tcsh infinite loop patch

[G John Lapeyre]

I added one line to the tcsh code, which seems to fix the infinite
loop problem.  I sent the patch to the author and the Debian maintainer 24
hrs.  ago.  I expect to hear something over the next few days.
Perhaps the author will make a better fix.
If someone can't wait that long, mail me

John

John Lapeyre <[EMAIL PROTECTED]>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: NPR piece on Linux

[G John Lapeyre]

Just played in AZ, US. 
I had to listen to 50 minutes of crap about presidents and big
companies ! They should have put Linux first.
Stallman may be a bit of a crank, but I'd be upset too about GNU
being downplayed out of existence. 

John Lapeyre <[EMAIL PROTECTED]>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Compiling xtide: C++ error

[adavis]

I am running a pretty much up to date hamm system with egcs g++, etc.
Compiling the most recent snapshot of xtide2 (found at
http://www.universe.digex.net/~dave/files) resulted in the following errors.  
The author of the program instantly revised the source file.  I am attaching
his comments.  The copy of xtide-2.0dev.tar.gz now at the site probably has
these changes already in place.  This is beta software, by the way.  Very
interesting beta software.


> c++ -O2 -fno-strength-reduce-I/usr/X11R6/include -I/include  -Dlinux
> -D__i386__ -D_POSIX_C_SOURCE=199309L -D_POSIX_SOURCE -D_XOPEN_SOURCE=500L
> -D_BSD_SOURCE -D_SVID_SOURCE -D_REENTRANT   -c Station.cc -o Station.o
> Station.cc: In method class Timestamp Station::next_high_or_low_tide(class
> Timestamp, int &, enum Station::Direction)':
> Station.cc:340: no matching function for call to Station::next_zero
> (Timestamp &, PredictionValue (Station::)(Timestamp, unsigned int), int &,
> Amplitude, Amplitude, Station::Direction &)'
> Station.cc:215: candidates are: Station::next_zero(Timestamp,
> PredictionValue (Station::*)(Timestamp, unsigned int), int &, Amplitude,
> Amplitude, Station::Direction)
> Station.cc: In method class Timestamp Station::find_mark_crossing(class
> Timestamp, class Timestamp, int &, enum Station::Direction)':
> Station.cc:375: no matching function for call to Station::find_zero
> (Timestamp &, Timestamp &, PredictionValue (Station::)(Timestamp, unsigned
> int), Station::Direction &)'
> Station.cc:108: candidates are: Station::find_zero(Timestamp, Timestamp,
> PredictionValue (Station::*)(Timestamp, unsigned int), Station::Direction)
> make: *** [Station.o] Error 1


The author's comments:

> I think this is fixable.  The problem is with the function being
> passed as an argument.  I have seen this problem before; GCC quietly
> hides the type mismatch while other compilers choke on it.  (Usually
> Sparcworks chokes on it, but this time it didn't.)
>  
> Please try variations on the syntax in lines 340 and 375 (actually 339
> and 375) and let me know what works.  Here is the original 339:
>   
>  Timestamp thilo = next_zero (t, f_hiorlo, rising,
>  
>  Things to try:
>   
>Timestamp thilo = next_zero (t, &f_hiorlo, rising,
>Timestamp thilo = next_zero (t, (PredictionValue 
> (Station::*)(Timestamp,
> +unsigned int))(f_hiorlo), rising,
>Timestamp thilo = next_zero (t, (PredictionValue 
> (Station::*)(Timestamp,
>   +unsigned int))(&f_hiorlo), rising,
>
> If none of these work, it can still be fixed by making f_hiorlo a
> regular old not-in-any-class type of function.  This is what I already
> did for all of the X toolkit callbacks to avoid exactly this issue,
> and I should probably just do that for these as well.
> 
> What compiler are you using now, anyway?


I am now recompiling.  Station.cc compiled ok.

I post this information for the benefit of those who may be able to make use
of it, not because I understand a single bit of it.

Alan 





  


-- 
Alan E. Davis   Marianas High School (Science Department)   
  
AAA196, Box 10001[EMAIL PROTECTED]   http://www.saipan.netpci.com/~adavis   
Saipan, MP  9695015.16oN 145.7oEGMT+10   Northern Mariana Islands


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Marcus Brinkmann]

On Thu, Apr 09, 1998 at 05:15:59PM -0600, Jason Gunthorpe wrote:
> 
> On Thu, 9 Apr 1998, Marcus Brinkmann wrote:
> 
> > Exception handling is a powerful feature, and makes other global error
> > strategies mostly unnecessary. Therefore the size of compiled and well
> > written C++ programs will not be larger than an equivalent C program. *And*
> > the source code will be much cleaner, as you don't have to nest if()
> > statements or such things.
> 
> Actually egcs just has a gross implementation of exceptions, the overhead
> added for the stack unwinding is horribly high, I have't looked too deeply
> but it may be a fixed overhead per-function and then some added stuff
> depending on the function's content so if you have lots of small functions
> you get burned really badly.

This is a problem of egcs (gcc). C++ on Linux *is* horrible at the moment.
But this does not say anything about the language feature in the standard
and how it *could* be implemented. Not using exceptions is curing the
symptoms but not the disease.
 
> As I said before, the exception handling information doubles the size of
> my binaries (+100k, + 240k, etc) which is pretty bad.

I still think that without using exception handling you have to write error
handling code by yourself. If you do it good, you double the size of your
program either way (approx. 50% of C code is error handling, when the
programmer cares about erro handling, esp. in large projects).

Marcus


-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Packages depending on essential/required packages

[Jason Gunthorpe]

On Wed, 8 Apr 1998, Adam Heath wrote:

> There are also other packages that have dependencies on essential packages.
> It was my understanding that this doesn't have to be done.

There is meaning for depending on specific versions of essential packages
because the package may require some new functions in the Essential
Package. In truth all available essential packages should be installed and
kept at their newest versions...

Jason


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Jason Gunthorpe]

On Thu, 9 Apr 1998, Marcus Brinkmann wrote:

> Exception handling is a powerful feature, and makes other global error
> strategies mostly unnecessary. Therefore the size of compiled and well
> written C++ programs will not be larger than an equivalent C program. *And*
> the source code will be much cleaner, as you don't have to nest if()
> statements or such things.

Actually egcs just has a gross implementation of exceptions, the overhead
added for the stack unwinding is horribly high, I have't looked too deeply
but it may be a fixed overhead per-function and then some added stuff
depending on the function's content so if you have lots of small functions
you get burned really badly.

As I said before, the exception handling information doubles the size of
my binaries (+100k, + 240k, etc) which is pretty bad.

Jason


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

intent to package: coda (+ copyright question)

[Anders Hammarquist]

I'm looking into packaging CMU's coda distributed filesystem. It is based on 
AFS, with enhancements to allow disconnected use. There are kernel drivers for 
it in the 2.1.x series and they are available as patches for the 2.0.x series.

Seeing that there are other kernel module packages available for the 
distributed 2.0.x kernel, would it be useful to package up a kernel-module of 
coda for 2.0? (I'll probably look in to that as well... Though I tend to be on 
the bleeding edge myself.)

Now, on to the copyright question. The following is an excerpt from one file 
(coda-src/vice/srvproc.cc to be exact). The CMU licence does not seem to pose 
a problem, but the IBM notice is rather unclear. The only thing that is 
obvious is that CMU can distribute the source. It would also appear that 
derrived works are allowed. Thoughts please...

Regards,
/Anders

--->8---

/*

Coda: an Experimental Distributed File System
 Release 4.0

  Copyright (c) 1987-1996 Carnegie Mellon University
 All Rights Reserved

Permission  to  use, copy, modify and distribute this software and its
documentation is hereby granted,  provided  that  both  the  copyright
notice  and  this  permission  notice  appear  in  all  copies  of the
software, derivative works or  modified  versions,  and  any  portions
thereof, and that both notices appear in supporting documentation, and
that credit is given to Carnegie Mellon University  in  all  documents
and publicity pertaining to direct or indirect use of this code or its
derivatives.

CODA IS AN EXPERIMENTAL SOFTWARE SYSTEM AND IS  KNOWN  TO  HAVE  BUGS,
SOME  OF  WHICH MAY HAVE SERIOUS CONSEQUENCES.  CARNEGIE MELLON ALLOWS
FREE USE OF THIS SOFTWARE IN ITS "AS IS" CONDITION.   CARNEGIE  MELLON
DISCLAIMS  ANY  LIABILITY  OF  ANY  KIND  FOR  ANY  DAMAGES WHATSOEVER
RESULTING DIRECTLY OR INDIRECTLY FROM THE USE OF THIS SOFTWARE  OR  OF
ANY DERIVATIVE WORK.

Carnegie  Mellon  encourages  users  of  this  software  to return any
improvements or extensions that  they  make,  and  to  grant  Carnegie
Mellon the rights to redistribute these changes without encumbrance.
*/

static char *rcsid = "$Header: /afs/cs/project/coda-src/cvs/coda/coda-src/vice/
s
rvproc.cc,v 4.10 1998/01/12 23:35:42 braam Exp $";
#endif /*_BLURB_*/


/*

 IBM COPYRIGHT NOTICE

  Copyright (C) 1986
 International Business Machines Corporation
 All Rights Reserved

This  file  contains  some  code identical to or derived from the 1986
version of the Andrew File System ("AFS"), which is owned by  the  IBM
Corporation.This  code is provded "AS IS" and IBM does not warrant
that it is free of infringement of  any  intellectual  rights  of  any
third  party.IBM  disclaims  liability of any kind for any damages
whatsoever resulting directly or indirectly from use of this  software
or  of  any  derivative work.  Carnegie Mellon University has obtained
permission to distribute this code, which is based on Version 2 of AFS
and  does  not  contain the features and enhancements that are part of
Version 3 of AFS.  Version 3 of  AFS  is  commercially  available  and
supported by Transarc Corporation, Pittsburgh, PA.

*/

--->8---
-- 
 -- This message specially handmade by the imps of the Couterweight Continent
Anders Hammarquist, Mandolingatan 5, S-421 40 Västra Frölunda
[EMAIL PROTECTED]



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Marcus Brinkmann]

On Wed, Apr 08, 1998 at 02:58:39PM +0300, Amos Shapira wrote:
> On Wed, Apr 08, 1998 at 01:01:46AM +0200, Marcus Brinkmann wrote:
> > However, you can give the compiler a hint that a function does not throw any
> > exceptions by adding throw() at the right place:
> > 
> > class ABC {
> >   ABC (int theInt) throw();
> > }
> 
> Shouldn't the compiler still handle eceptions in functions which call
> functions which throw exceptions?

The compiler will abort() if an exception is not caught. I don't know if the
exception handling code is in the function that calls or in the function
that is called. Note that I'm not at all sure if the above code will lead to
shorter code, as it is mostly for compile-time checking etc. Violations
against the above declaration are possible and in large complex programs
unavoidable.
 
> If so, this probably means that you should have exceptions handled in
> any binary which uses a function which throws exceptions (e.g. almost
> any STL container).

Exception handling is a powerful feature, and makes other global error
strategies mostly unnecessary. Therefore the size of compiled and well
written C++ programs will not be larger than an equivalent C program. *And*
the source code will be much cleaner, as you don't have to nest if()
statements or such things.

Voting for Exceptions,
Marcus

-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Marcus Brinkmann]

On Tue, Apr 07, 1998 at 11:14:17PM +, Falk Hueffner wrote:
> It seems that programs are larger even if they do not use exceptions
> at all (possibly even C programs). For those, it seems totally
> resonable to disable exceptions. It should probably even added to the
> policy, since it saves space.

Well, the rule is that every possible exception will be thrown that could be
thrown in the function ("assume the worst"). You have to declare the thrown
exceptions explicitely if you want to restrict the list (which does make
sense, doesn't it?).

C programs should be compiled with a C compiler.

Exception handling is broken enough yet. Please don't make it even more
unsafe by disabling it in some parts of the programming environment.

Note that by ANSI C++ even the new operator may throw an exception ( bad_alloc()
), and this operator is used in nearly every C++ program. For compatibility
with C functions, you can disable this with new(nothrow).

Hoever, I thought about what I said: I don't think that the program will be
smaller when the exceptions are declared, but some compile-time checking can
be done.

Note that the saving of space is misleading: Notably, a C++ program that
uses exceptions for error handling instead convential methods can even be
smaller, and the source code is *much* cleaner and shorter. It is not
uncommon to have half of the code provided for error handling. With
exceptions, this shrinks down significant. So, the bigger size is more a
sign of unefficient C++ programming style. Only to use exceptions in a few
places and use other global error handling strategies most of the time is
IMHO unnecessary.

Marcus
who prefers safety over brevity.


-- 
"Rhubarb is no Egyptian god."Debian GNU/Linuxfinger brinkmd@ 
Marcus Brinkmann   http://www.debian.orgmaster.debian.org
[EMAIL PROTECTED]for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/   PGP Key ID 36E7CD09


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Packages depending on essential/required packages

[Ben Pfaff]

   On Tue, 7 Apr 1998, Ben Pfaff wrote:

   >  .
   >* need mktemp so depend on debianutils >= 1.8
   > 

   I notice several packages in my available file that depend on debianutils.  
As
   it is essential/required, this doesn't need to be done.  This partitcular
   package isn't in my available yet(still in Incoming), but several others are.

In this case at least, you're wrong.  mktemp was introduced in
debianutils 1.8.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: NPR piece on Linux

[Bdale Garbee]

In article <[EMAIL PROTECTED]> you wrote:
: Anyone have a digitized copy of this?  :)

Look on www.npr.org later tonight.  The 'current' page still has the pieces
from the 7th...  All the interesting stuff on NPR becomes available via
RealAudio shortly after broadcast.

Bdale


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: NPR piece on Linux

[Douglas Bates]

Dale Scheetz <[EMAIL PROTECTED]> writes:

> OK, it has played here in the East. Those of you out west still have
> some time to find a radio ;-) I think it is the second piece on "All
> Things Considered" after the news.
> 
> This was a very good presentation for a public forumn. They pointed to all
> the right facts; talked about Linux clusters outperforming supercomputers,
> along with the fact that nothing M$ produces come close to the perfomance.
> They let Linus have a lot of personal air time, and he said all the right
> things (of course). Even Richard S. got a spoken line or two. I was
> impressed by the factual quality of the presentation.

Check tomorrow at
  http://www.npr.org/programs/atc/archives/1998/current.html
to see if they have that segment available in RealAudio.  They only
have that site updated to the April 7 (Tuesday) edition of All Things
Considered as of a few minutes ago.

This may even induce me to install the RealAudio player!  Then I'm
sure never to get any work done :-)


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Packages depending on essential/required packages

[Adam Heath]

On Tue, 7 Apr 1998, Ben Pfaff wrote:

>  .
>* need mktemp so depend on debianutils >= 1.8
> 

I notice several packages in my available file that depend on debianutils.  As
it is essential/required, this doesn't need to be done.  This partitcular
package isn't in my available yet(still in Incoming), but several others are.

There are also other packages that have dependencies on essential packages.
It was my understanding that this doesn't have to be done.

Should I file bugs?

Adam



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: NPR piece on Linux

[Jeff Noxon]

Anyone have a digitized copy of this?  :)

Thanks,

Jeff

On Wed, Apr 08, 1998 at 05:58:26PM -0400, Dale Scheetz wrote:
> OK, it has played here in the East. Those of you out west still have
> some time to find a radio ;-) I think it is the second piece on "All
> Things Considered" after the news.
> 
> This was a very good presentation for a public forumn. They pointed to all
> the right facts; talked about Linux clusters outperforming supercomputers,
> along with the fact that nothing M$ produces come close to the perfomance.
> They let Linus have a lot of personal air time, and he said all the right
> things (of course). Even Richard S. got a spoken line or two. I was
> impressed by the factual quality of the presentation.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

NPR piece on Linux

[Dale Scheetz]

OK, it has played here in the East. Those of you out west still have
some time to find a radio ;-) I think it is the second piece on "All
Things Considered" after the news.

This was a very good presentation for a public forumn. They pointed to all
the right facts; talked about Linux clusters outperforming supercomputers,
along with the fact that nothing M$ produces come close to the perfomance.
They let Linus have a lot of personal air time, and he said all the right
things (of course). Even Richard S. got a spoken line or two. I was
impressed by the factual quality of the presentation.

Waiting is,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Uploaded mc-4.1.28-1 (source i386) to master

[Chris Fearnley]

'Paul Seelig wrote:'
>
>-BEGIN PGP SIGNED MESSAGE-
>
>Format: 1.5
>Date: Mon, 06 Apr 1998 17:23:15 +0100
>Source: mc
>Binary: mc
>Architecture: source i386
>Version: 4.1.28-1
>Distribution: frozen unstable
>Urgency: low
>Maintainer: Paul Seelig <[EMAIL PROTECTED]>
>Description: 
> mc - Midnight Commander - A powerful file manager.
>Changes: 
> mc (4.1.28-1) frozen unstable; urgency=low
> .
>   * New upstream release.

No new upstream releases into frozen please!  Isn't this policy?  (I
guess we should allow exceptions in the case of important bugs, but no
such issue is mentioned here.)

-- 
Christopher J. Fearnley  |  Linux/Internet Consulting
[EMAIL PROTECTED]   |  Design Science Revolutionary
http://www.netaxs.com/~cjf   |  Explorer in Universe
ftp://ftp.netaxs.com/people/cjf  |  "Dare to be Naïve" -- Bucky Fuller


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: egcs and libc6-dev

[Elie Rosenblum]

Why should libc6-dev depend on gcc, exactly? Would it not make more sense
for it to Conflict: with gcc < 2.7.2.3-1, and let the compilers duke it 
out for themselves?

-- 
Elie Rosenblum That is not dead which can eternal lie,
And with strange aeons even death may die.
Developer / Mercenary / System Administrator - _The Necromicon_


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: sleep contains crypto stuff?

[James Troup]

Vincent Renardias <[EMAIL PROTECTED]> writes:

[ sleep is linked with libcrypt ]

> That's weird since it's in fact linked with libcrypt, but doesn't
> seem to use _any_ function/symbols from this lib:

Everything in shellutils is linked with libcrpyt, build it from source
and see.
 
-- 
James


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: egcs and libc6-dev

[Scott Ellis]

On Wed, 8 Apr 1998, Nuno Ferreira wrote:

>  Is there any reason for libc6-dev to depend on gcc and not on any 
> c-compiler ? 
> As it is now I can't keep egcs as my only compiler without breaking
> dependencies. Either libc6-dev should depend on c-compiler or egcs provide
> gcc. Am I missing something here??

Read the discussion on the bug tracking system on bug 19849
http://www.debian.org/Bugs/db/19/19849.html

It includes what I believe is the latest discussion on the subject.

-- 
Scott K. Ellis <[EMAIL PROTECTED]> http://www.gate.net/~storm/


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: egcs and libc6-dev

[Dale Scheetz]

On Wed, 8 Apr 1998, Nuno Ferreira wrote:

>  Is there any reason for libc6-dev to depend on gcc and not on any 
> c-compiler ? 
> As it is now I can't keep egcs as my only compiler without breaking
> dependencies. Either libc6-dev should depend on c-compiler or egcs provide
> gcc. Am I missing something here??

Several somethings.

Most important: it is still difficult to get egcs and gcc to reside on the
same machine in a friendly manner (some directories and files are in the
same place, some are not)

Secon most important: gcc is the compiler that currently builds the
distribution. While egcs currently seems to have better C++ support than
gcc does at the moment, development proceeds on both, but until they can
work together better, using them both is a problem.

When either of these compilers can be used to build the distribution
without large problems, then the advance to using "c-compiler" as a
virtual package will become possible.

If you really want egcs on your machine and don't want gcc, you can
probably safely force the installation of libc6-dev (or egcs, which ever
gets the complaint)

This is currently to much of a can of worms to allow the change in
libc6-dev at this point.

Waiting is,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

NPR piece on Linux

[Dale Scheetz]

I just got a call from a friend who said they had just heard that NPR was
doing a piece on Linux. It should be on right after the news, so I hope
this gets through the mail server pretty fast ;-)

Later,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

BEWARE

[Bernd Eckenfels]

Hello,

the broken grep (I think it is filed as a Bug already) will do a lot of
damage to your system. It will kill your Windowmanger -list if you install a
Windowmanager, and it will make the /etc/X11/config not work
(user-xsession).

Greetings
Bernd
-- 
  (OO)  -- [EMAIL PROTECTED] --
 ( .. )  [EMAIL PROTECTED],linux.de,debian.org} http://home.pages.de/~eckes/
  o--o *plush*  2048/93600EFD  [EMAIL PROTECTED]  +497257930613  BE5-RIPE
(OO)   If privacy is outlawed only Outlaws have privacy


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Aiding the constitutional procedure [was Re: Automation of the constitutional procedure]

[Dale Scheetz]

On Wed, 8 Apr 1998, James A.Treacy wrote:



> This seemed a natural and orderly extension of implementing the
> constitution. If Ian feels that this is unacceptable for some
> reason, I will make a formal amendment to the constitution
> stating that it be allowed.
> 
While I agree with the merrits of your previous arguments, I don't see
what this has to do with the constitution. The secretary has "powers"
which allow the secretary to execute that office. 

I would prefer that such issues be left of to the secretary to decide. If
there is a need for automated processes, the secretary only needs to
decide that it is necessary.

The constitution should be a minimal document. If we specify every
possible action by each officer and member, the document will get very
large, and even more confusing than it is now.

I would suggest that when an officer is given the power to execute their
office, that is all the constitution need declare. It is then the
responsibility of the office holder to define how the office will be
administered. Giving detailed descriptions of what is "the right way" to
do this task is, in the long run, counter productive.

Remember the KISS principle.

Waiting is,

Dwarf
--
_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz   Phone:   1 (850) 656-9769
  Flexible Software  11000 McCrackin Road
  e-mail:  [EMAIL PROTECTED] Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Aargh - mail

[Adrian Bridgett]

my fault - I was messing with emacs and something went screwy.  Something
moved my mail spool to ~/RMAIL.  

Phew -  while :; do echo "I *will* keep backups" ;done

A relieved Adrian

email: [EMAIL PROTECTED]   | Debian Linux - www.debian.org
http://www.poboxes.com/adrian.bridgett   | 2.0 release soon - over 1800
PGP key available on public key servers  | packages on a stable OS


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

egcs and libc6-dev

[Nuno Ferreira]

 Is there any reason for libc6-dev to depend on gcc and not on any 
c-compiler ? 
As it is now I can't keep egcs as my only compiler without breaking
dependencies. Either libc6-dev should depend on c-compiler or egcs provide
gcc. Am I missing something here??

-- 
Nuno Ferreira
[EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Aiding the constitutional procedure [was Re: Automation of the constitutional procedure]

[James A . Treacy]

When you have been through hundreds of meetings, one of the things
you learn is that communication is extremely important. It is
critical to know what people's intentions are. Critical to know what
the current state of affairs are. It is the normally the job of
the chair of a meeting to make decisions regarding protocol and
the secretary to take notes and count votes. In an electronic
setting, modifications need to be made to account for the distributed
(both in time and space) nature of things.

Ian's proposed constitution does a good job of setting down rules
that deal well with the fact that we are spacially distributed.
It is the job of the secretary to maintain order when things go
wrong and to see that elections are fair.

But there are also temporal problems with our being distributed all
over the world. Developers may only read mail once a day or less.
Going through hundreds of pieces of mail on the list archives is a
waste when you simply want to find the latest version of something.
What happens if the secretary is unavailable for 12 hours? For 3 days?
A week? It is important that the system be set up so that things will
continue to function, as well as possible, under these circumstances. Also,
to keep the job of secretary doable, it shouldn't involve hours every
day when things are hectic. There should be a method so everyday
events are handled without requiring human intervention.

Ian has questioned the wisdom of allowing certain aspects of the
proposal process to handled by scripts. The system proposed is
similar in some ways to the bug tracking system, but more flexible.
Proposals, formal amendments and seconds would all have to be
submitted to this system, which would then make an announcement
on the mailing list. The collection and counting of votes could
also be automated. The secretary has access to all the (ascii) files
and can override anything done by developers (except for change their
vote - they will be signed using PGP). To keep the system informal
everything else is done using the mailing lists.

No one has to guess what the current version of a proposal or amendment
is. The web page generated will have it. Anything else on the
mailing list is simply being discussed. No one has to guess whether
amendments are simply suggestions or when they have become formal.
If there is no web page for an amendment, it is still beind
discussed (hopefully in a friendly manner).

All this is done without the intervention of the secretary. It
is always clear what the status of proposals and amendments are.
Whether there are enough seconders; what is needed for a quorum;
what the supermajority is. When a disagreement arises, it is clear
to the secretary very shortly where the problem lies and can easily
fix it.

This seemed a natural and orderly extension of implementing the
constitution. If Ian feels that this is unacceptable for some
reason, I will make a formal amendment to the constitution
stating that it be allowed.

Jay Treacy


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: dinstall and PGP

[Vincent Renardias]

On Wed, 8 Apr 1998, Marco d'Itri wrote:

> Can someone hack dinstall to install packages which are not PGP signed
> but has been copied to incoming? If the UID of the files is the one of a
> developer we can know who did upload the package.

Definatly not an option, since people uploading anonymously to chiark
would be able to upload whatever in the distribution since the files
arrive in Incoming/ with IanJ's UID (also hold for other upload queues).

Anyway, I fail to see WHY we should allow non PGP signed packages.

Cordialement,

-- 
- Vincent RENARDIAS [EMAIL PROTECTED],pipo.com,debian.org} -
- Debian/GNU Linux:   Pipo:WAW:   -
- http://www.fr.debian.orghttp://www.pipo.com  http://www.waw.com -
---
- "La fonctionnalite Son Visuel vous delivre des avertissements visuels." -
-  [Message durant l'installation de Windows95]:wq 



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: dinstall and PGP

[Enrique Zanardi]

On Wed, Apr 08, 1998 at 08:23:48PM +0200, Marco d'Itri wrote:
> Can someone hack dinstall to install packages which are not PGP signed
> but has been copied to incoming? If the UID of the files is the one of a
> developer we can know who did upload the package.

No. We know which account the uploader used. (Even that is not true. The
uploader may have changed the UID if he obtained root privileges, but
then he can bypass dinstall). And what about packages uploaded to chiar
or erlangen?

We should be talking about improving our security instead (by signing the 
packages, and not the .changes file). One of these days we will find
trojan horses in Debian packages at compromised mirror sites, and will
have to hear all that "But, RPM packages are PGPsigned..." stuff again
and again.

--
Enrique Zanardi[EMAIL PROTECTED]
Dpto. Fisica Fundamental y Experimental Univ. de La Laguna


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

I18N (was Re: package pre-selections tool)

[David Frey]

On Tue, 07 Apr 1998 08:22:55 -0600 Anthony Fok wrote:
>  ( ) I18N/L10N (?) -- packages for the world.  :-)
>   * European languages
>   * Japanese packages (kon2, wnn, emacs20-mule (?), canna, kterm,
> doc-jp-linux, etc.)
>   * Chinese packages (xcin, xfntbig5p-cmex24m, doc-zh-linux,
> and in the future: yact, cjklatex, bcs, chdrv, CXWin? etc.)
I think, we should try hard to make Debian 2.0 8bit clean and
internationalized at least for ``Latin + Cyrillic'' languages per default
(no offense to the Asian people intended). So the first point should be
obsolete.

Then we could offer Japanese/Chinese/Korean/Arab/... languages (with many
glyphs and larger space needs) as extra packages.

Anyway, what I wanted to say is: internationalization/localization and
applications are orthogonal issues.

David


-- 
David Frey (B98D36A9) = 51F35923114FC864 7D05FF173C61EFDE
Those who do not understand Unix are condemned to reinvent it, poorly.
  -- Henry Spencer




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

dinstall and PGP

[Marco d'Itri]

Can someone hack dinstall to install packages which are not PGP signed
but has been copied to incoming? If the UID of the files is the one of a
developer we can know who did upload the package.

-- 
ciao,
Marco


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Automation of constitutional procedures

[James A . Treacy]

> This is precisely the aspect I disagree with.  There are a number of
> problems with this kind of thing.
> 
> Firstly, with an automated system developers' abilities to do things
> will be dependent on the bot's interpretation of what is allowed - the
> bot becomes the governor of the procedure rather than the Secretary or
> the developers together.
> 
How so? As I've stated a number of times only certain things will be
done through they system. The secretary can and should (and will have
time to since they don't have to deal with silly things like keeping
track of seconders) take care of everything else. The secretary is
always the final arbiter and can easily change anything they want to.
In fact, in mail with the secretary, they said that they were thinking
of doing something very much like this.

> Secondly, it is often necessary in procedures such as this to make
> `conformal changes' to a document or an amendment to ensure the
> consistency of the whole document.  Since a bot cannot tell what is
> merely a conformal change and what is a substantive change this will
> become impossible, and a human will have to step in anyway to ensure
> that what gets voted on makes sense.
> 
What's stopping people from doing this? When you make changes to the
constitution you post a new version. I'm not suggesting anything different.
All discussion leading up to that change was held on a public list.
Still no different.

> Thirdly, with a reasonably complicated procedure like the one in the
> proposed constitution it will be necessary to make it as easy and
> informal as possible for people to take actions like proposing and
> seconding resolutions, amendments, &c.
> 
Exactly. Ok, it's not as informal when you wish to second or make an
amendment, but it is still quite easy.

> Fourthly, I don't think developers should be required to learn how to
> drive such a mechanism before being able to take part in the
> decisionmaking process.
> 
Remember, it is only if they wish to second or to be a proposer that
they need to use it at all.

A comparison with the bug tracking system is fair. Are you suggesting
that it is too complicated for developers to use? Take a look at the web
page and you'll see it works along very similar lines. In fact it is
easier than the bug system since all discussion takes place in a single
place (or on multiple lists, as the case may be).

Also, you seem to be ignoring some issues. Developers should at any time
be able to tell what the current proposals are, who the seconders are and
what the current version is. Every time someone wants to find the latest
version of the constitution they must search through old mail to find the
URL. This will be standardized under such a system.

Doing everything by hand is also too dependent on the secretary dealing
with issues daily. Can you tell me who the seconders were for the constitution?
Sure, as leader you don't need any, but that's not the point. Why don't
we get some feedback from the secretary on this.

You seem to be turning this into something out of Brave New World.
Look at http:/www.easynet.on.ca/~treacy/

Jay Treacy


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

new packages

[Alex Romosan]

i finally got around to applying to become a debian developer. so i
want to announce here my intention to package the following programs:

f77reorder

  f77 compiler script calling f2c/gcc. Handles some fortran 77
  extensions by calling f77reorder to change the code.

vat

  The LBNL audio tool, vat, is a real-time, multi-party, multimedia
  application for audio conferencing over the Internet. Vat is based on
  the Draft Internet Standard Real-time Transport Protocol (RTP)
  developed by the IETF Audio/Video Transport working group. RTP is an
  application-level protocol implemented entirely within vat -- you need
  no special system enhancements to run RTP. Although vat can be run
  point-to-point using standard unicast IP addresses, it is primarily
  intended as a multiparty conferencing application. To make use of the
  conferencing capabilities, your system must support IP Multicast, and
  ideally, your network should be connected to the IP Multicast Backbone
  (MBone).
  .
  Vat provides only the audio portion of a multimedia conference; video,
  whiteboard, and session control tools are implemented as separate
  applications. Our video tool is called vic and our whiteboard tool wb,
  UCL developed the session directory tool sdr Other related
  applications include ISI's Multimedia Conference Control, mmcc, the
  Xerox PARC Network Video tool, nv and the INRIA Video-conferencing
  System, ivs.

xmmix

  Xmmix is an audio mixer utility for the X window system using the
  Motif graphical user interface. It operates the input and output
  mixer section on many PC sound cards.

xrn

  X11 USENET news reader, based on rn.

nt

  NTE is a shared text editor designed for use on the Mbone. It is not a
  word processor (it is not clear that word processing is a useful task
  to share) and it is not a whiteboard - if you want a whiteboard, wb
  from LBL is a much better whiteboard.
  .
  Using NTE can be very interactive - unless you lock a block of text,
  anyone else in your session can edit that text or delete it. This is
  intentional. Many people can (if they wish) edit the same document
  simultaneously. Many people can even edit the same block of text
  simultaneously, but if more than one person tries to edit the same
  line at one time, a conflict will occur, which results in only one of
  the changes being preserved.

and some of the packages which need an smotif/dmotif version and don't
have a maintainer yet (xmg, ddd, etc).

all these are already available by anonymous ftp from caliban.lbl.gov
in /pub/debian.

--alex--

-- 
| I believe the moment is at hand when, by a paranoiac and active |
|  advance of the mind, it will be possible (simultaneously with  |
|  automatism and other passive states) to systematize confusion  |
|  and thus to help to discredit completely the world of reality. |


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Default /etc/profile

[LeRoy D. Cressy]

Hi All,

I've been following the discussion on the default profile
prompt.  When I first loaded Linux on my system I used a
Slackware version packaged with a book a couple of years ago.

Though Slackware had no upgradability at the time, I switched 
debian.  But I noticed that Slackware provided for various
shells in use with the following:

if [ "$SHELL" = "/bin/pdksh" -o "$SHELL" = "/bin/ksh" ]; then
 PS1="! $ "
elif [ "$SHELL" = "/bin/zsh" ]; then
 PS1="%m:%~%# "
elif [ "$SHELL" = "/bin/ash" ]; then
 PS1="$ "
else
 PS1='\h:\w\$ '

This method gives a default prompt for users using different 
shells that read the /etc/profile.  Why couldn't Debian use
a simular approach?

Have a good day :-)
-- 
  0 0  L & R Associates
   "   Home Page:http://www.netaxs.com/~ldc/
___ooO ~ Ooo___

LeRoy D. Cressy  /\_/\  [EMAIL PROTECTED]
Computer Consulting ( o.o ) Phone (215) 535-4037
 > ^ <  Fax   (215) 535-4285


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: sleep contains crypto stuff?

[Vincent Renardias]

On Wed, 8 Apr 1998, Martin Schulze wrote:

> could anybode explain this to me?
> 
> kuolema!root(ttyp0):/tmp/initrd# ldd /bin/sleep
> libcrypt.so.1 => /lib/libcrypt.so.1 (0x4000f000)
> libc.so.6 => /lib/libc.so.6 (0x4003c000)
> /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x4000)
> kuolema!root(ttyp0):/tmp/initrd# dpkg -l shellutils
> 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  shellutils  1.16-6 The GNU shell programming utilities.

That's weird since it's in fact linked with libcrypt, but doesn't seem to
use _any_ function/symbols from this lib:

thor# nm -D /bin/sleep
08049d1c A _DYNAMIC
08049cd4 A _GLOBAL_OFFSET_TABLE_
08049db8 B _IO_stderr_
08049db4 A __bss_start
 U __dcgettext
08049e0c B __environ
 U __gmon_start__
 U __libc_init_first
08049db4 A _edata
08049e4c A _end
08049e0c W _environ
08048ab8 A _etext
08048ac0 ? _fini
08048550 ? _init
08049e08 B _nl_msg_cat_cntr
08048670 T _start
 U atexit
 U bindtextdomain
08049e0c W environ
 U error
 U exit
 U fprintf
 U getopt_long
 U printf
 U puts
 U setlocale
 U sleep
 U textdomain
thor#

-- 
- Vincent RENARDIAS [EMAIL PROTECTED],pipo.com,debian.org} -
- Debian/GNU Linux:   Pipo:WAW:   -
- http://www.fr.debian.orghttp://www.pipo.com  http://www.waw.com -
---
- "La fonctionnalite Son Visuel vous delivre des avertissements visuels." -
-  [Message durant l'installation de Windows95]:wq 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: mailto:debian-devel@lists.debian.org

[Lalo Martins]

On Apr 07, [EMAIL PROTECTED] decided to present us with:
> hello i was wondering if anyone knew how to cross compile from 
> linux for dos/win95.  i wrote a lot of c code using the curses.h
> library and it does not port to borland or turbo c.  is there 
> some way i can tell gcc to compile it for a different 
> operating system??

You may use the djgpp compiler to compile your code on DOS, and
cywin32 to compile it for win95. We used to have a cigwyn32
cross-compiler package, dunno what happened to it. About djgpp,
I'm working on packaging a cross-compiler right now.

djgpp: http://www.delorie.com
cygwin32: http://www.cygnus.com somewhere (I don't go there
since 95) ;-)

[]s,
   |alo
   +
--
   Howling to the moonlight on a hot summer night...
http://www.webcom.com/lalo  mailto:[EMAIL PROTECTED]
 pgp key in the web page

Free Software Union   --   http://www.fslu.org
Debian GNU/Linux   --http://www.debian.org


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Intent to create sfio package

[Dennis L. Clark]

Hello everyone,

This is a short note to let people know that I'm in the process of
packaging the Safe/Fast I/O library, otherwise known as sfio. It
provides an alternate to stdio as a C API for performing I/O. It's main
claim to fame is the ability to attach filters -- known as disciplines --
to I/O streams within the program.

So far i've make the needed changes to it to compile it correctly under
glibc2, and most of the packaging has been done. Currently undergoing
fine-tuning is the packing of example code. Not available is the stdio
binary-compatibility library, as it depends on a lot of libc internals
that have changed in glibc2. This shouldn't concern anyone as I doubt it
would perform better than glibc2's stdio.

I had high hopes for sfio at one stage, particularly as it provided a
neat way of pre-processing and post-processing I/O, and that a Perl
linked against sfio can use the FCGI.pm module for FastCGI support (a
huge performance-booster for Perl/CGI), but now with over a year since
the last upstream release, the fact that the code is bloated with so
many static variables as to make thread-safety impossible without a
total re-write, and that FCGI.pm may soon support glibc2, it's no longer
as promising as before.

Still, someone asked for it, I shall provide :-)

Cheers,

-- 
   Dennis Clark   [EMAIL PROTECTED]   PGP/MIME spoken here
---
Today's entry from the Jargon File:

   dec /dek/ /v./  Verbal (and only rarely written) shorthand
  for decrement, i.e. `decrease by one'.  Especially used by
  assembly programmers, as many assembly languages have a `dec'
  mnemonic.  Antonym: {inc}.


pgpkqknwGRVm1.pgp
Description: PGP signature

Aargh - no mail

[Adrian Bridgett]

I've just lost my email file (/var/spool/mail/bridgett) for some unknown
reason.  My sorted mail in /home/bridgett/mail/ is fine. 

a) Warning - backup your mail file
b) has anyone else had this?

A Seriously PO Adrian

email: [EMAIL PROTECTED]   | Debian Linux - www.debian.org
http://www.poboxes.com/adrian.bridgett   | 2.0 release soon - over 1800
PGP key available on public key servers  | packages on a stable OS


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: ksymoops packaged ? System.map issues ...

[Yann Dirson]

Bernd Eckenfels writes:
 > If you have installed a current MAP File for the kernel (which is done
 > automatically by the kernel installer package), syslogd/klogd will be able
 > to read this and will display all Ptr references in lines from the kernel
 > with the resolved symbol.

It doesn't contain the *modules* symbols.

 > This is a good start for debugging, have u ever
 > looked into /var/log/messages?
 > 
 > Apr  6 21:19:28 lina syslogd 1.3-3#25: restart.
 > Apr  6 21:19:28 lina kernel: klogd 1.3-3, log source = /proc/kmsg started.
 > Apr  6 21:19:29 lina kernel: Cannot find map file.
 > Apr  6 21:19:29 lina kernel: Error seeking in /dev/kmem
 > Apr  6 21:19:29 lina kernel: Error adding kernel module table entry.
 > 
 > -> no support for translation, because the /System.map file was not found.

What kernel version do you use ?  Do you use kernel flavours ?

 > >From klogd(8)
 > 
 > #   As  a  convenience  klogd  will  attempt to resolve kernel
 > #   numeric addresses to their symbolic forms if a kernel sym-
 > #   bol  table is available at execution time.  A symbol table
 > #   may be specified by using the -k  switch  on  the  command
 > #   line.   If  a  symbol file is not explicitly specified the
 > #   following filenames will be tried:
 > #
 > #   /boot/System.map
 > #   /System.map
 > #   /usr/src/linux/System.map

The manpage is out of date.  It also looks for
"/boot/System.map-$(uname -r)".

-- 
Yann Dirson  <[EMAIL PROTECTED]>  | Stop making M$-Bill richer & richer,
alt-email: <[EMAIL PROTECTED]>  | support Debian GNU/Linux:
debian-email:   <[EMAIL PROTECTED]>  | more powerful, more stable !
http://www.a2points.com/homepage/3475232 | Check 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Automation of constitutional procedures

[Ian Jackson]

James A.Treacy writes ("Re: Constitution - formal proposal (v0.6)"):
> [Ian:]
> > I disapprove of the idea of automation.
> > 
> > s4.2(5):
> >   Proposals, seconds, amendments, calls for votes and other formal
> >   actions are made by announcement on a public-readable electronic
> >   mailing list designated by the Project Leader's Delegate(s); any
> >   developer may post there.
> > 
> > Are you proposing a change to this ?
...
> The only aspect that could possibly be interpreted as being in conflict with
> the constitution is that I was planning on having developers send proposals,
> seconds and amendments, etc directly to the system which would then make
> the announcement for them if everything was in order.

This is precisely the aspect I disagree with.  There are a number of
problems with this kind of thing.

Firstly, with an automated system developers' abilities to do things
will be dependent on the bot's interpretation of what is allowed - the
bot becomes the governor of the procedure rather than the Secretary or
the developers together.

Secondly, it is often necessary in procedures such as this to make
`conformal changes' to a document or an amendment to ensure the
consistency of the whole document.  Since a bot cannot tell what is
merely a conformal change and what is a substantive change this will
become impossible, and a human will have to step in anyway to ensure
that what gets voted on makes sense.

Thirdly, with a reasonably complicated procedure like the one in the
proposed constitution it will be necessary to make it as easy and
informal as possible for people to take actions like proposing and
seconding resolutions, amendments, &c.

Fourthly, I don't think developers should be required to learn how to
drive such a mechanism before being able to take part in the
decisionmaking process.

I may be able to think of more reasons later.

Ian.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Call for recommendations for the Technical Committee

[Ian Jackson - Debian Project Leader]

As I said, I want to appoint the Technical Committee.  However, I
would like to gather some input from the developers before I decide on
my preferred committee.

So: if you feel someone is technically excellent and likely otherwise
to be overlooked (for example, because they've not been active very
much during the times when I have, or in different threads/lists)
please mail me at the address above.

In support of such recommendations I'd be particularly interested in
reports of instances where the person was in dispute about some
technical matter with someone else.  A reference to the discussion
would be especially good.

Please do not discuss this on the list; I think that would be likely
to reopen old discussions and/or generate more heat than light.

Ian.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: source packaging format (was Re: Questions about plans for Emacs 20...)

[Lalo Martins]

On Apr 07, Falk Hueffner decided to present us with:
> On Sun, 5 Apr 1998 22:43:28 +0100, Enrique Zanardi <[EMAIL PROTECTED]>
> wrote:
> 
> >Well, it's not broken, but we have no way to use pristine sources with
> >multiple-tar packages, so it may be improved (and it should).
> 
> And there are *still* no source dependencies yet... Are there any
> plans about this in 2.x versions?

I would really appreciate some work on the source package
format, and I could lend a hand - see my problem with djgpp:

To compile the djgpp cross-compiler to DOS, you need to unpack
the gcc and binutils sources somewhere, then unpack the djcrx
stuff over it. Go figure the size of my .orig.tar.gz... OTOH, if
we made some minor changes to the dsc file, it could know it has
to unpack gcc-whatever.orig.tar.gz binutils-whatever.orig.tar.gz
and then djgpp-whatever.orig.tar.gz. Maybe we could even come up
with a way to apply the .diff's...

[]s,
   |alo
   +
--
   Howling to the moonlight on a hot summer night...
http://www.webcom.com/lalo  mailto:[EMAIL PROTECTED]
 pgp key in the web page

Free Software Union   --   http://www.fslu.org
Debian GNU/Linux   --http://www.debian.org


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Intent to package: Quinn Diff

[James Troup]

Hi,

I plan on packaging up Quinn Diff despite reservations I have, because
enough people have asked me to.  Quinn Diff is a program for comparing
the Packages files of two architectures to see what needs recompiled
on the secondary architecture.  See

http://thor.lib.chalmers.se/~jamest/quinn-diff/>

for more details.  My reservations are basically that the basic
functionality of Quinn Diff can be done very easily in
perl/python/whatever (Quinn Diff is written in C), as proved by at
least three other people.  However no one has extended their perl
versions as much as I have done with quinn diff (which says more about
my lack of life than anything else), or packaged theirs, so I'll
upload mine.  It'll be section devel, I guess, and priority extra
(it's main use is for people with exotic (i.e. non-i386) hardware,).

Any objections? 

-- 
James


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[Santiago Vila]

-BEGIN PGP SIGNED MESSAGE-

On Wed, 8 Apr 1998, Raul Miller wrote:

> Riku Voipio <[EMAIL PROTECTED]> wrote:
> > The point is new users. 
> 
> Then we should be talking about /etc/skel/, rather than /etc/profile

Well, if we talk about /etc/skel, then we could ask:
Is there any other shell which reads .bash_profile?

-BEGIN PGP SIGNATURE-
Version: 2.6.3ia
Charset: latin1

iQCVAgUBNSunAyqK7IlOjMLFAQG0FQP/aTSFQtO6/8d5ucgEXZt9uMiE31DdMF55
+9474zPRsBmQgb4LTNSj480KqYfExh+ckOhwVkRvJnCoT8ikxu6aJw3iyHqkz6c6
uStppxRiEpTpU/XHtNJhn28XzDkuX5VUUsSK4fKf7vzKlP7iC02I8fG6thqVovP1
H6PT+vXwt3w=
=+E8L
-END PGP SIGNATURE-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: New APT Version

[Juergen Menden]

Jason Gunthorpe <[EMAIL PROTECTED]> wrote:
> 
> deb http://ftp.de.debian.org/debian-non-US stable/binary-i386/
 ^
better use 
   nonus.debian.org

jjm

-- 
Juergen Menden 
at work: [EMAIL PROTECTED]  tel: +49 (89) 289 - 22387  
private: [EMAIL PROTECTED] tel: +49 (89) 89 712 743

Support the anti-Spam amendment.  Join at http://www.cauce.org/


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Project Secretary appointed

[Juergen Menden]

Manoj Srivastava <[EMAIL PROTECTED]> wrote:
> Hi,
> 
>   Does the maintainers file also contain the people on the qa
>  and testing teams and others who do not upload packages? 

no, and it does not contain the porters to the various 
non-intel architectures.

IIRC there is a developers database built by Christian Schwarz.
why not useing that?

jjm

-- 
Juergen Menden 
at work: [EMAIL PROTECTED]  tel: +49 (89) 289 - 22387  
private: [EMAIL PROTECTED] tel: +49 (89) 89 712 743

Support the anti-Spam amendment.  Join at http://www.cauce.org/


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[Alex Romosan]

>OK, I've finished building an xteddy Debian Package and I'll take over
>maintainance when finished the procedure to become an official Debian
>maintainer.  Did you plan to maintain XTeddy too or was your package only for
>private uses.  I'm not intended to catch your job ;-).

good job. my xteddy package was intended for my private use only. i've
been using it for many years and i couldn't imagine not having it on
my linux desktop, so i made a package myself. i keep meaning to apply
to become a maintainer, but i can never seem to find the time. i don't
care who maintains it, and if you want to do it, that's fine with me.
i think the last release was some three years ago, so it shouldn't be
that much work to maintain. i don't know if you looked in my area, but
there are some other packages i made for myself which people should
feel free to use/adopt. some of them have become part of the
distribution, some not.

--alex--

-- 
| I believe the moment is at hand when, by a paranoiac and active |
|  advance of the mind, it will be possible (simultaneously with  |
|  automatism and other passive states) to systematize confusion  |
|  and thus to help to discredit completely the world of reality. |


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: policy about Qmail's Maildir support?

[Jaldhar H. Vyas]

On Wed, 8 Apr 1998, Amos Shapira wrote:

> I was thinking more of things like IMAPD, qpopper, and the
> c-client library, which have patches for maildir support
> available for them, they just have to be integrated into the
> "official" debian package (e.g. it looks like the procmail support
> was taken through the same page which points to the patches
> I mentioned - http://qmail.org).

I've found the patch you mentioned however it says it will not work with
IMAP4.1 which is what we use.  On the IMAP list archives I saw a message
from someone who was developing something that will work.  He said (in
February) that the patch would be available "soon."  As soon as it is I
will add it in.

-- 
Jaldhar H. Vyas <[EMAIL PROTECTED]>





--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Debian Beowulf

[aw]

-BEGIN PGP SIGNED MESSAGE-

In article <[EMAIL PROTECTED]>,
Camm Maguire  <[EMAIL PROTECTED]> wrote:
>what hardware would be best for this project.  Is the Alpha
>distribution ready for prime-time?  What about SMP boards?  We'd like
>to get the most bang for our buck, but don't have the manpower to
>handle extensive software debugging/instability.  Also, if any
>knowledgeable reader could comment on any performance surprises which
>may be in store for us, that would be most helpful.  The Beowulf
>systems I've seen described so far claim to peak at around 1.2 GFLOPS
>for 16 machines, which is about 2 to 3 times faster than to 400 MFLOPS
>we get with the Power Challenge, but I'm sure this is task dependent
>and I'd like to get a feel for how this performance will degrade as
>the granularity of the parallelism decreases.

We use a Linux cluster here, too. It was upgraded recently and consists
now of 32 PII/300 machines (upgradable with a second PII). We will upgrade
to Debian soon (we are using Slackware in the moment). We do not use
Beowulf in the moment. A web page of the cluster project is at
http://bigbrother.ag1.mpi-sb.mpg.de/.

Please compare with my previous mail about 'Unattended package install'

cu

AW

-BEGIN PGP SIGNATURE-
Version: 2.6.3i
Charset: noconv

iQCVAwUBNSuX/8CaUV487z8dAQFzpAP/UzDKrhfTknB/IR3PeY8fZqG2Lvmj4FeW
dNRxPlAU71DuYavPTB2mzdwAmopGXnCx7WotuBmk1wJPMBRkKA6Ycm5ebceFuCVs
1bLDMOmeW8F6VMilxePKQrZQwLTKJKAjq89NAjTicxGIdp4AVacjnpVwoXQPh4uW
WHOB6tAkoQQ=
=8/v9
-END PGP SIGNATURE-


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[Raul Miller]

Riku Voipio <[EMAIL PROTECTED]> wrote:
> The point is new users. 

Then we should be talking about /etc/skel/, rather than /etc/profile

-- 
Raul


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: in.ftpd refused to show directory (addition)

[tibor simko]

> "andreas" == Andreas Tille <[EMAIL PROTECTED]> writes:

andreas> I installed wu-ftpd now, but all is the same :-(((

running libc6 hamm/frozen, right?  this is a already reported and
known bug in the `addftpuser' script (see e.g. #12236, #17110, etc).
the point is that the script copies libc5 libraries instead of libc6
ones to the home ftp location.  hence the simple manual workaround
goes like this:

  $ cd /home/ftp/lib
  $ ldd ../bin/ls
libc.so.6 => /lib/libc.so.6 (0x4000f000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x4000)
  $ cp /lib/ld-linux.so.2 .
  $ cp /lib/libc.so.6 .

and ls should work now

cheers
-- 
[EMAIL PROTECTED]





--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

hostname resolution for libc5-compat apps

[Thomas Gebhardt]

Hi,

when I start ddd (from ddd-smotif) on a remote hamm system it cannot
resolve the hostname given in the DISPLAY variable:

_X11TransSocketINETConnect: Can't get address for pcrz64 

If I set the DISPLAY to the IP address, it is ok.

Other X11 applications don't show this problem.

I argue that this problem is due to the fact that ddd is a libc5-compat
application.

My nsswitch.conf says:

hosts:  files dns 

Is this a bug or is it my fault?

Cheers, Thomas



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: lftp segfaults sometimes on filename completion

[Richard Braakman]

Remco Blaakmeer wrote:
> [...] I think it should at least have severity
> 'important', so I first want to be sure that this is a real bug.

It is always a bug if a program segfaults.

However, I don't think this one is 'important'.  I'd say the
distribution is better off with lftp than without, even if it has
this bug.

I recommend that you submit your mail as a bugreport for lftp.

Richard Braakman


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: lftp segfaults sometimes on filename completion

[jdassen]

On Wed, Apr 08, 1998 at 04:20:07PM +0200, Remco Blaakmeer wrote:
> I am using lftp 0.14.3.980328-1

Upgrade.

>From the new changelog:
lftp (0.14.3.980402-1) frozen unstable; urgency=low

  * New upstream release:
 * null dereference in rglob.cc fixed (very nasty bug! lftp was giving
   coredump by pressing twice the tab key)

HTH,
Ray
-- 
Cyberspace, a final frontier. These are the voyages of my messages, 
on a lightspeed mission to explore strange new systems and to boldly go
where no data has gone before. 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

lftp segfaults sometimes on filename completion

[Remco Blaakmeer]

Hi,

I am having problems with lftp's filename completion. When I press ,
the program segfaults in a somewhat predictable manner. Does anyone have
the same problem? If there are more people with this problem, should I
file a bug report against lftp? I think it should at least have severity
'important', so I first want to be sure that this is a real bug.

I am using lftp 0.14.3.980328-1

This is the output from 'ldd /usr/bin/lftp':
libreadline.so.2 => /lib/libreadline.so.2 (0x4001)
libncurses.so.3.4 => /lib/libncurses.so.3.4 (0x4003b000)
libresolv.so.2 => /lib/libresolv.so.2 (0x4008)
libnsl.so.1 => /lib/libnsl.so.1 (0x4008e000)
libc.so.6 => /lib/libc.so.6 (0x40094000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x4000)

These libraries come from these packages:
libreadlineg2 2.1-8
ncurses3.4 1.9.9g-8
libc6 2.0.7pre1-4

Below are some transcripts to illustrate what I mean ("" means I
pressed the TAB key):

1 [EMAIL PROTECTED]:~]$ lftp snt
lftp snt:~> ls
total 30
drwxr-xr-x   9 root root 1024 Dec 21 10:51 .
drwxr-xr-x   9 root root 1024 Dec 21 10:51 ..
d--x--x--x   2 root root 1024 Mar 31 16:28 bin
d--x--x--x   2 root root 1024 Nov 25 10:47 etc
drwxrwxrwx   3 ftpadm   ftpadm   1024 Apr  7 11:48 incoming
d--x--x--x   2 root root 1024 Jun  8  1997 lib
drwxr-x---   4 ftpadm   local1024 Feb 12 20:34 local
drwxr-xr-x   2 root root12288 Jul  4  1997 lost+found
drwxrwxr-x  10 root ftpadm   1024 Apr  8 03:30 pub
-rw-r--r--   1 ftpadm   ftpadm784 Dec 18 12:30 welcome.msg
-rw-r--r--   1 ftpadm   ftpadm   9022 Apr  8 04:05 xferstat.txt
lftp snt:~> cd aSegmentation fault (core dumped)

2 [EMAIL PROTECTED]:~]$ lftp snt
lftp snt:~> cd pub/
cd ok, cwd=/pub
lftp snt:/pub> ls
total 1759
drwxrwxr-x  10 root ftpadm   1024 Apr  8 03:30 .
drwxr-xr-x   9 root root 1024 Dec 21 10:51 ..
drwxrwxr-x  12 florian  ftpadm   1024 Dec 14 16:19 dos
drwxrwxr-x  14 ftpadm   ftpadm   1024 Apr  2 21:52 linux
-rw-r--r--   1 ftpadm   ftpadm1716862 Apr  8 03:31 ls-lR
drwxrwxr-x  31 x10  ftpadm   1024 Dec 14 16:20 macintosh
-rw-rw-r--   1 mark ftpadm560 Mar 31 16:36 mirrors.txt
drwxrwxr-x   6 ftpadm   ftpadm   1024 Mar 31 23:11 misc
-rw-rw-r--   1 1119 ftpadm   6707 Apr  7 12:58 newfiles
-rw-rw-r--   1 ftpadm   ftpadm   8165 Aug 29  1996 newfiles.1995
-rw-rw-r--   1 ftpadm   ftpadm  36575 Nov  5 19:01 newfiles.1996
-rw-rw-r--   1 ftpadm   ftpadm   9422 Jan  1 17:21 newfiles.1997
drwxrwxr-x   6 ftpadm   ftpadm   1024 Dec 14 16:19 nt4
drwxrwxr-x   9 fornerod ftpadm   1024 Dec  6  1996 os2
-rw-rw-r--   1 ftpadm   ftpadm792 Dec 21 16:58 uploads.txt
drwxrwxr-x  13 ftpadm   ftpadm   1024 Mar 26 23:21 win3
drwxrwxr-x  14 ftpadm   ftpadm   1024 Mar 15 23:12 win95
lftp snt:/pub> cd win9Segmentation fault (core dumped)

3 [EMAIL PROTECTED]:~]$ lftp snt
lftp snt:~> cd pub/linux/debSegmentation fault (core
dumped)

4 [EMAIL PROTECTED]:~]$ lftp snt
lftp snt:~> ls pSegmentation fault (core dumped)

5 [EMAIL PROTECTED]:~]$ lftp snt
lftp snt:~> cd pub/
cd ok, cwd=/pub
lftp snt:/pub> ls linux 
total 17
drwxrwxr-x  14 ftpadm   ftpadm   1024 Apr  2 21:52 .
drwxrwxr-x  10 root ftpadm   1024 Apr  8 03:30 ..
-rw-rw-r--   1 ftpadm   ftpadm433 Jan 14 16:30 00index.txt
drwxrwxr-x  14 ftpadm   ftpadm   1024 Mar 29 02:11 Slackware
drwxrwxr-x   3 ftpadm   ftpadm   1024 Apr  2 21:57 X
drwxrwxr-x   2 ftpadm   ftpadm   1024 Feb 11  1997 anti4us
drwxrwxr-x  12 ftpadm   ftpadm   1024 Apr  8 02:02 debian
drwxrwxr-x   2 1117 ftpadm   1024 Feb  4 16:49 devel
-rw-rw-r--   1 ftpadm   ftpadm   1187 Oct 16  1996 ftp-sources
drwxrwxr-x   5 ftpadm   ftpadm   1024 Dec 14 00:51 kernels
drwxrwxr-x   2 ftpadm   ftpadm   1024 Apr  2 22:00 network
drwxrwxr-x   2 1117 ftpadm   1024 Apr  2 21:52 programs
drwxrwxr-x   5 ftpadm   ftpadm   1024 Apr  8 02:09 redhat
drwxrwxr-x   5 ftpadm   ftpadm   1024 Feb 23 03:35 redhat-contrib
drwxrwxr-x   2 ftpadm   ftpadm   1024 Apr  2 22:00 samba
lrwxrwxrwx   1 root root   10 Jul  4  1997 slackware -> Slackware/
drwxrwxr-x   5 ftpadm   ftpadm   1024 Oct 16  1996 www


What happens is this:

1. In the first case, there is nothing that starts with the letter 'a'. I
press  twice and get a segfault.

2. In the second case, there is one thing that starts with a 'p' and it
gets completed normally. In /pub, there are two things that start with
'wi', win3 and win95. I type 'wi', press , it becomes 'win', I type
'9' and press  again and then I get the segfault.

3. In this case, I type 'plidb' and then get the
segfault. In /pub/linux there are two things that start with 'd', 'debian'
and 'devel'. So, when I have '/pub/linux/de', I type 'b' and then get
a segfault.

4. Now I am using another command. I type 'ls p' and get th

Re: [?] egcs increases C++ binary size dramatically

[Dale Martin]

[EMAIL PROTECTED] (Falk Hueffner) writes:

> It seems that programs are larger even if they do not use exceptions
> at all (possibly even C programs). For those, it seems totally
> resonable to disable exceptions. It should probably even added to the
> policy, since it saves space.
> 
>   Falk

That would make it unsafe to link code that has exceptions to libraries
that don't support compiled in.  (There is code added to be able to safely
unwind the stack, even if the library in question doesn't have any
exception code in it.)

Later,
Dale
-- 
+  finger for pgp public key  -+
| Dale E. Martin | University of Cincinnati Savant Research Laboratory |
| [EMAIL PROTECTED]| http://www.ececs.uc.edu/~dmartin   |
+--+


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[jdassen]

On Wed, Apr 08, 1998 at 02:18:07PM +0200, Andreas Tille wrote:
> But who is xcaptainblueeye???

If I remember correctly, "Captain Blueeye" is the name of the entity
depicted in the Debian logo.

Come to think of it, xtux might be a better name than xpenguin (as the Linux
Penguin is called "Tux").

Ray
-- 
UNFAIR  Term applied to advantages enjoyed by other people which we tried 
to cheat them out of and didn't manage. See also DISHONESTY, SNEAKY, 
UNDERHAND and JUST LUCKY I GUESS. 
- The Hipcrime Vocab by Chad C. Mulligan  


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[Andreas Tille]

On Wed, 8 Apr 1998 [EMAIL PROTECTED] wrote:

> Well, you could make it a coding exercise by modifying it to be able to
> choose a pixmap through the commandline. "xpenguin" or "xcaptainblueeye" :-)
Adding xpenguin is a very nice idea.  I will try to include it into
xteddy.deb.  But who is xcaptainblueeye???

Regards

Andreas. 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[jdassen]

On Tue, Apr 07, 1998 at 07:45:23PM +0200, Andreas Tille wrote:
>Xteddy is a cuddly teddy bear for your X Windows desktop. It is more or
>less an excersise for package bundling and maintaining.

Well, you could make it a coding exercise by modifying it to be able to
choose a pixmap through the commandline. "xpenguin" or "xcaptainblueeye" :-)

Ray
-- 
Obsig: developing a new sig


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Amos Shapira]

On Wed, Apr 08, 1998 at 01:01:46AM +0200, Marcus Brinkmann wrote:
> However, you can give the compiler a hint that a function does not throw any
> exceptions by adding throw() at the right place:
> 
> class ABC {
>   ABC (int theInt) throw();
> }

Shouldn't the compiler still handle eceptions in functions which call
functions which throw exceptions?

If so, this probably means that you should have exceptions handled in
any binary which uses a function which throws exceptions (e.g. almost
any STL container).

Cheers,

--Amos
-- 
--Amos Shapira  | "Of course Australia was marked for
|  glory, for its people had been chosen
[EMAIL PROTECTED] |  by the finest judges in England."
| -- Anonymous


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: package pre-selections tool

[Charles Briscoe-Smith]

In article <[EMAIL PROTECTED]>,
Steve Dunham <[EMAIL PROTECTED]> wrote:
[...]
>Finally, it installs all the selected packages without asking the user
>20 times in a row if they prefer XV (or which dictionary they prefer).
[...]
>(The dictionary issue isn't as important since fewer people install
>English, German and French dictionaries.)

(But even if you only install one dictionary, it'll still ask you
a question.)

The other wordlist package maintainers and I have been discussing this for
a while now.  We're going to move over to using update-alternatives for
Debian 2.1 -- it's really too far into the freeze now to do it for 2.0.

For now, all the wordlist packages are being updated to use some
slightly more intelligent scripts.  When people upgrade bo -> hamm, they
will get asked which dictionary they want to be the default, but after
that, all further upgrades should be questionless.

If you want to have a script install these packages without questions,
(let's say you want to make wenglish the default dictionary) first
make the symlink /etc/dictionary -> /usr/dict/english, then install
the package.  Just make sure you unpack the package containing the
file /etc/dictionary points to before any of the other dictionaries get
configured, though, otherwise the 'broken' symlink will be moved out of
the way!

Hope this helps,

-- 
Charles Briscoe-Smith
White pages entry, with PGP key: http://alethea.ukc.ac.uk/wp?95cpb4>
PGP public keyprint: 74 68 AB 2E 1C 60 22 94  B8 21 2D 01 DE 66 13 E2


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[Riku Voipio]

On Tue, Apr 07, 1998 at 08:39:40PM -0700, Guy Maor wrote:
> [EMAIL PROTECTED] (Gregory S. Stark) writes:

> > Am I the only one who thinks the only correct prompts would be '$ ' and '# 
> > '?
> > Barring that I suggest leaving the defaults, 'bash$' et. al.
 
> You're not the only one.  I also prefer to leave the defaults.
 
> Any prompt in /etc/profile would be overridden by 99% of the users
> anyway so what's the point?

The point is new users. I can't imagine how it cuold hurt anyone to
see where they are. It makes working a million times easier, if you
can see all the time where you are. The first thing I allways do when
installing debian to a friend is setting the the prompt in /etc/profile.
Why should specifically make it hard for new users to get accustemed
with Debian?

Riku


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: 19980407 Work-Needing and Prospective Packages

[Christoph Martin]

[EMAIL PROTECTED] writes:
 >
 >Packages someone could package for Debian 2.1
 >  * A pdftex package, for a version of TeX/LaTeX which directly
 >generates PDF.
 >

pdftex is a part of tetex-0.9 which just made its way to master.

Christoph


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: in.ftpd refused to show directory

[Joop]

> Hello,
> 
> yesterday I updated all my installed hamm files.  However, the following
> error occured in former times, too.
> 
> If I login on my local the ftp server
> 
> ftp://bridge.physik.uni-halle.de
> 
> I can't get any directory information via `dir` or `ls` command.
> I can switch directories (`cd pub` for instance) and check the
> currend directorie via `pwd` also get files via `get`.  But I
> can't get any information about the directory contents.
> By the way: Midnight Commander shows the same behaviour when
> accessing this server via `cd ftp://bridge.physik.uni-halle.de`.
> The window remains blank, but you can fairly switch directories
> via cd.
> 
> I want to use only in.ftpd from
> 
> Package: netstd
> Status: install ok installed
> Priority: standard
> Section: net
> Installed-Size: 1299
> Maintainer: Peter Tobias <[EMAIL PROTECTED]>
> Version: 3.03-1
> Depends: cpp, libc6, libreadlineg2 (>= 2.1-4), ncurses3.4
> Pre-Depends: netbase (>= 3.00)
> Suggests: wu-ftpd, dip, mail-transport-agent
> 
> because it is only 486/20MB box and a featureful server like wu-ftpd
> isn't required.  I will temporarily switch to wu-ftpd, to get rid of
> this boring error (hopefully this will work) but want to use the
> smallest beast which would fit the need of getting some files.
> 
> Any hints?
> 


Yep, have a look in /home/ftp/bin. Is the program
called 'ls' there? If yes, is it a dynamically 
linked version (noticable by a small-sized ls) ?

If it is a dynamically linked version, you have 
to put associated libraries in /home/ftp/lib.
Libs can be found with 'ldd ls'. Normally, this would
be libc.so.x and ld.so.x I think.

Check out that you actually have the libs that
ls needs, so it is no use to have ls use libc5,
if it is a libc6 executable...

You are better of finding a statically linked ls
or compile one yourself
and put it in ~bin. Then you do not need all these
libraries...

> Regards
> 
>  Andreas
> 
> 
> --
> To UNSUBSCRIBE, email to [EMAIL PROTECTED]
> with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> 
> 

Joop

[EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

in.ftpd refused to show directory (addition)

[Andreas Tille]

I installed wu-ftpd now, but all is the same :-(((

I want to make some files available (my prepackaged Debian files)
but I'm affraid that potential downloaders will be confused by
this behaviour.

Please help

  Andreas.



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

in.ftpd refused to show directory

[Andreas Tille]

Hello,

yesterday I updated all my installed hamm files.  However, the following
error occured in former times, too.

If I login on my local the ftp server

ftp://bridge.physik.uni-halle.de

I can't get any directory information via `dir` or `ls` command.
I can switch directories (`cd pub` for instance) and check the
currend directorie via `pwd` also get files via `get`.  But I
can't get any information about the directory contents.
By the way: Midnight Commander shows the same behaviour when
accessing this server via `cd ftp://bridge.physik.uni-halle.de`.
The window remains blank, but you can fairly switch directories
via cd.

I want to use only in.ftpd from

Package: netstd
Status: install ok installed
Priority: standard
Section: net
Installed-Size: 1299
Maintainer: Peter Tobias <[EMAIL PROTECTED]>
Version: 3.03-1
Depends: cpp, libc6, libreadlineg2 (>= 2.1-4), ncurses3.4
Pre-Depends: netbase (>= 3.00)
Suggests: wu-ftpd, dip, mail-transport-agent

because it is only 486/20MB box and a featureful server like wu-ftpd
isn't required.  I will temporarily switch to wu-ftpd, to get rid of
this boring error (hopefully this will work) but want to use the
smallest beast which would fit the need of getting some files.

Any hints?

Regards

 Andreas


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Processed: foo

[Ian Jackson]

Processing commands for [EMAIL PROTECTED]:

> reassign 20739 xbase
Bug#20739: xbase: Buffer overruns in xterm
Bug assigned to package `xbase'.

> reassign 20740 xbase
Bug#20740: xbase: TIOCSLTC in xterm on alpha
Bug assigned to package `xbase'.

> reassign 20741 xserver-vga16
Bug#20741: xserver-vga16: XF86Setup blows away existing /etc/X11/XF86Config
Bug assigned to package `xserver-vga16'.

> reassign 20742 xserver-vga16
Bug#20742: xservers: Kensington Thinking Mouse and Expert Mouse support
Bug assigned to package `xserver-vga16'.

> close 20793
Bug#20793: Low Cost Sales SKILL Testing Available for Small Business
Bug closed, ack sent to submitter - they'd better know why !

> close 20823
Bug#20823: hi
Bug closed, ack sent to submitter - they'd better know why !

> reassign 20734 general
Bug#20734: autoup.sh
Bug assigned to package `general'.

> thanks
Stopping processing here.

Please contact me if you need assistance.

Ian Jackson
(administrator, Debian bugs database)


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: policy about Qmail's Maildir support?

[Amos Shapira]

[EMAIL PROTECTED] wrote:
> 
> On Wed, Apr 08, 1998 at 09:43:08AM +0300, Amos Shapira wrote:
> > Looking for support for Qmail's maildir format in Debian packages, I came
> > up with empty hands.
> 
> Look more closely. Mutt handles maildir.

Thanks.

> > Would it be possible to add this to the debian policy to have Maildir
> > support in packages like mailx, pine and imapd?
> 
> AFAIK there is no DFSG-free MTA that supports Maildir. Therefore I don't
> think that Maildir support should be obligatory via Debian policy.
> 
> I think we should encourage maintainers to add Maildir support to
> mail-related packages, but this should be a suggestion, not an obligation
> under Debian policy.

Fine with me.
 
> > Is there another mechanism to make debian support this format?
> 
> I vaguely recall that procmail has some support for it; if that's the case,
> I think it should be fairly easy to have a "formail" script to convert to
> mbox or mh format; that could serve as a temporary form of support.

I was thinking more of things like IMAPD, qpopper, and the
c-client library, which have patches for maildir support
available for them, they just have to be integrated into the
"official" debian package (e.g. it looks like the procmail support
was taken through the same page which points to the patches
I mentioned - http://qmail.org).

Thanks,

--Amos
-- 
--Amos Shapira | "Of course Australia was marked for
   |  glory, for its people had been chosen
[EMAIL PROTECTED]|  by the finest judges in England."
   |   -- Anonymous


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: policy about Qmail's Maildir support?

[jdassen]

On Thu, Apr 09, 1998 at 01:47:53AM -0600, Jason Gunthorpe wrote:
> On Wed, 8 Apr 1998 [EMAIL PROTECTED] wrote:
> > AFAIK there is no DFSG-free MTA that supports Maildir. Therefore I don't
> > think that Maildir support should be obligatory via Debian policy.
> 
> Exim does.

In that case I'm in favour of making an addition to Debian policy about
Maildir along the lines of "if a package has support for Maildir, this
should be enabled".

> > I vaguely recall that procmail has some support for it; if that's the
> > case, I think it should be fairly easy to have a "formail" script to
> > convert to mbox or mh format; that could serve as a temporary form of
> > support.
> 
> I think debian's procmail package lacks the maildir patch :<

I didn't know it was a patch; if you know the location, please file a bug
about this (wishlist for now, until policy covers Maildir support).

Ray
-- 
PATRIOTISM  A great British writer once said that if he had to choose 
between betraying his country and betraying a friend he hoped he would
have the decency to betray his country.  
- The Hipcrime Vocab by Chad C. Mulligan 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

HEALTH WARNING: bash_2.01.1-1 dies on install

[Philip Hands]

Hi folks,

This has been reported as a bug (#20572), but since it passed me by, I thought 
I'd mention it here and save people some aggravation. 

bash_2.01.1-1 is lacking a pre-depends on libreadlineg2_2.1-8, and will start 
segfaulting if you just let dselect do an upgrade, which kills all subsequent 
package installs.

The fix is to upgrade to libreadlineg2_2.1-8 by hand before running dselect.

Cheers, Phil.




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: does not support iso9660

[Paul Slootman]

On Tue 07 Apr 1998, [EMAIL PROTECTED] wrote:
> 
>   After I booted  Linux, I ran dselect but I got an error
> during mounting a CD:
>   mount -t iso9660 /dev/sda /cdrom

SCSI cdrom devices are /dev/scd0, /dev/scd1, ...
Also -oro (for readonly) should be given, else you might get a warning.

Please, in the future, use debian-user for this type of question which
doesn't have anything to do with the development of debian. Thanks.


Paul Slootman
-- 
home: [EMAIL PROTECTED] | work: [EMAIL PROTECTED]
http://www.wurtel.demon.nl | Murphy Software, Enschede, the Netherlands
Support Randal Schwartz!
See http://www.lightlink.com/fors/ or send empty email to [EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: policy about Qmail's Maildir support?

[Jason Gunthorpe]

On Wed, 8 Apr 1998 [EMAIL PROTECTED] wrote:

> > Would it be possible to add this to the debian policy to have Maildir
> > support in packages like mailx, pine and imapd?
> 
> AFAIK there is no DFSG-free MTA that supports Maildir. Therefore I don't
> think that Maildir support should be obligatory via Debian policy.

Exim does.
 
> I vaguely recall that procmail has some support for it; if that's the case,
> I think it should be fairly easy to have a "formail" script to convert to
> mbox or mh format; that could serve as a temporary form of support.

I think debian's procmail package lacks the maildir patch :<

Jason


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: policy about Qmail's Maildir support?

[jdassen]

On Wed, Apr 08, 1998 at 09:43:08AM +0300, Amos Shapira wrote:
> Looking for support for Qmail's maildir format in Debian packages, I came
> up with empty hands.

Look more closely. Mutt handles maildir.

> Would it be possible to add this to the debian policy to have Maildir
> support in packages like mailx, pine and imapd?

AFAIK there is no DFSG-free MTA that supports Maildir. Therefore I don't
think that Maildir support should be obligatory via Debian policy.

I think we should encourage maintainers to add Maildir support to
mail-related packages, but this should be a suggestion, not an obligation
under Debian policy.

> Is there another mechanism to make debian support this format?

I vaguely recall that procmail has some support for it; if that's the case,
I think it should be fairly easy to have a "formail" script to convert to
mbox or mh format; that could serve as a temporary form of support.

Ray
-- 
ART  A friend of mine in Tulsa, Okla., when I was about eleven years old. 
I'd be interested to hear from him. There are so many pseudos around taking 
his name in vain. 
- The Hipcrime Vocab by Chad C. Mulligan 


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[Andreas Tille]

On Tue, 7 Apr 1998, Bob Hilliard wrote:

>  I have been working on this, and expect to release this package
> soon.  (Actually, it is two binary packages from one source.)
> 
>  wordnet (which you mentioned in your message) is one of the 7
> dictionary databases made available by the DICT group in a format
> suitable for use by the dictd client/server software.  I have
> downloaded the wordnet source package from project/orphaned, but
> haven't had a chance to look at it.  I don't know if that package
> includes any more functionality than the dict package.
I don't know anything about the DICT group and their dict package.
All I know is that wordnet 1.5 is fairly outdated and you should
download 1.6 from the wordnet ftp-server.  If you are working on
a Debian wordnet package I would gladly support you with the automake
stuff I wrote to compile wordnet easily.  If not I will spend my effort
on building the new upstream release in the same way the old one looked
like.  Wordnet 1.6 comes with an advanced tcl/tk GUI which is easy to
use and works much better than the old interface.

Regards

 Andreas



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Questions for maintaining several packages

[Andreas Tille]

On Tue, 7 Apr 1998, Alex Romosan wrote:

> you can get an xteddy debian package by anonymous ftp from
> caliban.lbl.gov in /pub/debian. i am happy to see this finally become
> part of the distribution.
OK, I've finished building an xteddy Debian Package and I'll take over
maintainance when finished the procedure to become an official Debian
maintainer.  Did you plan to maintain XTeddy too or was your package only for
private uses.  I'm not intended to catch your job ;-).

Regards

 Andreas.



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

policy about Qmail's Maildir support?

[Amos Shapira]

Hello,

Looking for support for Qmail's maildir format in Debian packages,
I came up with empty hands.

Would it be possible to add this to the debian policy to have
Maildir support in packages like mailx, pine and imapd?  Is there
another mechanism to make debian support this format?

Thanks,

--Amos

-- 
--Amos Shapira | "Of course Australia was marked for
   |  glory, for its people had been chosen
[EMAIL PROTECTED]|  by the finest judges in England."
   |   -- Anonymous


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why isn't /bin/sh managed with alternatives?

[Adam Heath]

On Tue, 7 Apr 1998, Robert Woodcock wrote:

> Another idea I got on IRC was providing a --background flag to
> start-stop-daemon so that daemons could be started in parallel - this
> might have quite an effect on SMP systems, and DNS misconfigs would be
> more treatable if sendmail started in the background instead of waiting a 
> few minutes timing out on stuff before anything else could run.

Already done.  Here is the wrapper.  Rename start-stop-daemon to
start-stop-daemon~.

Adam

===
#!/bin/sh
back=0
msg=1
rest=" "
while [ $# -gt 0 ];do
[ "$1" = "-B" -o "$1" = "--background" ] && back=1 && shift && continue
[ "$1" = "-N" -o "$1" = "--nobackmsg" ] && msg=0 && shift && continue
opt=$1;shift
rest="$rest $opt"
case $opt in
--) rest="$rest $*";shift $#;;
esac
done
if [ $back = 1 ];then
[ $msg = 1 ] && echo -n "(background)" 1>&2
bk="&"
fi

eval /sbin/start-stop-daemon~ $rest $bk
===



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

does not support iso9660

[omill]

I need some advice in installing Debian 1.3.
I have booted the machine with a SCSI bus V30PCI using a rescue
floppy. The hard disk and a NEC cdrom were recognized without any problem.
Then I followed all installation instructions.
After I booted  Linux, I ran dselect but I got an error
during mounting a CD:
mount -t iso9660 /dev/sda /cdrom
Unable to identify a CD rom format.
 mount : wrong fs type, bad option, bad superblock in
/dev/sda or too many mounted files
I tried to mount a CD during a boot time, and I got an error
message saying that a CD Rom file system is not supported by a kernel.
I have tried 2 different CDs to make sure that CDs I am using are not
damaged.
So, how do I make it to recognize iso9660 if I have a bare
file system installed on Linux with nothing else ?

One thing I did not configure -- was a CD during an installation because
there was not any NEC CD ROM on a list, and I did not choose one presuming
that
it has been already recognized and necessary drivers installed. I did not
find any iso9660 option among file systems available during install.
sincerely,
Olga Mill
[EMAIL PROTECTED]

___
Olga Mill
^^^ * ^^^ [EMAIL PROTECTED] ^^^ * ^^^

Eastern Washington University,  WA,  USA

__



--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: was default prompt debate - system defaults

[David Welton]

On Tue, Apr 07, 1998 at 10:24:42PM -0500, Manoj Srivastava wrote:
> Hi,
> >>"David" == David Welton <[EMAIL PROTECTED]> writes:
> 
> David> If we won't even set a default prompt, what business do we have
> David> doing things like: 
> [Setting up IP spoofing protection...]
> 
>   Well, maybe we are closer to achieving a consensus about what
>  one should do wrt ip spoofing, and one does not seem to come to an
>  consensus about prompts.

Good point.  I still feel that it is missleading, or at the very
least, not detailed enough.

The script tells us:
echo "Setting up IP spoofing protection."
but what it does is:
# deny incoming packets pretending to be from 127.0.0.1
and
# deny incoming packets pretending to be from our own system.
but only if you uncomment it.

I think that we should at the very least be a bit more descriptive of
what we are doing:

if [ -e /proc/ip_input ]
 then 
  echo "Denying incoming packets with spoofed address 127.0.0.1"
fi

Especially since many people will still recompile their kernels and
possibly not realize that this feature has been disabled.  I think a
phrase such as the above is a bit more honest with hour users.  To
many new users who have heard of the advanced and stable networking of
linux, 'spoofing protection' might mean any number of things.  I think
we should be clear about this.

Incidentally, if we are decided to put this sort of thing in, it might
not be a bad idea to set up filters against spoofed packets going
*out* of the computer, to thwart attempts by people who have installed
linux as a quick way to launch an arsenal of nasties against other
people on the net.

I'd prefer to just see the whole thing commented out though...

Ciao,
-- 
David Welton  http://www.efn.org/~davidw 

Debian GNU/Linux - www.debian.org


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[Guy Maor]

[EMAIL PROTECTED] (Gregory S. Stark) writes:

> Am I the only one who thinks the only correct prompts would be '$ ' and '# '?
> 
> Barring that I suggest leaving the defaults, 'bash$' et. al.

You're not the only one.  I also prefer to leave the defaults.

Any prompt in /etc/profile would be overridden by 99% of the users
anyway so what's the point?


Guy


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: http mirror (was Re: New APT Version)

[Anand Kumria]

On Tue, 7 Apr 1998, Steve Hsieh wrote:

> On 5 Apr 1998, Jason Gunthorpe wrote:
> 
> > > Oh I'm all for switching to HTTP. Can we convince all our mirrors to
> > > switch?
> >
> > I'm going through the mirror list and building a sources.list of all the
> > possible sources. I have 8 sites already
> 
> 
> I have enabled http transfers on our mirror as well...
>http://linux.eecs.umich.edu/pub/linux/debian
> 

As have I ...

http://ftp.progsoc.uts.edu.au/pub/Linux/debian-non-US/

Anand.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: was default prompt debate - system defaults

[Manoj Srivastava]

Hi,
>>"David" == David Welton <[EMAIL PROTECTED]> writes:

David> If we won't even set a default prompt, what business do we have
David> doing things like: 
[Setting up IP spoofing protection...]

Well, maybe we are closer to achieving a consensus about what
 one should do wrt ip spoofing, and one does not seem to come to an
 consensus about prompts.

I like different prompts for xterms and the onsole, as I use
 color and title in Xterms ;-)

manoj
-- 
 "Tell the truth and run." Yugoslav proverb
Manoj Srivastava  <[EMAIL PROTECTED]> 
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05  CC 2D 27 12 1D F5 E8 6E


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Is my hard drive FUBAR, or is it dpkg?

[Martin Schulze]

Ben Gertzfield writes:
> I've started getting messages like the following from upgrading
> packages:
> 
> Do you want to install the files fetched [y]:
> Installing files...
> (Reading database ... 28597 files and directories currently installed.)
> Preparing to replace xbase 3.3.2-2 (using .../x11/xbase_3.3.2-3.deb) ...
> xdm and xfs not running, continuing with xbase installation.
> Unpacking replacement xbase ...
> dpkg: error processing 
> debian/dists/frozen/main/binary-i386/x11/xbase_3.3.2-3.deb (--install):
>  error flushing `usr/X11R6/lib/X11/x11perfcomp/fillblnk': No space left on 
> device
> dpkg-deb: subprocess paste killed by signal (Broken pipe)
> 
> but the /usr partition has over 400M free:
> 
> [EMAIL PROTECTED]:/usr/X11R6/lib/X11/x11perfcomp]# df .
> 7:09PM
> Filesystem 1024-blocks  Used Available Capacity Mounted on
> /dev/hda21486371  931336   478226 66%   /usr

What about inodes?  df -i

Regards,

Joey

-- 
  / Martin Schulze  *  [EMAIL PROTECTED]  *  26129 Oldenburg /
 / http://home.pages.de/~joey/
/ The only stupid question is the unasked one   /


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why isn't /bin/sh managed with alternatives?

[James LewisMoss]

> On Tue, 7 Apr 1998 18:18:57 -0700 (PDT), Robert Woodcock <[EMAIL 
> PROTECTED]> said:

 Robert> On Sun, April 5, Shaleh <[EMAIL PROTECTED]> wrote:
 >> My assumption is that /bin/sh is VERY important to the system.
 >> All server scripts use it.  A dangling symlink could be hazardous.
 >> Also, some systems initially mount only certain directories.  And
 >> /etc is sometimes on a different partition entirely.  /bin should
 >> stand on its own.

 Robert> *NO* dynamically linked binary (this includes /sbin/init and
 Robert> /bin/sh) will run without /etc/ld.so.cache:

This is untrue.  if no ld.so.cache file exists the dynamic linker will 
default to manually searching /lib and /usr/lib (in fact it does this
if the ld.so.cache doesn't contain the library that is needed).

> strace ls
open("/etc/ld.so.preload", O_RDONLY)= -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY)  = -1 ENOENT (No such file or directory)
open("/lib/libc.so.6", O_RDONLY)= 3

Dres
-- 
@James LewisMoss <[EMAIL PROTECTED]> |  Blessed Be!
@http://www.dimensional.com/~dres   |  Linux is cool!
@"Argue for your limitations and sure enough, they're yours." Bach


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Is my hard drive FUBAR, or is it dpkg?

[Ben Pfaff]

   I noticed in /var/log/kernel.log that there are lots of messages
   like the following:

   Free blocks count corrupted for block group 17
   Apr  7 18:57:49 everybody kernel: EXT2-fs error (device 03:02): 
ext2_new_block:
   Free blocks count corrupted for block group 17
   Apr  7 18:58:22 everybody kernel: EXT2-fs error (device 03:02): 
ext2_new_block:
   Free blocks count corrupted for block group 17
   Apr  7 19:06:10 everybody kernel: EXT2-fs error (device 03:02): 
ext2_new_block:
   Free blocks count corrupted for block group 17
   Apr  7 19:06:34 everybody last message repeated 4 times
   Apr  7 19:08:27 everybody kernel: EXT2-fs error (device 03:02): 
ext2_new_block:
   Free blocks count corrupted for block group 17

   which seems to suggest SOMETHING horrible is wrong.

You need to run e2fsck.  Do `touch /forcefsck' as root, then reboot.
After e2fsck gets run you can remove /forcefsck.

It is possible that there is a bad block, as well.  You might want to
run `badblocks' on the drive.  If it is an IDE or modern SCSI drive,
then this may mean that the drive is dying, as they normally remap bad
blocks onto unused good blocks reserved for that purpose.  If more bad
blocks start appearing, then you'd better back up *soon*, as you've
probably got less than 24 hours to total FUBAR mode.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Is my hard drive FUBAR, or is it dpkg?

[Ben Gertzfield]

I've started getting messages like the following from upgrading
packages:

Do you want to install the files fetched [y]:
Installing files...
(Reading database ... 28597 files and directories currently installed.)
Preparing to replace xbase 3.3.2-2 (using .../x11/xbase_3.3.2-3.deb) ...
xdm and xfs not running, continuing with xbase installation.
Unpacking replacement xbase ...
dpkg: error processing 
debian/dists/frozen/main/binary-i386/x11/xbase_3.3.2-3.deb (--install):
 error flushing `usr/X11R6/lib/X11/x11perfcomp/fillblnk': No space left on 
device
dpkg-deb: subprocess paste killed by signal (Broken pipe)

but the /usr partition has over 400M free:

[EMAIL PROTECTED]:/usr/X11R6/lib/X11/x11perfcomp]# df .
7:09PM
Filesystem 1024-blocks  Used Available Capacity Mounted on
/dev/hda21486371  931336   478226 66%   /usr

and the file doesn't appear to be corrupted:

[EMAIL PROTECTED]:/usr/X11R6/lib/X11/x11perfcomp]# ls -la .
7:09PM
total 5
drwxr-xr-x   2 root root 1024 Apr  7 19:08 .
drwxr-xr-x  21 root root 1024 Apr  7 19:08 ..
-rwxr-xr-x   1 root root  424 Mar 25 20:58 fillblnk
-rwxr-xr-x   1 root root  784 Mar 25 20:58 perfboth
-rwxr-xr-x   1 root root  723 Mar 25 20:58 perfratio

so what's going on? This is on a 2.0.33 machine. Should I downgrade
the kernel? Is the hard drive dying?

I noticed in /var/log/kernel.log that there are lots of messages
like the following:

Free blocks count corrupted for block group 17
Apr  7 18:57:49 everybody kernel: EXT2-fs error (device 03:02): ext2_new_block:
Free blocks count corrupted for block group 17
Apr  7 18:58:22 everybody kernel: EXT2-fs error (device 03:02): ext2_new_block:
Free blocks count corrupted for block group 17
Apr  7 19:06:10 everybody kernel: EXT2-fs error (device 03:02): ext2_new_block:
Free blocks count corrupted for block group 17
Apr  7 19:06:34 everybody last message repeated 4 times
Apr  7 19:08:27 everybody kernel: EXT2-fs error (device 03:02): ext2_new_block:
Free blocks count corrupted for block group 17

which seems to suggest SOMETHING horrible is wrong.

What do I do?

Ben

-- 
Brought to you by the letters I and T and the number 18.
"Porco ga daisuki!" -- Fio, "Porco Rosso"
Ben Gertzfield  Finger me for my public
PGP key. I'm on FurryMUCK as Che, and EFNet and YiffNet IRC as Che_Fox.


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

sleep contains crypto stuff?

[Martin Schulze]

Howdy,

could anybode explain this to me?

kuolema!root(ttyp0):/tmp/initrd# ldd /bin/sleep
libcrypt.so.1 => /lib/libcrypt.so.1 (0x4000f000)
libc.so.6 => /lib/libc.so.6 (0x4003c000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x4000)
kuolema!root(ttyp0):/tmp/initrd# dpkg -l shellutils
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  shellutils  1.16-6 The GNU shell programming utilities.

Regards,

Joey

-- 
  / Martin Schulze  *  [EMAIL PROTECTED]  *  26129 Oldenburg /
 / http://home.pages.de/~joey/
/ The only stupid question is the unasked one   /


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[James R. Van Zandt]

>However, I'm willing to set default root's prompt in base-files to
>'\h:\w\$ ' if enough people prefer it to '[EMAIL PROTECTED]:\w\$ '.

I would prefer '\h:\w\$ '

- Jim Van Zandt


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Why isn't /bin/sh managed with alternatives?

[Robert Woodcock]

On Sun, April 5, Shaleh <[EMAIL PROTECTED]> wrote:
> My assumption is that /bin/sh is VERY important to the system.  All
> server scripts use it.  A dangling symlink could be hazardous.  Also,
> some systems initially mount only certain directories.  And /etc is
> sometimes on a different partition entirely.  /bin should stand on its
> own.

*NO* dynamically linked binary (this includes /sbin/init and /bin/sh) will
run without /etc/ld.so.cache:

mercury:~$ strace init
execve("/sbin/init", ["init"], [/* 27 vars */]) = 0
brk(0)  = 0x804dde8
open("/etc/ld.so.preload", O_RDONLY)= -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY)  = 3
fstat(3, {st_mode=S_IFREG|0644, st_size=9333, ...}) = 0
mmap(0, 9333, PROT_READ, MAP_PRIVATE, 3, 0) = 0x4000c000
close(3)= 0
open("/lib/libc.so.6", O_RDONLY)= 3

For the curious, it retrieved that "/lib/libc.so.6" string from
ld.so.cache.

By the way, is ash any faster at sh script processing than bash?
If we made ash the default /bin/sh, would it speed the bootup process any?

Another idea I got on IRC was providing a --background flag to
start-stop-daemon so that daemons could be started in parallel - this
might have quite an effect on SMP systems, and DNS misconfigs would be
more treatable if sendmail started in the background instead of waiting a 
few minutes timing out on stuff before anything else could run.

Or we could just use & to do it.
--
Robert Woodcock - [EMAIL PROTECTED]
All I want is a warm bed and a kind word and unlimited power.
-- Ashleigh Brilliant


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: base-files 1.6 (source all) uploaded to master

[Avery Pennarun]

On Tue, Apr 07, 1998 at 07:18:33AM -0400, Gregory S. Stark wrote:

> > Anyway, I remember a Slackware trick to set the default prompt for many
> > different shells. Could not we do the same?
> 
> Am I the only one who thinks the only correct prompts would be '$ ' and '#
> '?

I hope so.  I know prompts are a religious issue, but they should at least
tell you _something_.  $ vs # tells you whether you're the superuser or not,
but that's pretty dull.  bash$ just provides useless information -- 99% of
the time, I'm in bash; otherwise I'm in tcsh, which uses a % anyway.  (I use
Solaris at work and Linux at home.)

I can live with having a hostname in the prompt, but I prefer to
colour-code my hosts instead :)  I know this one is definitely personal
preference.

I usually find that having my username in the prompt is pretty useless,
given that I'm almost always either me or root, and I can use $/# to
determine which.  People with a lot of sub-accounts would like to have the
username in their prompt, but I think this is uncommon (?).

Having my path in the prompt is absolutely essential.  I have a tendency
to open tonnes of shell windows and forget where I am in each one --
displaying it in the prompt is a great help.

I also like to have a newline (\n) as the first character of my prompt. 
This prevents it popping up at the end of an existing line, and does a good
job of separating each command on the screen so I don't lose my bearings.

If TERM==xterm or rxvt, I add a code that places the current hostname and
directory in the window title -- that way I can remember which of my many
minimized rxvt's I'm looking for.  I tend to log into many computers
simultaneously in the course of my work, so having the hostname/directory
show up automatically in the icon is a great help.

Maybe we should have a vote on popular prompt features, then try to
implement them on a shell-by-shell basis :)  For reference, mine looks like
this:

[newline]
/usr/local/bin $
^^^^
 | |
  Yellow Bold yellow
  
In my frame of reference, "yellow" means my laptop.  I have a handy login
script that chooses a colour based on hostname.

I think we should stay away from colours and xterm codes in the default
prompt, but just bash$ and # are pretty useless.

IMHO :)

Avery


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: [?] egcs increases C++ binary size dramatically

[Falk Hueffner]

On Wed, 8 Apr 1998 01:01:46 +0200, Marcus Brinkmann
<[EMAIL PROTECTED]> wrote:

>> It's related to the fact that egcs does exception handling - add
>> -fno-exceptions to your CFLAGS, and you'll get a shorter binary.
>
>I think this is not the right way to think of C++ programming. If you don't
>use exceptions, you are free to disable them, but they are really an
>integrated feature in the standard. GNU g++ had poorly to none exception
>handling, egcs does a better job here, but maybe it is not optimized for
>size yet.

It seems that programs are larger even if they do not use exceptions
at all (possibly even C programs). For those, it seems totally
resonable to disable exceptions. It should probably even added to the
policy, since it saves space.

Falk


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: Immutable flag and packages

[bear]

> 1) Newbie will ask "why I can't removed file X? I do all that they say
> in the *Linux is Great - 101 Book* (hoping not to enfrange any
> copyright here).

Probaby not just newbies. :-)
 
> 2) If, as a SysAdmin, I decided to put some file immutable, I
> certainly don't want a program like dpkg (whom task is package
> maintenance, not security administration) make it obseleted because
> it, it changed it. Even, maybe I set it immutable because I don't want
> dpkg to changed it!

This is an annoyingly good point.  One possibility is the use of
other bits in the ext2 extended attributes in the package maintenance
program; another is the introduction of a second immutable bit 
specifically for package tools.  

> Briefly, most people don't need it or will misused it (see the pgp
> manual about false security impression) 

The fact that locks *can* be picked isn't an argument to never
use one.  As most cops will readily admit, you often only need
a lock just good enough to convince the bad guys to move on down
the street.

Setting critical system files immutable won't stop a dedicated
attacker, but it should (from what I understand) be enough to
stop a casual attack from someone without root access.

> and people who really need it,
> don't need dpkg to bother with them.

I think that point is debatable.  I'm not ignorant about security
issues, but there are so many variables today that it's virtually
impossible for someone to keep track of them all.   On the other
hand, it's not unreasonable for part of the final release process
to be a fairly through security check and any problems either fixed
or documented.  

Bear Giles
[EMAIL PROTECTED]


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Re: source packaging format (was Re: Questions about plans for Emacs 20...)

[Falk Hueffner]

On Sun, 5 Apr 1998 22:43:28 +0100, Enrique Zanardi <[EMAIL PROTECTED]>
wrote:

>Well, it's not broken, but we have no way to use pristine sources with
>multiple-tar packages, so it may be improved (and it should).

And there are *still* no source dependencies yet... Are there any
plans about this in 2.x versions?

Falk Hueffner




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]