the permissions must be set in symbolic
> >mode; octal won't work. Second, the execute bit for the group owner must be
> >turned on for the setgid permission to work. With setgid properly setup, the
> >ls -l output will look like this
> >
> > rwx r-sr-x
>
> Not quite and there is no reason you cannot use octals. The directory
> should be chmod 2775 so it looks like: drwxrwsr-x
>
> This will automatically give all files created there the group of the
> directory. However, users that have 022 umask would still need to do a
> chmod g+w on files for the group to be able to write to them.
>
> >Note the lowercase 's' in the group owner column. This indicates the setgid
> >bit has been properly set. If you forget to ensure the execute permission is
> >set, there will be an upper case 'S' in the group owner, indicating an
> >undefined state.
** What do you mean undefined? You very clearly stated that "S" indicates the sgid
bit IS on and "x" bit is OFF. That is very well defined. Now, a SGID set and a X bit
not set will not propogate the group - but that IS the defined state.
> >
> >
> >Andre Quinkertz wrote:
> >
> >> Hi there,
> >>
> >> How can I tell the system to always set the same user of a created file in
> >> a specific directory (no matter who creates this file). In the smbd.conf
> >> there is an option called 'force user', which is exactly what I'd need. How
> >> do I do that if the file is not created via samba, but on the normal linux
> >> level ? Can anybody help ?
> >>
> >> Many thanx in advance,
> >>
> >> Andre
> >>
> >> ____________________________________
> >> Andre Quinkertz, Dipl.Biol.
> >> Institut fuer Entwicklungsbiologie
> >> Universitaet zu Koeln
> >> Gyrhofstrasse 17
> >> 50931 Koeln
> >> Germany
> >> email: [EMAIL PROTECTED]
> >> lab phone: +49 221 470-4327
> >> lab fax: +49 221 470-5164
> >>
> >> Support PGP!
>
> --
> David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
> http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
> http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:17:33 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: general kernel compile question
[EMAIL PROTECTED] wrote:
>
> What is the benefit of running "make clean"
** gets rid of ???.o files so you are 'starting clean' when kernel is built.
in between "make dep"
> and "make bzImage"?
>
> I ask cause I've seen this on a redhat tutorial, but is not stated in
> the kernel readme.
>
> Sent via Deja.com
> http://www.deja.com/
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:20:06 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: general kernel compile question
"H.Bruijn" wrote:
>
> On Wed, 24 Jan 2001 16:39:04 GMT, [EMAIL PROTECTED] allegedly wrote:
> >What is the benefit of running "make clean" in between "make dep"
> >and "make bzImage"?
> >
> >I ask cause I've seen this on a redhat tutorial, but is not stated in
> >the kernel readme.
>
> It cleans up the files from previous make round. Normally you don't want
> to do that, as leaving them will speed the compilation process up. Make
> is designed to only recompile those sections of source code that have
> changed since the previous time make was used, or the sections which are
> effected by a change in configuration. Normally make will insure that
> when you fi add a module to support a new network interface card, you
> won't waste CPU cycles by recompiling your audio modules. In principle.
** Is this really true? It seems that one of the makes goes around and 'touches'
every blasted piece of source so it looks like somebody updated every routine and I go
thru an ENTIRE COMPILE of the kernel even if I change only one thing! Maybe it
is the "make clean" that is doing that?
>
> Now sometimes the dependancies in the makefile aren't 100% tight so that
> after a change make would not recompile every piece of source code
> effected by that change. That is were the make clean comes in. That
> cleans up all previously compiled sections, and makes sure everything is
> compiled and no stale .o files are included.
>
> So make clean may be usefull, but may also force you to longer compile
> times.
> --
> If a trainstation is the place where trains stop, what is a workstation?
> ------------------------------------------------------------------------
> Herman Bruijn mail: [EMAIL PROTECTED]
> The Netherlands website: http://hermanbruijn.com
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:28:25 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.hardware,comp.os.linux.networking
Subject: Re: Help, please, post install...
Greg wrote:
>
> I�m installing a Linux server with RH6.2. After formatting and installing
> packages,
> the system shows the message: "Performing post install configuration..." and
> nothing more happens... 10 minutes, 20...
>
> What could be the reason for this to happen??
** on occassions the graphic install hangs at this point. Try booting again from the
cd or floppy and at the bottom of the first 'install' screen, enter "text" (no quotes)
and do a text install. This WILL work.
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:43:37 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Intel 810 chipset probs
Sri Panyam wrote:
>
> HI
>
> Could you tell me how you got your X working with the intel 810 chip set
> drivers. Currently I am stuck as it cannot find the I810 drivers and is
> defaulting to a lousy resolution of 320x200!!
** in 6.2, use vga16 - at least 640*480
Go to 7.0. Detected and runs 1200*1000*16
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
Date: Thu, 25 Jan 2001 17:50:21 -0500
From: Norman Levin <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Line editor in bash
David wrote:
>
> Eric Wertman wrote:
> >
> > All-
> >
> > Could anyone tell me how to make vi the default line editor in bash?
> > set -o vi works great on the command line, but no matter where I put it
> > (be it /etc/profile, /etc/bashrc, or any of the config files in the user
> > home directory .bashrc, etc.. ) I can't make it take effect
> > automatically. There must be something simple that I am missing, but I
> > can't figure out what. Any help greatly appreciated. thanks!
** this should be a FAQ. Here is is,
every line entered is going thru /etc/inputrc (defined in the variable INPUTRC)
so bash can handle command recall like doskey. THis S****
1. Either in /etc/profile (if you own the system)
or $HOME/.bash_profile, have the line "unset INPUTRC" after it is defined. (never
know when you might want it later.
2. edit $HOME/.bashrc and as the last line, add "set -o vi"
that's it.
btw - .bashrc calls /etc/bashrc which ONLY sets up a prompt. If you define your own
prompt in .bashrc, just comment out the if statement (or remove it)
>
> If your trying to make it the default editor in X for the user. For
> Gnome look in the Gnome configuration tool for a "Default editor"
> choice. In Helix-Gnome it is under the "Document Handlers" section.
>
> --
> 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 98.992% of seti users. +/- 0.01%
--
Norman Levin
"In 1555, Nostradamus wrote: 'Come the millennium, month 12, in the home of
greatest power, the village idiot will come forth to be acclaimed the
leader.'"
------------------------------
** 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
******************************
