The f-word in kernel source (Was: Re: Emacs, RMS and sex(6))

[Remco Blaakmeer]

On Mon, 17 May 1999, Edward Betts wrote:

> All food for thought. Finally on a slightly serious note, sex(6) is quite
> explicit and some might find it offensive. We have fortune-off separate, so
> people can leave it uninstalled. I suppose the kernel source does include the
> word f*** a good few times.

Just for the fun of it, I just counted how many times this word (or any
derivative of it) is mentioned in the 2.2.9 kernel source. I found 24
files, with a total of 32 occurrences of it.

Remco
-- 
rd1936:  1:40am  up 1 day,  6:24,  3 users,  load average: 1.27, 1.57, 1.44

Re: jdk not working in potato, working jdk removed from incoming, license problem

[Phillip R. Jaenke]

On Mon, 17 May 1999, Seth M. Landsman wrote:

>   What is wrong with distributing an installation package like is
> done with netscape and realaudio?

Hrm. You know, that didn't occur to me. As long as it contains NOTHING of
JDK, that's good. :)
 
>   For the record, kaffe is *NOT* as good as the blackdown JDK.  I
> have used both, and, as it is, kaffe crashes before my research system
> loads, yet the blackdown jdk works flawlessly.

Well, I can't say I've used Kaffe, so whatever works for you(tm). It's
GPL'd, which is better than JDK currently. ;P

-prj 

Re: Bug#37606: /var/spool/texmf/ls-R unwritable

[Julian Gilbey]

> The main reason I didn't want to have mktex{mf,tfm,pk} be setuid is
> because they run all sorts of different programs - metafont, gsftopk,
> etc. - which can (IIRC) be replaced by the user.  Even if they can't,
> their inputs can, and the inputs are turing-complete macro languages.
> If mktex{mf,tfm,pk} drop privileges before invoking the real generator
> programs, I'll be happy.

I don't think it would work to drop privileges before starting up the
generator programs -- that would defeat the point.  But what must be
done is: clear the environment, reset PATH to something known and
secure and setuid(geteuid()).  The combined effect of the first and
third of these would also result in the texmf search paths being
unaffected by anything that the user might do, which is crucially
important.  Resetting PATH prevents the user from getting their own
programs into the works.  And the fact that this will run as a
dedicated user (tex) means that if there were any security holes, the
worst that could be done is to interfere with the generated fonts,
which would be hardly worse than the present situation.  And
hopefully, the result will be secure, and then we are a lot better
off.

> I would also rather not install suidperl if it can be avoided.

I had realised that from other people's postings on another issue.
It's something I'm thinking about, but my ideas on how to write these
scripts as setuid scripts (even with a wrapper) are still in pre-alpha
stage.  Part of the difficulty is that the Web2C system allows the
binaries to be installed anywhere.  I have to ensure that the PATH
contains the correct directory if both (1) the script is running
setuid and (2) the directory of the script is not /usr(/local)/bin.
I'm thinking about it

   Julian

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

  Julian Gilbey, Dept of Maths, QMW, Univ. of London. [EMAIL PROTECTED]
 Debian GNU/Linux Developer.  [EMAIL PROTECTED]
   -*- Finger [EMAIL PROTECTED] for my PGP public key. -*-

Re: better /etc/init.d/network

[Erik]

On Sun, May 16, 1999 at 10:15:48PM +0200, Massimo Dal Zotto wrote:
> Hi,
> 
> I have written a generic network interface management command, net, which
> can be used to start/stop/show/configure network interfaces, and a smarter
> replacement for the /etc/init.d/network script.
> 
> The net command makes use of configuration files stored in /etc/network/
> which contain the various interface options. For example my eth0 is:
> 
>   # /etc/network/eth0
>   IPADDR=192.168.0.1
>   NETMASK=255.255.255.0
>   NETWORK=192.168.0.0
>   BROADCAST=192.168.0.255
>   GATEWAY=192.168.0.1
> 
Howabout instead of having eth0, eth1, etc. have like home, work, etc.
the files could then have an extra section, called DEVICE or something, that
would be eth0, eth1, etc. It could also have multiple DEVICE sections, so that
it would setup all the adapters related to that network.
This would be most usefull on laptops, but usefull on desktop machines too.
I know some people take their desktop machines arround with them every once
in awhile(I take mine to the local LUG every other month or so).  You could
then add the ability to do like, net start home eth0, to start individual parts
of your home network, while net stop eth0 would still disable eth0.

> 
> The advantage is that you can now start/stop specific interfaces with simple
> commands using predefined configs, while the old script could only be used
> to start the entire network and couldn't stop or restart it or part of it.
> 
> The new /etc/init.d/network script just calls the /usr/sbin/net command,
> which does all the real work, with the proper args, just start or stop, and
> all the configuration options are now stored as separate config files.
> 
> The package can be installed over an slink system because the preinst script
> can convert automatically the old network file to the new eth0 config.
> 
> The package is available at the following location:
> 
>   http://www.cs.unitn.it/~dz/debian/net_1.0-1_all.deb
> 
> Please have a look and see if it can added to the main debian distribution.
> 
> -- 
> Massimo Dal Zotto

Overall it sounds pretty good to me, something just a little better, to make
things just a little easier.

Erik Bernhardson
[EMAIL PROTECTED]

---
"[T]he last thing I want to do is spread fear, uncertainty and doubt
in [the users'] minds."
- Don Jones, Microsoft's Y2K Product Manager



pgpUKZrgmaOeR.pgp
Description: PGP signature

Re: jdk not working in potato, working jdk removed from incoming, license problem

[Seth M. Landsman]

> Basically, we're in BLATANT violation of the license currently. It states
> quite clearly that redistribution is prohibited. So, plain and simple,
> we're shit out of luck. As someone else pointed out, Kaffe is just as
> good, with better response. But either way, we have to lose jdk or
> convince Sun to grant us special permission to redistribute.

What is wrong with distributing an installation package like is
done with netscape and realaudio?

For the record, kaffe is *NOT* as good as the blackdown JDK.  I
have used both, and, as it is, kaffe crashes before my research system
loads, yet the blackdown jdk works flawlessly.

-Seth

--
"It is by will alone I set my mind in motion"

Re: (FINISH) Correct non-US solution

[Chris Waters]

Jonathan Walther <[EMAIL PROTECTED]> writes:

> No.  The scheme makes us less liable than we already are, since it shows
> that we are "trying".

Excuse me?  Are you a lawyer, or have you consulted with competent
legal advisors in order to arrive at this *theory*?  I suspect not,
and I suspect that you are sadly mistaken.  (But IANAL.)  It could put
us in the position of *seeming* to be offering legal advice, and could
open us up to accusations of misrepresentation and practicing law
without a license.

I find it touching that you have such an innocent view of the world's
legal systems, but in many cases, "trying" is worse than doing
nothing.  It's a sad but true fact that if you try and fail to save
someone's life (and maybe even if you succeed), the family (or the
state) may sue you for practicing medicine without a license.
(Esp. if you happen to be in the US, where lawsuits are a Way Of
Life.)

I'm not saying we shouldn't do this, period, I'm saying we shouldn't
do this without legal consultation first.  Intuition and the Law are
*often* directly opposed, so it would be foolish for us to be guided
by pure intuition here.  Ignorance of the law is not a valid legal
excuse for doing something.  Being earnest, having puppy-eyes, and
protesting, "I was only trying to help," doesn't cut the mustard.

-- 
Chris Waters   [EMAIL PROTECTED] | I have a truly elegant proof of the
  or[EMAIL PROTECTED] | above, but it is too long to fit into
http://www.dsp.net/xtifr | this .signature file.

Re: Old Library dependencies Re: Release Plans (19990513)

[Richard Braakman]

Hartmut Koptein wrote:
> > >   Remove as many dependencies on old libraries as possible, this
> > >   includes:
> > > 
> > > libjpegg6a, libncurses3.4, newt0.25, libpgsql, tk4.2, tcl7.6,
> > > libwraster1, libpng0g
> > > 
> > >   and various older gtk/gnome libraries.
> > 
> > Lintian has a "depends-on-obolete-package" tag.  I will add newt0.25,
> > libpgsql, tk4.2, tcl7.6, libwraster1, and libpng0g to its list.
> > This way, the Lintian page for that tag will form a useful index.
> 
> Why not  libjpegg6a and libncurses3.4?

I didn't have to add them, they were already there :)

> What about  xbase, emacs19, altgcc, freetype1, libc5, libg++27, libpaper,
> tclx76  and possible some more?

Most of those aren't obsolete in this sense.  emacs19 is still standard!
There's also no point in adding packages that aren't in the distribution
anymore, since they will show up in the dependency check.  But I'll add
xbase.

Richard Braakman

Re: jdk not working in potato, working jdk removed from incoming, license problem

[Phillip R. Jaenke]

On Mon, 17 May 1999 [EMAIL PROTECTED] wrote:

> Hi
> I am given to understand that someone has found a problem in the license of
> jdk, to the point that same person finds that debian cannot distribute
> the jdk at all. I was told that the problem found in the license has
> existed for a long time. 

This is true. I did confirm this on IRC today. The JDK specifically states
that you may not redistribute it in part, or in it's entirety, outside a
single organization, with or without any other software, flat out NO,
Period.

> If this is the case, 
> WHY is a jdk that doesn't even work in potato? By precisely the same token, 
> why is there a jdk in ANY debian dist?? Is there a difference in the
> license between versions? Has anyone talked to Sun?

I don't know whether or not there is; if somebody could confirm/deny this,
it would be appreciated.

> If this is NOT the case,
> Can this be resolved quickly please? I would imagine that it is the
> intent of  Sun that Java in its pure form would make it big. I have a
> few java projects  that I'm taking off the back burner presently, and
> now this.
> Inquiring, jdk-using minds want to know.

Well. Unfortunately, after checking things over with our lawyer at work
(he went over the license briefly) this is the only viable solution:

1) change the license.
2) *IMMEDIATELY* remove *ALL* jdk packages from *ALL* maintained Debian
   distributions.

Basically, we're in BLATANT violation of the license currently. It states
quite clearly that redistribution is prohibited. So, plain and simple,
we're shit out of luck. As someone else pointed out, Kaffe is just as
good, with better response. But either way, we have to lose jdk or
convince Sun to grant us special permission to redistribute.

-Phillip R. Jaenke, Head Unix Guru, Unicent Telecom
 216-344-2603 / ~9a->~5p Eastern -> Pester Me!
 --- I WILL NOT BE IN MY OFFICE ON MEMORIAL DAY. :P ---

Re: VX Chipsets and 2.2.5

[Phillip R. Jaenke]

On Sun, 16 May 1999, Robert Woodcock wrote:

> Michael Beattie wrote:
> >I have been told by an aquaintance that linux 2.1.x or greater kernels are
> >unlikely to boot on a Motherboard that uses a VX Chipset. I have a
> >VXPro...
> Well at least your friends know where to get good crack.

Aye.
 
> 2.2.5 runs fine on VX boards, and oh, BTW, the VXPro is absolutely not a VX.
> It's a clone with a similar feature set.

Former VXpro owner here. It's a similar feature set, yes. But not
implemented anything like VX. Be prepared for a nightmare of IDE resets
(which are a *REAL* bad thing) and kernel Oopses every few days. I'd
suggest replacing the board. Super7's are really cheap now; FIC VA-503+ is
down to about $60.

-prj

Emacs, RMS and sex(6)

[Edward Betts]

I was looking at the new version of miscfiles, in the changelog, I noticed
that the previous version (the one before this one)'s changelog entry read
like this

miscfiles (1.1-6) unstable; urgency=low

  * Removed duplication of the GNU Manifesto. closes: BUG#29565

 -- Manoj Srivastava <[EMAIL PROTECTED]>  Wed, 25 Nov 1998 02:18:23 -0600

I immediately thought of Emacs. I remembered that keypress C-h C-p would display
the GNU-manifesto, so I thought the miscfiles changelog might be linked with
the fact that Emacs can display the GNU Manifesto without miscfiles installed.

But I was wrong, both include a copy of the manifesto, identical except the
miscfiles version is compressed.

$ ls -l /usr/share/misc/GNU-manifesto.gz /usr/share/emacs/20.3/etc/GNU
-rw-r--r--   1 root root26334 Oct  4  1998 
/usr/share/emacs/20.3/etc/GNU
-rw-r--r--   1 root root10459 May 14 21:56 
/usr/share/misc/GNU-manifesto.gz
$ zcat /usr/share/misc/GNU-manifesto.gz | diff /usr/share/emacs/20.3/etc/GNU -
$

I was going to file a bug, but which package? emacs20 or miscfiles, neither is
really at fault and there is no real solution. The only solution I can think
of is moving the GNU-manifesto to its own package and have emacs20 depend on
it, with a symlink, but it is only 27k, how many people want the GNU-manifesto
installed, but neither emacs20 or miscfiles. Nobody is worrying about the loss
of disk space either.

So, I was going to file a bug against general, but then I noticed the other
files in the /usr/share/emacs/20.3/etc/

For starters a copy of the GPL: 

$ diff COPYING /usr/doc/copyright/GPL
5c5
<  59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
---
>59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
$

So that one is a bit odd, and could be fixed with a symlink, does that pass
lintian?

But have a look at some of the other files:

CENSORSHIPCensoring my Software by Richard Stallman (thoughts on CDA)
COOKIES   Mrs Fields Cookies recipe (that one about paying $200 for it)
COPYING   The GNU General Public Licence (mentioned before)
GNU   The GNU Manifesto (mentioned above)
HELLO This is a list of ways to say hello in various languages.
INTERVIEW Richard Stallman discusses his public-domain (interview with byte)
JOKES Some gnu jokes
LINUX-GNU Linux and the GNU system
LPF   League for Programming Freedom
MAILINGLISTS  GNU Project Electronic Mailing Lists and gnUSENET Newsgroups
MOTIVATIONSTUDIES FIND REWARD OFTEN NO MOTIVATOR (why non-free is bad)
OTHER.EMACSES How is this Emacs different from all other Emacses
SERVICE   GNU Service Directory (Consultancy)
WHY-FREE  Why Software Should Not Have Owners by Richard Stallman
celibacy.1Joke man page in catman format
condom.1  Joke man page in catman format (a bit offensive)
copying.paper Why Software Should Be Free by Richard Stallman
ctags.1   A .so link to etags.1
echo.msg  The man page for GNU(1) echo phrased as an email
emacs.1   The emacs manual in roff format
emacstool.1   The emacstool manual in roff format
etags.1   The etags manual in roff format
foo.elI have no idea what this is
future-bugSubject: M-x search-backward-in-time broken... (another joke)
rgb.txt   Similar, but not identical to /usr/X11R6/lib/X11/rgb.txt
sex.6 Joke man page in catman format (quite offensive)
spook.lines   Words for `M-x spook' (look up NSA in the emacs manual)
tasks.texiThe GNU task list in texinfo format
termcap.src   A copy of the termcap database?
yow.lines Zippy quotations (from comic books and strips by Bill Griffith) 

I mean no wonder emacs is big, /usr/share/emacs/20.3/etc/ takes 5Mb, although
there are some other emacs docs there. Could any of these be compressed? or
symlinked from /usr/doc/emacs20, where I would expect to find them. Well,
really I would not. I halve think that sex.6, condom.1, and celibacy.1 are in
funny-manpages, I would look in Contents-I386.gz but I lent my slink CD to
somebody.

Should the emacs20 package be split, with some of RMS' `ideas' moved to a
gnu-philosophy package? What do we do with the GNU jokes? leave them there?

All food for thought. Finally on a slightly serious note, sex(6) is quite
explicit and some might find it offensive. We have fortune-off separate, so
people can leave it uninstalled. I suppose the kernel source does include the
word f*** a good few times.

-- 
I consume, therefore I am


pgpKgc1wktb3x.pgp
Description: PGP signature

Re: Compiling wordinspect for potato

[James Mastros]

On Mon, May 17, 1999 at 05:02:30PM -0400, Bob Hilliard wrote:
>  I installed libglib1.2_1.2.3-1.deb, libglib1.2-dev_1.2.3-1.deb,
> libgtk1.2_1.2.3-1.deb, and libgtk1.2-dev_1.2.3-1.deb, and built the
> package.  It compiled cleanly, and runs as it did in slink, except
> that every time it pops up a window it displays the following message:
> 
> > Gtk-WARNING **: gtk_scrolled_window_add(): cannot add non scrollable widget
> > use gtk_scrolled_window_add_with_viewport() instead
> 
>  Would another libgtk be a better choice to replace libgtk1.1,
> and possibly get away from this warning message, or will the source
> have to tweaked to avoid this?

You'll have to make some changes to the source; they aren't too difficult --
I don't remember specifics, though.  I'd stick with gtk (et al) 1.2; things
in potato depending on obselete libraries is a Bad Thing when at all
avoidable.

BTW, debian-gtk-gnome@lists.debian.org is the cannonical place to ask
debian-related gtk questions, and it's low-volume now that the Great GNOME
Copy is done.  (CCed/Reply-toed there.)

-=- James Mastros
-- 
First they came for the fourth amendment, but I said nothing because I
wasn't a drug dealer. Then they came for the sixth amendment, but I kept
quiet because I wasn't guilty. Finally they came for the first amendment,
and by then it was too late to say anything at all." 
-=- Nancy Lebowitz
cat /dev/urandom|james --insane=yes > http://www.rtweb.net/theorb/
ICQ: 1293899   AIM: theorbtwo  YPager: theorbtwo

Re: jdk not working in potato, working jdk removed from incoming, license problem

[Jules Bean]

[EMAIL PROTECTED] wrote:
> 
> Hi
> 
> I am given to understand that someone has found a problem in the license of
> jdk, to the point that same person finds that debian cannot distribute the jdk
> at all. I was told that the problem found in the license has existed for a
> long time.
> 
> If this is the case,
> 
> WHY is a jdk that doesn't even work in potato? By precisely the same token,
> why is there a jdk in ANY debian dist?? Is there a difference in the license
> between versions? Has anyone talked to Sun?
> 
> If this is NOT the case,
> 
> Can this be resolved quickly please? I would imagine that it is the intent of
> Sun that Java in its pure form would make it big. I have a few java projects
> that I'm taking off the back burner presently, and now this.
> 
> Inquiring, jdk-using minds want to know.

[This is intended to be a helpful comment - I hope it doesn't come
across as a troll]

Kaffe (www.kaffe.org) is, in my experience, a fast and efficient
replacement for the JDK (1.1, some of 1.2 is implemented).  It includes
a native AWT and a JIT, and whilst performance isn't excellent, it's as
good as jdk-interpreted, and it's open source.  Completely. 
Including a reimplementation of classes.zip, clean-room, ground-up, with
source code.

The Kaffe team are, in my experience, swift to response to, and fix,
bugs (more than you can say for Sun).

So, if you have a Java project, at least assess whether or not it is
feasible to use kaffe (and support open source!).

Jules


-- 
/+---+-\
|  Jelibean aka  | [EMAIL PROTECTED] |  6 Evelyn Rd|
|  Jules aka |   |  Richmond, Surrey   |
|  Julian Bean   | [EMAIL PROTECTED]|  TW9 2TF *UK*   |
++---+-+
|  War doesn't demonstrate who's right... just who's left. |
|  When privacy is outlawed... only the outlaws have privacy.  |
\--/

Compiling wordinspect for potato

[Bob Hilliard]

 I have recently adopted wordinspect and am trying to compile it
for potato.  The slink version lists the following dependencies:
Depends: dict, libc6 (>= 2.0.7u), libglib1.1 (>= 1.1.3-2), libgtk1.1
(>= 1:1.1.\2-2), xlib6g (>= 3.3-5)

 libglib1.1 and libgtk1.1 are not in potato under those
names. libglib1.2 seems to be a logical replacement for libglib1.1, but
there are many libgtk* libraries.

 I installed libglib1.2_1.2.3-1.deb, libglib1.2-dev_1.2.3-1.deb,
libgtk1.2_1.2.3-1.deb, and libgtk1.2-dev_1.2.3-1.deb, and built the
package.  It compiled cleanly, and runs as it did in slink, except
that every time it pops up a window it displays the following message:

> Gtk-WARNING **: gtk_scrolled_window_add(): cannot add non scrollable widget
> use gtk_scrolled_window_add_with_viewport() instead

 Would another libgtk be a better choice to replace libgtk1.1,
and possibly get away from this warning message, or will the source
have to tweaked to avoid this?

Bob
-- 
   _
  |_)  _  |_   Robert D. Hilliard<[EMAIL PROTECTED]>
  |_) (_) |_)  Palm City, FL  USAPGP Key ID: A8E40EB9

Re: better /etc/init.d/network

[shaleh]

I looked at the code (have not run it yet).

Nice.  Well documented, clean.  The design seems sound.  An up/down section is
also handy.

Shipping a default config would be nice, maybe in the /usr/doc/net/examples?

A little heavy on the bash code for my liking, but I understand why.

Have you mailed the maintainer of the netbase package (which provides the 
functionality you are replacing, I believe)?  He may have suggestions.

Short term solution, remove the network script from your package.  Inform the
user that they must edit /etc/init.d/network, you can provide the script in
an examples directory in /usr/doc/net.

Re: better /etc/init.d/network

[Rene Mayrhofer]

Am Sun, 16 May 1999 schrieb Massimo Dal Zotto:
> Hi,
> 
> The /etc/init.d/network script created by the debian installation is very
> simple and not flexible enough if you need to manage complex networks with
> many interfaces.
> 
> I have written a generic network interface management command, net, which
> can be used to start/stop/show/configure network interfaces, and a smarter
> replacement for the /etc/init.d/network script.
I will try it out and maybe use it for a firewall with more than 10 network
cards. This could be a good test for the new script.

Rene

--
--
Rene Mayrhofer, ViaNova KEG NIC-HDL: RM1677-RIPE
Email: [EMAIL PROTECTED] Snail: Penz 217, A-4441 Behamberg

PGP(DSS): E661 2E45 9B7F B239 D422  0A90 A4C2 DA09 F72F 6EC5
PGP(D/H): B77F 51A8 B046 87A6 4D61  2C5D 742F F433 6732 E4DC
GPG:  D356 69B6 6A08 E033 257B  1872 6AEA 88FB C805 63BD
--

jdk not working in potato, working jdk removed from incoming, license problem

[jim]

Hi

I am given to understand that someone has found a problem in the license of
jdk, to the point that same person finds that debian cannot distribute the jdk
at all. I was told that the problem found in the license has existed for a
long time. 

If this is the case, 

WHY is a jdk that doesn't even work in potato? By precisely the same token, 
why is there a jdk in ANY debian dist?? Is there a difference in the license 
between versions? Has anyone talked to Sun?

If this is NOT the case,

Can this be resolved quickly please? I would imagine that it is the intent of
Sun that Java in its pure form would make it big. I have a few java projects
that I'm taking off the back burner presently, and now this.

Inquiring, jdk-using minds want to know.

-Jim

Re: better /etc/init.d/network

[Massimo Dal Zotto]

> On Sun, May 16, 1999 at 10:15:48PM +0200, Massimo Dal Zotto wrote:
> > 
> > Hi,
> > 
> > The /etc/init.d/network script created by the debian installation is very
> > simple and not flexible enough if you need to manage complex networks with
> > many interfaces.
> > 
> > I have written a generic network interface management command, net, which
> > can be used to start/stop/show/configure network interfaces, and a smarter
> > replacement for the /etc/init.d/network script.
> > ...
> 
> So what is the big difference between your tool and ifconfig? Seems you
> get the same results and you don't save a lot of work... Please provide
> more details on benifits of your tool.
> 

Obviously you can do the same things with ifconfig. The difference is that
now you don't need to put all the ifconfig and route commands for your 
network in one big network startup script, but instead you store only the
configuration parameters in separate config files which are used by the
new net script.

The big advantage is that you can now start, stop, and configure (that is
creating the configuration file, not running ifconfig) each interface
separately. This doesn't save a lot of work but provides much more
flexibility to network management.

The current network script can't do that, unless you create a different
script for each interface and all the needed links in rc*.d, and can only
start the network, not stopping it or a part of it.
So if you want to stop or restart your network or change the status of just
one interface you must type all the ifconfig and route commands by hand and
remember all the numbers each time.

With my solution you store the parameters for each interface and then just
type 'net stop' or 'net restart' or 'net start eth1' without needing to
remember all the arguments. If you need to add or remove an interface you
just create or delete its config file and the job is done because you don't
need to change the /etc/init.d/network, which just does a 'net start' or
'net stop'. Simple and easy.

This is not very useful if you have a simple network with one interface and
one ip, but is really handy if you have to manage complex networks with many
interfaces or ip aliases and complex routing tables, as is my case.

In general I believe it is better to separate the configurations from the
programs which do the work, which should be kept as general as possible.
My script does just that: it separates programs from configurations.

Note also that currently there isn't a network configuration program in
debian. The network script is created by dinstall and must be edited by
hand. My net script provides also a very simple configuration function.
Think of a novice user which can now just type 'net config eth0' and answer
a few simple questions.

-- 
Massimo Dal Zotto

+--+
|  Massimo Dal Zotto   email: [EMAIL PROTECTED]   |
|  Via Marconi, 141phone: ++39-0461534251  |
|  38057 Pergine Valsugana (TN)  www: http://www.cs.unitn.it/~dz/  |
|  Italy pgp: finger [EMAIL PROTECTED]  |
+--+

FAQ and call for help: Linuxconf and Debian

[Stefan Gybas]

Hi!

I'm getting a lot of mails full of questions about my Linuxconf Debian
package so I've put together a little FAQ (attached).

I also want to further intergate Linuxconf into Debian but this requires
a lot of work. If you want to help me with this please contact me.

If you want to reply to this mail please choose the right mailing list. For
general Linuxconf questions please use linuxconf@xc.org, for Debian specific
issues please use [EMAIL PROTECTED]

-- 
Stefan Gybas1. What is Linuxconf?

   Linuxconf is a program for Linux systems with three major functions:

   (a) configuration utility

   With Linuxconf you can do basic and advanced system administration
   and configuration. Linuxconf has some core functionality (like
   creating and managing users, groups and file systems) and several
   modules for other system components, e.g. bind, apache, sendmail,
   samba and squid. Currently there are over 20 modules available.

   The configuration is done using a GUI with lot of help texts that can
   either use a text, X11 or web interface.

   Linuxconf does not use a database to store the configuration, instead
   it just uses the normal system configuration files like /etc/fstab,
   /etc/hosts - while trying very hard to preserve their structure (like
   manually added comments). So you can always switch between a text
   editor and Linuxconf.
   
   (b) configuration activator and manager

   Linuxconf can keep track of changes made to configuration files
   (either using Linuxconf itself of some text editor) and then update
   the system to reflect those changes. An example might be

   vi /etc/inetd.conf
   vi /etc/apache/httpd.conf
   linuxconf --update # This will cause inetd and apache to
  # reload their configuration files

   Another feature is the ability to archive and restore configuration
   files (using RCS if available):

   linuxconf --archive
   linuxconf --diff
   linuxconf --extract
   
   (c) boot selector

   The third main feature of Linuxconf is called askrunlevel and
   this is exactly what it does. It shows a little menu during boot up
   where you can select the system runlevel and the profile.

   A profile is a name for an archived configuration (see section b)
   like "office" or "home", so you can e.g. have different IP addresses
   or XFree configurations with your laptop depending on your location.

   Each of these features can be turned on or off independently, that means
   you don't need to activate the boot selector if you just want to do some
   Samba configuration. And you can always De-install the Linuxconf package
   and everything is as it was before.


2. Where can I get the Linuxconf Debian package?

   I've split up Linuxconf into four seperate Debian packages:

   linuxconf The main Linuxconf binary (text and web interface
 only, with all modules)

   linuxconf-x   The X11 GUI for Linuxconf (this is in a sperate
 package so the main package does not depend on
 xlib6g, wxxt1 and xpm4g)

   linuxconf-locale  The foreign language files (all except English)

   linuxconf-bootThe boot selector (see 1c) - if you don't want to
 enable this features, simply don't install this
 package and Linuxconf will not make any grave
 modifications to your system.

   All packages are in the "experimental" distribution - available on
   all Debian mirrors in project/experimental/. They can be installed
   on a Debian 2.1 (slink) system but require a newer netbase package.


3. What is working on Debian GNU/Linux? And what is not?

   All of the configuration stuff (see 1a) except network configuration is
   working on Debian. The configuration activator (1b) and boot selector (1c)
   are partly working (see question 4).


4. What needs to be done in order to make the other features work?

   The problem with network configuration is that this is done differently
   on each Linux distribution and thus Linuxconf does not know where to store
   the host name, IP addresses and routing table. So distribution specific
   modules were written to handle this part, but unfortunately there is none
   available for Debian yet.

   For the configuration activator, Linuxconf needs to know which config file
   belongs to which service/daemon and how to make this daemon reload its
   configuration. Again, this is distribution specific but Linuxconf has
   a little bit basic knowledge here, e.g. it knows that inetd uses
   /etc/inetd.conf and can be restarted using "kill -HUP".

   But as you might probably know, Debian uses SysV init scripts in
   /etc/init.d/ that can also make a daemon reload its configuration, like
   in "/etc/init.d/proftpd reload". So it would be a good idea to tell
   Linux

Re: (COMPATIBILITY) Correct non-US solution

[Joseph Carter]

On Mon, May 17, 1999 at 04:33:14AM -0400, Havoc Pennington wrote:
> Simpler: instead of requiring people to add /etc/LEGAL, either add it by
> default or require them to add /etc/ILLEGAL. No reason to have illegal be
> the default, might get someone sued. (Actually, the whole scheme might be
> considered "hooks" for encryption and be illegal in some countries; beats
> me. Too bad only lawyers understand the law. :-( )

Bash is a crypto hook.  I suggest ignoring that bit of insanity. 
(Precedent in the mutt package, which was formerly mutt-i in non-us)

--
Joseph Carter <[EMAIL PROTECTED]>Debian GNU/Linux developer
PGP: E8D68481E3A8BB77 8EE22996C9445FBEThe Source Comes First!
-
 rit/ara:  There's something really demented about UNIX
underwear...


pgpT83pcMUhVs.pgp
Description: PGP signature

Re: (FINISH) Correct non-US solution

[Jonathan Walther]

No.  The scheme makes us less liable than we already are, since it shows
that we are "trying". It puts us ahead of every other Linux distribution out
there.  Certainly we only distinguish non-US stuff right now.  But the laws
of France and Russia are equally clear and well known.

We don't increase our liability... How can increasing our compliance to the
law make us more liable?  Sheesh.  No, I don't propose making us more
RESPONSIBLE for following the law, but propose making us better ABLE to
follow the law.

As is currently the case, a package wouldn't have restrictions unless it was
brought to a Debian maintainers attention what the law was.  Why hunt out
trouble?  Let it come to you.  For heavens sakes.

We aren't lawyers, and everyone knows that.  This can be seen as "reasonable
effort" to cover our asses.

Jonathan

On 17 May 1999, Manoj Srivastava wrote:
> Hi,
> 
> I like the idea. But are we then in the position of practicing
>  law (giving legal advice)? Would we be liable for these decisions? We
>  may not be any worse off than we are now, but so far we only make
>  decisions about what is and is not legally exportable from the US,
>  and that is very public knowledge. Extending this to other areas may
>  take us into grayer areas.
> 
> manoj
> look before we leap

Configurator Panel

[Diego Delgado Lages]

I'm making something like a Control Panel for Linux (for Debian), and I
would like you to test me and send me comments.

It's somewhat very very alpha (some programs don't work), but you can see
what it will do.

I'll be working on that, and the Network and the Printer configurators
should be working till the Weekend.


To get the sources : http://linuxlabs.lci.ufrj.br/~lages/cpanel


Diego Delgado Lages

Re: 3c5x9setup and isapnptools

[Edward Betts]

On Sun, 16 May, 1999, Ben Pfaff wrote:
> Dale Scheetz <[EMAIL PROTECTED]> writes:
> 
>After my recent experience gettying my new 3COM EtherLink III cards to
>work, I would like to suggest that 3c5x9setup be included in the
>isapnptools package. It is composed of a single .c source file with an
>embedded copyright notice licensing it under the GPL. I would be willing
>to write a man page from the .html file provided, if Frederic Lepied would
>be willing to include the two in the isapnptools package. I think this is
>a better place for this than trying to build another package around this
>simple program.
> 
> The corresponding program for configuring Western Digital and SMC
> Ethernet cards (wdsetup) is in netstd.  Perhaps this is the approved
> place for such tools?

$ sudo wdsetup -v 3
Setup for Western Digital and SMC ethercards, version 0.6a
lm_get_at_config returned 2, cnfg_val=0
verifyparams returned 87, cnfg_val=68
configuration error: 68
$

Well that did not work :-(

-- 
I consume, therefore I am


pgpaCeMUlVaG4.pgp
Description: PGP signature

Re: GPG as a PGP replacement

[Steve Haslam]

On Mon, May 17, 1999 at 11:27:39AM +0200, Alexander N. Benner wrote:
> Ship's Log, Lt. Michael Meskes, Stardate 140599.1439:
> > Which version do you use? I don't have that script.
> 
> 
> Was it removed in the l8est Version ?
> Cannot check the changelog as I still have it ;-)

The gpg-pgp script and pgp2 compatibility hackage is in gpg-rsaidea, not
gnupg (afaicr).

SRH
-- 
Steve Haslam   Debian GNU/Linux   [EMAIL PROTECTED]
gnome-libs, gnome-core, gnome-control-center, gdm, p3nfs.what, me worry?


pgpcT0Xe3qbfU.pgp
Description: PGP signature

Re: a Chinese version of X-window system for Linux available

[Anthony Wong]

On Sun, May 16, 1999 at 09:13:23PM -0400, Daniel Martin wrote:
|liug <[EMAIL PROTECTED]> writes:
|
|> Dear Sir,
|> I am not sure whether this is the right place to post
|> this mail.
|> We have developed a Chinese version of X-window system
|> several years ago, and now we have developed one for
|> Linux, 
|
|> I am wondering whether our product could be integrated
|> into or bundled with
|> the new Debian Linux release.
|> We also have some document and screen shot of our
|> product.
|> Please do not hesitate to contact us if we can help.
|> We are looking forward to your answer.
|
|Has anyone contacted these people, or forwarded the message on to the
|Debian-Chinese people?

Yep, I've contacted him already, and also redirected him to -chinese
for further discussion if he sees fit. But he has not got back to me
yet.

-- 
Anthony Wong

Re: Old Library dependencies Re: Release Plans (19990513)

[Josip Rodin]

On Mon, May 17, 1999 at 11:36:41AM +0200, Hartmut Koptein wrote:
> > >   Remove as many dependencies on old libraries as possible, this
> > >   includes:
> > > 
> > > libjpegg6a, libncurses3.4, newt0.25, libpgsql, tk4.2, tcl7.6,
> > > libwraster1, libpng0g
> > > 
> > >   and various older gtk/gnome libraries.
> > 
> > Lintian has a "depends-on-obolete-package" tag.  I will add newt0.25,
> > libpgsql, tk4.2, tcl7.6, libwraster1, and libpng0g to its list.
> > This way, the Lintian page for that tag will form a useful index.
> 
> Why not  libjpegg6a and libncurses3.4?
> 
> What about  xbase, emacs19, altgcc, freetype1, libc5, libg++27, libpaper,
> tclx76  and possible some more?

Aren't most of these already on that list? libghttp0 could be added, too.

-- 
enJoy -*/\*- http://jagor.srce.hr/~jrodin/

intend to package Perl module Text::Format

[Ardo van Rangelrooij]

Hi,

I intend to package the Perl module Text::Format which provides the
following functions to format (text) paragraphs:

format()Format text into a paragraph.  Text is first broken into
words and then joined back together to make up the
paragraph.  There are numerous attributes you can set to
your liking.

paragraphs()Treats each element of the array passed in as a separate
paragraph and passes them to format() for formatting.
The list returned will be then formatted into separate
paragraphs.

center()Centers all the lines that were passed in.

expand()Expand tabs into spaces.

unexpand()  Turn spaces into tabs.  First calls expand() to expand
tabs into spaces and then turns tabstop number of spaces
into tabs, you can set tabstop size with tabstop().

Thanks,
Ardo
-- 
Ardo van Rangelrooij
home email: [EMAIL PROTECTED], [EMAIL PROTECTED]
home page:  http://www.tip.nl/users/ardo.van.rangelrooij
PGP fp: 3B 1F 21 72 00 5C 3A 73  7F 72 DF D9 90 78 47 F9

Re: Potato compiling environment on otherwise slink system

[Sven LUTHER]

On Mon, May 17, 1999 at 10:55:33AM -0400, Peter S Galbraith wrote:
> 
> Sven LUTHER wrote:
> 
> > Alternatively, you could install a chrooted potatyo environment, just to
> > compile your stuff, and not touch your actual slink stuff.
> 
> Any docs on how to do this?
> sh-utils.info doesn't say much, and nothing this specific of
> course.

basically, you install a new system (by untaring the base tarball on a second
partition or a directory on a partition with some place in it.) then you do
chroot directory, and you are now in an environment that is as if you booted
into this new partition. You install apt-get, upgrade all the stuff you want,
and play with it.

Once you are finished you quite it with exit, i think.

read man chroot.

Friendly,

Sven LUTHER

Re: Potato compiling environment on otherwise slink system

[Peter S Galbraith]

Sven LUTHER wrote:

> Alternatively, you could install a chrooted potatyo environment, just to
> compile your stuff, and not touch your actual slink stuff.

Any docs on how to do this?
sh-utils.info doesn't say much, and nothing this specific of
course.

Thanks,

Peter

Re: Potato compiling environment on otherwise slink system

[Peter S Galbraith]

I wrote:
 
> I run Slink at work and at home, but decided to install potato's
> gcc and g++ on my home box to recompile the potato packages that
> I maintain
> 
> # apt-get -d -u install gcc g++
> Reading Package Lists... Done
> Building Dependency Tree... Done
> The following extra packages will be installed:
>   libfltk-dev libstdc++2.9-glibc2.1-dev cpp libstdc++2.9-glibc2.1 libc6-dev
>   libfltk1 libc6 mesag3 
> The following packages will be REMOVED:
>   libstdc++2.9-dev timezone 
> The following NEW packages will be installed:
>   libstdc++2.9-glibc2.1-dev libstdc++2.9-glibc2.1 libfltk1 mesag3 
> The following packages will be upgraded
>   libfltk-dev g++ cpp gcc libc6-dev libc6 
> 6 packages upgraded, 4 newly installed, 2 to remove and 265 not upgraded.
> Need to get 7524kb of archives. After unpacking 7823kb will be used.
> Do you want to continue? [Y/n] 
> 
> 
> Anything wrong with this?  Or must I upgrade _everything_ to
> potato.

Petr Cech wrote:

> No. 

No? there's nothing wrong with this?

> More or less, yes.

But yes? I must upgrade _everything_ to potato?
I'm sorry, I'm confused.

>The reason for recompiling is, that packages depend
> on glibc-2.1 (libc6 (>=2.1)) - so you must install those libraries. The
> compiler alone won't help. Not mentioning, that gcc in potato in linked with
> libc-2.1.

I understand this.  That's why I'm using apt--get to catch all
the dependencies, including the new glibc-2.1.  I'm just worried
that this could leave my system in an unusable state.

> > This will upgrade libc6.  Will the rest of the system (slink)
> > still function correctly?  
>
> again - almost all. The problem is with shared libraries. When I upgraded
> to glibc-2.1 one machine, I forgot about libreadlineg2 - bash worked almost
> corrently, that is, when you hit  it core-dumped.

So I might have a few problems, some of which are already known?
Has anyone done this and taken notes?

Thanks, 
Peter

Re: (FINISH) Correct non-US solution

[Manoj Srivastava]

Hi,

I like the idea. But are we then in the position of practicing
 law (giving legal advice)? Would we be liable for these decisions? We
 may not be any worse off than we are now, but so far we only make
 decisions about what is and is not legally exportable from the US,
 and that is very public knowledge. Extending this to other areas may
 take us into grayer areas.

manoj
look before we leap

>>"Jonathan" == Jonathan Walther <[EMAIL PROTECTED]> writes:

 Jonathan> Since the main (but not exclusive) use of non-US right now
 Jonathan> is for crypto software, we might want to create a
 Jonathan> Crypto-Regulations package which contains references to
 Jonathan> which countries restrict import and export of crypto, and
 Jonathan> how, with references to appropriate legislation and
 Jonathan> documentation.

 Jonathan> This would save every maintainer of a crypto package from
 Jonathan> having to go hunting down the information, and in fact
 Jonathan> could be made a part of Debian Policy for crypto packages
 Jonathan> to respect it.

 Jonathan> Don't think of this as added work: think of it as doing
 Jonathan> right something which has only been half-done up to now.
 Jonathan> The US is not the only country with restrictions on
 Jonathan> software.  Our source of possible countries in which to
 Jonathan> host our non-US mirrors seems to shrink every year.


-- 
 Q: What do you call a blind, deaf-mute, quadraplegic Virginian? A:
 Trustworthy.
Manoj Srivastava   <[EMAIL PROTECTED]>  
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

Re: (LONG) Correct non-US solution

[Manoj Srivastava]

Hi,
>>"Jonathan" == Jonathan Walther <[EMAIL PROTECTED]> writes:

 Jonathan> Changes to a packages control file:
 Jonathan> -- Two new fields are added
 Jonathan> to the control file, Import-Restricted and
 Jonathan> Export-Restricted.  These fields take a comma delimited
 Jonathan> list of countries.

 Jonathan> For example,

 Jonathan> Package: ssh
 Jonathan> Export-Restricted: United States
 Jonathan> Import-Restricted: Russia, France

I like the idea, but I am concerned about 
 a) Potential liability placed on the maintainer (is he liable for
being wrong or not having the complete list of countries in eithe
category? 
 b) the additional burden placed on developers (I have no idea what
laws strange countries may have about even what is, to me,
innocuous pieces of software).

I think we may need a repository (which may just be a file)
 that people contribute to, which would be a resource that developers
 can look at. Since more hands are doing the work, there are fewer
 chances of errors. On the other hand that may make the project, or at
 least all contributors, partly liable as well.

manoj
-- 
 So this is it.  We're going to die.
Manoj Srivastava   <[EMAIL PROTECTED]>  
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E

Re: Potato compiling environment on otherwise slink system

[Sven LUTHER]

On Mon, May 17, 1999 at 04:14:57PM +0200, Petr Cech wrote:
> On Mon, May 17, 1999 at 09:23:03AM -0400 , Peter S Galbraith wrote:
> > 
> > I run Slink at work and at home, but decided to install potato's
> > gcc and g++ on my home box to recompile the potato packages that
> and libc6_2.1. Only compiling with new gcc won't have desired effect.
> > I maintain (keeping work box on slink for stability).
> > 
> > Since my bandwidth is at work, I doing the following to download
> > what I need (and then I'll sneaker-net everything home on a Zip):
> > 
> > # apt-get -d -u install gcc g++
> > Reading Package Lists... Done
> > Building Dependency Tree... Done
> > The following extra packages will be installed:
> >   libfltk-dev libstdc++2.9-glibc2.1-dev cpp libstdc++2.9-glibc2.1 libc6-dev
> >   libfltk1 libc6 mesag3 
> > The following packages will be REMOVED:
> >   libstdc++2.9-dev timezone 
> > The following NEW packages will be installed:
> >   libstdc++2.9-glibc2.1-dev libstdc++2.9-glibc2.1 libfltk1 mesag3 
> > The following packages will be upgraded
> >   libfltk-dev g++ cpp gcc libc6-dev libc6 
> > 6 packages upgraded, 4 newly installed, 2 to remove and 265 not upgraded.
> > Need to get 7524kb of archives. After unpacking 7823kb will be used.
> > Do you want to continue? [Y/n] 
> > 
> > 
> > Anything wrong with this?  Or must I upgrade _everything_ to
> No. More or less, yes. The reason for recompiling is, that packages depend
> on glibc-2.1 (libc6 (>=2.1)) - so you must install those libraries. The
> compiler alone won't help. Not mentioning, that gcc in potato in linked with
> libc-2.1.

Alternatively, you could install a chrooted potatyo environment, just to
compile your stuff, and not touch your actual slink stuff.

Friendly,

Sven LUTHER

Re: Potato compiling environment on otherwise slink system

[Petr Cech]

On Mon, May 17, 1999 at 09:23:03AM -0400 , Peter S Galbraith wrote:
> 
> I run Slink at work and at home, but decided to install potato's
> gcc and g++ on my home box to recompile the potato packages that
and libc6_2.1. Only compiling with new gcc won't have desired effect.
> I maintain (keeping work box on slink for stability).
> 
> Since my bandwidth is at work, I doing the following to download
> what I need (and then I'll sneaker-net everything home on a Zip):
> 
> # apt-get -d -u install gcc g++
> Reading Package Lists... Done
> Building Dependency Tree... Done
> The following extra packages will be installed:
>   libfltk-dev libstdc++2.9-glibc2.1-dev cpp libstdc++2.9-glibc2.1 libc6-dev
>   libfltk1 libc6 mesag3 
> The following packages will be REMOVED:
>   libstdc++2.9-dev timezone 
> The following NEW packages will be installed:
>   libstdc++2.9-glibc2.1-dev libstdc++2.9-glibc2.1 libfltk1 mesag3 
> The following packages will be upgraded
>   libfltk-dev g++ cpp gcc libc6-dev libc6 
> 6 packages upgraded, 4 newly installed, 2 to remove and 265 not upgraded.
> Need to get 7524kb of archives. After unpacking 7823kb will be used.
> Do you want to continue? [Y/n] 
> 
> 
> Anything wrong with this?  Or must I upgrade _everything_ to
No. More or less, yes. The reason for recompiling is, that packages depend
on glibc-2.1 (libc6 (>=2.1)) - so you must install those libraries. The
compiler alone won't help. Not mentioning, that gcc in potato in linked with
libc-2.1.
> potato.
> 
> This will upgrade libc6.  Will the rest of the system (slink)
> still function correctly?  
again - almost all. The problem is with shared libraries. When I upgraded
to glibc-2.1 one machine, I forgot about libreadlineg2 - bash worked almost
corrently, that is, when you hit  it core-dumped.
> 
> Thanks!
> -- 
> Peter Galbraith, research scientist  <[EMAIL PROTECTED]>

Petr Čech
--
[EMAIL PROTECTED]
Debian GNU/Linux  -  maintainer & administrator

Re: Solaris NFS problems with potato / Release Notes

[Tim \(Pass the Prozac\) Sailer]

On Mon, May 17, 1999 at 03:29:45PM +0300, Rainer Dorsch wrote:
> 
> I forget to say, that I am running 
> 
>  # uname -a
>  Linux rai16 2.2.7 #1 Mon May 10 15:53:20 CEST 1999 i686 unknown
> 
> Where do I get these NFS 3 patches from?

ftp.varesearch.com

I forwarded you the announcement just now.

Tim

-- 
 (work) [EMAIL PROTECTED] / (home) [EMAIL PROTECTED] - http://www.buoy.com/~tps
  "The quality of accurate observation is commonly called 
 cynicism by those who have not got it."
  G.B. Shaw
** Disclaimer: My views/comments/beliefs, as strange as they are, are my own.**

Re: Solaris NFS problems with potato / Release Notes

[Rainer Dorsch]

> On Mon, May 17, 1999 at 10:30:52AM +0300, Rainer Dorsch wrote:
> > 
> > It is nearly impossible to compile a large software package on a NFS 
> > mounted 
> > partion on potato, when it is exported by Solaris 2.6 (UltraSPARC). A 
> > search 
> > in Deja News (Solaris NFS patch in comp.os.linux.*) found a couple of 
> > controversal postings. After installing the patch recommended by Linus
> > 
> >  105379-05
> > 
> > (& stopping and starting the NFS-server on Solaris) the problems did not 
> > disappear.
> > 
> > To illustrate the problem, I use the cvs source package and compile it on a 
> > local disc. Works well.
> > 
> > Now I do it on an NFS mounted partition:
> > 
> > $debuild
> > [...]
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c vers_ts.c
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c subr.c
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c filesubr.c
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c run.c
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c version.c
> > gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c error.c
> > gcc  -I. -I.. -I. -I../lib -I../zlib -DHAVE_CONFIG_H  -g -O2 -c ./zlib.c
> > gcc add.o admin.o buffer.o checkin.o checkout.o classify.o client.o 
> > commit.o 
> > create_adm.o cvsrc.o diff.o edit.o entries.o expand_path.o fileattr.o 
> > find_names.o hardlink.o hash.o history.o ignore.o import.o lock.o log.o 
> > login.o logmsg.o main.o mkmodules.o modules.o myndbm.o no_diff.o 
> > parseinfo.o 
> > patch.o rcs.o rcscmds.o recurse.o release.o remove.o repos.o root.o rtag.o 
> > scramble.o server.o status.o tag.o update.o watch.o wrapper.o vers_ts.o 
> > subr.o 
> > filesubr.o run.o version.o error.o zlib.o ../lib/libcvs.a ../diff/libdiff.a 
> > -lz -lcrypt   -o cvs
> > checkin.o: file not recognized: File truncated
> > collect2: ld returned 1 exit status
> > make[2]: *** [cvs] Error 1 
> > 
> > The error message is missleading, it is nearly random per run!
> > 
> > Is anybody using potato with NFS exported Solaris partitions? Which patches 
> > did you apply? Did you try to compile large programs? Hello world programs 
> > work!
> 
> We were having the same problem, but with NFS AIX. Here at [EMAIL PROTECTED],
> we are using AIX and Solaris for NFS servers. Trying to compile trivial
> programs worked, but something like cvs source failed with the same
> errors  you have there only on the AIX servers running NFSv3. We
> went to 2.2.7 with the NFS 3 patches, the problem went away...
> 

I forget to say, that I am running 

 # uname -a
 Linux rai16 2.2.7 #1 Mon May 10 15:53:20 CEST 1999 i686 unknown

Where do I get these NFS 3 patches from?

Thanks.

-- 
Rainer Dorsch
Abt. Rechnerarchitektur  e-mail:[EMAIL PROTECTED]
Uni StuttgartTel.: 0711-7816-215

-- 
Rainer Dorsch
Abt. Rechnerarchitektur  e-mail:[EMAIL PROTECTED]
Uni StuttgartTel.: 0711-7816-215

Re: Solaris NFS problems with potato / Release Notes

[Tim \(Pass the Prozac\) Sailer]

On Mon, May 17, 1999 at 10:30:52AM +0300, Rainer Dorsch wrote:
> 
> It is nearly impossible to compile a large software package on a NFS mounted 
> partion on potato, when it is exported by Solaris 2.6 (UltraSPARC). A search 
> in Deja News (Solaris NFS patch in comp.os.linux.*) found a couple of 
> controversal postings. After installing the patch recommended by Linus
> 
>  105379-05
> 
> (& stopping and starting the NFS-server on Solaris) the problems did not 
> disappear.
> 
> To illustrate the problem, I use the cvs source package and compile it on a 
> local disc. Works well.
> 
> Now I do it on an NFS mounted partition:
> 
> $debuild
> [...]
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c vers_ts.c
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c subr.c
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c filesubr.c
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c run.c
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c version.c
> gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c error.c
> gcc  -I. -I.. -I. -I../lib -I../zlib -DHAVE_CONFIG_H  -g -O2 -c ./zlib.c
> gcc add.o admin.o buffer.o checkin.o checkout.o classify.o client.o commit.o 
> create_adm.o cvsrc.o diff.o edit.o entries.o expand_path.o fileattr.o 
> find_names.o hardlink.o hash.o history.o ignore.o import.o lock.o log.o 
> login.o logmsg.o main.o mkmodules.o modules.o myndbm.o no_diff.o parseinfo.o 
> patch.o rcs.o rcscmds.o recurse.o release.o remove.o repos.o root.o rtag.o 
> scramble.o server.o status.o tag.o update.o watch.o wrapper.o vers_ts.o 
> subr.o 
> filesubr.o run.o version.o error.o zlib.o ../lib/libcvs.a ../diff/libdiff.a 
> -lz -lcrypt   -o cvs
> checkin.o: file not recognized: File truncated
> collect2: ld returned 1 exit status
> make[2]: *** [cvs] Error 1 
> 
> The error message is missleading, it is nearly random per run!
> 
> Is anybody using potato with NFS exported Solaris partitions? Which patches 
> did you apply? Did you try to compile large programs? Hello world programs 
> work!

We were having the same problem, but with NFS AIX. Here at [EMAIL PROTECTED],
we are using AIX and Solaris for NFS servers. Trying to compile trivial
programs worked, but something like cvs source failed with the same
errors  you have there only on the AIX servers running NFSv3. We
went to 2.2.7 with the NFS 3 patches, the problem went away...

Tim

-- 
 (work) [EMAIL PROTECTED] / (home) [EMAIL PROTECTED] - http://www.buoy.com/~tps
  "The quality of accurate observation is commonly called 
 cynicism by those who have not got it."
  G.B. Shaw
** Disclaimer: My views/comments/beliefs, as strange as they are, are my own.**

Potato compiling environment on otherwise slink system

[Peter S Galbraith]

I run Slink at work and at home, but decided to install potato's
gcc and g++ on my home box to recompile the potato packages that
I maintain (keeping work box on slink for stability).

Since my bandwidth is at work, I doing the following to download
what I need (and then I'll sneaker-net everything home on a Zip):

# apt-get -d -u install gcc g++
Reading Package Lists... Done
Building Dependency Tree... Done
The following extra packages will be installed:
  libfltk-dev libstdc++2.9-glibc2.1-dev cpp libstdc++2.9-glibc2.1 libc6-dev
  libfltk1 libc6 mesag3 
The following packages will be REMOVED:
  libstdc++2.9-dev timezone 
The following NEW packages will be installed:
  libstdc++2.9-glibc2.1-dev libstdc++2.9-glibc2.1 libfltk1 mesag3 
The following packages will be upgraded
  libfltk-dev g++ cpp gcc libc6-dev libc6 
6 packages upgraded, 4 newly installed, 2 to remove and 265 not upgraded.
Need to get 7524kb of archives. After unpacking 7823kb will be used.
Do you want to continue? [Y/n] 


Anything wrong with this?  Or must I upgrade _everything_ to
potato.

This will upgrade libc6.  Will the rest of the system (slink)
still function correctly?  

Thanks!
-- 
Peter Galbraith, research scientist  <[EMAIL PROTECTED]>
Maurice Lamontagne Institute, Department of Fisheries and Oceans Canada
P.O. Box 1000, Mont-Joli Qc, G5H 3Z4 Canada. 418-775-0852 FAX: 775-0546
6623'rd GNU/Linux user at the Counter - http://counter.li.org/ 

Re: Bug#33262: xlib6g now depends on xfree86-common (?)

[Peter Van Eynde]

On Thu, Feb 11, 1999 at 08:49:06PM -0500, Branden Robinson wrote:
> Package: cmucl-clx
> Version: 2.4.9
> 
> On Fri, Feb 12, 1999 at 12:04:19AM +0100, Pierre Mai wrote:
> > This has in fact already happened some time ago, as can be witnessed
> > by CLX, which is an implementation of the X protocol for and in Common 
> > Lisp, and has been the de-facto standard for Common Lisp X programs
> > for quite some time (see X11 contrib tapes).  And indeed it has
> > already been packaged for Debian, as part of Peter Van Eynde's CMU CL
> > packages (cmucl-clx).
> 
> Huh, I'll be damned.  I knew a Common LISP equivalent of Xlib existed, but
> I thought it was long dead and not packaged for Debian.  I shouldn't
> underestimate my own distribution like that.  :)
> 
> Can I ask that cmucl-clx *please* depend on xfree86-common for frozen and
> unstable?


Done in 2.4.13.

Groetjes, Peter

-- 
It's logic Jim, but not as we know it. | [EMAIL PROTECTED] for pleasure,
"God, root, what is difference?",Pitr  | [EMAIL PROTECTED] for more pleasure!

Re: Hey, Y'all, check out my new improved "Free Software Research Paper Project" web site!

[R. Brock Lynn]

Adam Di Carlo wrote:
> 
> > "brock" == R Brock Lynn <[EMAIL PROTECTED]> writes:
> 
> brock> And you really can't argue with a FEELING.
> 
> Oh -- well, I guess not.  In fact, that kinda points out that there's
> no really reason for carrying on any further debate.

Well the debate may end, but you can then extend a bit into a "feeling" state.
Where you let your mind attempt to come to harmony with the natural
surroundings... And then whatever route shows itself as a good or best way to
go, that's the route you follow. Sounds a bit funky I know but hell why
not? (If you are having trouble choosing a particular route to journey through
in the future, then follow your heart...)

Why did the mountain climber climb the mountain? Because it was *there*.

I think the Asian cultures have mastered the non-verbal / feeling / emotional
communication to levels never achieved in the west. I have adapted over time
after making lasting friendships with many Asian friends, mostly Japanese. They
frequently communicate on the street with facial expression, body language, and
grunts, rather than with words alone. There is a term in Japanese called
"harahgeh", or "belly language". The meaning of this word is the ability to
communicate emotionally, mostly by non-verbal expression. Sometimes logic is
used to eventually cause a change of heart or mood. Well, I say why not just
change the heart itself directly, rather than go through logic to eventually
persuade you into a certain feeling, when I could just impart that feeling upon
you immediately?

> brock> Yes, I don't doubt you have much experience... but only in
> brock> "your neck of the woods". I feel like I'm seeing things from a
> brock> bit more of a grander perspective, and you are allowing the
> brock> local trees to blind your view of the forest overall.  All I
> brock> ask is for you to stretch yourself, let yourself grow just a
> brock> little bit more. Allow yourself to be just a little more open
> brock> minded. Please try! That's *all* I ask.

> This is just insulting.  I've contributed untold hours to free
> software over the past 8 years (they still credit me for some NCSA
> httpd stuff), I have spearheaded the release of software written and
> paid for by my company (onShore) under the GPL.  What have *you* done?

Sorry. I didn't mean to sound derogatory, condescending, or insulting. :(
But instead to recognize people like yourself who have helped to get us where we
are today! But this is not the end of the story. There is more to this tale to
unfold in the coming years. As much great things as you have done, there will
always be more that can be done. All I'm saying is don't let it end here, but
rather help the "next generation" of "hackers" to learn the craft and the
perfect it even further!

> >> You really should take up this discussion at the FSB (free software
> >> business) list; a very good, well-balanced, knowledgable list.

Ok. I subscribed. I'll see what I can learn from them. Real life can tell you
some very interesting tales. Indeed, truth can be stranger than fiction. ;^)

> brock> Interesting, what's the address to subscribe and such?
> 
><[EMAIL PROTECTED]>:
>Receive future messages sent to the mailing list.
> 
><[EMAIL PROTECTED]>:
>Stop receiving messages.
> 
> These guys really know their shit -- they'll roast you even harder
> than I have.  Or ignore you, which is maybe what I should have done.

I'd assume that getting roasted is better than being ignored?! (=:]

hrmmm... :)

--Brock

-  PGP key ID: FED76A3D <[EMAIL PROTECTED]> 4 / 5 / 1999

   __ _Debian GNU   R. Brock Lynn <[EMAIL PROTECTED]>
  / /(_)_ __  _   ___  __   http://www.debian.org/ irc.openprojects.net
 / / | | '_ \| | | \ \/ /  'Free Software'
/ /__| | | | | |_| |>  <   Remember that's "Free" as in Freedom, not "Free" as
\/_|_| |_|\__,_/_/\_\   in price!   Debian's the Greatest!

Re: PostgreSQL INC Press Release

[Michael Meskes]

On Sun, May 16, 1999 at 07:28:23PM +0200, Martin Schulze wrote:
> I have received this, you'd know better what to do.
> 
> Regards,
> 
>   Joey
> 
> Jeff MacDonald wrote:
> > Greetings,
> > 
> >Today we (PostgreSQL INC.) made our Initial Press Release at
> > http://www.pgsql.com/release.html
> > Regarding the beginning of techincal support etc.
> > 
> >Also we are anticipating the release of PostgreSQL 6.5 on June
> > 1st. On behalf of PostgreSQL INC. and the PostgreSQL Global Development
> > Group, we are curious if your group would be interested in including
> > PostgreSQL with your next distribution release.


No need to worry. AFAIK they start offering consulting. The software itself
won't go non-free.

Michael
-- 
Michael Meskes | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz| Go Rhein Fire!
Tel.: (+49) 2431/72651 | Use Debian GNU/Linux!
Email: [EMAIL PROTECTED]  | Use PostgreSQL!

Re: better /etc/init.d/network

[Erick Kinnee]

On Sun, May 16, 1999 at 10:15:48PM +0200, Massimo Dal Zotto wrote:
> 
> Hi,
> 
> The /etc/init.d/network script created by the debian installation is very
> simple and not flexible enough if you need to manage complex networks with
> many interfaces.
> 
> I have written a generic network interface management command, net, which
> can be used to start/stop/show/configure network interfaces, and a smarter
> replacement for the /etc/init.d/network script.
> 
> The net command makes use of configuration files stored in /etc/network/
> which contain the various interface options. For example my eth0 is:
> 
>   # /etc/network/eth0
>   IPADDR=192.168.0.1
>   NETMASK=255.255.255.0
>   NETWORK=192.168.0.0
>   BROADCAST=192.168.0.255
>   GATEWAY=192.168.0.1
> 
> The command can be used in the following ways (more info in the man page):
> 
>   # Start the network, typically from init(8)
>   net start
> 
>   # Start some specific interfaces
>   net start eth0 eth1:0
> 
>   # Start an interface with command-line options
>   net start -i eth0 -a 192.168.1.1 -n 192.168.1.0 \
>   -b 192.168.1.255 -m 255.255.255.0
> 
>   # Stop all the interfaces, typically from init(8)
>   net stop
> 
>   # Stop some specific interfaces
>   net stop lo eth0
> 
>   # Restart the network
>   net restart
> 
>   # Restart a specific interface
>   net restart eth1
> 
>   # Show network status
>   net status
> 
>   # Same as above (default action)
>   net
> 
>   # Show the status of a specific interface
>   net status eth1
> 
>   # Create or modify an interface configuration file
>   net config eth0
> 
> The advantage is that you can now start/stop specific interfaces with simple
> commands using predefined configs, while the old script could only be used
> to start the entire network and couldn't stop or restart it or part of it.
> 
> The new /etc/init.d/network script just calls the /usr/sbin/net command,
> which does all the real work, with the proper args, just start or stop, and
> all the configuration options are now stored as separate config files.
> 
> The package can be installed over an slink system because the preinst script
> can convert automatically the old network file to the new eth0 config.
> 
> The package is available at the following location:
> 
>   http://www.cs.unitn.it/~dz/debian/net_1.0-1_all.deb
> 
> Please have a look and see if it can added to the main debian distribution.
> 
> -- 
> Massimo Dal Zotto
> 
> +--+
> |  Massimo Dal Zotto   email: [EMAIL PROTECTED]   |
> |  Via Marconi, 141phone: ++39-0461534251  |
> |  38057 Pergine Valsugana (TN)  www: http://www.cs.unitn.it/~dz/  |
> |  Italy pgp: finger [EMAIL PROTECTED]  |
> +--+
> 

So what is the big difference between your tool and ifconfig? Seems you
get the same results and you don't save a lot of work... Please provide
more details on benifits of your tool.

-- 
Even more amazing was the realization that God has Internet access.  I
wonder if He has a full newsfeed?
-- Matt Welsh


pgpUTDIkThwH9.pgp
Description: PGP signature

Intent to package asmon

[Brian E. Ermovick]

Hrmm -- I maintain wmsysmon, but am always looking around the dockapps,
and ran across asmon the other day, which looks kinda like a combination
of wmmon and wmsysmon -- it has the best parts of each... Needless to say,
it's the one on my wmaker desktop now *grin* -- so unless somebody else
is working on it, I'll probably get it uploaded later on this week.

Re: Ethernet newbee failure

[Goswin Brederlow]

On Fri, 14 May 1999, Dale Scheetz wrote:

> From: Dale Scheetz <[EMAIL PROTECTED]>
> To: debian-devel@lists.debian.org
> Subject: Ethernet newbee failure
> 
> I have added Ethernet cards to two machines, one my Linux box, the other
> my partner's Win'95 machine. To reduce the configuration problems, I
> installed Debian on the second drive of my partner's machine, reducing the
> problem to two Linux machines connected through the same hardware.
> 
> Machine one is 10.1.1.10 and machine two is 10.1.1.20.
> 
> >From either machine I can ping that machine but not the other one. Both
> cards seem to be working, but I have no network connection. I set them
> both up with ifconfig and route add, and when I do an ifconfig, I get the
> following:
>...

Looks ok. A bit unfamiliar, but ok. You forgot to tell the route.
Must be a problem there.

> didn't set them. It isn't clear that this is the failure either.
> 
> I'm pretty ignorant about this stuff, but I think I did everything I need
> to have a LAN but it doesn't work. Any ideas?
> 
> All informative flames appreciated ;-)

You should probably use 192.168.1.10 and 192.168.1.20 as ip adresses and use
a netmask of 255.255.255.0. Also please try

ftp://mirjam.informatik.uni-tuebingen.de/pub/debian/network*.deb

(you also need the dpkgconfig*.deb file there) in two or three days time.

May the Source be with you.
Goswin

Re: Old Library dependencies Re: Release Plans (19990513)

[Hartmut Koptein]

> >   Remove as many dependencies on old libraries as possible, this
> >   includes:
> > 
> > libjpegg6a, libncurses3.4, newt0.25, libpgsql, tk4.2, tcl7.6,
> > libwraster1, libpng0g
> > 
> >   and various older gtk/gnome libraries.
> 
> Lintian has a "depends-on-obolete-package" tag.  I will add newt0.25,
> libpgsql, tk4.2, tcl7.6, libwraster1, and libpng0g to its list.
> This way, the Lintian page for that tag will form a useful index.

Why not  libjpegg6a and libncurses3.4?

What about  xbase, emacs19, altgcc, freetype1, libc5, libg++27, libpaper,
tclx76  and possible some more?


Thnx,

   Hartmut

Re: GPG as a PGP replacement

[Alexander N. Benner]

Hi

Ship's Log, Lt. Michael Meskes, Stardate 140599.1439:
> Which version do you use? I don't have that script.
> ii  gnupg   0.9.6-1GNU privacy guard - a free PGP replacement.
> 

I have:
Desired=Unknown/Install/Remove/Purge
| Status=Not/Installed/Config-files/Unpacked/Failed-config/Half-installed
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ NameVersionDescription
+++-===-==-
ii  gnupg   0.9.5-1GNU privacy guard - a free PGP replacement.


Was it removed in the l8est Version ?
Cannot check the changelog as I still have it ;-)

Greetings
-- 
Alexander N. Benner   [EMAIL PROTECTED]   #Ixthys #Darmstadt #LinuxGer
The grit in your eye soon enters your heartAnne Clark
And all that was strength is just falling apart
We're jumping from one bed and into anotherSELF DESTRUCT
Searching for something that we'll never discover  Joined Up Writing

Re: Old Library dependencies Re: Release Plans (19990513)

[Richard Braakman]

Steve Dunham wrote:
> Library dependencies:
> 
>   Remove as many dependencies on old libraries as possible, this
>   includes:
> 
> libjpegg6a, libncurses3.4, newt0.25, libpgsql, tk4.2, tcl7.6,
> libwraster1, libpng0g
> 
>   and various older gtk/gnome libraries.

I hesitate to add this as a release goal, except for dependencies
on packages which have been removed or which no longer have source.
For other packages it should be worked on, but it is not critical for potato.

Lintian has a "depends-on-obolete-package" tag.  I will add newt0.25,
libpgsql, tk4.2, tcl7.6, libwraster1, and libpng0g to its list.
This way, the Lintian page for that tag will form a useful index.

Richard Braakman

Re: Release Plans (1999-05-10)

[Richard Braakman]

Adam Di Carlo wrote:
> Richard Braakman <[EMAIL PROTECTED]> writes:
> 
> > We have a long history of overly optimistic freeze dates :-)  I'd like
> > to try something else this time.  I note, though, that if we do manage
> > to freeze on July 1, we'll be able to have a release in time for the
> > Linuxworld Expo in August.  That would be cool.
> 
> I'd like to point out that expecting freeze to be shorter than 10
> weeks is lunacy.  We have 5 architectures now Consider that
> archive changes at any point in freeze imply changes in boot floppies
> (well, for anything in base) and in the CD system.

>From freeze to release, one month.  We won't freeze at all until we
have a plan that allows this.  The plan must have room for delays,
and the ripple effects caused by changes.

Richard Braakman

Re: Release Plans (1999-05-10)

[Richard Braakman]

Adam Di Carlo wrote:
> David Bristel <[EMAIL PROTECTED]> writes:
> > there's been a major package release since the dist went frozen.  If
> > the developer wants to make a slink version, because of either
> > personal reasons, or because of requests, then, once the new
> > package(s) have been tested, let them be added into "updates".
> 
> Um, we already have this.  It's called 'stable-updates' or
> 'proposed-updates'.

Not really.  Packages are rejected from there, if they are not accepted
for stable.  It's not a permanent repository for unstable packages compiled
for stable.

Richard Braakman

Re: (LONG) Correct non-US solution

[Craig Brozefsky]

Richard Braakman <[EMAIL PROTECTED]> writes:

> Jonathan Walther wrote:
> > Mirroring Software:
> > ---
> > Im not sure what software is currently used for synchronizing mirrors,
> > however, it will need to take the above policies into account.  Hopefully
> > our additions to the policy will make it so much easier to "stay legal" and
> > avoid worries about legalities that the maintainers will wish to use such
> > software.
> 
> Here's the problem.  We have no control over what software most mirror
> sites run, and most of them run a standard mirroring program that handles
> all their mirrors.  They will not want to make an exception for Debian.
> You will have to find a scheme that can be handled by generic tools.

Well, nearly every mirroring script I have ever seen takes a list of
files to exclude.  Since every mirror would have a set of the metadata
for all packages, they could generate the list of excluded packages
which they should not get.  So a script which took the package
metadata, and the ISO country code would be able to generate a script
for mirrors in any given country.  These could be put into a well
known location on the master server and all servers would mirror them
just like another file.

Then at mirror time, the exclusion list appropriate to their country
would be given to their mirroring software.  Only one mirror per
country (the root mirror prefereably) need do this, as the rest could
just mirror them straight up like they always have.

Am I missing anything?


-- 
Craig Brozefsky<[EMAIL PROTECTED]>
Less matter, more form!  - Bruno Schulz
ignazz, I am truly korrupted by yore sinful tzourceware. -jb
The Osmonds! You are all Osmonds!! Throwing up on a freeway at dawn!!!

Re: (COMPATIBILITY) Correct non-US solution

[Michel LESPINASSE]

> Simpler: instead of requiring people to add /etc/LEGAL, either add it by
> default or require them to add /etc/ILLEGAL. No reason to have illegal be
> the default, might get someone sued. (Actually, the whole scheme might be
> considered "hooks" for encryption and be illegal in some countries; beats
> me. Too bad only lawyers understand the law. :-( )

The more politically correct way to handle this would be to have
/etc/LEGAL-RESTRICTIONS or something like that, that would include a
single line like "France" or "United States". If this file is not present,
the system would rightfully assume that you live in a free country.

--
Michel "Walken" LESPINASSE - Development Engineer at Wind River Systems
 [EMAIL PROTECTED] - http://www.via.ecp.fr/~walken/
DNA is the software of Life.
Did you realize you can have that much fun for a code merge ?

Re: (LONG) Correct non-US solution

[Ben Bell]

On Mon, May 17, 1999 at 12:41:04AM -0700, Jonathan Walther wrote:
> For example,
> 
> Package: ssh
> Export-Restricted: United States
> Import-Restricted: Russia, France
Can I suggest that we use ISO country codes instead?

> The user can do a `touch /etc/LEGAL` to make apt respect Import-Restricted.
It should be the other way around surely, otherwise Debian's default install
would be illegal. There's also the question of whether an official Debian
should provide tools for breaking the law.

> We add a file called "country" which contains the name of the country the
This would be useful for other reasons too - default languages, timezones
etc. could be guessed from this (as mentioned by someone else on this list).

There have been concerns about the mirroring implications raised, and there
may be some dependency problems that arise from packages suddenly
disappearing but this is an area worth consideration (IMHO)

Cheers,
Ben

-- 
+-Ben Bell - "A song, a perl script and the occasional silly sig.-+
  ///  email: [EMAIL PROTECTED]www: http://www.deus.net/~bjb/
  bjb Snail: 20 Guildford Road West, Farnborough, GU14 6PU
  \_/"Open Source - It's easier to create than to destroy."

Developers in Dresden, Potsdam, Greifswald

[Christian T. Steigies]

Hi,
I will be going to the esa/pac conference in Potsdam (31.5. - 3.6.).
On my way to there I will be in Dresden/Weissig on the 29/30.5. and in
Greifswald from the 3.6. till the 5th or 6th.
Id like to meet developers there, towns nearby (Berlin?) or on my way
from/to Kiel as time permits. Signing keys and showing me the local beer
specialities :*)

Ciao,
Christian.

Re: Star Office 5.0

[Jean Charles Delepine]

"M. Robert Tomasch" <[EMAIL PROTECTED]> writes:

> Obviously so5.0 doesn't work with stock potato.  Has any started to work
> on a hack around the glibc problems yet or no?  Also has anyone
> contacted Star  Division about this?

You might have a look to :

http://www.linux-france.org/article/appli/StarOffice/StarOffice_glibc_2.1.txt

It's a french text but you should understand the shell script :-)
(Anyway, I can try a translation if needed).

Jean Charles

(MIRROR ADMINS) Correct non-US solution

[Jonathan Walther]

Another concern noted was that if we require special mirroring software to
mirror Debian, many hardnosed sysadmins will take some convincing to use our
script.

That is not our problem. Either they use our mirroring script, or use their
regular script to mirror from a debian mirror in the SAME country as it is,
which DOES run our script.

If they do not, they are not an official mirror.  I don't see this as losing
us any mirrors.  Most `official' mirrors we have now care about Debian at
least somewhat, or they wouldn't have became "official" in the first place.
Only the official mirrors go in the masters.list

And the security of knowing that only software that is legal to be on the
server IS on it should be very reassuring to most site admins out there.
We just must take care to INFORM the admins of this when politely requesting
them to use our software.


Jonathan

Re: (LONG) Correct non-US solution

[Richard Braakman]

Jonathan Walther wrote:
> Mirroring Software:
> ---
> Im not sure what software is currently used for synchronizing mirrors,
> however, it will need to take the above policies into account.  Hopefully
> our additions to the policy will make it so much easier to "stay legal" and
> avoid worries about legalities that the maintainers will wish to use such
> software.

Here's the problem.  We have no control over what software most mirror
sites run, and most of them run a standard mirroring program that handles
all their mirrors.  They will not want to make an exception for Debian.
You will have to find a scheme that can be handled by generic tools.

Richard Braakman

Re: y2k compliance - release goal for potato ???

[Richard Braakman]

John Lines wrote:
> Should we make year 2000 compliance a goal for potato ?- i.e. everyone should
> check their packages for year 2000 compliance, and have the Debian web page
> updated to confirm this (I am sure there are many packages which are
> compliant, but where the maintainer has just not got around to saying so)

Making this a release goal will be counterproductive, since such a project
would delay the release until sometime in 2000.

Richard Braakman

Re: Release Plans (1999-05-10)

[Richard Braakman]

Brian Almeida wrote:
> On Thu, May 13, 1999 at 01:02:10PM +0200, Richard Braakman wrote:
> > Hmm... then why isn't it used on my system?  devpts is mounted, I
> > have /dev/ptmx, but /dev/pts is empty.

> Perhaps you aren't using anything that uses unix98 ptys?  Not everything
> uses them by default, you know. Sometimes patches are required.  Try ssh'ing 
> to 
> localhost, or install Eterm, or grab some of the packages from 
> ftp.espy.org/pub/debian (I think that's the correct path).  In there is 
> patched 
> packages for screen, telnet, etc to use Unix98 ptys.

Ah, I was using old xterms.  The new ones do use /dev/pts.  Thanks
for clearing up my confusion :)

Richard Braakman

Re: Release Plans (1999-05-10)

[Richard Braakman]

Anthony Towns wrote:
> Could you please add "IPv6 Support" as a Release Goal? I'm willing to
> act as a sponsor for this (especially since it looks like everyone else
> is more than happy to do the actual work :).

Certainly.  How much time do you expect this to take?

> We're aiming, more or less, to get the base system IPv6 capable for potato,
> and, I guess, most of the networky programs of priority Standard or higher
> for woody [0].

It's the first part that interests me, of course :)  How do you define
"the base system"?

> Still to do for potato:
> 
> * ftp, ftpd
> * finger, fingerd
> * identd
> * tcpdump
> * lynx
> * ssh

This looks like a rather large base system, you see.  Which parts would
be the release goal for potato?

Richard Braakman

Re: (COMPATIBILITY) Correct non-US solution

[Havoc Pennington]

On Mon, 17 May 1999, Jonathan Walther wrote:
> 
> The concern has been raised about people using older versions of apt
> suddenly unknowingly breaking the law. I propose that the new mirroring
> scheme only apply to those distributions (potato? the one after?) which
> implement the policy.  All the older ones would continue to be mirrored as
> before.
> 

Simpler: instead of requiring people to add /etc/LEGAL, either add it by
default or require them to add /etc/ILLEGAL. No reason to have illegal be
the default, might get someone sued. (Actually, the whole scheme might be
considered "hooks" for encryption and be illegal in some countries; beats
me. Too bad only lawyers understand the law. :-( )

Havoc

Solaris NFS problems with potato / Release Notes

[Rainer Dorsch]

It is nearly impossible to compile a large software package on a NFS mounted 
partion on potato, when it is exported by Solaris 2.6 (UltraSPARC). A search 
in Deja News (Solaris NFS patch in comp.os.linux.*) found a couple of 
controversal postings. After installing the patch recommended by Linus

 105379-05

(& stopping and starting the NFS-server on Solaris) the problems did not 
disappear.

To illustrate the problem, I use the cvs source package and compile it on a 
local disc. Works well.

Now I do it on an NFS mounted partition:

$debuild
[...]
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c vers_ts.c
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c subr.c
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c filesubr.c
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c run.c
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c version.c
gcc  -I. -I.. -I. -I../lib -DHAVE_CONFIG_H  -g -O2 -c error.c
gcc  -I. -I.. -I. -I../lib -I../zlib -DHAVE_CONFIG_H  -g -O2 -c ./zlib.c
gcc add.o admin.o buffer.o checkin.o checkout.o classify.o client.o commit.o 
create_adm.o cvsrc.o diff.o edit.o entries.o expand_path.o fileattr.o 
find_names.o hardlink.o hash.o history.o ignore.o import.o lock.o log.o 
login.o logmsg.o main.o mkmodules.o modules.o myndbm.o no_diff.o parseinfo.o 
patch.o rcs.o rcscmds.o recurse.o release.o remove.o repos.o root.o rtag.o 
scramble.o server.o status.o tag.o update.o watch.o wrapper.o vers_ts.o subr.o 
filesubr.o run.o version.o error.o zlib.o ../lib/libcvs.a ../diff/libdiff.a 
-lz -lcrypt   -o cvs
checkin.o: file not recognized: File truncated
collect2: ld returned 1 exit status
make[2]: *** [cvs] Error 1 

The error message is missleading, it is nearly random per run!

Is anybody using potato with NFS exported Solaris partitions? Which patches 
did you apply? Did you try to compile large programs? Hello world programs 
work!



Bob,

I think, this issue should go into the release notes, because it may break 
systems running slink for months without any problems, until the patch was 
applies on the Solaris NFS server. Since the admin of the linux box might not 
be able to do the solaris patch it should go into the

 If the user doesn't read the upgrade notes for this package
   *before* upgrading, the upgrade will not run smoothly
 => the maintainer needs to clean up their package so this is not
the case, but if this is impossible, the situation should be
described on the main release notes page

And the problem is ugly, because as normal user you will not note immediately, 
that something is going wrong, but you can possibly corrupt you NFS mounted 
data (usually the home directory!).

I will send you a summary, as soon I get the problem solved.

--Rainer.

(COMPATIBILITY) Correct non-US solution

[Jonathan Walther]

The concern has been raised about people using older versions of apt
suddenly unknowingly breaking the law. I propose that the new mirroring
scheme only apply to those distributions (potato? the one after?) which
implement the policy.  All the older ones would continue to be mirrored as
before.

If you want to upgrade to a new major release, you have to expect things to
change and shouldn't be using an older version of apt.

Jonathan

Re: (FINISH) Correct non-US solution

[J.H.M. Dassen]

On Mon, May 17, 1999 at 01:12:20 -0700, Jonathan Walther wrote:
> Since the main (but not exclusive) use of non-US right now is for crypto
> software, we might want to create a Crypto-Regulations package which
> contains references to which countries restrict import and export of
> crypto, and how, with references to appropriate legislation and
> documentation.

Most of this is already available at the Crypto Law Survey,
http://cwis.kub.nl/~frw/people/koops/lawsurvy.htm

Ray
-- 
LEADERSHIP  A form of self-preservation exhibited by people with auto-
destructive imaginations in order to ensure that when it comes to the crunch 
it'll be someone else's bones which go crack and not their own.   
- The Hipcrime Vocab by Chad C. Mulligan

(FINISH) Correct non-US solution

[Jonathan Walther]

Since the main (but not exclusive) use of non-US right now is for crypto
software, we might want to create a Crypto-Regulations package which
contains references to which countries restrict import and export of crypto,
and how, with references to appropriate legislation and documentation.

This would save every maintainer of a crypto package from having to go
hunting down the information, and in fact could be made a part of Debian
Policy for crypto packages to respect it.

Don't think of this as added work:  think of it as doing right something
which has only been half-done up to now.  The US is not the only country
with restrictions on software.  Our source of possible countries in which to
host our non-US mirrors seems to shrink every year.

Jonathan

posix.1b semaphores?

[Vladimir G Stanishev]

hello.  

I am writing a paper on real time and linux for a class I am taking this
semester and so am looking at some of the posix.1b features available.
I've installed libc6-dev 2.1.1-3, more to browse through the files than
for anything else.  In the  file the #define
_POSIX_SEMAPHORES 1 is commented out with teh comment that "we are not
quite there yet", and rigth above that line another comment says that only
non-shared semaphores are supported. Is this the reason why this line is
commented out? I am just puzzled a little that's all.


- Vladimir

Putting distributed-net back up for adoption

[Adam Klein]

When Joey Hess decided to get rid of all his non-free packages, I snapped
up distributed-net, but I was a bit too hasty.  It has a rather large
number of bugs filed against it, and it's been exhibiting some weird
behavior (apparently caused by the glibc2.1 move).  And being a binary-only
package, it's a bit hard to debug.  So, I'm putting it back up for adoption.

adam
-- 
no joy killed a yam

(SHORT) Correct non-US solution

[Jonathan Walther]

the only difference is that non-US disappears, and formerly non-US software
goes on every mirror where its legal to go.

Jonathan

better /etc/init.d/network

[Massimo Dal Zotto]

Hi,

The /etc/init.d/network script created by the debian installation is very
simple and not flexible enough if you need to manage complex networks with
many interfaces.

I have written a generic network interface management command, net, which
can be used to start/stop/show/configure network interfaces, and a smarter
replacement for the /etc/init.d/network script.

The net command makes use of configuration files stored in /etc/network/
which contain the various interface options. For example my eth0 is:

  # /etc/network/eth0
  IPADDR=192.168.0.1
  NETMASK=255.255.255.0
  NETWORK=192.168.0.0
  BROADCAST=192.168.0.255
  GATEWAY=192.168.0.1

The command can be used in the following ways (more info in the man page):

  # Start the network, typically from init(8)
  net start

  # Start some specific interfaces
  net start eth0 eth1:0

  # Start an interface with command-line options
  net start -i eth0 -a 192.168.1.1 -n 192.168.1.0 \
-b 192.168.1.255 -m 255.255.255.0

  # Stop all the interfaces, typically from init(8)
  net stop

  # Stop some specific interfaces
  net stop lo eth0

  # Restart the network
  net restart

  # Restart a specific interface
  net restart eth1

  # Show network status
  net status

  # Same as above (default action)
  net

  # Show the status of a specific interface
  net status eth1

  # Create or modify an interface configuration file
  net config eth0

The advantage is that you can now start/stop specific interfaces with simple
commands using predefined configs, while the old script could only be used
to start the entire network and couldn't stop or restart it or part of it.

The new /etc/init.d/network script just calls the /usr/sbin/net command,
which does all the real work, with the proper args, just start or stop, and
all the configuration options are now stored as separate config files.

The package can be installed over an slink system because the preinst script
can convert automatically the old network file to the new eth0 config.

The package is available at the following location:

  http://www.cs.unitn.it/~dz/debian/net_1.0-1_all.deb

Please have a look and see if it can added to the main debian distribution.

-- 
Massimo Dal Zotto

+--+
|  Massimo Dal Zotto   email: [EMAIL PROTECTED]   |
|  Via Marconi, 141phone: ++39-0461534251  |
|  38057 Pergine Valsugana (TN)  www: http://www.cs.unitn.it/~dz/  |
|  Italy pgp: finger [EMAIL PROTECTED]  |
+--+

(LONG) Correct non-US solution

[Jonathan Walther]

We are here to make software free.  We can make it free, or we can drive
thorns into our flesh trying to change the minds of uncaring governments.

Our current situation with the non-US section of our distribution is akin to
a form of fruitless martyrdom.  Its painful to us, but doesn't really affect
the policy of any governments involved.

I would like to propose a solution that makes the distribution of
export/import restricted software both painless to us, and as hard as
possible to any collection of entitites to stamp out.  And, for citizens who
wish to respect the law of their country, I propose that the same
measures will add simple, automatic facilities for keeping their systems
"legal", configurable dynamically.

The proposal calls for the folding of non-US into the other three
distributions so it disappears without a trace, like the morning dew in the
afternoon sun.  The changes that would make this feasible follow:

Changes to a packages control file:
--
Two new fields are added to the control file, Import-Restricted and
Export-Restricted.  These fields take a comma delimited list of countries.

For example,

Package: ssh
Export-Restricted: United States
Import-Restricted: Russia, France

Import-Restricted lists countries where its illegal to install the software.
The user can do a `touch /etc/LEGAL` to make apt respect Import-Restricted.
Someone might also want to write a "legalize" program to deinstall illegal
software should the feds come a'knocking.  `rm /etc/LEGAL` would allow full
access again.

Export-Restricted determines which mirrors will accept the package for
redistribution.

Changes to /etc
---
We add a file called "country" which contains the name of the country the
box is in.  This lets the package software keep the system conformant to the
laws for the particular country its in.  It also will allow a maintainer to
easily see if a configuration works for a particular country in conjunction
with the "legalize" program.

As mentioned before, there is the "LEGAL" file, which makes the package
software respect the laws of the country its in if its present, and if
absent, the software ignores the Import-Restricted field.

Change to dupload and dinstall:
---
If the maintainer of a package is in one of the Export-Restricted countries,
refuses upload the package.  If the server specified is in one of the
Import-Restricted countries, refuses to upload the package.

A package may be uploaded to any of the "official" servers that allow it, by
a maintainer, however the .dsc and .changes file will be uploaded to one
central server (probably master.debian.org) automatically by the script,
from which the Packages files will be generated and Mirrored.

Dinstall will be modified to account for the fact that a package may be on
another server, but the security implications of having an untrusted server
are minimal, given we have md5sums and a rejected Package won't show up in
the Packages file, thus being invisible, should a mirror maintainer decide
to unilaterally move something from Incoming to its appropriate directory
themselves.

The mirroring software will be modified to check its current packages
against the Packages list, and hunt down and download any package it is
allowed to (which it is not Export or Import restricted from) that has
changed.

Thus, server foo in France will not download the ssh package, but if the
maintainer of ssh always uploads to the Incoming on a canada.debian.org, all
mirrors that are allowed to will hit every server in the master.list that
might have the package until it finds the one (canada.debian.org) that has
it.

Changes to apt and dpkg:
---
Respect the presence or absence of /etc/LEGAL.  If a selected package is
Import-Restricted, it won't download or install it unless /etc/LEGAL is
missing.

Packages files:  are the same on every mirror, are NOT generated locally. If
a package isn't found on one server, apt automatically hunts for it first,
on servers in sources.list, then on servers in master.list
 
/etc/apt/sources.list will now just be taken as hints: downloads and
Packages updates will be attempted from the sites in the file, but failure
of those servers is no longer fatal; downloads will be attempted from
master.list

/usr/share/apt/master.list will contain a list of all official debian
mirrors in the same format as sources.list, with the exception that the name
of the country the server is in will be prepended to each line.  However,
the meaning of the entries are slightly different; it is a "what I provide
and where to find it" entry, as opposed to a "look here for this" entry.

This:
canada deb http://http.ca.debian.org/debian bo main
is the entry for a Canadian server that just provides the main section of
the bo release.

This:
france deb http://http.fr.debian.org/t/debian unstable main contrib non-free
france deb http://http.fr.debian.org/gin/borsch stable main con

Re: y2k compliance - release goal for potato ???

[Joseph Carter]

On Mon, May 17, 1999 at 07:55:00AM +0100, John Lines wrote:
> Should we make year 2000 compliance a goal for potato ?- i.e. everyone should
> check their packages for year 2000 compliance, and have the Debian web page
> updated to confirm this (I am sure there are many packages which are
> compliant, but where the maintainer has just not got around to saying so)
> 
>   John Lines
> 
> p.s. Anyone who is running Debian in a corporate environment is probably
> being nagged by their management about year 2000 issues. (all our Windows
> NT stuff is Year 2000 compliant - it must be because they keep releasing
> updates to make it be ;-)

Linux is almost completely y2k compliant BY DEFAULT.  Any y2k issues that
haven't been fixed by now will get fixed in SLINK, not just potato.  We
don't know of any that remain however.  If something binary only in
non-free breaks, you are on your own unless someone else fixes it.

hmm, psychic sig generator strikes again!

--
Joseph Carter <[EMAIL PROTECTED]>Debian GNU/Linux developer
PGP: E8D68481E3A8BB77 8EE22996C9445FBEThe Source Comes First!
-
While the year 2000 (y2k) problem is not an issue for us, all Linux
implementations will impacted by the year 2038 (y2.038k) issue. The
Debian Project is committed to working with the industry on this issue
and we will have our full plans and strategy posted by the first quarter
of 2020.


pgpvTbJpWm9nE.pgp
Description: PGP signature

qt2beta available

[Heiko Schlittermann]

[ Please CC to me personally since I was unsubscribed here due to too
much `uploaded' traffic. And I don't know if my re-subscription went
through. ]

I've made available a qt2beta package on

http://master.debian.org:~heiko/qt2/


NOTES:  

o This stuff is BETA (the QT as well as the packaging)

o The package doesn't conflict with any regular QT package
  (at least this is my intention -- again: BETA!).

o You must not send any bug report regarding the packaging to any of
  the QT mailing lists or maintainers!

o You have to point you compiler/linker to the directories
  /usr/lib/qt2beta/{lib,include,bin} -- probably best by
  using something like QTDIR=/usr/lib/qt2beta ...

  e.g.:

QTDIR=/usr/lib/qt2beta make

  For a Makefile example see 
/usr/doc/qt2beta/examples/aclock/Makefile

  ADDITIONAL NOTE:  If you take the examples Makefile your binary 
  will be built with the `rpath' linker option.  This makes the
  binary aware of the library location in /usr/lib/qt2beta/lib and  
  may produce bizarre effects if different versions of the QT2 library 
  exist in /usr/lib _and_ /usr/lib/qt2beta/lib!

o The package is built on a slink system, this might produce
  some library problems on potato (why?  Aren't we compatible??)  
  -- Anyway, you might be forced to re-build the package on a 
  potato system.

  [ Can anybody give me ssh access to a potato system to do this job
  myself? ]


Best Regards from Dresden/Germany
Viele Gruesse aus Dresden
Heiko Schlittermann
-- 
[internet & unix support - Heiko Schlittermann]
[http://debian.schlittermann.de/";> Debian 2.1 CD ]
[Heiko Schlittermann HS12-RIPE finger:[EMAIL PROTECTED] -]
[pgp: A1 7D F6 7B 69 73 48 35  E1 DE 21 A7 A8 9A 77 92 ---]

y2k compliance - release goal for potato ???

[John Lines]

Should we make year 2000 compliance a goal for potato ?- i.e. everyone should
check their packages for year 2000 compliance, and have the Debian web page
updated to confirm this (I am sure there are many packages which are
compliant, but where the maintainer has just not got around to saying so)

John Lines

p.s. Anyone who is running Debian in a corporate environment is probably
being nagged by their management about year 2000 issues. (all our Windows
NT stuff is Year 2000 compliant - it must be because they keep releasing
updates to make it be ;-)

Re: Hey, Y'all, check out my new improved "Free Software Research Paper Project" web site!

[Adam Di Carlo]

> "brock" == R Brock Lynn <[EMAIL PROTECTED]> writes:

brock> And you really can't argue with a FEELING.

Oh -- well, I guess not.  In fact, that kinda points out that there's
no really reason for carrying on any further debate.

brock> Yes, I don't doubt you have much experience... but only in
brock> "your neck of the woods". I feel like I'm seeing things from a
brock> bit more of a grander perspective, and you are allowing the
brock> local trees to blind your view of the forest overall.  All I
brock> ask is for you to stretch yourself, let yourself grow just a
brock> little bit more. Allow yourself to be just a little more open
brock> minded. Please try! That's *all* I ask.

This is just insulting.  I've contributed untold hours to free
software over the past 8 years (they still credit me for some NCSA
httpd stuff), I have spearheaded the release of software written and
paid for by my company (onShore) under the GPL.  What have *you* done?

>> You really should take up this discussion at the FSB (free software
>> business) list; a very good, well-balanced, knowledgable list.

brock> Interesting, what's the address to subscribe and such?

   <[EMAIL PROTECTED]>:
   Receive future messages sent to the mailing list.

   <[EMAIL PROTECTED]>:
   Stop receiving messages.

These guys really know their shit -- they'll roast you even harder
than I have.  Or ignore you, which is maybe what I should have done.

--
.Adam Di [EMAIL PROTECTED]http://www.onShore.com/>

Re: Upload queue software?

[Roman Hodek]

> Well, it's about time I upgraded from the fairly ancient version of
> this that I'm using on www.uk.debian.org, and making a package will
> probably only add a minor overhead to the procedure, so if you like,
> I'll look at packaging it.

Sure, go ahead; I won't mind :-)

Roman

Developer DB, gpg and stuf..

[Jason Gunthorpe]

Hi all,

As part of the Developer database project we are going to be having an
email gateway that allows modification of many of the fields based on
signed messages. To test the basic setup of this I have setup an address
called [EMAIL PROTECTED]

If you send a pgp signed email to that address you will receive a response
like I have included below indicating your public ldap record. In future
adresses will be provided to allow you to reset a lost password via pgp
signed email, change some fields and maybe a few other things as they come
up.

How the service works is it accepts the signed message and feeds it
through GPG. GPG then tells it who signed it which is then used to
determine which LDAP record to use. It then connects to the database and
sends the text message back from it. The ping address ignores the content
of the mail, others will require a particular format. The response will be
sent instantly.

New for our PGP type services is a replay cache. Particularly for things
like changing passwords you don't want someone re-sending your messages so
the server protects against this. It should never accept the same message
twice - please test this :> Part of that also means your clock will have
to be accurate to at least one week.

Since it uses GPG it will accept basically any signature you throw at it,
including the DSA ones (a GPG key).

You can email me if there is some [important] incorrect information I'll
update things by hand for the moment.

Thanks,
Jason

-

Hello Jason Gunthorpe <[EMAIL PROTECTED]>!

Here is a list of all the public fields associated with your LDAP entry:

c: ca
cn: Jason
createtimestamp: 19990425035220Z
creatorsname: uid=admin2,ou=users,dc=debian,dc=org
emailforward: [EMAIL PROTECTED]
gecos: Jason Gunthorpe
gidnumber: 800
homedirectory: /debian/home/jgg
ircnick: Culus
keyfingerprint: 64BE1319CCF6D393BF87FF9358A6D4EE
l: Edmonton, Alberta
labeledurl: http://www.debian.org/~jgg
loginshell: /bin/bash
modifiersname: uid=jgg,ou=users,dc=debian,dc=org
modifytimestamp: 19990503040420Z
shadowlastchange: 10568
shadowmax: 9
shadowmin: 0
sn: Gunthorpe
supplementarygid: adm
supplementarygid: distmnt
uid: jgg
uidnumber: 1083

Please email [EMAIL PROTECTED] if you have any questions.

Re: Hurray! and thanks! Re: Update ... Re: Ethernet newbee failure

[Adam Di Carlo]

> "Dale" == Dale Scheetz <[EMAIL PROTECTED]> writes:

Dale> Thanks to Tony Mancill for pointing out:
Dale> http://cesdis.gsfc.nasa.gov/linux/setup/3c5x9setup.html

Dale> The 3c5x9setup program takes the 509 out of PnP mode and lets
Dale> you set the base and IRQ addresses and write them into the
Dale> eeprom.

Is this packaged?  I have a 3c509, and it's reliable as hell, but I
hate the DOS boot disk to configure aspect of it.

--
.Adam Di [EMAIL PROTECTED]http://www.onShore.com/>

Re: VX Chipsets and 2.2.5

[Robert Woodcock]

Michael Beattie wrote:
>I have been told by an aquaintance that linux 2.1.x or greater kernels are
>unlikely to boot on a Motherboard that uses a VX Chipset. I have a
>VXPro...

Well at least your friends know where to get good crack.

2.2.5 runs fine on VX boards, and oh, BTW, the VXPro is absolutely not a VX.
It's a clone with a similar feature set.

Also, check http://www.debian.org/releases/slink/running-kernel-2.2 for
important information about running 2.2 kernels on slink.
-- 
Robert Woodcock - [EMAIL PROTECTED]
"Now don't you think that's better than some quadrupally redundant,
electronic, Microsoft software control system?"
  -- Burt Rutan on the crashworthiness of the Proteus rocket module

Re: Homapages in list of maintainers

[Atsuhito Kohda]

Hello, I am a member of Debian JP and wish to become an official
maintainer of Debian.

It is a great news and very welcomed in Debian JP that Mr. Ukai
declared to volunteer to take charge of New Maintainers Interview
in Japan.

And many members in Debian JP are now waiting the time it is realized.
Of course, I expect it is realized as fast as possible.

If it is realized, I will try to be the first official maintainer
from Debian JP among the maintainers interviewed by Mr. Ukai :-)

There is no objections at all untill now but in spite of the question
by Mr. Sano:

From: Taketoshi Sano <[EMAIL PROTECTED]>
Subject: Re: Homapages in list of maintainers
Date: 15 May 1999 03:49:47 +0900

> Are there any objections for Ukai's taking charge of 
> New Maintainers Interview in Japan ?
> 
> If not, Please tell me the address to contact about this.
> Is [EMAIL PROTECTED] the correct address to contact ?

it seems no answer is given yet.

Is something more needed for Ukai's proposal to be realised ?
I wish a fast and practical answer from one of the authorized members
and it is realised as fast as possible.

Or something is going on already where I do not know ?

Thanks in advance,  1999.5.17

--
 Debian JP Developer - much more I18N of Debian
 Atsuhito Kohda <[EMAIL PROTECTED]>
 Department of Math., Tokushima Univ.

Re: Abacus Portsentry License

[UnderGrid Founder]

I've actually had several contacts with the author of portSentry 
(whom also wrote hostSentry & logcheck) and have also beta test'd hostSentry
for him... From my communications I've found him to be quite responsive to
suggestions... If there is any changes that need to be made to portSentry
I would make them to him and chances are they would be made... For one I've
talk'd with him before about possibly packaging portSentry for Debian but had
not registered at the time and mention'd the fact that config files would 
need to be changed for Debian and he mention'd that he had been planning to 
move the config files to a different hierarchy... So chances are he would be
willing to move the config files and thus removing that need for modifications
on the part of Debian...
From my experience portSentry is a very nice utility for intrusion 
detection coupled with both hostSentry and logcheck... This could also lend 
itself to modules for portSentry as I had a external module that would get
call'd when portSentry would be triggered... 
Before deciding to just accept portSentry as being non-free on the 
basis of the current wording of the license I would suggest having an indepth
discussion with the author with proposed changes and see if they can't be
made in a public release or if he is willing to modify the license for a 
future release which would lend itself more to the DFSG and be able to be a
part of main...


Respectfully,
Jeremy T. Bouse
Software Engineer
Netsurfer, Inc.

Ben Pfaff decided to waste my bandwidth saying:
> Rene Mayrhofer <[EMAIL PROTECTED]> writes:
> 
>As I am new to creating Debian packages, I am not sure if Abacus 
> Portsentry's
>license allows it to be put in the main (or if it has to go into non-free)
>section. The program is free to use by anybody and can be distributed in
>any form, the only problem is that the author prohibits modifications he
>is not aware of.
> 
> That sounds non-free to me.
> 
>Please could somebody check it (at www.psionic.com) ?  
>I have a written statement by the Author that he allows packages (including
>the needed minor modifications on scripts and Makefiles) to be made and 
> that
>Portsentry can be included in a distribution if the distribution is not 
> sold
>because of Portsentry.
> 
> Please post the full license and the author's statement to
> [EMAIL PROTECTED]  By doing that you'll get everyone
> who's interested in licensing issues to read through it and comment.

-- 
,-,
| Jeremy T. Bouse  -  UnderGrid Network Services, LLC  -   www.UnderGrid.net  |
| PGP ID/Fingerprint: 1024/E83D9AE5/4ACC03F098D78198 19D0593E50E597E9 |
|   Public PGP key available via PGP keyserver at http://pgp.UnderGrid.net|
| [EMAIL PROTECTED]  -  NIC Whois: JB5713  -  [EMAIL PROTECTED] |
|  promotion, n.: New title, new salary, new office, same old crap.   |
`-'


pgpwke6zaO9aT.pgp
Description: PGP signature

Re: Hey, Y'all, check out my new improved "Free Software Research Paper Project" web site!

[R. Brock Lynn]

Adam Di Carlo wrote:
> 
> "R. Brock Lynn" <[EMAIL PROTECTED]> writes:
> 
> > Adam Di Carlo wrote:
> 
> > >   * stressing sharing vs hoarding as the fundamental issue is
> > > reductionistic; it flattens other problems and issues, such as
> > > economic issues, social issues, etc.
> 
> > Sure there are exceptions. Notice I did not say absolute sharing and
> > no hoarding... but more sharing than hoarding in the most general
> > sense is better for all of us.
> >
> > If you can expound on your reasons behind why the reverse might be
> > better, I'd like to hear them.
> 
> Um, no, you misread me.  I object to the two sides of the picture
> being portrayed as "sharing" versus "hoarding".  Sure, for slogans or
> sound bytes, that's a great dichotomy; but for reasoned debate, it's a
> straw man.

Hmm, I disagree. Sometimes one loses sight of the forest for all the trees. I
believe I'm seeing a bigger picture, whilst you are more focused on a few groups
of trees. Try to expand your vision, is what I ask of you. I'm not being Naive.
I have seen how software businesses work on a small scale, but now it's time to
see how the whole software "economy" can work on a grand scale.

Free Software is the best way for most, but not all.

I firmly believe that.

And you really can't argue with a FEELING.

FEELINGS are inherently "illogical" and "unprovable", except maybe in a sort of
"emotional logic" system.

> Personally, I think the issue is "programming for hack value, fun, or
> altruism"" versus "programming for cash".

Yes, but by writing, and providing service and support for Free Software, I
FIRMLY believe that THAT WILL bring you more money in the long run, to those
skilled in doing those things, than can writing and providing service and
support for proprietary software systems. BUT... of course only time will tell.
But I certainly have a very STRONG FEELING that I'm thinking in the right
direction.

> Now, put in those terms,
> it's a little more clear why there are probably 200 commercial
> programmers for every free one.  In fact, it starts to be difficult to
> try to explain to a programmer why they *shouldn't* be coding for cash
> (which seems to be RMS' issue -- "no code for cash").  Personally, I
> like to find a balance.  If a client wants to pay for code to be
> written, great.  If I can convince the client to free the source for
> all, even better (and there are benefits for many situations to this).
> However, in some cases, the client would rather "hoard" the code and
> slap a non-disclosure on me as well.

Well, as they will probably find in the longer term, trying to restrict others
from using their software, THEY will get HURT more than benefit from this
behavior pattern. And it will then get so painful on down the road, that they
will be forced to change to a "Free Software" method. It's only the natural best
thing to do, and most beneficial to all parties, including the software
developers and the people who paid to have the software developed. Sometimes
peoples' hard headedness can only be corrected by the school of hard knocks...
(I know, I'm pretty hard headed myself. And the school of hard knocks is indeed
a very effective teacher.) but sometimes that's just the way it goes. So be it.
But I say that we try to EDUCATE all affected parties of software technology
that "GPL Free Software" is the BEST way to go for most, but not all software
projects. One exception would be software developed for military use that has
more strictly military use, and very little use by the general populace. In that
case it would be of National Security Concern to keep such software development
projects under wraps. Sometimes you have to make trade offs in the fight for
Freedom and Human Rights. Sometimes you MUST use FORCE, when you, or the noble
values you believe in are threatened. Power Hungry people whose sole goals in
life are to strive for more an more power and wealth, strictly as ends in
themselves MUST BE KEPT IN CHECK... for all our sakes.

"And the meek shall inherit the Earth." :)

> I know what I'm talking about -- this is reality.  I've been
> programming and supervising programing in the commercial (and
> academic) context for over 10 years now.

Yes, I don't doubt you have much experience... but only in "your neck of the
woods". I feel like I'm seeing things from a bit more of a grander perspective,
and you are allowing the local trees to blind your view of the forest overall.
All I ask is for you to stretch yourself, let yourself grow just a little bit
more. Allow yourself to be just a little more open minded. Please try! That's
*all* I ask.

> So you can see, when people start getting all frothy about "sharing
> versus hoarding", that's all well and good, it just doesn't speak to
> the reality of the actual world.

> Personally, I think the client has the right to expect certain terms
> on their code.  Some code represents a business edge -- they can't
> free that logic and expect to retain 

VX Chipsets and 2.2.5

[Michael Beattie]

I have been told by an aquaintance that linux 2.1.x or greater kernels are
unlikely to boot on a Motherboard that uses a VX Chipset. I have a
VXPro...

Comments?

 Michael Beattie ([EMAIL PROTECTED])

   PGP Key available, reply with "pgpkey" as subject.
 -
   Cat Game #1:  Hah - made you look!
 -
Debian GNU/Linux  Ooohh You are missing out!

Re: a Chinese version of X-window system for Linux available

[Daniel Martin]

liug <[EMAIL PROTECTED]> writes:

> Dear Sir,
> I am not sure whether this is the right place to post
> this mail.
> We have developed a Chinese version of X-window system
> several years ago, and now we have developed one for
> Linux, 

> I am wondering whether our product could be integrated
> into or bundled with
> the new Debian Linux release.
> We also have some document and screen shot of our
> product.
> Please do not hesitate to contact us if we can help.
> We are looking forward to your answer.

Has anyone contacted these people, or forwarded the message on to the
Debian-Chinese people?