First, while subscribed to a number of debian lists, I am not
currently subscribed to debian-user so please reply to me or
cc me. Thank you.
I have tried to build (yet another new kernel) for a machine
that has the now usual pri/sec ide and a new Adaptec 1505
scsi card. The card is recognized
I have NOT used dpkg-split but looking at the man page (man dpkg-split)...
You probably want to create a directory (as root): mkdir /var/lib/dpkg/parts
Copy the part files from the floppies to this location. If there are MSDOS
floppies and you already have mtools then cd /var/lib/dpkg/parts
This is a case where you most definitely want to read the HOW-TOs!
The normal distribution kernels are compiled without sound support.
The reasons are probably multiple, such as smaller size but in addition
sound hardware is so non-standard that it is probably counter productive
to even try to
Well since you are grasping at straws...
1. You did _run_ lilo after changing /etc/lilo.conf, yes?
2. What does rdev /boot/vmlinuz-2.0.36 give you?
(although your root=/dev/hda1 should over-ride)
3. Should you have the argument compact in your lilo.conf?
4. You don't have one of those
Hi Michelle;
I don't believe that Linux (any distribution) and XFree will both fit
on your ZIP disk.
You can not install X without graphic card dependencies. That is,
to display X sessions on your computer there must be an xserver that
knows how to speak to your graphics hardware.
OTOH, you
While thinking in terms of absorbing is probably not too bad an idea,
it is also misleading. The terminating resistors do indeed absorb
energy (as does any shunting resistance). The SCSI bus terminating
resistance is the same as the Thin-net coaxial cable ethernet
termination.
The problem is
I have been using potato for many months on a couple of machines but
unfortunately they have NOT been upgraded since the perl changes.
The machines have NOT crashed even once in many months of operation.
One has a 2.2.5, another a 2.2.6, and yet another a 2.2.10 kernel.
On Tue, Jul 27, 1999 at
Andrew;
I suppose that my first thought would be to BIOS differences...
Have you already checked for memory holes and shadow memory?
What is the size difference between the two kernels?
While I am ignorant of the majority of the differences between 2.0-
and 2.2- kernels is there any chance that
I too am woefully ignorant of apt...
You should not need however to put on hold any package installed using
dpkg if the package that was obtained is a later release than listed in
the available.
I have done this nearly countless times without ever using hold and no
problem has ever come up.
On
Someone has posted a sequence of steps to use the build a kernel and in
looking them over they do appear correct however...
Install the kernel build package called kernel-package (in slink it is
debian/dists/slink/main/binary-i386/misc/kernel-package_6.05.deb for the
PC).
This package is THE
The reason that you should not remove the auth option in the ppp/options
file is that any and every upgrade will overwrite your changes.
The proper location for the noauth entry is in the /etc/ppp/peers/
provider (or whatever name you are using for the particular provider
with whom you are trying
I'm in danger of displaying and attitude here... I use netscape but
consider it to be the worst piece of software that I have on any Linux
box of mine. GUI web browsers that I have used on Linux are all junk
in my opinion!
Lynx OTOH is a typical UNIX program. I have never seen Lynx crash,
Well, that is because exim (nor sendmail, smail, etc.) does not do
POP2/3 (or IMAP). The MTAs (exim, etc.) use SMTP or the UNIX pipe
to process mail.
To pick up mail from an ISP when you are an intermittent connection
calls for a protocol such as POP and that function is supplied by
another
This is true but also please try to appreciate the nature and the
magnitude of the problem. Almost every distribution of Linux (and
for that matter UNIX) have critical differences in the fine details
of how certain tasks are accomplished. There is almost NO task that
has only one right way to
Yes; but /usr/local/bin should precede /usr/bin and /bin.
On Sat, Jul 17, 1999 at 02:49:29PM -0600, David Karlin wrote:
Better write your own poff version, which does the stuff you want and
then calls poff and place it into /usr/local/bin.
If you change poff directly, the changes will be
Hi Doug;
I would suggest that you do through the debian-user and debian-devel
message archive lists (ie: on debian.org).
You have probably been hit by the perl problem.
An apt-get upgrade may or may not (at this point) bring things to
a sane point.
Those of us that are using potato in
If you are using exim you can edit /etc/exim.conf and insert:
In the TRANSPORTS CONFIGURATION section:
procmail_pipe
driver = pipe
command = /usr/bin/procmail -d ${local_part}
In the DIRECTORS CONFIGURATION section:
procmail:
driver = localuser
transport = procmail_pipe
I _believe_ that
Hi Max;
Happy to hear that you got things working.
I can tell you that I have certainly turned the air blue a few times
while trying to do something with Linux/Unix (mostly with Unix come to
think of it).
One of the biggest and most frustrating things that you will likely
experience in the
Read the installation HOWTOs!
One suggestion that I can think of is that there are machines with BIOS
related problems. The installation HOWTO tells you about some BIOS
setting that might cause problems. Note that some of these problem may
only affect some kernel versions and configurations.
As usual, there is more than one way to do things in Unix/Linux...
Have you looked at cron? (man cron)
crontab -l lists any current cron job schedules and
crontab -e allows you to edit the crontab (cron table).
Basically, you would write a script to do the ftp operation,
mark the script as
No flames to you:-)
My personal experience differs from yours. However, recognize that X is a
network GUI as opposed to Windowz which is an integral GUI.
Even on a single workstation X is running as a client/server model.
Essentially every keystroke, mouse event, screen draw action, etc.
Hi Max;
Yes, if you want to use /dev/hda5 as the /usr directory then you must
choose to initialize another partition during the installation
otherwise all of the files normally destined for /usr (quite a few)
will be copied to /usr on the root filesystem.
If you were to later initialize
Peter;
Don't know what is going on with your machines but (at least) two
of my potato based machine DO have the host information present for
logins from remote machines.
On Mon, Jul 05, 1999 at 04:29:15PM +1200, Peter Dobcsanyi wrote:
Hi,
After I moved from 'slink' to 'potato' I have noticed
Hi George!
Been awhile! Nice to see that you are still around and still providing
the wonderful, thoughtful and kind assistance that I remember from the
past!
As a contrary comment... IF newbie question could be directed to a
newbie group then I believe that the suggestion would be useful.
Chris;
I upgraded a machine and installed a new win98 on my network. No special
action on the Linux side was required. I suppose that I should say NO action
of any kind was required on the Linux side of things (the new machine had
been defined before it was added). I do use encrypted passwords.
I have purchased several hub kits at 100Mb that sold for the same
price as 10Mb kits (same number of ports/same brand). I suspect that
from a manufacturing standpoint that the 10Meg stuff is obsolete.
OTOH I still agree with you in that there is no point in paying a
premium for a 100Meg hub for
Exim does permit header manipulations including rewrite.
But Art, what I am talking about is a message with _NO_ headers
even so much as hinting at the _destination_ for the message.
Fetchmail, errored out (rightfully so as far as I am concerned)
because is was presented with a message that had no destination!
I say 'rightfully so' since it is pretty
Manjo;
I am interpreting what you are saying as meaning that _every time_
I run dselect, I have to choose hold (=) to prevent dselect from
upgrading automagically. Now since I KNOW that this is not correct,
ie: I frequently run dselect and I do not 'rechoose' hold to prevent
updating what I
Steve, it is not unreasonable to want dselect (or apt) to be
configurable such that you can choose to have it not do anything that
you do not explicitely request it to do.
I have maintained a partial mirror of debian for quite some time now and
still would find it handy if dselect were to have a
I had my first 'fetchmail failure'. Fetchmail error-ed out when
retrieving mail from my ISP night before last. The fetchmail error
was something along the lines of no valid domain name for sender.
Upon starting netscape and configuring it to POP mail from my ISP,
I let netscape pull down the
I had essentially the same symptoms with an Adaptec 2840 (I believe
that is the correct model number). I did try a kernel compiled with
only the AIC7XXX SCSI support and had no change in symptom.
I tried the 1.3.1 disks and the hamm disks (the aic7xxx only kernel
disk was tried only with the
There are several. Be aware that there is a condition where you can
loose mail using the 'supress duplicate email' scripts (mentioned in
the man pages for procmail IIRC).
On Thu, May 07, 1998 at 11:21:23AM -0700, Mike Schmitz wrote:
Does anyone have a procmail recipe to supress duplicate email?
be the hostname of your system.
I don't think that there is any interaction between PAP or
CHAP and your account password (as in /etc/passwd).
On Thu, May 07, 1998 at 04:17:11PM +1100, James Whitwell wrote:
On 6/5/98 1:18 AM Bill Leach wrote:
My question would be, if Mr. Whitwell's machine is using PAP
Of course the machine's own ROMs determine if it can even boot without
a keyboard (that is not AFAIK not a Linux issue). Given that you can
boot without a keyboard, then all you have to do is set up to run a
'getty' type daemon on the serial port.
Amiga computers (m68k) for example will boot
I show crt1.o in '/usr/lib' and in '/usr/i486-linuxlibc/'.
and:
bash-2.01$ dpkg -S crt1.o
libc5-altdev: /usr/i486-linuxlibc1/lib/crt1.o
libc6-dev: /usr/lib/crt1.o
(deleted lines where 'crt1.o' was a substring)
On Tue, May 05, 1998 at 06:27:20AM +0700, Michael Acklin wrote:
Hello,
I
The file '/etc/init.d/netbase' has the commands for setting up you
IP-Masquerading. The defaults that I have seen are always to deny.
I have looked and not found any reference to a configuration tool so
I just added the necessary commands directly to the file.
In any event, check what you
Well Art don't feel too silly, what you said is essentially correct (except
for the chatscript stuff of course).
The ppp protocol does not itself have a 'host/user' concept, it is a peer to
peer protocol. In practice there typically are differences when PAP or CHAP
are involved but again the
You have to have a kernel compiled with IP Forwarding and Masquerading
enabled (the default for most kernels is _disabled_).
Then configure ipfwadm (in /etc/init.d/netbase), see man 8 ipfwadm.
On Tue, May 05, 1998 at 12:14:25PM +0800, Paul Guidera wrote:
Can anyone please point me in the right
This should be relatively easy...
Partition the drive with fdisk (cfdisk or whatever) under linux.
'fdisk /dev/sda' probably
You might want to think about putting a swap partition on this
new drive also. It _should_ be a noticable improvement if you
swap much. You probably do not do a great
Hi Gerald;
Though I noticed your original posting I did not then comment for it
being too much of a 'blind leading the blind' situation.
What I did notice is that the lines you quoted:
On Tue, May 05, 1998 at 12:25:40AM -0700, G. Crimp wrote:
[snip]
if hash encaps 2 /dev/null; then \
Correction...
I have not found any reference to a configuration tool that would work
on my system!
I did try 'dotfile ipfwadm' a couple of time but it did not work for
me and I have not yet attempted to find out why not.
On Tue, May 05, 1998 at 08:01:00AM -0600, Rick Macdonald wrote:
Bill
location
for the additional rules.
If not, I'd rather like to know why not as well as where they should
be placed.
On Wed, May 06, 1998 at 12:11:14AM +1000, Hamish Moffatt wrote:
On Tue, May 05, 1998 at 09:44:56AM -0400, Bill Leach wrote:
The file '/etc/init.d/netbase' has the commands for setting
Hi Keith;
I don't remember the 'mapping' for the device numbers now but I
think that 03:01 is the first partition on the first IDE drive.
Is that really where your root partition is located?
If not, then you can patch the correct boot device/partition into
the kernel with the 'rdev' command.
It is in the package 'bin86'
On Tue, May 05, 1998 at 08:36:53PM +0200, Christian Herold wrote:
Hi!
I've got a problem concerning Debian 1.3.1:
I wanted to compile the kernel (2.0.9) and while making zImage, the
following error occured: as86 not found
Could you please explain me where
I am more than just a little amazed at parts of this thread...
The 'Official Debian' position is not only clear it is also the ONLY
rational possition given the terms of Debian's own license.
Encouraging CD producers to review the licenses in non-free and make
their own decisions is sane.
As
No, follow the kernel-package instructions (though I think that your
doing a make dep won't actually hurt anything, you'll just be doing
something manually that make-kpkg clean will do anyway).
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL
It sounds a lot like memory that is not fast enough (or not enough wait
states).
It might also be some sort of addressing error that does not show up
in Win.
Also, did you tell lilo that you have more than 64M of memory?
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
Did you turn off fetchmail's rewrite option?
Also you can tell exim:
sender_unqualified_hosts = localhost
in it's configuration file (see the fetchmail FAQ)
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996 Micro$loth ad
You are probably using debian version 1.3 or earlier and have a ps2
style mouse.
Your message is not clear to me but...
If you move the mouse when in a console, does a mouse curson appear on
the screen?
Where did the error message you listed come from? If you received that
message when
I think that you need the file 'lmemroot.bin' to do the installation
(see the intallation text files).
Also, please turn off your 'html' mail feature.
Regis Hautière wrote:
My system(386 with 6Mo of RAM) stop during the installation from the
rescue disk after the floppy drive detection.
John is probably right (insmod psaux).
Where exactly did the 'device in use' message show up?
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996 Micro$loth ad campaign:
The less you know about computers the more you want
The 'normal' default for dpkg is --force-overwrite. This is disabled
in hamm (frozen) currently to make detection of just the sort of error
that you encounted easier to find.
Check the bug list and report if it is not already there. You can use
dpkg --install --force-overwrite package name or
After you do the 'insmod' then do o a 'cat /proc/modules' and you should
see an entry like:
psaux 11
which indicate that the mouse module is indeed loaded and known to the
kernel.
For gpm to work you need either:
/dev/psaux (or /dev/mouse and a link from /dev/mouse to
Maybe you just need to remove the symlinks. I notice that my
/usr/X11R6/lib/i486-linuxaout directory is empty (hamm).
Randy Edwards wrote:
When it rains, it pours! :-)
Does anyone know what package contains the various libraries found
in /usr/X11R6/lib/i486-linuxaout? I'm missing
I'm sure I dont' understand your question...
2) Where can I look for documantation about customizing the X
windows (not neccesarily the window manager) ?
It seems to me that most of the documantation is about the
instalation of Xfree86 and the window manager customization.
fvwm2 is,
Let me start by saying that I have no particular love for AOL. AOL has
proven to be a real PITA for me several times (including necessitating
my putting together a Windoz box to access one of their accounts for
business purposes).
However, having said that, I believe that AOL is a 'responsible'
Someone else will correct me if I have not really understood this
correctly. As I understand it, the issue is that the kernel header
files are pretty volatile but that the overwhelming majority of
applications written deal with pretty 'standard' kernel functions.
The header files were 'breaking'
it.
[EMAIL PROTECTED] wrote:
On Sun, 29 Mar 1998, Bill Leach wrote:
: In addition, I have personally received many messages from AOL
: concerning abuse reports and while I can not prove the assertions, it
: certainly looks to me as though AOL does try to prevent abuse of their
: system
bash: /usr/X11R6/bin/XF86_SVGA: unable to exec
cannot find file or directory
That sounds like it is a bash error report and not an error reported by
XF86_SVGA. I am not sure why you are not getting a better message
though it might have some to do with the amount of indirection involved
in
With a ps2 style mouse, it is either use gpm for both consoles and X or
kill gpm to run X (for whatever reasons they will not co-exist on a ps2
style mouse).
I've never seen the advantages of using /dev/gpmdata as I have never
had any problems getting my mouse to work correctly in X11. YMMV
There are probably better ways to do this (so I'll be watching for other
answers too) but you can just do a 'more /var/lib/dpkg/available' and
take a look at the 'depends' and 'recommends' sections for 'xbase' and
the appropriate 'xserver'.
You can also just run dselect. Make sure that it is a
The short answer is 'yes'.
One way would be to set up a ppp link on both the windoz box and the
linux box (easy on Linux, don't know about windoz).
Basically though, I would guess that if you can tell the Windoz dialer
to make a connection without it issuing modem command then it should
also be
I don't even know 'how far back' this was (bo, rexx?) but I do know that
at one time, I had to 'kill' gpm to run X and there was a note somewhere
in the gpm stuff about the problem.
Really, this has never been a problem for me. The only problem I had with
X11 and a ps2 mouse is related to
Unless I am misunderstanding something here (easily possible) the only
difference between zImage and bzImage kernels is the inclusion of the
boot sector code at the beginning of the file.
If you use lilo or loadlin then you don't want the bzImage but if you
cat or dd the kernel image to a floppy
Humm, I don't see xlib6_ at all.
Many of the packages that are 'missing' can be found in:
debian/project/orphaned
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996 Micro$loth ad campaign:
The less you know about computers the
Thanks to Nathan and Scott for the correct information. Though I have
used both zImage and bzImage kernels I had forgotten the comments (re
bzImage). I also hope that my comments did not cause anyone a problem.
The offset mentioned is just a (once handy anyway) means of putting a
kernel on a
As usual YMMV but yes Autoup.sh is safe to run on a running machine.
Personnally, I would suggest that you do 'shut everything down' since
replacing something as critical as libc might be a bit 'chancy' while
you have the possibility of tasks being launched.
Janos Bujtar wrote:
Hello !
I
Well I just copied resc1440.bin to a floppy last night.
Did you use?: dd if=resc1440.bin of=/dev/fd0
Mike Santner wrote:
I was attempting to install Debian from floppy but the RESC1440.BIN is
too large to fit on a 1.44 disk. This seems to make installation
impossible. Am I doing something
You are right in that it call for an option in the docs (though the
option is 'bs=512'). However, I have pretty much always just specified
the 'if', 'of' and they work.
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996
I think that it is mostly historical (just about every Unix system since
about 1970 has had 'more').
There are some other reasons that are now at least mostly not
applicable. For example, 'more' will work on a printing terminal but I
believe that 'less' will not. There might be other VDTs that
The 'bug' mentioned in 'more' when used with man pages is IMHO not a
bug. More does 'back-scroll' when viewing man pages because it is not
possible to 'back-scroll' a pipe. 'Less' does not 'show' this
characteristic behaviour of pipes because less creates its' own buffer
containing the entire
Need help. I set up Samba to let me network a Windoz 3.11 box (yuck)
with my main debian Linux machine:
Pentium 48MB
Kernel version 2.0.30
smbd Version 1.9.16p11
Passwords do not seem to work. If I set /etc/samb.conf so that the
various services are 'public' then those services are available
Did you look on a NON-US mirror site? The pgp stuff can not be posted
on U.S. based sites because the U.S. Govt. is paranoid that somehow the
U.S. defense department will be destroyed if a cryptographic system
(available worldwide) is imported into the U.S. and then exported.
Norbert Veber
Please everyone! The advice to use minicom or seyon is of course good.
Recognize however that with at least some ISPs that use CHAP, you can be
mislead by the results that you see with a term program.
The problem is that some ISPs are monitoring the port such that a
carriage return (or newline)
Yes they are related. Look at the autoup.sh script or the Libc5 -
Libc6 upgrade HOWTO.
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996 Micro$loth ad campaign:
The less you know about computers the more you want Micro$oft!
Chris, nice job!
re: http://schwarz.developer.debian.org/
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED]
from a 1996 Micro$loth ad campaign:
The less you know about computers the more you want Micro$oft!
See! They do get some
I don't think that what you are getting is exactly a bug in mirror.
While not certain about this, I believe that it is a problem experienced
by running into the maximum number of ftp users on the mirror's source
site.
--
best,
-bill
[EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL
Ok, I'm sure we can work through this...
Your /etc/nsswitch.conf file probably needs to have something like this
(this is from the hamm distribution but I believe that bo is similar):
# /etc/nsswitch.conf
#
# Example configuration of GNU Name Service Switch functionality.
# Information about this
This is just a bit of guessing here. It sounds like your messages have
to do with trying to compile crt1.o. However, there are lots of other
possibilties. It could be that the makefile uses command/syntax not
recognized by GNUmake
Do you know what the function of crt1.o is? It is possible
Yes, the first part of what you have done should be correct for a floppy
boot system. The entries in /etc/fstab do HAVE to reference the correct
device/partitions.
If you are using lilo on the boot floppy then that has to be corrected
for the new location also.
Hunter H Marshall wrote:
I
Well on my hamm system they ARE the same file (hard link):
bash-2.01$ ls -li /sbin/mke2fs
19716 -rwxr-xr-x 2 root root 17116 Jan 29 12:56 /sbin/mke2fs
bash-2.01$ ls -li /sbin/mkfs.ext2
19716 -rwxr-xr-x 2 root root 17116 Jan 29 12:56 /sbin/mkfs.ext2
Bob Hilliard wrote:
I have
Hi Helmut;
I am not really much of an expert in this sort of problem but would like
to see if I can help you to have a little better understanding of what
is going on with your system (as best as I understand anyway)...
The ROM portion of DOS loads and executes the mbr, including accessing
the
I don't remember the specific meaning of lilo stopping at LI either but
as mentioned it is explained in the LILO HOWTO.
Answering (at least partially) the question that I think that you are
asking...
You probably DO NOT need to reinstall. The problem is likely just a
lilo configuration problem
I presume that you installed the 'bo' or stable release (ver 1.3.1) and
that you installed close to the latest packages...
ppp-on and ppp-off were replaced with /usr/bin/pon and /usr/bin/poff.
Have you determined if your ISP uses PAP or CHAP?
Do you have to login with a 'username' or 'account'
Unless I am hopelessly dense (a distinct possibility) the 1024 cylinder
'limit' is a BIOS/ROM specific issue and that there are now motherboards
with BIOS/ROMs that no longer impose that limit (and that this situation
has existed for a long enough period of time that virtually all newer
machines
If you have successfully run autoup.sh then you need to run dselect
(though do read the upgrade howto's, etc.).
Autoup.sh is designed to get you to the point where deselect CAN do the
upgrade. One of the problems is that with the libc5 -- libc6 upgrade
many of the things that dselect/dpkg rely
I'm not so sure that it doesn't. That is certainly well within the
capabilities of make itself...
Joost Kooij wrote:
On Tue, 3 Mar 1998, joost witteveen wrote:
[A very interesting and informative expose, thanks Joost!]
As it works now (as I understand) the rules makefile effectively
I don't know the answer to that one but one possibility is that the
package was present but that the 'Packages' had not yet been replaced.
I did not look at the time stamps so I have no idea if that could
possibly have been the problem but only that during 'archive
maintenance' the various
FIRST--READ the Linux-Win95 HOWTOs!!
I do have, nor have I ever had Win95 on any machine that I own or
control however...
When you re-installed Win95, it appears that Win95 rewrote the mbr
(Master Boot Record) on your hard disk. This means that the code that
lilo installed there was 'blown
Pine is clearly one of the better documented programs in the world of
computing. If you press the '?' key from almost anywhere other than
when in 'compose' then pine gives you help.
As Remco mentioned the 'h' command is the one that you want to toggle
between 'all headers' and 'normal headers'.
In ftp: mget *
However, if you are running Linux on anything then I strongly suggest
that you get one of the mirror programs so that dates and symlink are
preserved. mget * will copy symlinks by copying the files. So, for
example the 6 meg+ kernel source packages will appear multiple times in
From the xconf menu:
CONFIG_MAX_16M
This is for some buggy motherboards which cannot properly deal
with the memory above 16M. If you have more than 16MB of
RAM and experience weird problems, you might want to try Y,
everyone else says N. Note for machines with more that (sic) 64MB of
RAM: in
All this makes sense, of course. What I don't understand is
how something got written to the MBR and why my floppy drive
is ignored now. To clear my SCSI drives 1 and 3, I'll have to
do a low-level format.
From an earlier message, I gathered that you DID tell the installation
program to
This really should be:
dpkg -i pathname/filename1.deb ... pathname/filenameX.deb
(or cd to pathname first).
Ben Pfaff wrote:
How can i install the both packages in Linux with the other hard disk
partition mounted or how can install the packages from floppies?
dpkg -i FILENAME1.deb
Thanks for your posting!!
I also just now received a bunch of 'returned mail messages' for
messages that I have posted to the debian-users list and have already
been sent to my mail account from debian-users.
In my case the specific postings for which the 'returned' messages were
sent were
For the 'bo' distribution (1.3.1r6) look in:
ftp://ftp.debian.org/pub/debian/bo-updates/
For hamm look in:
ftp://ftp.debian.org/pub/debian/hamm/hamm/binary-all/devel/
or:
ftp://ftp.debian.org/pub/debian/hamm/hamm/binary-i386/devel/
or for that matter:
The install and rescue floppies are indeed supposed to use the ramfs.
Obviously the install floppy has to since at least most of the time
there will be no other linux filesystem available. However, the rescue
floppy does also because it has to be able to boot fully even if the
root partition is
Normally you would boot from a rescue floppy for such maintenance
activities and mount your filesystem read only. I haven't had to do
this in a long time but it also seems to me that you can boot to single
user and remount root as read only.
FizzyPop wrote:
I keep getting {DriveReady
This certainly isn't any sort of answer to your question but does anyone
know of hand if Linux frees shared code (libraries, maybe fonts) as soon
as there is no code using it or does it wait until the memory is
actually needed). i did look through some of the loader code and
discussions about
1 - 100 of 162 matches
Mail list logo
