debian-user-digest Digest Volume 2003 :
Issue 305
Today's Topics:
Sort of OT: network logins [ Neal Lippman <nl@lippman.org> ]
Lag test. [ Mike Dresser
<mdresser_l@windsormac ]
Re: M$ Curse [ Andy <list@firman.us> ]
Re: about RUNLEVELs -- was "debian r [ Colin Watson
<cjwatson@debian.org> ]
Re: do i need stable in my sources.l [ Seneca
<seneca-cunningham@rogers.co ]
Re: setting up mysql-server [ Matthew Daubenspeck
<matt@oddproces ]
Re: sit (compressed files) [ will trillich
<will@serensoft.com> ]
Re: do i need stable in my sources.l [ Colin Watson
<cjwatson@debian.org> ]
exim - retry time not reached for an [ Jerome Acks Jr
<jracksjr@bellatlant ]
Re: what's fstype 83? "Linux&qu [ will trillich
<will@serensoft.com> ]
redirecting output to /dev/null on c [ "louie miranda"
<lmiranda@chikka.co ]
Switching groups - not working [ Ian Melnick
<dazed@vonsteuben.cps.k ]
ntp ha no installation candidate (wo [ will trillich
<will@serensoft.com> ]
Re: active programs overview [ Nathan E Norman
<nnorman@incanus.ne ]
Keyboard Lockup [ Harshu <crazyharshu@yahoo.com> ]
Re: cdrecord and BIG DISKS - Might h [ Pigeon
<jah.pigeon@ukonline.co.uk> ]
Re: kernel-2.4.18 module mis-install [ "Jonathan Brandmeyer"
<jbrandmeyer@ ]
Re: Backup Consensus? [ Pigeon
<jah.pigeon@ukonline.co.uk> ]
Re: Speculation: Debian GNU/Watch [ Nathan E Norman
<nnorman@incanus.ne ]
Date: 27 Jan 2003 20:44:55 -0500
From: Neal Lippman <nl@lippman.org>
To: Debian-User <debian-user@lists.debian.org>
Subject: Sort of OT: network logins
Message-Id: <[🔎] 1043718295.892.18.camel@gandalf>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
The subject doesn't really explain what I am looking for, but I couldn't
think of a better two-word summary. I know this is a bit off topic, but
I was hoping the collective expertise would provide me with some ideas.
I am looking for an approach to the problem of having multiple
installations of debian on each computer on my lan that I use. While it
is certainly reasonable to have a minimal install on each system,
consisting of a basic debian system, it seems counterproductive to have
to install each program that I use on each workstation, rather than
having such software "served" by a central applications server.
My present setup consists of a fileserver which exports various
directories via nfs, including both a network-wide data store (called
/share, for lack of a better idea), and /home. /home of course contains
all my home directories (for myself and everyone else using our
systems), and on each local workstation I have a full debian install
with all software, and in the /home directory the actual files are
symlinks to the appropriate nfs share.
By way of example, the workstation mounts server:/home onto /nfs; my
home directory on the workstation (/home/nl) is a symlink to /nfs/nl.
This way, no matter which workstation I log into, I have my global
/home/nl directory. Network-wide logins are handled by nis.
Currently, I install all software onto each workstation. It would be far
easier, however, to install it once onto an application server and have
it available to each workstation.
I've thought of two possible solutions:
1) Somehow get apt to install the software to /opt on the server, and
nfs mount /opt to each workstation;
2) Install as usual to the server, and have each workstaiton mount /usr
via nfs from the server;
Since /etc would be local to each workstation, the same install could
conceivably be used by each system with it operating differently because
of different config files (X comes to mind here, since hardware may
differ).
Another issue is that I use KDE. On the plus side, if I edit my kde
menus on one system to point to the appropriate places in /usr or /opt,
then since menus are stored in my home directory, I'll have the right
stuff whereever I log in. A problem, however, is that (as far as I can
tell) KDE does not understand multiple simultaneous logins, and
therefore I risk file corruption (or worse?) if I log in twice to my
account at the same time.
I had thought of solving this latter problem by implementing a login
script to copy /home/<user>/.kde from the server to local storage, and
then a logout script to sync it back onto the server at logout.
Theoretically, I would need to do this for any porgrams that cannot
sucessfully sync shared storage (like evolution), however - so this
isn't really a good overall solution. I am also unsure how to make kde
run a script at session start and end (or if there is even a way to make
this happen under KDE).
Any advice, pointers to references, etc, thoughts greatly appreciated.
nl
Date: Mon, 27 Jan 2003 21:01:39 -0500 (EST)
From: Mike Dresser <mdresser_l@windsormachine.com>
To: <debian-user@lists.debian.org>
Subject: Lag test.
Message-ID:
<[🔎] Pine.LNX.4.33.0301272101100.28564-100000@router.windsormachine.com>
Content-Type: TEXT/PLAIN; charset=US-ASCII
The current time here is 8:58, January 27th, and if it gets bad enough,
2003.
Just curious what the current lag is.
Mike
Date: Mon, 27 Jan 2003 17:39:37 -0900
From: Andy <list@firman.us>
To: debian-user@lists.debian.org
Subject: Re: M$ Curse
Message-Id: <[🔎] 200301271739.37657.list@firman.us>
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
> > Just boot from your CD and fix lilo.
>
> "fix lilo" is a large black box with "pandora" written all over
> it.
>
> that's not a helluva lot of information. it's a good idea to
> choose one of two options:
>
> 1) respond with assistance or inquiries,
> or
> 2) don't respond.
>
> given the nature of the question, your response wouldn't be
> likely to help much, i'm guessing.
Understood and my apologies to the debian-user list.
At the time, I did not see any respones to his question so I thought that by
letting him know his Debian CD was bootable, this could get him started.
(probably should have just emailed him directly)
Then others followed up with some excellent advice beyond what I could offer.
I am quite new and was helping out to the best of my ability with good
intentions. I understand your point totally and will try to make more
informative and helpful posts in the future.
andy
Date: Tue, 28 Jan 2003 02:28:46 +0000
From: Colin Watson <cjwatson@debian.org>
To: debian-user@lists.debian.org
Subject: Re: about RUNLEVELs -- was "debian rookie"
Message-ID: <[🔎] 20030128022846.GA15372@riva.ucam.org>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Mon, Jan 27, 2003 at 05:20:20PM -0600, will trillich wrote:
> update-rc.d is a good way to keep the debian package system
> up-to-date on your intentions. (sure, you may manually rename or
> remove enough files to stop the service from auto-starting, but a
> future `apt-get upgrade` may -- or may not -- undo all your hard
> work, if you work outside the packaging system.)
If you move files aside in /etc/rc*.d, a future 'apt-get upgrade' will
not undo your changes provided that you have left at least one link
there. update-rc.d isn't magic in this regard: if you remove all the
links with update-rc.d, your changes will be overwritten on future
upgrades. So 'update-rc.d <name> remove' doesn't tell the package
management system anything special, contrary to what your expectations
might be; it just removes the links, and they'll be put back on the next
upgrade. update-rc.d's remove action therefore requires an extra -f
(force) option if the target of the symlink is still there.
This is one of the reasons I tend to suggest moving the links around by
hand rather than using update-rc.d. The semantics of the latter are just
too confusing, whereas the former is relatively easily explained.
Cheers,
--
Colin Watson [cjwatson@flatline.org.uk]
Date: Mon, 27 Jan 2003 21:52:37 -0500
From: Seneca <seneca-cunningham@rogers.com>
To: debian-user@lists.debian.org
Subject: Re: do i need stable in my sources.list?
Message-ID: <[🔎] 20030128025237.GN15260@localhost>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Tue, Jan 28, 2003 at 01:43:32AM +0000, iain d broadfoot wrote:
> testing/unstable, do i still need the stable entries?
No.
--
Seneca
seneca-cunningham@rogers.com
Date: Mon, 27 Jan 2003 21:36:11 -0500
From: Matthew Daubenspeck <matt@oddprocess.org>
To: debian-user@lists.debian.org
Subject: Re: setting up mysql-server
Message-ID: <[🔎] 20030128023611.GA20068@caustic.oddprocess.org>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Mon, Jan 27, 2003 at 05:22:51PM -0600, Ray wrote:
> i seem to be missing part of the setup on this. i can't remotely
connect to
> mysqld and it seems to have the TCP port disabled, but the config files
seem
> to say it should be using the default tcp port.
>
> >a few lines from /etc/mysql/my.cnf
> [mysqld]
> user = mysql
> pid-file = /var/run/mysqld/mysqld.pid
> socket = /var/run/mysqld/mysqld.sock
> port = 3306
>
> >a few lines from /var/log/mysql.log
> /usr/sbin/mysqld, Version: 3.23.49-log, started with:
> Tcp port: 0 Unix socket: /var/run/mysqld/mysqld.sock
>
> local connections to the database work just fine.
Comment out skip-networking and restart the mysql.
Date: Mon, 27 Jan 2003 20:36:25 -0600
From: will trillich <will@serensoft.com>
To: debian-user <debian-user@lists.debian.org>
Subject: Re: sit (compressed files)
Message-ID: <[🔎] 20030128023625.GJ13740@mail.serensoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
short version: http://stuffit.com/expander
On Mon, Jan 27, 2003 at 01:27:38PM -0500, Emma Jane Hogbin wrote:
> My silly designer just sent me an .sit file. Does anyone know if this can
> be expanded in debian?
those silly designers seem to like their silly macs. (i know i
sure do...)
aladdin systems is the proud owner of the stuffit compression
algorithm, one of the oldest in the business (unless it's been
reworked from the ground up, i think it's still based on raymond
lau's code from the mid eighties... http://raylau.com/StuffIt.html)
you can grab it from http://aladdinsys.com/ or (more precisely)
http://stuffit.com/expander and they certainly have one for
linux.
NOW your problem will become, did your silly designer send you
a stuffit archive of mac-only quark or photoshop docs -- or did
she send you an industry-standard collection of jpegs and pdfs?
--
I use Debian/GNU Linux version 3.0;
Linux server 2.4.20-k6 #1 Mon Jan 13 23:49:14 EST 2003 i586 unknown
DEBIAN NEWBIE TIP #75 from USM Bish <bish@nde.vsnl.net.in>
:
Do you want to have MUTT IGNORE PGP-SIGNED MESSAGES? To have
mutt to *not* verify PGP-signed messages, you can shut it off
by including
set pgp_verify_sig = no
in your ~/.muttrc Or you could use:
set pgp_verify_sig = ask-no
to have mutt prompt you each time a signed message comes up,
with the default being not to verify.
Also see http://newbieDoc.sourceForge.net/ ...
Date: Tue, 28 Jan 2003 03:03:50 +0000
From: Colin Watson <cjwatson@debian.org>
To: debuser <debian-user@lists.debian.org>
Subject: Re: do i need stable in my sources.list?
Message-ID: <[🔎] 20030128030350.GA15702@riva.ucam.org>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Tue, Jan 28, 2003 at 01:43:32AM +0000, iain d broadfoot wrote:
> testing/unstable, do i still need the stable entries?
No. testing started out as a copy of stable. The exceptions are if you
happen to want something that's been removed from testing since the last
stable release; even then you probably didn't want it anyway, and if you
do you can always get it by hand.
--
Colin Watson [cjwatson@flatline.org.uk]
Date: Mon, 27 Jan 2003 22:04:40 -0500
From: Jerome Acks Jr <jracksjr@bellatlantic.net>
To: debian-user@lists.debian.org
Subject: exim - retry time not reached for any host
Message-ID: <[🔎] 20030128030440.GB21887@xanadu.home.l>
Content-Type: multipart/signed; micalg=pgp-sha1;
protocol="application/pgp-signature"; boundary="LZvS9be/3tNcYl/X"
Content-Disposition: inline
For mail I use a fetchmail+procmail+mutt+exim combination. Not every
message I send but a few don't get sent to my ISP's SMTP server.
When I run eximon and check the message log of messages awaiting
delivery, I will see entries like:
2003-01-27 20:09:01 ButchersHill@yahoogroups.com T=remote_smtp defer
(-44): retry time not reached for any host
2003-01-27 20:23:01 ButchersHill@yahoogroups.com T=remote_smtp defer
(-44): retry time not reached for any host
2003-01-27 20:38:02 ButchersHill@yahoogroups.com T=remote_smtp defer
(-44): retry time not reached for any host
2003-01-27 20:53:01 ButchersHill@yahoogroups.com T=remote_smtp defer
(-44): retry time not reached for any host
2003-01-27 21:08:02 ButchersHill@yahoogroups.com T=remote_smtp defer
(-44): retry time not reached for any host
The message will stay in the queue until I tell exim to send it.
When I check the retry rule for the SMTP server, I get:
$ /usr/sbin/exim -brt outgoing.verizon.net
Retry rule: * F,2h,15m; G,16h,1h,1.5; F,4d,8h;
From the retry rule, I expect exim to try to resend the message every
15 minutes for the first two hours. From the above log entries, the retry
rule does not appear to work as I expect.
Any suggestions on how I can fix this?
--
Jerome
Date: Mon, 27 Jan 2003 20:58:00 -0600
From: will trillich <will@serensoft.com>
To: debian-user@lists.debian.org
Subject: Re: what's fstype 83? "Linux"?
Message-ID: <[🔎] 20030128025800.GK13740@mail.serensoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Mon, Jan 27, 2003 at 04:22:21PM -0800, nate wrote:
> will trillich said:
> > ideas? (i think this was my slink disk drive -- i'd like to
> > use it to alleviate some space pressure on my woody
> > server...)
>
> what does e2fsck say for those drives you cannot mount? Try
> running a read-only pass on them. I can't imagine why the
> newer kernel would be unable to mount a slink partition(though
> I can see it happening the other way around), though I haven't
> personally tried it.
root: /mnt# e2fsck /dev/hdb1
e2fsck 1.27 (8-Mar-2002)
Couldn't find ext2 superblock, trying backup blocks...
e2fsck: Bad magic number in super-block while trying to open /dev/hdb1
root: /mnt# e2fsck /dev/hdb5
e2fsck 1.27 (8-Mar-2002)
Couldn't find ext2 superblock, trying backup blocks...
e2fsck: Bad magic number in super-block while trying to open /dev/hdb5
N.B. an earlier thread noticed "bad magic" mentioned at or
before LILO, so it may have been this type of thing
(certainly not the file-detector 'magic number' theory)...
files on /dev/hdb2 have modification times no later than
september 2000 -- pre-ext3 by a long shot. and i'm *positive*
i've never even tried reiserfs, certainly not two-and-a-half
years ago. wasn't ext2 the default for formatting under the
potato or slink install? (as i recall, potato would start out as
ext2 and then offered an ext3 option later... nope, ext3 didn't
work either.)
> and partition type 83 is linux yes, but it's just a partition type,
> many kinds of filesystems can reside in there.
racking my brain (what there is left of it) i stir no memory of
anything unusual, file-system-wise. i'm just about certain that
all three of these partitions would be the same file system.
yet /dev/hdb2 mounts like a charm.
--
I use Debian/GNU Linux version 3.0;
Linux server 2.4.20-k6 #1 Mon Jan 13 23:49:14 EST 2003 i586 unknown
DEBIAN NEWBIE TIP #76 from USM Bish <bish@nde.vsnl.net.in>
:
To SEARCH THE CONTENTS OF A TAR.GZ file without having to
extract everything:
tar -tzf file.tar.gz | grep something
Also try zcat.
Also see http://newbieDoc.sourceForge.net/ ...
Date: Tue, 28 Jan 2003 11:06:39 +0800
From: "louie miranda" <lmiranda@chikka.com>
To: <debian-user@lists.debian.org>
Subject: redirecting output to /dev/null on cron not working!
Message-ID: <[🔎] 005b01c2c67a$46fe35d0$0b00000a@nocpc3>
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
I have a script that is on a cron basis, It runs every hour.
I have read a document that if you dont want any output.
You can add >/dev/null 1>&2 to redirect it to /dev/null
But i still received email about those output, is this syntax
im trying to add on my cron for debian correct?
/scripts/cron_inactivityalert.sh >/dev/null 1>&2
Please advise
--
thanks,
louie miranda
Date: Mon, 27 Jan 2003 21:08:32 -0600
From: Ian Melnick <dazed@vonsteuben.cps.k12.il.us>
To: debian-user@lists.debian.org
Subject: Switching groups - not working
Message-ID: <[🔎] 20030128030832.GA24172@vonsteuben.cps.k12.il.us>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Hello, everyone
I want certain users to be able to switch into different groups. So I
did a usermod -g/-G on them and checked the group file to make sure it
did it right (and it looks like it did). So then I try out the sg and
newgrp commands. Problem is, these groups don't have passwords. In the
group file, the password field is an 'x', and in the group shadow file
it's a '*'. But when I just press return at the switch group password
prompt, it doesn't accept it.
What am I doing wrong?
Thanks
Ian
Date: Mon, 27 Jan 2003 21:08:03 -0600
From: will trillich <will@serensoft.com>
To: debian-user@lists.debian.org
Subject: ntp ha no installation candidate (woody)
Message-ID: <[🔎] 20030128030803.GL13740@mail.serensoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
after revisiting the longest-thread-of-the-millennium again, i
thought i'd be a good citizen and get ntpd going instead of
ntpdate.
# apt-get update
<snip>
Fetched 257kB in 4s (62.2kB/s)
Reading Package Lists... Done
Building Dependency Tree... Done
root: /mnt# apt-get install ntpd
Reading Package Lists... Done
Building Dependency Tree... Done
Package ntpd has no available version, but exists in the database.
This typically means that the package was mentioned in a
dependency and
never uploaded, has been obsoleted or is not available with the
contents
of sources.list
E: Package ntpd has no installation candidate
root: /mnt# apt-cache show ntpd
root: /mnt#
odd! (suggestions welcome.)
--
I use Debian/GNU Linux version 3.0;
Linux server 2.4.20-k6 #1 Mon Jan 13 23:49:14 EST 2003 i586 unknown
DEBIAN NEWBIE TIP #11 from Will Trillich <will@serensoft.com>
:
Which COMMANDS pertain to <xyz>? Try "apropos <xyz>",
"info <xyz>", and "man -k <xyz>".
Also see http://newbieDoc.sourceForge.net/ ...
Date: Mon, 27 Jan 2003 21:24:56 -0600
From: Nathan E Norman <nnorman@incanus.net>
To: debian user <debian-user@lists.debian.org>
Subject: Re: active programs overview
Message-ID: <[🔎] 20030128032456.GA27325@incanus.net>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Fri, Jan 24, 2003 at 01:11:41PM -0600, Ron Johnson wrote:
> On Fri, 2003-01-24 at 09:52, Joris Huizer wrote:
> > Hello,
> >
> > What is the program to use when I want to know which
> > programs are running ?
>
> $ ps -ax
> $ ps -axf
> $ ps -axf --cols=`echo $COLUMNS`
I hate to nitpick, but all your examples should leave off the "dash"
(the '-' character): ps has been hacked so that it exhibits system V
behavior when provided options preceded by a dash, and bsd behavior
otherwise (but it also tries the bsd behavior when the system V option
you've provided don't make sense).
$ ps -V
procps version 2.0.7
$ ps -x
Bad syntax, perhaps a bogus '-'?
PID TTY STAT TIME COMMAND
3553 ? S 0:00 /usr/sbin/sshd
3554 pts/0 S 0:00 -bash
3590 pts/0 R 0:00 ps -x
$ ps -ef
UID PID PPID C STIME TTY TIME CMD
root 1 0 0 Jan17 ? 00:00:17 init
root 2 1 0 Jan17 ? 00:00:00 [keventd]
[ snip ]
nnorman 3554 3553 0 21:18 pts/0 00:00:00 -bash
nnorman 3591 3554 0 21:21 pts/0 00:00:00 ps -ef
$ ps ax
PID TTY STAT TIME COMMAND
1 ? S 0:17 init
2 ? SW 0:00 [keventd]
[ snip ]
3554 pts/0 S 0:00 -bash
3592 pts/0 R 0:00 ps ax
Otherwise a good post; I find 'ps fax' especially useful when I'm
trying to figure out which process needs killing.
--
Nathan Norman - Incanus Networking mailto:nnorman@incanus.net
When you're in command, command.
-- Adm. Chester W. Nimitz
Date: Mon, 27 Jan 2003 19:28:34 -0800 (PST)
From: Harshu <crazyharshu@yahoo.com>
To: debian-user@lists.debian.org
Subject: Keyboard Lockup
Message-ID: <[🔎] 20030128032834.72193.qmail@web10508.mail.yahoo.com>
Content-Type: text/plain; charset=us-ascii
Hi Folks,
I am having this wierd problem. When I launch a X session from the
prompt and exit the keyboard locks up and become unresponsive. This
happens only when I am the root user. The freeze up also happens when I
hit Cntrl+Alt+ Fn to switch to the terminal when running a X session.
The display card is ATI 3D Rage.(ati driver) The readme for ATI under
xserver-xfree86 says that lockup can occur when a wrong mouse is
choosen. I have a ps/2 mouse with scroll wheel (disabled) and it is
Dell Dimension 4300. I have ps/2 as the mouse driver. I am using the
unstable tree.
I would appreciate some pointers
Thank you
regards
Harshu
=====
Never underestimate the predictibitly of stupidity!
__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com
Date: Mon, 27 Jan 2003 23:08:58 +0000
From: Pigeon <jah.pigeon@ukonline.co.uk>
To: debian-user@lists.debian.org
Subject: Re: cdrecord and BIG DISKS - Might help someone
Message-ID: <[🔎] 20030127230858.GA469@pigeon.pigeonloft>
Content-Type: text/plain; charset=unknown-8bit
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
On Mon, Jan 27, 2003 at 11:47:37AM +0800, csj@mindgate.net wrote:
> On Sat, 25 Jan 2003 21:28:16 -0800,
> Marc Wilson wrote:
> >
> > On Sun, Jan 26, 2003 at 01:34:17AM +0000, Pigeon wrote:
> > > cdrecord -v -speed=32 dev=x,y,z -dao -isosize filename.raw
> >
> > The hitch was the instruction to cdrecord to write the disc in
> > DAO mode. Many many recorders cannot deal with the CUE sheet
> > they're sent in DAO mode unless they're told the total size of
> > the image.
>
> What works for me is option "-ignsize":
>
> echo "burn 800MB" ; cdrecord -v dev=lite8 -dao -ignsize cdrom.iso
Hmm, didn't work for me.
>From man cdrecord:
-ignsize
Ignore the known size of the medium. This options
should be used with extreme care, it exists only
for debugging purposes don't use it for other rea
sons. It is not needed to write disks with more
than the nominal capacity.
Very odd.
Pigeon
Date: Mon, 27 Jan 2003 22:26:52 -0500
From: "Jonathan Brandmeyer" <jbrandmeyer@earthlink.net>
To: <debian-user@lists.debian.org>
Subject: Re: kernel-2.4.18 module mis-install
Message-ID: <[🔎] 00cd01c2c67d$1ac75d40$6400a8c0@nomadic.ncsu.edu>
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
> "Jonathan Brandmeyer" <...> writes:
> > Installed the stock kernel-image-2.4.18-i686 debian package, with
dselect.
> > Installed and unpacked kernel-source-2.4.18, with dselect and tar.
> > Downloaded and unpacked nvidia-kernel-source and nvidia-glx-source, with
> > apt-get and tar.
> > Built nvidia-kernel-source the Debian Way:
> > (in /usr/src/kernel-source-2.4.18 )
> > make oldconfig
> > kpkg-make modules
>
> You need to make sure you get the .config file that goes with the
> kernel, which isn't going to be the default one in the kernel-source
> package. Try copying /boot/config-2.4.18-686 to .config, and then
> running 'make-kpkg modules-image' again.
Thanks. It worked perfectly.
> (The Righter Way to do this, from what I can tell, is even more of a
> pain; you can get the kernel-image-$KVERS-i386 source package, which
> build-depends on the kernel-source-$KVERS binary package; you need to
> emulate its behavior to build custom modules, though. Ick. You can
> also do things with kernel-build-* packages, but only in unstable, and
> again not terribly cleanly.)
>
> --
> David Maze dmaze@debian.org http://people.debian.org/~dmaze/
> "Theoretical politics is interesting. Politicking should be illegal."
> -- Abra Mitchell
>
Date: Mon, 27 Jan 2003 23:22:37 +0000
From: Pigeon <jah.pigeon@ukonline.co.uk>
To: debian-user@lists.debian.org
Subject: Re: Backup Consensus?
Message-ID: <[🔎] 20030127232237.GB469@pigeon.pigeonloft>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Mon, Jan 27, 2003 at 11:03:47PM +1100, bob parker wrote:
> On Mon, 27 Jan 2003 08:32, Grant Bowman wrote:
> > Is there a place where a general consensus has been reached on exactly
> > what is necesary to backup a Debian system? I'm sure this has been
> > asked and answered many times before, so I am looking for URLs to where
> > this has been discussed in the past.
> >
> > I apologize in advance, but I'm not a subscriber of this list. Please
> > cc me on replies.
> >
> > Thank you very much,
>
> FWIW I first of all dump my postgres databases into $HOME, then make a
list
> of my installed (debian) packages, also in $HOME.
>
> I then backup $HOME excluding browser cache files, /etc and /usr/local.
>
> My idea is that after a disaster I'd make a minimum debian install,
restore
> $HOME, /usr/local. After that I'd reinstall my packages from the list I
> gathered and then selectively restore /etc to get my configs back the
way I
> had them.
>
> I've never used this in anger and I'd welcome any suggestions from the
wiser
> heads out there.
Well, this is almost exactly what I did when I found I couldn't
straightforwardly upgrade from slink to woody, and needed to install
woody but keep all my settings from slink. Main difference is I didn't
save /usr/local, since it was full of stuff I'd need to recompile
anyway to work with the upgraded libc6. It worked... very well. See
thread "Can't upgrade from slink to woody" or similar. The only
problems I had were specific to the upgrade situation and wouldn't
arise when restoring the original version.
Couple of suggestions:
- might be an idea to save /var as well
- if you're into kernel customising, make sure you have a rescue
kernel with built-in support for all your critical hardware (ie, not
as modules). I NEARLY got caught without a kernel that would recognise
my Initio 9100UW SCSI card, but managed to find one. (Phew!)
Pigeon
Date: Mon, 27 Jan 2003 21:31:08 -0600
From: Nathan E Norman <nnorman@incanus.net>
To: debuser <debian-user@lists.debian.org>
Subject: Re: Speculation: Debian GNU/Watch
Message-ID: <[🔎] 20030128033108.GB27325@incanus.net>
Content-Type: multipart/signed; micalg=pgp-sha1;
protocol="application/pgp-signature"; boundary="EVF5PPMfhYS0aIcm"
Content-Disposition: inline
On Fri, Jan 24, 2003 at 01:27:33PM -0500, Mark L. Kahnt wrote:
> Overcoming your typical limitations of watches, the Debian GNU/Watch is
> a new, open source solution to portable timekeeping. A multi-user
> solution, multiple people can elect to wear the same watch, and for one
> person, it could be analog, for another digital, for a third a sundial,
> for a fourth, a stopwatch. With wireless networking, its time server
> ability allows it to correct the time of any clocks or vcrs the user
> passes.
>
> Avoid the problems of the Microsoft WinWatch, with it's virus problems,
> and tendencies to try to read 13:00 AM/PM, at which point it crashes to
> a BSoD.
You need to subscribe to debian-curiosa :-)
--
Nathan Norman - Incanus Networking mailto:nnorman@incanus.net
prepBut nI vrbLike adjHungarian! qWhat's artThe adjBig nProblem?
-- alec flett @netscape