From undefine at aramin.one.pl  Tue Oct  1 23:26:46 2002
From: undefine at aramin.one.pl (undefine at aramin.one.pl)
Date: Tue, 1 Oct 2002 23:26:46 +0200
Subject: [MBT] new entry in =?iso-8859-2?Q?pkg_?=
	=?iso-8859-2?Q?gdm_=22gdm_uruchamiany_z_rc=2Ed_nie_dzia=B3a=22?=
In-Reply-To: <Z2Rt.720@bugs.pld.org.pl>
References: <Z2Rt.388@bugs.pld.org.pl> <Z2Rt.720@bugs.pld.org.pl>
Message-ID: <20021001212646.GA15029@aramin.one.pl>

On Tue, Oct 01, 2002 at 06:32:49PM -0000, bugs at pld.org.pl wrote:
> Date: 2002-10-01 20:32:49+02	Author: Michal/ Margula (AlchemyX) <alchemyx at pld.org.pl> 
> Title:         gdm uruchamiany z rc.d nie dzia?a
> Ticket ID:     #205
> Ticket URL:    http://bugs.pld.org.pl/?bug=205
> Package:       gdm-1:2.2.5.4-5
> Distribution:  
> Category:      doesn't work as it should
> Current state: opened
> Text:
> 
> A czy system jest w pe?ni zaktualizowany? gdm-2.2.5.4-8 i pozosta?e programy uaktualnione (chodzi o kolejno?? startowania) i ca?o?? dzia?a OK.
wszystko najnowsze ;)
problem polega na tym ?e rc startuje skrypty z /etc/rc.d/rc5.d/ (
for i in /etc/rc.d/rc$runlevel.d/S*; do ) w kolejno?ci alfabetycznej.
Numerek 99 (czyli ?adowane na ko?cu) b?d? trzy skrypty:
S99allowlogin
S99local
S99gdm (xdm, ...)

A poniewa? wykonuj? si? alfabetycznie to wpierw b?dzie allowlogin, potem
gdm a nast?pnie local.

Je?eli masz pusty local r??nicy nie zauwazysz - zanim gdm podniesie X-y
local si? zako?czy i inittab podniesie konsole u?ytkownik?w.
ale spr?b?j w local da? jak?? bardziej czasoch?onn? operacj? (u mnie
jest tam uruchamianie paru skrypcik?w z sieci kt?re mi?dzy innymi
updatuj? automagicznie niekt?re elementy systemu). Efekt b?dzie taki ?e
uruchomi si? gdm z ixami(na 2 konsoli - przynajmniej tak jest w logach),
a po chwili dopiero odpal? si? mingetty... Doprowadza to do zwisu.
Testowane na kilkudziesi?ciu r??nych komputerach. Po wywaleniu moich
drobiazg?w z local wszystko ruszy?o...

Rozwi?zania mog? by? chyba dwa:
1. przeniesienie local, allowlogin np na 98 albo okolice. Albo nawet
local na 97, allowlogin na 98 a ?dm na 99 - bo teraz allow login odpala
si? przed local...
2. odpalanie gdm-a ju? po zako?czeniu wszystkich skrypt?w. z specjalnego
miejsca w rc-scriptach.

Andrzej


From malekith at pld-linux.org  Tue Oct  1 23:29:44 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Tue, 1 Oct 2002 23:29:44 +0200
Subject: [MBT] new entry in =?iso-8859-2?Q?pkg_?=
	=?iso-8859-2?Q?gdm_=22gdm_uruchamiany_z_rc=2Ed_nie_dzia=B3a=22?=
In-Reply-To: <20021001212646.GA15029@aramin.one.pl>
References: <Z2Rt.388@bugs.pld.org.pl> <Z2Rt.720@bugs.pld.org.pl> <20021001212646.GA15029@aramin.one.pl>
Message-ID: <20021001212944.GA19186@roke.freak>

On Tue, Oct 01, 2002 at 11:26:46PM +0200, undefine at aramin.one.pl wrote:
> On Tue, Oct 01, 2002 at 06:32:49PM -0000, bugs at pld.org.pl wrote:
> > Date: 2002-10-01 20:32:49+02	Author: Michal/ Margula (AlchemyX) <alchemyx at pld.org.pl> 
> > Title:         gdm uruchamiany z rc.d nie dzia?a
> > Ticket ID:     #205
> > Ticket URL:    http://bugs.pld.org.pl/?bug=205
> > Package:       gdm-1:2.2.5.4-5
> > Distribution:  
> > Category:      doesn't work as it should
> > Current state: opened
> > Text:
> > 
> > A czy system jest w pe?ni zaktualizowany? gdm-2.2.5.4-8 i pozosta?e programy uaktualnione (chodzi o kolejno?? startowania) i ca?o?? dzia?a OK.
> wszystko najnowsze ;)
> problem polega na tym ?e rc startuje skrypty z /etc/rc.d/rc5.d/ (
> for i in /etc/rc.d/rc$runlevel.d/S*; do ) w kolejno?ci alfabetycznej.
> Numerek 99 (czyli ?adowane na ko?cu) b?d? trzy skrypty:
[snip]

Resend this to pld-devel-_pl_@ and next time look at mail headers.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From wrobell at ite.pl  Sun Oct  6 20:48:41 2002
From: wrobell at ite.pl (wrobell)
Date: Sun, 6 Oct 2002 20:48:41 +0200
Subject: SPECS: dia.spec (HEAD) [wrobell]
In-Reply-To: <dia.spec.1.71@cvs.pld.org.pl>
References: <dia.spec.1.70@cvs.pld.org.pl> <dia.spec.1.71@cvs.pld.org.pl>
Message-ID: <20021006184841.GA14921@pred.dk.org>

On Sun, Oct 06, 2002 at 08:22:33PM +0200, PLD CVS wrote:
> Module name:	SPECS
> Changes by:	wrobell	02/10/06 20:22:31
> 
> Modified files:
> 	dia.spec
> 
> Log message:
> - cvs snapshot 20021002-0723
And DO NOT put it in Ra.

    wrobell <wrobell at ite.pl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021006/7708df0f/attachment.sig>

From kloczek at rudy.mif.pg.gda.pl  Mon Oct  7 20:58:58 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Mon, 7 Oct 2002 20:58:58 +0200 (CEST)
Subject: SPECS: dia.spec (HEAD) [wrobell]
In-Reply-To: <20021006184841.GA14921@pred.dk.org>
References: <dia.spec.1.70@cvs.pld.org.pl> <dia.spec.1.71@cvs.pld.org.pl>
 <20021006184841.GA14921@pred.dk.org>
Message-ID: <Pine.LNX.4.44L.0210072057050.5270-100000@rudy.mif.pg.gda.pl>

On Sun, 6 Oct 2002, wrobell wrote:

> On Sun, Oct 06, 2002 at 08:22:33PM +0200, PLD CVS wrote:
> > Module name:	SPECS
> > Changes by:	wrobell	02/10/06 20:22:31
> > 
> > Modified files:
> > 	dia.spec
> > 
> > Log message:
> > - cvs snapshot 20021002-0723
> And DO NOT put it in Ra.

Eaaasssyyy .. finish 1.0 is very close so put some new stuff now on HEAD 
isn't wrong.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From wrobell at ite.pl  Mon Oct  7 21:36:21 2002
From: wrobell at ite.pl (wrobell)
Date: Mon, 7 Oct 2002 21:36:21 +0200
Subject: SPECS: dia.spec (HEAD) [wrobell]
In-Reply-To: <Pine.LNX.4.44L.0210072057050.5270-100000@rudy.mif.pg.gda.pl>
References: <dia.spec.1.70@cvs.pld.org.pl> <dia.spec.1.71@cvs.pld.org.pl> <20021006184841.GA14921@pred.dk.org> <Pine.LNX.4.44L.0210072057050.5270-100000@rudy.mif.pg.gda.pl>
Message-ID: <20021007193621.GA3683@pred.dk.org>

On Mon, Oct 07, 2002 at 08:58:58PM +0200, Tomasz K?oczko wrote:
> On Sun, 6 Oct 2002, wrobell wrote:
> 
> > On Sun, Oct 06, 2002 at 08:22:33PM +0200, PLD CVS wrote:
> > > Module name:	SPECS
> > > Changes by:	wrobell	02/10/06 20:22:31
> > > 
> > > Modified files:
> > > 	dia.spec
> > > 
> > > Log message:
> > > - cvs snapshot 20021002-0723
> > And DO NOT put it in Ra.
> 
> Eaaasssyyy .. finish 1.0 is very close so put some new stuff now on HEAD 
> isn't wrong.
Yep... but I know you, crazy man. ;-]

    wrobell <wrobell at ite.pl>


From migo at supernet.com.pl  Tue Oct  8 15:53:25 2002
From: migo at supernet.com.pl (Bartosz Jakubski)
Date: Tue, 8 Oct 2002 15:53:25 +0200
Subject: Two (or more) packages from one spec file (with different prefixes)
Message-ID: <20021008135324.GA8845@anorak.localdomain>

Does anybody has an idea (or can point me to a spec file which does
this thing) how the <Subject> can be achieved in a nice way?

I have splitted ethereal because I was unable to find satisfactory
solution. Now I'm looking at another package that could be splitted but
I really don't want to split spec file into two (or more) files...

-- 
-- .- Bartosz Jakubski  ------------------- Sosnowiec -. --
-- |  Free Software is a matter of liberty, not price  | --
-- `---------------(http://www.fsf.org/philosophy)-----' --


From wolf42 at wp.pl  Tue Oct  8 17:54:47 2002
From: wolf42 at wp.pl (Bartosz Taudul)
Date: Tue, 8 Oct 2002 17:54:47 +0200
Subject: Two (or more) packages from one spec file (with different prefixes)
In-Reply-To: <20021008135324.GA8845@anorak.localdomain>
References: <20021008135324.GA8845@anorak.localdomain>
Message-ID: <20021008155447.GA13725@bajzel>

On Tue, Oct 08, 2002 at 03:53:25PM +0200, Bartosz Jakubski wrote:
> Does anybody has an idea (or can point me to a spec file which does
> this thing) how the <Subject> can be achieved in a nice way?
uade.spec

wolf
-- 
  Bartek   .  - Zwr??my uwag? na wielorako?? i r??norodno?? tego ruchu.
  Taudul   :  
(temp. sig):....................................... magiczne teksty nauczycieli:
    w o l f <at> p l d . o r g . p l             .:. http://4f.lo14.szczecin.pl


From undefine at aramin.one.pl  Tue Oct  8 20:58:16 2002
From: undefine at aramin.one.pl (undefine at aramin.one.pl)
Date: Tue, 8 Oct 2002 20:58:16 +0200
Subject: [MBT] new ticket for pkg quota "edquota+vim = no changes"
In-Reply-To: <cXVvdGE=.753@bugs.pld.org.pl>
References: <cXVvdGE=.753@bugs.pld.org.pl>
Message-ID: <20021008185816.GB24185@aramin.one.pl>

On Tue, Oct 08, 2002 at 07:53:09AM -0000, bugs at pld.org.pl wrote:
> Date: 2002-10-08 09:53:08+02	Author: Daniel Bauke (bonkey) <bonkey at pld.org.pl> 
> Title:         edquota+vim = no changes
> Ticket ID:     #377
> Ticket URL:    http://bugs.pld.org.pl/?bug=377
> Package:       quota-1:3.06-2
> Distribution:  PLD-Ra.main PLD-1.0.devel.main
> Category:      doesn't work as it should
> Current state: opened
> Text:
> 
> problem similar to the one with updating
> crontabs in vixie-cron:
> $ EDITOR=vi edquota user
> won't update changes from the temporary file
> $ EDITOR=joe edquota user
> will do.
but "bug" is on vim which create new file instead of changing old.


From migo at supernet.com.pl  Tue Oct  8 21:32:39 2002
From: migo at supernet.com.pl (Bartosz Jakubski)
Date: Tue, 8 Oct 2002 21:32:39 +0200
Subject: Two (or more) packages from one spec file (with different prefixes)
In-Reply-To: <20021008155447.GA13725@bajzel>
References: <20021008135324.GA8845@anorak.localdomain> <20021008155447.GA13725@bajzel>
Message-ID: <20021008193239.GA28546@anorak.localdomain>

Dnia 08 pa? o godzinie 17:54:47 +0200, Bartosz Taudul napisa?(a):
> On Tue, Oct 08, 2002 at 03:53:25PM +0200, Bartosz Jakubski wrote:
>> Does anybody has an idea (or can point me to a spec file which does
>> this thing) how the <Subject> can be achieved in a nice way?
> uade.spec
> 

Are you sure? There's only one prefix used there...

Problem is with packages, that contain some files that should use
/usr/X11R6 as its prefix and some that should use /usr. (manuals,
executables)

But the package uses autoconf and there you may only specify one
'--prefix' which is used for all files...




-- 
-- .- Bartosz Jakubski  ------------------- Sosnowiec -. --
-- |  Free Software is a matter of liberty, not price  | --
-- `---------------(http://www.fsf.org/philosophy)-----' --


From wolf42 at wp.pl  Wed Oct  9 17:12:49 2002
From: wolf42 at wp.pl (Bartosz Taudul)
Date: Wed, 9 Oct 2002 17:12:49 +0200
Subject: Two (or more) packages from one spec file (with different prefixes)
In-Reply-To: <20021008193239.GA28546@anorak.localdomain>
References: <20021008135324.GA8845@anorak.localdomain> <20021008155447.GA13725@bajzel> <20021008193239.GA28546@anorak.localdomain>
Message-ID: <20021009151249.GA2845@bajzel>

On Tue, Oct 08, 2002 at 09:32:39PM +0200, Bartosz Jakubski wrote:
> Are you sure? There's only one prefix used there...
Ah, *this* prefix...

> Problem is with packages, that contain some files that should use
> /usr/X11R6 as its prefix and some that should use /usr. (manuals,
> executables)
> 
> But the package uses autoconf and there you may only specify one
> '--prefix' which is used for all files...
Maybe some sedding on makefiles?

wolf
-- 
  Bartek   .  - Pan d?ugopisem stawia?
  Taudul   :  - A co si? b?d? chrzani?.
(temp. sig):....................................... magiczne teksty nauczycieli:
    w o l f <at> p l d . o r g . p l             .:. http://4f.lo14.szczecin.pl


From migo at supernet.com.pl  Wed Oct  9 21:24:20 2002
From: migo at supernet.com.pl (Bartosz Jakubski)
Date: Wed, 9 Oct 2002 21:24:20 +0200
Subject: About gpm upgrade...
Message-ID: <20021009192420.GA2409@anorak.localdomain>

So we have gpm-1.20.1rc1 in Ra.
blues writes in spec's changelog:
NOTE: if you have scroll mouse and it crashes while going X -> console
change mouse type to exps2. It works...

Last sentence isn't exactly true in my case. If I set exps2 in
/etc/sysconfig/mouse then I'll get non-working mouse after system boot.
I _have to_ run X and leave to make mouse work properly under console.

I have popular A4Tech mouse - 2 wheels, 3 buttons.

-- 
-- .- Bartosz Jakubski  ------------------- Sosnowiec -. --
-- |  Free Software is a matter of liberty, not price  | --
-- `---------------(http://www.fsf.org/philosophy)-----' --


From wiget at pld.org.pl  Thu Oct 10 10:54:39 2002
From: wiget at pld.org.pl (Artur Frysiak)
Date: Thu, 10 Oct 2002 10:54:39 +0200
Subject: About gpm upgrade...
In-Reply-To: <20021009192420.GA2409@anorak.localdomain>
References: <20021009192420.GA2409@anorak.localdomain>
Message-ID: <20021010085439.GC3366@localhost.localdomain>

On Wed, Oct 09, 2002 at 09:24:20PM +0200, Bartosz Jakubski wrote:
> So we have gpm-1.20.1rc1 in Ra.
> blues writes in spec's changelog:
> NOTE: if you have scroll mouse and it crashes while going X -> console
> change mouse type to exps2. It works...
> 
> Last sentence isn't exactly true in my case. If I set exps2 in
> /etc/sysconfig/mouse then I'll get non-working mouse after system boot.
> I _have to_ run X and leave to make mouse work properly under console.
> 
> I have popular A4Tech mouse - 2 wheels, 3 buttons.

I confirm this. I have A4Tech WWW-35 (2 wheels, 5 buttons).
With imps2 mouse work after system boot, but stop working after leaving
X.
With exps2 doesn't work after system boot, but work after leaving X.

-- 
Artur Frysiak
http://www.pld-linux.org/


From blues at ds6.pg.gda.pl  Thu Oct 10 11:17:29 2002
From: blues at ds6.pg.gda.pl (Blues)
Date: Thu, 10 Oct 2002 11:17:29 +0200 (CEST)
Subject: About gpm upgrade...
In-Reply-To: <20021010085439.GC3366@localhost.localdomain>
References: <20021009192420.GA2409@anorak.localdomain>
 <20021010085439.GC3366@localhost.localdomain>
Message-ID: <Pine.LNX.4.44L.0210101116510.19034-100000@piorun.ds.pg.gda.pl>

On Thu, 10 Oct 2002, Artur Frysiak wrote:
> > So we have gpm-1.20.1rc1 in Ra.
> > blues writes in spec's changelog:
> > NOTE: if you have scroll mouse and it crashes while going X -> console
> > change mouse type to exps2. It works...
> > 
> > Last sentence isn't exactly true in my case. If I set exps2 in
> > /etc/sysconfig/mouse then I'll get non-working mouse after system
> > boot. I _have to_ run X and leave to make mouse work properly under
> > console.
> > 
> > I have popular A4Tech mouse - 2 wheels, 3 buttons.
> 
> I confirm this. I have A4Tech WWW-35 (2 wheels, 5 buttons). With imps2
> mouse work after system boot, but stop working after leaving X. With
> exps2 doesn't work after system boot, but work after leaving X.

Please, make bugreport to gpm developers... 

-- 
---------------------------------
pozdr.  Pawe? Go?aszewski        
---------------------------------
CPU not found - software emulation...


From agaran at ep09.kernel.pl  Thu Oct 10 21:56:11 2002
From: agaran at ep09.kernel.pl (Maciej 'Agaran' Pijanka)
Date: Thu, 10 Oct 2002 21:56:11 +0200
Subject: About gpm upgrade...
In-Reply-To: <20021010085439.GC3366@localhost.localdomain>
References: <20021009192420.GA2409@anorak.localdomain> <20021010085439.GC3366@localhost.localdomain>
Message-ID: <20021010195610.GA31638@ep09.kernel.pl>

On Thu, Oct 10, 2002 at 10:54:39AM +0200, Artur Frysiak wrote:
> Date: Thu, 10 Oct 2002 10:54:39 +0200
> From: Artur Frysiak <wiget at pld.org.pl>
> To: pld-devel-en at pld.org.pl
> Subject: Re: About gpm upgrade...
> Mail-Followup-To: pld-devel-en at pld.org.pl
> 
> On Wed, Oct 09, 2002 at 09:24:20PM +0200, Bartosz Jakubski wrote:
> > So we have gpm-1.20.1rc1 in Ra.
> > blues writes in spec's changelog:
> > NOTE: if you have scroll mouse and it crashes while going X -> console
> > change mouse type to exps2. It works...
> > 
> > Last sentence isn't exactly true in my case. If I set exps2 in
> > /etc/sysconfig/mouse then I'll get non-working mouse after system boot.
> > I _have to_ run X and leave to make mouse work properly under console.
> > 
> > I have popular A4Tech mouse - 2 wheels, 3 buttons.
> 
> I confirm this. I have A4Tech WWW-35 (2 wheels, 5 buttons).
> With imps2 mouse work after system boot, but stop working after leaving
> X.
> With exps2 doesn't work after system boot, but work after leaving X.

logitech one wheel three buttons.. i confirm above problem

> 
> -- 
> Artur Frysiak
> http://www.pld-linux.org/
> 
> ________________________
> http://lists.pld.org.pl/

-- 
agaran at ep09 dot kernel dot pl


From blues at ds6.pg.gda.pl  Fri Oct 11 07:52:45 2002
From: blues at ds6.pg.gda.pl (Blues)
Date: Fri, 11 Oct 2002 07:52:45 +0200 (CEST)
Subject: About gpm upgrade...
In-Reply-To: <20021010195610.GA31638@ep09.kernel.pl>
References: <20021009192420.GA2409@anorak.localdomain>
 <20021010085439.GC3366@localhost.localdomain> <20021010195610.GA31638@ep09.kernel.pl>
Message-ID: <Pine.LNX.4.44L.0210110752050.30751-100000@piorun.ds.pg.gda.pl>

On Thu, 10 Oct 2002, Maciej 'Agaran' Pijanka wrote:
> > > So we have gpm-1.20.1rc1 in Ra.
> > > blues writes in spec's changelog:
> > > NOTE: if you have scroll mouse and it crashes while going X -> console
> > > change mouse type to exps2. It works...
> > > 
> > > Last sentence isn't exactly true in my case. If I set exps2 in
> > > /etc/sysconfig/mouse then I'll get non-working mouse after system
> > > boot. I _have to_ run X and leave to make mouse work properly under
> > > console.
> > > 
> > > I have popular A4Tech mouse - 2 wheels, 3 buttons.
> > I confirm this. I have A4Tech WWW-35 (2 wheels, 5 buttons). With imps2
> > mouse work after system boot, but stop working after leaving X. With
> > exps2 doesn't work after system boot, but work after leaving X.
> logitech one wheel three buttons.. i confirm above problem

Well... I have got such a mouse and it simply works...

-- 
---------------------------------
pozdr.  Pawe? Go?aszewski        
---------------------------------
CPU not found - software emulation...


From lubos at klokner.sk  Fri Oct 11 10:19:18 2002
From: lubos at klokner.sk (lubos klokner)
Date: Fri, 11 Oct 2002 10:19:18 +0200
Subject: vim colors
Message-ID: <20021011081918.GC3719@creon.profinet.sk>

hello,

i have problem with colors in vim.
when i export TERM=screen, vim show colors,
but if i have TERM=xterm (this is default when i ssh to host)
vim is only in b&w.

i'm sorry but i don't know all the magic of term and ncurses so
i can not more explore this problem :-(

thank you for any help.

best regards
-- 
lubos klokner <lubos at klokner.sk>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021011/9a722cab/attachment.sig>

From qboosh at pld.org.pl  Fri Oct 11 10:20:18 2002
From: qboosh at pld.org.pl (Jakub Bogusz)
Date: Fri, 11 Oct 2002 10:20:18 +0200
Subject: vim colors
In-Reply-To: <20021011081918.GC3719@creon.profinet.sk>
References: <20021011081918.GC3719@creon.profinet.sk>
Message-ID: <20021011082018.GD17559@gruby.cs.net.pl>

On Fri, Oct 11, 2002 at 10:19:18AM +0200, lubos klokner wrote:
> hello,
> 
> i have problem with colors in vim.
> when i export TERM=screen, vim show colors,
> but if i have TERM=xterm (this is default when i ssh to host)
> vim is only in b&w.

"xterm" is mono terminal. Use TERM=xterm-color if you want colors.


-- 
Jakub Bogusz


From orzech at pld.org.pl  Fri Oct 11 10:32:44 2002
From: orzech at pld.org.pl (Tomek Orzechowski)
Date: Fri, 11 Oct 2002 10:32:44 +0200
Subject: vim colors
In-Reply-To: <20021011081918.GC3719@creon.profinet.sk>
References: <20021011081918.GC3719@creon.profinet.sk>
Message-ID: <20021011083244.GB2217@alpha.hq.cs.net.pl>

lubos klokner wrote:

>hello,
>
>i have problem with colors in vim.
>when i export TERM=screen, vim show colors,
>but if i have TERM=xterm (this is default when i ssh to host)
>vim is only in b&w.

You are using putty, am i right?
If so, change TERM to xterm-color in connection options.

-- 
  _  _._. _  .|_
 (_)|  /_(/.(.| )
Tomek  Orzechowski


From lubos at klokner.sk  Fri Oct 11 15:21:09 2002
From: lubos at klokner.sk (lubos klokner)
Date: Fri, 11 Oct 2002 15:21:09 +0200
Subject: vim colors
In-Reply-To: <20021011083244.GB2217@alpha.hq.cs.net.pl>
References: <20021011081918.GC3719@creon.profinet.sk> <20021011083244.GB2217@alpha.hq.cs.net.pl>
Message-ID: <20021011132109.GA28562@creon.profinet.sk>

* Tomek Orzechowski <orzech at pld.org.pl> (11.10.2002):
> lubos klokner wrote:
> 
> >i have problem with colors in vim.
> >when i export TERM=screen, vim show colors,
> >but if i have TERM=xterm (this is default when i ssh to host)
> >vim is only in b&w.
> 
> You are using putty, am i right?
> If so, change TERM to xterm-color in connection options.

no, i'm using konsole from kde.

-- 
lubos klokner <lubos at klokner.sk>


From kloczek at rudy.mif.pg.gda.pl  Fri Oct 11 15:33:57 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Fri, 11 Oct 2002 15:33:57 +0200 (CEST)
Subject: vim colors
In-Reply-To: <20021011081918.GC3719@creon.profinet.sk>
References: <20021011081918.GC3719@creon.profinet.sk>
Message-ID: <Pine.LNX.4.44L.0210111527190.26967-100000@rudy.mif.pg.gda.pl>

On Fri, 11 Oct 2002, lubos klokner wrote:

> hello,
> 
> i have problem with colors in vim.
> when i export TERM=screen, vim show colors,
> but if i have TERM=xterm (this is default when i ssh to host)
> vim is only in b&w.

It is correct because:

$ TERM=xterm tput colors
-1

which informs TERM=xterm is mono terminal (and also X11 specifications 
specifiessx TERM=xterm is mono :). This is why all our [x]term 
emulation applications negotiates TERM=xterm-colors which:

$ TERM=xterm-color tput colors
8

which informs: this terminal is color capable :)
(look at terminfo(5) man page on colors sequence description).

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From lubos at klokner.sk  Fri Oct 11 15:35:16 2002
From: lubos at klokner.sk (lubos klokner)
Date: Fri, 11 Oct 2002 15:35:16 +0200
Subject: vim colors
In-Reply-To: <20021011083244.GB2217@alpha.hq.cs.net.pl>
References: <20021011081918.GC3719@creon.profinet.sk> <20021011083244.GB2217@alpha.hq.cs.net.pl>
Message-ID: <20021011133516.GB28562@creon.profinet.sk>

* Tomek Orzechowski <orzech at pld.org.pl> (11.10.2002):
> lubos klokner wrote:
> 
> >hello,
> >
> >i have problem with colors in vim.
> >when i export TERM=screen, vim show colors,
> >but if i have TERM=xterm (this is default when i ssh to host)
> >vim is only in b&w.
> 
> You are using putty, am i right?
> If so, change TERM to xterm-color in connection options.

i didn't find how to change terminal in konsole, so i fix it in .bashrc:

if [ "$TERM" == "xterm" ]; then
	export TERM="xterm-color"
fi

:-)

-- 
lubos klokner <lubos at klokner.sk>


From kloczek at rudy.mif.pg.gda.pl  Fri Oct 11 15:49:17 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Fri, 11 Oct 2002 15:49:17 +0200 (CEST)
Subject: vim colors
In-Reply-To: <20021011132109.GA28562@creon.profinet.sk>
References: <20021011081918.GC3719@creon.profinet.sk> <20021011083244.GB2217@alpha.hq.cs.net.pl>
 <20021011132109.GA28562@creon.profinet.sk>
Message-ID: <Pine.LNX.4.44L.0210111534170.26967-100000@rudy.mif.pg.gda.pl>

On Fri, 11 Oct 2002, lubos klokner wrote:

> * Tomek Orzechowski <orzech at pld.org.pl> (11.10.2002):
> > lubos klokner wrote:
> > 
> > >i have problem with colors in vim.
> > >when i export TERM=screen, vim show colors,
> > >but if i have TERM=xterm (this is default when i ssh to host)
> > >vim is only in b&w.
> > 
> > You are using putty, am i right?
> > If so, change TERM to xterm-color in connection options.
> 
> no, i'm using konsole from kde.

Are you talking about konsole from PLD kdebase ?
Our konsole is patched (kdebase-konsole-TERM.patch) for negotiate by 
default TERM=xterm-color.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From malekith at pld-linux.org  Fri Oct 11 22:02:36 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 11 Oct 2002 22:02:36 +0200
Subject: libtool/pthread problem
Message-ID: <20021011200236.GA23107@roke.freak>

If libtool is called like:

/bin/sh ../libtool --mode=link i686-pld-linux-gcc  -O2 -march=athlon
-Wall -pthread  -s
-L/home/users/malekith/tmp/glib2-2.1.0-root-malekith/usr/lib -o
libgthread-2.0.la -rpath /usr/lib -version-info 100:0:100
-export-dynamic   gthread-impl.lo -pthread  ../glib/libglib-2.0.la 

Shouldn't it link resulting library against -lpthread, or at least
mention pthread in generated .la file?

Sorry, but I'm by no means autocrap wizard and have to ask.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Fri Oct 11 22:09:03 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 11 Oct 2002 22:09:03 +0200
Subject: libtool/pthread problem
In-Reply-To: <20021011200236.GA23107@roke.freak>
References: <20021011200236.GA23107@roke.freak>
Message-ID: <20021011200903.GB23668@roke.freak>

On Fri, Oct 11, 2002 at 10:02:36PM +0200, Michal Moskal wrote:
> If libtool is called like:
> 
> /bin/sh ../libtool --mode=link i686-pld-linux-gcc  -O2 -march=athlon
> -Wall -pthread  -s
> -L/home/users/malekith/tmp/glib2-2.1.0-root-malekith/usr/lib -o
> libgthread-2.0.la -rpath /usr/lib -version-info 100:0:100
> -export-dynamic   gthread-impl.lo -pthread  ../glib/libglib-2.0.la 
> 
> Shouldn't it link resulting library against -lpthread, or at least
> mention pthread in generated .la file?
> 
> Sorry, but I'm by no means autocrap wizard and have to ask.

Or maybe I should do s/-pthread/& -lpthread/ on Makefiles?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Fri Oct 11 23:37:36 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 11 Oct 2002 23:37:36 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
Message-ID: <20021011213736.GA29968@roke.freak>

Glibc references some <asm/*.h> files. They are in kernel-headers
package. Shouldn't therefore glibc-devel require kernel-headers?

Additionally, do I really need 20 megs of includes for alpha, ppc, i960
etc? Maybe there should be just asm-i386 in kernel-headers for x86,
asm-ppc for ppc etc?

And yet another idea. Files outside <asm/*.h> are more or less private 
to the kernel. Maybe separate /usr/src/linux/include/asm into separate
package and make glibc-devel require it?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Sat Oct 12 03:25:44 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Sat, 12 Oct 2002 03:25:44 +0200
Subject: rpm 4.1 problem
Message-ID: <20021012012544.GA8464@roke.freak>

[malekith at roke RPMS]$ rpm -e libpng1
error: Failed dependencies:
	libpng = 1.2.4 is needed by (installed) libpng-devel-1.2.4-1
	...
	
[malekith at roke SPECS]$ rpm -q libpng1 libpng libpng-devel
libpng1-1.0.14-1
libpng-1.2.4-1
libpng-devel-1.2.4-1
[malekith at roke SPECS]$ rpm -q --provides libpng 
libpng12.so.0  
libpng = 2:1.2.4-1
[malekith at roke SPECS]$ rpm -q --provides libpng1
libpng = 1.0.14
libpng.so.2  
libpng1 = 2:1.0.14-1
[malekith at roke SPECS]$ 

It assumes, that libpng1 is the only package providing libpng and
screams.  I had the same problem with qt2/qt packages.

Is the problem known?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From wrobell at ite.pl  Sat Oct 12 13:39:44 2002
From: wrobell at ite.pl (wrobell)
Date: Sat, 12 Oct 2002 13:39:44 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021011213736.GA29968@roke.freak>
References: <20021011213736.GA29968@roke.freak>
Message-ID: <20021012113944.GA17127@pred.dk.org>

On Fri, Oct 11, 2002 at 11:37:36PM +0200, Michal Moskal wrote:
> Glibc references some <asm/*.h> files. They are in kernel-headers
> package. Shouldn't therefore glibc-devel require kernel-headers?
No. There are some people who use non-distro kernel.

[...]

    wrobell <wrobell at ite.pl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021012/6b67dcb7/attachment.sig>

From mmazur at kernel.pl  Sat Oct 12 13:53:20 2002
From: mmazur at kernel.pl (Mariusz Mazur)
Date: Sat, 12 Oct 2002 13:53:20 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021011213736.GA29968@roke.freak>
References: <20021011213736.GA29968@roke.freak>
Message-ID: <200210121353.20929.mmazur@kernel.pl>

On Friday 11 October 2002 23:37, Michal Moskal wrote:
> Glibc references some <asm/*.h> files. They are in kernel-headers
> package. Shouldn't therefore glibc-devel require kernel-headers?

I thought about it too (I stumbled across this yesterday when something didn't 
want to build although all BuildRequires were met). Wouldn't it be ok to just 
provide a kernel-fake.noarch.rpm for people not using disto kernel?


-- 
Ka?dy cz?owiek, kt?ry naprawd? ?yje, nie ma charakteru, nie mo?e go mie?.
Charakter jest zawsze martwy, otacza ci? zgni?a struktura przeniesiona z 
przesz?o?ci. Je?eli dzia?asz zgodnie z charakterem wtedy nie dzia?asz w og?le
- jedynie mechanicznie reagujesz.                 { Osho }


From lubos at klokner.sk  Sat Oct 12 15:04:40 2002
From: lubos at klokner.sk (lubos klokner)
Date: Sat, 12 Oct 2002 15:04:40 +0200
Subject: vim colors
In-Reply-To: <Pine.LNX.4.44L.0210111534170.26967-100000@rudy.mif.pg.gda.pl>
References: <20021011081918.GC3719@creon.profinet.sk> <20021011083244.GB2217@alpha.hq.cs.net.pl> <20021011132109.GA28562@creon.profinet.sk> <Pine.LNX.4.44L.0210111534170.26967-100000@rudy.mif.pg.gda.pl>
Message-ID: <20021012130439.GA23313@creon.profinet.sk>

* Tomasz K?oczko <kloczek at rudy.mif.pg.gda.pl> (11.10.2002):
> > no, i'm using konsole from kde.
> 
> Are you talking about konsole from PLD kdebase ?
> Our konsole is patched (kdebase-konsole-TERM.patch) for negotiate by 
> default TERM=xterm-color.

no, i have mandrake on my workstation... i'm using pld only on servers.

-- 
lubos klokner <lubos at klokner.sk>


From malekith at pld-linux.org  Sat Oct 12 15:07:31 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Sat, 12 Oct 2002 15:07:31 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <200210121353.20929.mmazur@kernel.pl>
References: <20021011213736.GA29968@roke.freak> <200210121353.20929.mmazur@kernel.pl>
Message-ID: <20021012130731.GA24101@roke.freak>

On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > Glibc references some <asm/*.h> files. They are in kernel-headers
> > package. Shouldn't therefore glibc-devel require kernel-headers?
> 
> I thought about it too (I stumbled across this yesterday when something didn't 
> want to build although all BuildRequires were met). Wouldn't it be ok to just 
> provide a kernel-fake.noarch.rpm for people not using disto kernel?

I guess this is better idea then hacks with %{?_without_dist_kernel:...}
all over the specs.

Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, or
somewhere). Wouldn't that be a good idea?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From kloczek at rudy.mif.pg.gda.pl  Sat Oct 12 15:30:21 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Sat, 12 Oct 2002 15:30:21 +0200 (CEST)
Subject: vim colors
In-Reply-To: <20021012130439.GA23313@creon.profinet.sk>
References: <20021011081918.GC3719@creon.profinet.sk> <20021011083244.GB2217@alpha.hq.cs.net.pl>
 <20021011132109.GA28562@creon.profinet.sk> <Pine.LNX.4.44L.0210111534170.26967-100000@rudy.mif.pg.gda.pl>
 <20021012130439.GA23313@creon.profinet.sk>
Message-ID: <Pine.LNX.4.44L.0210121510000.20859-100000@rudy.mif.pg.gda.pl>

On Sat, 12 Oct 2002, lubos klokner wrote:

> * Tomasz K?oczko <kloczek at rudy.mif.pg.gda.pl> (11.10.2002):
> > > no, i'm using konsole from kde.
> > 
> > Are you talking about konsole from PLD kdebase ?
> > Our konsole is patched (kdebase-konsole-TERM.patch) for negotiate by 
> > default TERM=xterm-color.
> 
> no, i have mandrake on my workstation... i'm using pld only on servers.

MDK/RH developers ignores fact TERM=xterm is mono and some packages have
exist hacks for this. Example in mc: mc-4.5.27-xtermcolor.patch (mc runed
on TERM=xterm-color starts in color mode by default [1]). Also after
swithing to TERM=xterm-color many other programs starts using colors
without any addytional changes (lynx, jed, slrn, mutt ..). In PLD also was
added by me few patches which removes injection in eterm, gnome-terminal
enviroment variable $COLORTERM which using as next ingnores TERM=xterm is
mono terminal tries inform application about terminal color abilities 
(ignoring terminfo/termcap).

Summarize: in PLD all this kind hacks was removed or not applayed and
all programs uses *only* terminfo avalaible terminal descriptions. Result:
much more term applications uses colors without any additional
manipulations.

[1] IMHO it will be good simplify mc by remove code which andles
variable color_terminals in [Colors] section in ~/.mc/ini file.
Afer swithing to preffer TERM=xterm-color this can be performed without
remove any valueable mc functionalities.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From wrobell at ite.pl  Sat Oct 12 17:51:28 2002
From: wrobell at ite.pl (wrobell)
Date: Sat, 12 Oct 2002 17:51:28 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021012130731.GA24101@roke.freak>
References: <20021011213736.GA29968@roke.freak> <200210121353.20929.mmazur@kernel.pl> <20021012130731.GA24101@roke.freak>
Message-ID: <20021012155128.GA27865@pred.dk.org>

On Sat, Oct 12, 2002 at 03:07:31PM +0200, Michal Moskal wrote:
> On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> > On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > > Glibc references some <asm/*.h> files. They are in kernel-headers
> > > package. Shouldn't therefore glibc-devel require kernel-headers?
> > 
> > I thought about it too (I stumbled across this yesterday when something didn't 
> > want to build although all BuildRequires were met). Wouldn't it be ok to just 
> > provide a kernel-fake.noarch.rpm for people not using disto kernel?
> 
> I guess this is better idea then hacks with %{?_without_dist_kernel:...}
> all over the specs.
> 
> Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, or
> somewhere). Wouldn't that be a good idea?

No, because it will overwrite used asm headers from non-distro
kernel.

    wrobell <wrobell at ite.pl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021012/ce9f2819/attachment.sig>

From malekith at pld-linux.org  Sat Oct 12 17:42:44 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Sat, 12 Oct 2002 17:42:44 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021012155128.GA27865@pred.dk.org>
References: <20021011213736.GA29968@roke.freak> <200210121353.20929.mmazur@kernel.pl> <20021012130731.GA24101@roke.freak> <20021012155128.GA27865@pred.dk.org>
Message-ID: <20021012154244.GA31284@roke.freak>

On Sat, Oct 12, 2002 at 05:51:28PM +0200, wrobell wrote:
> On Sat, Oct 12, 2002 at 03:07:31PM +0200, Michal Moskal wrote:
> > On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> > > On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > > > Glibc references some <asm/*.h> files. They are in kernel-headers
> > > > package. Shouldn't therefore glibc-devel require kernel-headers?
> > > 
> > > I thought about it too (I stumbled across this yesterday when something didn't 
> > > want to build although all BuildRequires were met). Wouldn't it be ok to just 
> > > provide a kernel-fake.noarch.rpm for people not using disto kernel?
> > 
> > I guess this is better idea then hacks with %{?_without_dist_kernel:...}
> > all over the specs.
> > 
> > Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, or
> > somewhere). Wouldn't that be a good idea?
> 
> No, because it will overwrite used asm headers from non-distro
> kernel.

I guess one should use kernel headers that was used during glibc
compilation?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From wrobell at ite.pl  Sat Oct 12 18:35:56 2002
From: wrobell at ite.pl (wrobell)
Date: Sat, 12 Oct 2002 18:35:56 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021012154244.GA31284@roke.freak>
References: <20021011213736.GA29968@roke.freak> <200210121353.20929.mmazur@kernel.pl> <20021012130731.GA24101@roke.freak> <20021012155128.GA27865@pred.dk.org> <20021012154244.GA31284@roke.freak>
Message-ID: <20021012163556.GA31245@pred.dk.org>

On Sat, Oct 12, 2002 at 05:42:44PM +0200, Michal Moskal wrote:
> On Sat, Oct 12, 2002 at 05:51:28PM +0200, wrobell wrote:
> > On Sat, Oct 12, 2002 at 03:07:31PM +0200, Michal Moskal wrote:
> > > On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> > > > On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > > > > Glibc references some <asm/*.h> files. They are in kernel-headers
> > > > > package. Shouldn't therefore glibc-devel require kernel-headers?
> > > > 
> > > > I thought about it too (I stumbled across this yesterday when something didn't 
> > > > want to build although all BuildRequires were met). Wouldn't it be ok to just 
> > > > provide a kernel-fake.noarch.rpm for people not using disto kernel?
> > > 
> > > I guess this is better idea then hacks with %{?_without_dist_kernel:...}
> > > all over the specs.
> > > 
> > > Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, or
> > > somewhere). Wouldn't that be a good idea?
> > 
> > No, because it will overwrite used asm headers from non-distro
> > kernel.
> 
> I guess one should use kernel headers that was used during glibc
> compilation?
I do not think so.

For example take PLD Ra situation where 2.2 kernel
is used. I had been using Ra glibc very long time 
with kernel 2.4.x.

There are some situations when you should recompile
your glibc, but they are rare.

    wrobell <wrobell at ite.pl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021012/24a61127/attachment.sig>

From ankry at green.mif.pg.gda.pl  Sat Oct 12 21:21:54 2002
From: ankry at green.mif.pg.gda.pl (Andrzej Krzysztofowicz)
Date: Sat, 12 Oct 2002 21:21:54 +0200 (CEST)
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021012163556.GA31245@pred.dk.org> from "wrobell" at Oct 12, 2002 06:35:56 PM
Message-ID: <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>

> 
> On Sat, Oct 12, 2002 at 05:42:44PM +0200, Michal Moskal wrote:
> > On Sat, Oct 12, 2002 at 05:51:28PM +0200, wrobell wrote:
> > > On Sat, Oct 12, 2002 at 03:07:31PM +0200, Michal Moskal wrote:
> > > > On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> > > > > On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > > > > > Glibc references some <asm/*.h> files. They are in kernel-headers
> > > > > > package. Shouldn't therefore glibc-devel require kernel-headers?
> > > > >=20
> > > > > I thought about it too (I stumbled across this yesterday when somet=
> hing didn't=20
> > > > > want to build although all BuildRequires were met). Wouldn't it be =
> ok to just=20
> > > > > provide a kernel-fake.noarch.rpm for people not using disto kernel?
> > > >=20
> > > > I guess this is better idea then hacks with %{?_without_dist_kernel:.=
> ..}
> > > > all over the specs.
> > > >=20
> > > > Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, =
> or
> > > > somewhere). Wouldn't that be a good idea?
> > >=20
> > > No, because it will overwrite used asm headers from non-distro
> > > kernel.
> >=20
> > I guess one should use kernel headers that was used during glibc
> > compilation?
> I do not think so.
> 
> For example take PLD Ra situation where 2.2 kernel
> is used. I had been using Ra glibc very long time=20
> with kernel 2.4.x.

... and you probably will not be able to use all 2.4 specyfic features 
with your glibc, I guess. 
Can you work with large files ?

> There are some situations when you should recompile
> your glibc, but they are rare.

Kernel 2.4+ headers are no longer intended to be used in userspace.
Eg., you must hack them in some places to use them with C++.
So providing own copy of hacked kernel-headers included into glibc-devel
seems to be the only choice when we switch into 2.4/2.6.

-- 
=======================================================================
  Andrzej M. Krzysztofowicz               ankry at mif.pg.gda.pl
  phone (48)(58) 347 14 61
Faculty of Applied Phys. & Math.,   Gdansk University of Technology


From wrobell at ite.pl  Sun Oct 13 01:33:31 2002
From: wrobell at ite.pl (wrobell)
Date: Sun, 13 Oct 2002 01:33:31 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
Message-ID: <20021012233331.GB31245@pred.dk.org>

On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > 
> > On Sat, Oct 12, 2002 at 05:42:44PM +0200, Michal Moskal wrote:
> > > On Sat, Oct 12, 2002 at 05:51:28PM +0200, wrobell wrote:
> > > > On Sat, Oct 12, 2002 at 03:07:31PM +0200, Michal Moskal wrote:
> > > > > On Sat, Oct 12, 2002 at 01:53:20PM +0200, Mariusz Mazur wrote:
> > > > > > On Friday 11 October 2002 23:37, Michal Moskal wrote:
> > > > > > > Glibc references some <asm/*.h> files. They are in kernel-headers
> > > > > > > package. Shouldn't therefore glibc-devel require kernel-headers?
> > > > > >=20
> > > > > > I thought about it too (I stumbled across this yesterday when somet=
> > hing didn't=20
> > > > > > want to build although all BuildRequires were met). Wouldn't it be =
> > ok to just=20
> > > > > > provide a kernel-fake.noarch.rpm for people not using disto kernel?
> > > > >=20
> > > > > I guess this is better idea then hacks with %{?_without_dist_kernel:.=
> > ..}
> > > > > all over the specs.
> > > > >=20
> > > > > Anyway I heard somewhere about moving <asm/*> to glibc-devel (in rh, =
> > or
> > > > > somewhere). Wouldn't that be a good idea?
> > > >=20
> > > > No, because it will overwrite used asm headers from non-distro
> > > > kernel.
> > >=20
> > > I guess one should use kernel headers that was used during glibc
> > > compilation?
> > I do not think so.
> > 
> > For example take PLD Ra situation where 2.2 kernel
> > is used. I had been using Ra glibc very long time=20
> > with kernel 2.4.x.
> 
> ... and you probably will not be able to use all 2.4 specyfic features 
> with your glibc, I guess. 
> Can you work with large files ?
Take for example, CD-ROM over SG 3 access. You need glibc only for
ioctl'ing device.  With 2.2.x headers in glibc-devel package you cannot
compile the program. You do need 2.4.x headers and there is no
requirement for glibc compiled with 2.4.x.
 
> > There are some situations when you should recompile
> > your glibc, but they are rare.
> 
> Kernel 2.4+ headers are no longer intended to be used in userspace.
Maybe... But above statement is told by glibc developers, I think. :-)

[...]

    wrobell <wrobell at ite.pl>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021013/2d16f92d/attachment.sig>

From malekith at pld-linux.org  Sun Oct 13 23:41:51 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Sun, 13 Oct 2002 23:41:51 +0200
Subject: bce capability in xterm-xfree86
Message-ID: <20021013214151.GA8714@roke.freak>

infocmp xterm-xfree86 shows bce capability. I have two questions:
  1) what terminal (beside gnome terminal from gnome2) use xterm-xfree86
  2) what does exactly bce mean (any terminfo wizard here?)

I ask it, bacause gnome-terminal doesn't seem to support it (at least
search option in otak, that is implemented using ncurses functions
only, clutters the screen iff the option is present).

I suggest removing it from our terminfo.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Mon Oct 14 00:05:34 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Mon, 14 Oct 2002 00:05:34 +0200
Subject: bce capability in xterm-xfree86
In-Reply-To: <20021013214151.GA8714@roke.freak>
References: <20021013214151.GA8714@roke.freak>
Message-ID: <20021013220534.GA9266@roke.freak>

On Sun, Oct 13, 2002 at 11:41:51PM +0200, Michal Moskal wrote:
> infocmp xterm-xfree86 shows bce capability. I have two questions:
>   1) what terminal (beside gnome terminal from gnome2) use xterm-xfree86

info: xterm, gnome-terminal from ra, multi-gnome-terminal from ra all
set TERM to xterm-color.


-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From kloczek at rudy.mif.pg.gda.pl  Mon Oct 14 02:44:14 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Mon, 14 Oct 2002 02:44:14 +0200 (CEST)
Subject: bce capability in xterm-xfree86
In-Reply-To: <20021013214151.GA8714@roke.freak>
References: <20021013214151.GA8714@roke.freak>
Message-ID: <Pine.LNX.4.44L.0210140239300.4075-100000@rudy.mif.pg.gda.pl>

On Sun, 13 Oct 2002, Michal Moskal wrote:

> infocmp xterm-xfree86 shows bce capability. I have two questions:
>   1) what terminal (beside gnome terminal from gnome2) use xterm-xfree86
>   2) what does exactly bce mean (any terminfo wizard here?)

RTFM: terminfo(5).

> I ask it, bacause gnome-terminal doesn't seem to support it (at least
> search option in otak, that is implemented using ncurses functions
> only, clutters the screen iff the option is present).
> 
> I suggest removing it from our terminfo.

Sorry .. removing what ? bce ?
Try use /dev/random for removing another term sequence from terminal
description :->

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From malekith at pld-linux.org  Mon Oct 14 02:49:14 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Mon, 14 Oct 2002 02:49:14 +0200
Subject: bce capability in xterm-xfree86
In-Reply-To: <Pine.LNX.4.44L.0210140239300.4075-100000@rudy.mif.pg.gda.pl>
References: <20021013214151.GA8714@roke.freak> <Pine.LNX.4.44L.0210140239300.4075-100000@rudy.mif.pg.gda.pl>
Message-ID: <20021014004914.GA23331@roke.freak>

On Mon, Oct 14, 2002 at 02:44:14AM +0200, Tomasz K?oczko wrote:
> On Sun, 13 Oct 2002, Michal Moskal wrote:
> 
> > infocmp xterm-xfree86 shows bce capability. I have two questions:
> >   1) what terminal (beside gnome terminal from gnome2) use xterm-xfree86
> >   2) what does exactly bce mean (any terminfo wizard here?)
> 
> RTFM: terminfo(5).

       back_color_erase              bce        ut        screen erased with 
       							  background color

Well... and what's that?

> > I ask it, bacause gnome-terminal doesn't seem to support it (at least
> > search option in otak, that is implemented using ncurses functions
> > only, clutters the screen iff the option is present).
> > 
> > I suggest removing it from our terminfo.
> 
> Sorry .. removing what ? bce ?
> Try use /dev/random for removing another term sequence from terminal
> description :->

If the only terminal that uses xterm-xfree86 doesn't support it...

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From kloczek at rudy.mif.pg.gda.pl  Mon Oct 14 08:48:09 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Mon, 14 Oct 2002 08:48:09 +0200 (CEST)
Subject: bce capability in xterm-xfree86
In-Reply-To: <20021014004914.GA23331@roke.freak>
References: <20021013214151.GA8714@roke.freak>
 <Pine.LNX.4.44L.0210140239300.4075-100000@rudy.mif.pg.gda.pl>
 <20021014004914.GA23331@roke.freak>
Message-ID: <Pine.LNX.4.44L.0210140839520.4075-100000@rudy.mif.pg.gda.pl>

On Mon, 14 Oct 2002, Michal Moskal wrote:

> On Mon, Oct 14, 2002 at 02:44:14AM +0200, Tomasz K?oczko wrote:
> > On Sun, 13 Oct 2002, Michal Moskal wrote:
> > 
> > > infocmp xterm-xfree86 shows bce capability. I have two questions:
> > >   1) what terminal (beside gnome terminal from gnome2) use xterm-xfree86
> > >   2) what does exactly bce mean (any terminfo wizard here?)
> > 
> > RTFM: terminfo(5).
> 
>        back_color_erase              bce        ut        screen erased with 
>        							  background color
> 
> Well... and what's that?

This indicates is erase to end of line or all lines for output clear space 
with correct color must be performed current color or default color.

This informs is errasing must/can be performed using clear end
of line or clear screen sequence or using some count spaces.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From wrobell at ite.pl  Mon Oct 14 11:13:06 2002
From: wrobell at ite.pl (wrobell)
Date: Mon, 14 Oct 2002 11:13:06 +0200
Subject: procps
Message-ID: <20021014091305.GA28281@pred.posexperts.com.pl>

[1] http://procps.sourceforge.net/

[2] http://surriel.com/procps/

Which is the better one? Why we have chosen the second fork?

    wrobell <wrobell at ite.pl>


From arekm at pld-linux.org  Mon Oct 14 11:34:46 2002
From: arekm at pld-linux.org (Arkadiusz Miskiewicz)
Date: 14 Oct 2002 11:34:46 +0200
Subject: procps
In-Reply-To: <20021014091305.GA28281@pred.posexperts.com.pl>
References: <20021014091305.GA28281@pred.posexperts.com.pl>
Message-ID: <87znthweh5.fsf@arm.t19.ds.pwr.wroc.pl>

wrobell <wrobell at ite.pl> writes:

> [1] http://procps.sourceforge.net/
> 
> [2] http://surriel.com/procps/
> 
> Which is the better one? Why we have chosen the second fork?
[1] is general cleanup
[2] is rather update to reflect latest kernels (2.5)

People are beliving in Rik version [2] ;-)

>     wrobell <wrobell at ite.pl>

-- 
Arkadiusz Mi?kiewicz    CS at FoE, Wroclaw University of Technology
arekm at sse.pl   AM2-6BONE, 1024/3DB19BBD, arekm(at)ircnet, PLD/Linux


From lubos at klokner.sk  Mon Oct 14 11:56:50 2002
From: lubos at klokner.sk (lubos klokner)
Date: Mon, 14 Oct 2002 11:56:50 +0200
Subject: php vs. http auth
Message-ID: <20021014095649.GA6105@creon.profinet.sk>

hello,

i have problem with php (4.2.3-3) from pld and http auth.
when i turn on safe_mode this code doesn't work:
	Header("WWW-Authenticate: Basic realm=\"INTERMEDIAL\"");
	Header("HTTP/1.0 401 Unauthorized");
i only get unauthorized access.
with safe_mode = off everything works ok.

i thing, that this bug was in some older version of php.

thank you for any help.
-- 
lubos klokner <lubos at klokner.sk>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: </mailman/pipermail/pld-devel-en/attachments/20021014/9d5f05d8/attachment.sig>

From qboosh at pld.org.pl  Tue Oct 15 20:58:54 2002
From: qboosh at pld.org.pl (Jakub Bogusz)
Date: Tue, 15 Oct 2002 20:58:54 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
Message-ID: <20021015185854.GE15227@satan.blackhosts>

On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > There are some situations when you should recompile
> > your glibc, but they are rare.
> 
> Kernel 2.4+ headers are no longer intended to be used in userspace.
> Eg., you must hack them in some places to use them with C++.

Where? I didn't change anything in 2.4 headers and don't see such
problems (after recompiling glibc on those headers).
But I did see strange type conflicts when I tried to use fresh kernel
headers with glibc compiled on older headers (that was in 2.2.1x era).
Those problems dissapeared after recompiling glibc.

> So providing own copy of hacked kernel-headers included into glibc-devel
> seems to be the only choice when we switch into 2.4/2.6.

I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
package for user space programs, and headers in %{_kernelsrcdir}/include
in kernel-headers package for distro kernel modules development.

I don't like the "kernel-fake package" idea - I used something like
this on RH (5.x ;)), and I don't want anymore.


-- 
Jakub Bogusz    http://www.cs.net.pl/~qboosh/


From malekith at pld-linux.org  Tue Oct 15 22:10:58 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Tue, 15 Oct 2002 22:10:58 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021015185854.GE15227@satan.blackhosts>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl> <20021015185854.GE15227@satan.blackhosts>
Message-ID: <20021015201058.GB30185@roke.freak>

On Tue, Oct 15, 2002 at 08:58:54PM +0200, Jakub Bogusz wrote:
> On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > There are some situations when you should recompile
> > > your glibc, but they are rare.
> > 
> > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > Eg., you must hack them in some places to use them with C++.
> 
> Where? I didn't change anything in 2.4 headers and don't see such
> problems (after recompiling glibc on those headers).
> But I did see strange type conflicts when I tried to use fresh kernel
> headers with glibc compiled on older headers (that was in 2.2.1x era).
> Those problems dissapeared after recompiling glibc.
> 
> > So providing own copy of hacked kernel-headers included into glibc-devel
> > seems to be the only choice when we switch into 2.4/2.6.
> 
> I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
                                                ^^^^^

If program does include <linux/...> it probably should
-I/usr/src/linux.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From arekm at pld-linux.org  Tue Oct 15 23:08:13 2002
From: arekm at pld-linux.org (Arkadiusz Miskiewicz)
Date: 15 Oct 2002 23:08:13 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021015185854.GE15227@satan.blackhosts>
References: <20021012163556.GA31245@pred.dk.org>
	<200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
	<20021015185854.GE15227@satan.blackhosts>
Message-ID: <87ptubl8aq.fsf@arm.t19.ds.pwr.wroc.pl>

Jakub Bogusz <qboosh at pld.org.pl> writes:

> On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > There are some situations when you should recompile
> > > your glibc, but they are rare.
> > 
> > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > Eg., you must hack them in some places to use them with C++.
> 
> Where? I didn't change anything in 2.4 headers and don't see such
> problems (after recompiling glibc on those headers).
Well generic idea is that kernel headers in /usr/include/{asm,linux}
should be exactly the same as when glibc was being compiled.

This is a TODO idem for 2.0 I think.

> Jakub Bogusz    http://www.cs.net.pl/~qboosh/

-- 
Arkadiusz Mi?kiewicz    CS at FoE, Wroclaw University of Technology
arekm at sse.pl   AM2-6BONE, 1024/3DB19BBD, arekm(at)ircnet, PLD/Linux


From wrobell at ite.pl  Wed Oct 16 11:46:55 2002
From: wrobell at ite.pl (wrobell)
Date: Wed, 16 Oct 2002 11:46:55 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021015201058.GB30185@roke.freak>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl> <20021015185854.GE15227@satan.blackhosts> <20021015201058.GB30185@roke.freak>
Message-ID: <20021016094655.GE17184@pred.posexperts.com.pl>

On Tue, Oct 15, 2002 at 10:10:58PM +0200, Michal Moskal wrote:
> On Tue, Oct 15, 2002 at 08:58:54PM +0200, Jakub Bogusz wrote:
> > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > There are some situations when you should recompile
> > > > your glibc, but they are rare.
> > > 
> > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > Eg., you must hack them in some places to use them with C++.
> > 
> > Where? I didn't change anything in 2.4 headers and don't see such
> > problems (after recompiling glibc on those headers).
> > But I did see strange type conflicts when I tried to use fresh kernel
> > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > Those problems dissapeared after recompiling glibc.
> > 
> > > So providing own copy of hacked kernel-headers included into glibc-devel
> > > seems to be the only choice when we switch into 2.4/2.6.
> > 
> > I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
>                                                 ^^^^^
> 
> If program does include <linux/...> it probably should
> -I/usr/src/linux.
-I/usr/src/linux/include :-)

But... which files will be included first when you specify
-I/usr/src/linux/include?  These in /usr/include or these in -I/usr/src...?

    wrobell <wrobell at ite.pl>


From malekith at pld-linux.org  Wed Oct 16 12:07:31 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Wed, 16 Oct 2002 12:07:31 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021016094655.GE17184@pred.posexperts.com.pl>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl> <20021015185854.GE15227@satan.blackhosts> <20021015201058.GB30185@roke.freak> <20021016094655.GE17184@pred.posexperts.com.pl>
Message-ID: <20021016100731.GA18769@ep09.kernel.pl>

On Wed, Oct 16, 2002 at 11:46:55AM +0200, wrobell wrote:
> But... which files will be included first when you specify
> -I/usr/src/linux/include?  These in /usr/include or these in -I/usr/src...?
             
 From /usr/src/linux/include.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From qboosh at pld.org.pl  Wed Oct 16 12:21:59 2002
From: qboosh at pld.org.pl (Jakub Bogusz)
Date: Wed, 16 Oct 2002 12:21:59 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021015201058.GB30185@roke.freak>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl> <20021015185854.GE15227@satan.blackhosts> <20021015201058.GB30185@roke.freak>
Message-ID: <20021016102159.GA30129@gruby.cs.net.pl>

On Tue, Oct 15, 2002 at 10:10:58PM +0200, Michal Moskal wrote:
> On Tue, Oct 15, 2002 at 08:58:54PM +0200, Jakub Bogusz wrote:
> > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > There are some situations when you should recompile
> > > > your glibc, but they are rare.
> > > 
> > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > Eg., you must hack them in some places to use them with C++.
> > 
> > Where? I didn't change anything in 2.4 headers and don't see such
> > problems (after recompiling glibc on those headers).
> > But I did see strange type conflicts when I tried to use fresh kernel
> > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > Those problems dissapeared after recompiling glibc.
> > 
> > > So providing own copy of hacked kernel-headers included into glibc-devel
> > > seems to be the only choice when we switch into 2.4/2.6.
> > 
> > I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
>                                                 ^^^^^
> If program does include <linux/...> it probably should
> -I/usr/src/linux.

I don't think so... user space programs usually include <linux/*> only
for ioctl numbers and similar things.
Also, many glibc headers and <asm/*> headers include <linux/*>.

Using -I%{_kernelsrcdir}/include to compile programs would require
recompiling glibc after each kernel headers upgrade (required to build
modules for new kernel).


-- 
Jakub Bogusz


From wrobell at ite.pl  Wed Oct 16 12:39:14 2002
From: wrobell at ite.pl (wrobell)
Date: Wed, 16 Oct 2002 12:39:14 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021016102159.GA30129@gruby.cs.net.pl>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl> <20021015185854.GE15227@satan.blackhosts> <20021015201058.GB30185@roke.freak> <20021016102159.GA30129@gruby.cs.net.pl>
Message-ID: <20021016103914.GG17184@pred.posexperts.com.pl>

On Wed, Oct 16, 2002 at 12:21:59PM +0200, Jakub Bogusz wrote:
> On Tue, Oct 15, 2002 at 10:10:58PM +0200, Michal Moskal wrote:
> > On Tue, Oct 15, 2002 at 08:58:54PM +0200, Jakub Bogusz wrote:
> > > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > > There are some situations when you should recompile
> > > > > your glibc, but they are rare.
> > > > 
> > > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > > Eg., you must hack them in some places to use them with C++.
> > > 
> > > Where? I didn't change anything in 2.4 headers and don't see such
> > > problems (after recompiling glibc on those headers).
> > > But I did see strange type conflicts when I tried to use fresh kernel
> > > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > > Those problems dissapeared after recompiling glibc.
> > > 
> > > > So providing own copy of hacked kernel-headers included into glibc-devel
> > > > seems to be the only choice when we switch into 2.4/2.6.
> > > 
> > > I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
> >                                                 ^^^^^
> > If program does include <linux/...> it probably should
> > -I/usr/src/linux.
> 
> I don't think so... user space programs usually include <linux/*> only
> for ioctl numbers and similar things.
new structures, etc... :-)

> Also, many glibc headers and <asm/*> headers include <linux/*>.
> 
> Using -I%{_kernelsrcdir}/include to compile programs would require
> recompiling glibc after each kernel headers upgrade (required to build
> modules for new kernel).

Take iptables (I mean user space programs, not modules) for example. For
several months I had been using glibc from Ra, used non-distro kernel,
compiled and used iptables without any problems.

    wrobell <wrobell at ite.pl>


From ankry at green.mif.pg.gda.pl  Wed Oct 16 12:53:42 2002
From: ankry at green.mif.pg.gda.pl (Andrzej Krzysztofowicz)
Date: Wed, 16 Oct 2002 12:53:42 +0200 (CEST)
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021015185854.GE15227@satan.blackhosts> from "Jakub Bogusz" at Oct 15, 2002 08:58:54 PM
Message-ID: <200210161053.g9GArgVv023237@green.mif.pg.gda.pl>

> On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > There are some situations when you should recompile
> > > your glibc, but they are rare.
> > 
> > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > Eg., you must hack them in some places to use them with C++.
> 
> Where? I didn't change anything in 2.4 headers and don't see such
> problems (after recompiling glibc on those headers).
> But I did see strange type conflicts when I tried to use fresh kernel
> headers with glibc compiled on older headers (that was in 2.2.1x era).
> Those problems dissapeared after recompiling glibc.

I don't remember exactly - look for "bug" reports in l-k archives.
But AFAIR, it was a conflict between C++ keywords and variable/field/type
names used in kernel headers.

In C you would probably be more lucky.

Note also removal of __KERNEL__ macro tests from kernel headers in some
places.

-- 
=======================================================================
  Andrzej M. Krzysztofowicz               ankry at mif.pg.gda.pl
  phone (48)(58) 347 14 61
Faculty of Applied Phys. & Math.,   Gdansk University of Technology


From wrobell at ite.pl  Wed Oct 16 13:07:18 2002
From: wrobell at ite.pl (wrobell)
Date: Wed, 16 Oct 2002 13:07:18 +0200
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <200210161053.g9GArgVv023237@green.mif.pg.gda.pl>
References: <20021015185854.GE15227@satan.blackhosts> <200210161053.g9GArgVv023237@green.mif.pg.gda.pl>
Message-ID: <20021016110718.GH17184@pred.posexperts.com.pl>

On Wed, Oct 16, 2002 at 12:53:42PM +0200, Andrzej Krzysztofowicz wrote:
> > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > There are some situations when you should recompile
> > > > your glibc, but they are rare.
> > > 
> > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > Eg., you must hack them in some places to use them with C++.
> > 
> > Where? I didn't change anything in 2.4 headers and don't see such
> > problems (after recompiling glibc on those headers).
> > But I did see strange type conflicts when I tried to use fresh kernel
> > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > Those problems dissapeared after recompiling glibc.
> 
> I don't remember exactly - look for "bug" reports in l-k archives.
> But AFAIR, it was a conflict between C++ keywords and variable/field/type
> names used in kernel headers.
So... kernel headers should be fixed, ain't it?
[...]

    wrobell <wrobell at ite.pl>


From ankry at green.mif.pg.gda.pl  Wed Oct 16 13:00:21 2002
From: ankry at green.mif.pg.gda.pl (Andrzej Krzysztofowicz)
Date: Wed, 16 Oct 2002 13:00:21 +0200 (CEST)
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021016110718.GH17184@pred.posexperts.com.pl> from "wrobell" at Oct 16, 2002 01:07:18 PM
Message-ID: <200210161100.g9GB0LLl023323@green.mif.pg.gda.pl>

> 
> On Wed, Oct 16, 2002 at 12:53:42PM +0200, Andrzej Krzysztofowicz wrote:
> > > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > > There are some situations when you should recompile
> > > > > your glibc, but they are rare.
> > > > 
> > > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > > Eg., you must hack them in some places to use them with C++.
> > > 
> > > Where? I didn't change anything in 2.4 headers and don't see such
> > > problems (after recompiling glibc on those headers).
> > > But I did see strange type conflicts when I tried to use fresh kernel
> > > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > > Those problems dissapeared after recompiling glibc.
> > 
> > I don't remember exactly - look for "bug" reports in l-k archives.
> > But AFAIR, it was a conflict between C++ keywords and variable/field/type
> > names used in kernel headers.
> So... kernel headers should be fixed, ain't it?
> [...]

Yes. Kernel developers say: "they should be fixed by glibc distributors
while enclosing necessary kernel data structure interface in glibc as
the `conflicting' code has nothing to do in userspace" ...
;)

-- 
=======================================================================
  Andrzej M. Krzysztofowicz               ankry at mif.pg.gda.pl
  phone (48)(58) 347 14 61
Faculty of Applied Phys. & Math.,   Gdansk University of Technology


From kloczek at rudy.mif.pg.gda.pl  Wed Oct 16 19:49:28 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Wed, 16 Oct 2002 19:49:28 +0200 (CEST)
Subject: Shouldn't glibc-devel r: kernel-headers?
In-Reply-To: <20021016094655.GE17184@pred.posexperts.com.pl>
References: <20021012163556.GA31245@pred.dk.org> <200210121921.g9CJLsCR013263@green.mif.pg.gda.pl>
 <20021015185854.GE15227@satan.blackhosts> <20021015201058.GB30185@roke.freak>
 <20021016094655.GE17184@pred.posexperts.com.pl>
Message-ID: <Pine.LNX.4.44L.0210161940430.13128-100000@rudy.mif.pg.gda.pl>

On Wed, 16 Oct 2002, wrobell wrote:

> On Tue, Oct 15, 2002 at 10:10:58PM +0200, Michal Moskal wrote:
> > On Tue, Oct 15, 2002 at 08:58:54PM +0200, Jakub Bogusz wrote:
> > > On Sat, Oct 12, 2002 at 09:21:54PM +0200, Andrzej Krzysztofowicz wrote:
> > > > > There are some situations when you should recompile
> > > > > your glibc, but they are rare.
> > > > 
> > > > Kernel 2.4+ headers are no longer intended to be used in userspace.
> > > > Eg., you must hack them in some places to use them with C++.
> > > 
> > > Where? I didn't change anything in 2.4 headers and don't see such
> > > problems (after recompiling glibc on those headers).
> > > But I did see strange type conflicts when I tried to use fresh kernel
> > > headers with glibc compiled on older headers (that was in 2.2.1x era).
> > > Those problems dissapeared after recompiling glibc.
> > > 
> > > > So providing own copy of hacked kernel-headers included into glibc-devel
> > > > seems to be the only choice when we switch into 2.4/2.6.
> > > 
> > > I agree - kernel headers in /usr/include/{asm,linux}, in glibc-devel
> >                                                 ^^^^^
> > 
> > If program does include <linux/...> it probably should
> > -I/usr/src/linux.
> -I/usr/src/linux/include :-)
> 
> But... which files will be included first when you specify
> -I/usr/src/linux/include?  These in /usr/include or these in -I/usr/src...?

IIRC if some package will require kernel headers during compile with
-D__KERNEL__ it will be kernel module. All not __KERNEL__ depedent code in
this headers probaly in near or slightly far fure will moved to *libc
headers.
Now searation between kernel and libc headres isn't complet so we still
must use headers from kernel and libc (glibc or other libc like uClibc or
dietlibc).

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From malekith at pld-linux.org  Fri Oct 18 11:24:20 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 18 Oct 2002 11:24:20 +0200
Subject: SPECS: mozilla.spec (HEAD) [wiget]
In-Reply-To: <mozilla.spec.1.183@cvs.pld.org.pl>
References: <mozilla.spec.1.182@cvs.pld.org.pl> <mozilla.spec.1.183@cvs.pld.org.pl>
Message-ID: <20021018092420.GA5774@roke.freak>

On Fri, Oct 18, 2002 at 10:26:22AM +0200, PLD CVS wrote:
> Module name:	SPECS
> Changes by:	wiget	02/10/18 10:26:20
> 
> Modified files:
> 	mozilla.spec
> 
> Log message:
> - enable old ABI wrappers
> - remove not builded libmai.so from %%files
> - release 0.11

Khm... 

[malekith at roke ~]$ rpm -qf /usr/X11R6/lib/libmai.so
mozilla-1.1-0.10
[malekith at roke ~]$ rpm -qi mozilla
Name        : mozilla                      Relocations: (not
relocateable)
Version     : 1.1                               Vendor: (none)
Release     : 0.10                          Build Date: Tue Oct 15 11:38:15 2002
Install date: Tue Oct 15 18:39:36 2002      Build Host: roke.freak
Group       : X11/Applications/Networking   Source RPM:
mozilla-1.1-0.10.src.rpm
Size        : 28219310                         License: GPL
Signature   : (none)
URL         : http://www.mozilla.org/
Summary     : Mozilla - web browser
Description :
Mozilla is an open-source web browser, designed for standards
compliance, performance and portability.
[malekith at roke ~]$ 

So this change should be %{!_with_gtk2:...} only I guess.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Fri Oct 18 11:32:34 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 18 Oct 2002 11:32:34 +0200
Subject: SPECS: mozilla.spec (HEAD) [wiget]
In-Reply-To: <20021018092420.GA5774@roke.freak>
References: <mozilla.spec.1.182@cvs.pld.org.pl> <mozilla.spec.1.183@cvs.pld.org.pl> <20021018092420.GA5774@roke.freak>
Message-ID: <20021018093234.GA5964@roke.freak>

On Fri, Oct 18, 2002 at 11:24:20AM +0200, Michal Moskal wrote:
> So this change should be %{!_with_gtk2:...} only I guess.

ergh... s/!/?/. Anyway done.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From malekith at pld-linux.org  Fri Oct 18 13:05:49 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 18 Oct 2002 13:05:49 +0200
Subject: problem with rpm: rename/unlink failed: is a directory
Message-ID: <20021018110549.GA8601@roke.freak>


[malekith at roke RPMS]$ sudo rpm -Uvh yelp-2.1.0-2.i586.rpm 
Preparing...                ########################################### [100%]
   1:yelp                   ########################################### [100%]
[malekith at roke RPMS]$ ls -l /usr/X11R6/bin/yelp
-rwxr-xr-x    1 root     root        90472 Oct 17 19:02 /usr/X11R6/bin/yelp*
[malekith at roke RPMS]$ sudo rpm -e yelp
[malekith at roke RPMS]$ sudo mkdir /usr/X11R6/bin/yelp
[malekith at roke RPMS]$ sudo rpm -Uvh yelp-2.1.0-2.i586.rpm
Preparing...                ########################################### [100%]
   1:yelp                   ########################################### [100%]
error: unpacking of archive failed on file /usr/X11R6/bin/yelp: cpio: rename failed - Is a directory[malekith at roke RPMS]$ ls -l /usr/X11R6/bin/yelp*         
-rw-r--r--    1 root     root        90472 Oct 18 12:48 /usr/X11R6/bin/yelp;3dafe6f2

/usr/X11R6/bin/yelp:
total 0
[malekith at roke RPMS]$ sudo rpm -e yelp
error: package yelp is not installed

I use rpm-4.1, but the same problem (with s/rename/unlink/ in error
message I guess) occurs in older versions.

The solution would be to (in shell terms) do rm -r <file> before rename,
not just rm <file> as rpm does.

However I'm not sure if that's so good idea. The problem occurs mainly
with some random junk left in filesystem by installation/reinstallation
or manual installation. However there can be also a problem if package
qux places its files in directory /foo/bar, that is symlink to say
/foo/bar-1.1 in foo package, and qux doesn't require qux. Now when
we install foo, when qux is already installed files from qux will go to
/dev/null.

However rpm by default overwrites files, so why shouldn't it overwrite
directories?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From wiget at pld.org.pl  Fri Oct 18 15:03:35 2002
From: wiget at pld.org.pl (Artur Frysiak)
Date: Fri, 18 Oct 2002 15:03:35 +0200
Subject: problem with rpm: rename/unlink failed: is a directory
In-Reply-To: <20021018110549.GA8601@roke.freak>
References: <20021018110549.GA8601@roke.freak>
Message-ID: <20021018130335.GA16173@localhost.localdomain>

On Fri, Oct 18, 2002 at 01:05:49PM +0200, Michal Moskal wrote:
> 
> However rpm by default overwrites files, so why shouldn't it overwrite
> directories?

rpm 4.1 install new files as name.randomstring and rename it to name at
end of successfull transaction (think 'commit') or remove name.randomstring (think
'rollback') if installation fail.

Regards
-- 
Artur Frysiak
http://www.pld-linux.org/


From malekith at pld-linux.org  Fri Oct 18 15:16:04 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 18 Oct 2002 15:16:04 +0200
Subject: problem with rpm: rename/unlink failed: is a directory
In-Reply-To: <20021018130335.GA16173@localhost.localdomain>
References: <20021018110549.GA8601@roke.freak> <20021018130335.GA16173@localhost.localdomain>
Message-ID: <20021018131604.GA2680@roke.freak>

On Fri, Oct 18, 2002 at 03:03:35PM +0200, Artur Frysiak wrote:
> On Fri, Oct 18, 2002 at 01:05:49PM +0200, Michal Moskal wrote:
> > 
> > However rpm by default overwrites files, so why shouldn't it overwrite
> > directories?
> 
> rpm 4.1 install new files as name.randomstring and rename it to name at
> end of successfull transaction (think 'commit') or remove name.randomstring (think
> 'rollback') if installation fail.

Ah, so yelp really wasn't installed? Ok then. But the problem with
directories remains.

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From qboosh at pld.org.pl  Fri Oct 18 16:50:41 2002
From: qboosh at pld.org.pl (Jakub Bogusz)
Date: Fri, 18 Oct 2002 16:50:41 +0200
Subject: problem with rpm: rename/unlink failed: is a directory
In-Reply-To: <20021018110549.GA8601@roke.freak>
References: <20021018110549.GA8601@roke.freak>
Message-ID: <20021018145041.GB29229@gruby.cs.net.pl>

On Fri, Oct 18, 2002 at 01:05:49PM +0200, Michal Moskal wrote:
> [malekith at roke RPMS]$ sudo rpm -Uvh yelp-2.1.0-2.i586.rpm
> Preparing...                ########################################### [100%]
>    1:yelp                   ########################################### [100%]
> error: unpacking of archive failed on file /usr/X11R6/bin/yelp: cpio: rename failed - Is a directory[malekith at roke RPMS]$ ls -l /usr/X11R6/bin/yelp*         
> -rw-r--r--    1 root     root        90472 Oct 18 12:48 /usr/X11R6/bin/yelp;3dafe6f2
[...]
> The solution would be to (in shell terms) do rm -r <file> before rename,
> not just rm <file> as rpm does.
> 
> However I'm not sure if that's so good idea. The problem occurs mainly
> with some random junk left in filesystem by installation/reinstallation
> or manual installation. However there can be also a problem if package
> qux places its files in directory /foo/bar, that is symlink to say
> /foo/bar-1.1 in foo package, and qux doesn't require qux. Now when
> we install foo, when qux is already installed files from qux will go to
> /dev/null.

IMO bad idea - directory may contain files from other packages not only
because of missing dependencies, but also when some directory is changed
to other directory with compatibility symlink left (remember automake
1.6 before no_versioned_dir patch?)...


-- 
Jakub Bogusz


From malekith at pld-linux.org  Fri Oct 18 17:54:04 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Fri, 18 Oct 2002 17:54:04 +0200
Subject: problem with rpm: rename/unlink failed: is a directory
In-Reply-To: <20021018145041.GB29229@gruby.cs.net.pl>
References: <20021018110549.GA8601@roke.freak> <20021018145041.GB29229@gruby.cs.net.pl>
Message-ID: <20021018155404.GA3869@roke.freak>

On Fri, Oct 18, 2002 at 04:50:41PM +0200, Jakub Bogusz wrote:
> On Fri, Oct 18, 2002 at 01:05:49PM +0200, Michal Moskal wrote:
> > [malekith at roke RPMS]$ sudo rpm -Uvh yelp-2.1.0-2.i586.rpm
> > Preparing...                ########################################### [100%]
> >    1:yelp                   ########################################### [100%]
> > error: unpacking of archive failed on file /usr/X11R6/bin/yelp: cpio: rename failed - Is a directory[malekith at roke RPMS]$ ls -l /usr/X11R6/bin/yelp*         
> > -rw-r--r--    1 root     root        90472 Oct 18 12:48 /usr/X11R6/bin/yelp;3dafe6f2
> [...]
> > The solution would be to (in shell terms) do rm -r <file> before rename,
> > not just rm <file> as rpm does.
> > 
> > However I'm not sure if that's so good idea. The problem occurs mainly
> > with some random junk left in filesystem by installation/reinstallation
> > or manual installation. However there can be also a problem if package
> > qux places its files in directory /foo/bar, that is symlink to say
> > /foo/bar-1.1 in foo package, and qux doesn't require qux. Now when
> > we install foo, when qux is already installed files from qux will go to
> > /dev/null.
> 
> IMO bad idea - directory may contain files from other packages not only
> because of missing dependencies, but also when some directory is changed
> to other directory with compatibility symlink left (remember automake
> 1.6 before no_versioned_dir patch?)...

Yes, you are right. But probably we can fix it to overwrite empty
directories?

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From ser at metalab.unc.edu  Sun Oct 20 22:21:59 2002
From: ser at metalab.unc.edu (Sergiusz Pawlowicz)
Date: Sun, 20 Oct 2002 22:21:59 +0200
Subject: CVSROOT [kloczek]: logdiff.pl (HEAD)
In-Reply-To: <logdiff.pl.1.29@cvs.pld.org.pl>
References: <logdiff.pl.1.28@cvs.pld.org.pl> <logdiff.pl.1.29@cvs.pld.org.pl>
Message-ID: <20021020202159.GG31567@szafa.it-zone.org>

--->[Quoting PLD CVS <pld-cvs-commit at pld.org.pl>:]

> Log message:
> - remove login from Subject.

Good idea, but why do not to put login in From: ?


cheers - Sergiusz
--
private: http://ibiblio.org/ser/  company: http://it-zone.org/ 
______________________________________________________________
on wallpaper: tldp.org, gnu.org, pld-linux.org, hyperreal.info


From krol at sigma.wsb-nlu.edu.pl  Mon Oct 21 16:51:20 2002
From: krol at sigma.wsb-nlu.edu.pl (Marcin Krol)
Date: Mon, 21 Oct 2002 16:51:20 +0200 (CEST)
Subject: [MBT] new entry in pkg oidentd "Buffer overflow..."
Message-ID: <Pine.LNX.4.44.0210211648030.29788-100000@sigma.wsb-nlu.edu.pl>

On 20 Oct 2002 bugs at pld.org.pl wrote:

> indent like all other indent damons do not fork().
> multiply indent instances can be foud in case when you are uses enteries in hosts.{allow,deny) like:
>
> <service>: All at ALL
>
> and on other sides also are used the same entries.
> Useing for ecample in hosts.allow:
>
> telnetd: ALL@<host>
>
> will allow additional indent lookup from tcpd. If on host someonn will have
>
> indent: ALL at ALL
>
> it will cause next indent from <host> .. classic indent ping-pong.
> How to avoid this ? Simple for do not use indent specyfic ruler for indent rntrues in hosts.{allow,deny}.
> Summarize: this not bug .. this feacture :)
>
>
>
> *** State changed to 'ignored'

OK. I know this. But... If there are no rules for ident in
hosts.{allow,deny}, and single ident request from some host/service
causes, lets say 7000 oidentd processes, this is not a classic ident
ping-pong. More than that, even when request is served, oidentd is still
multiplying with response for the same request infinitely, until machine
hangs due to "out of memory" problem. This is happening only on oidentd
2.x. I've never noticed that kind of behaviour on any other identd
software, including oidentd 1.x. And after upgrading to 2.x, this happens
at least one  time per week. Its irritating because if I'm away from
server when oidentd starts eating up memory, machine will hang and will
not respond until I'll reboot it. BTW: oidentd ping-pong will not eat
memory with speed >= 50 megabytes per second.

Marcin Kr?l


From malekith at pld-linux.org  Mon Oct 21 17:41:31 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Mon, 21 Oct 2002 17:41:31 +0200
Subject: problem with rpm 4.1
Message-ID: <20021021154131.GA30345@roke.freak>

poldek> install php-common-4.2.3-1 --force     
Processing dependencies...
There are 1 package to install:
I php-common-4.2.3-1
Executing sudo /bin/rpm --upgrade -vh --force --root / --noorder...
error: Failed dependencies:
	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
There were errors
poldek>

Of course php-common-4.2.3-1 also provides php-common = 4.2.3 so rpm is
wrong. I've seen few similar situations.


-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From mmazur at kernel.pl  Mon Oct 21 17:46:10 2002
From: mmazur at kernel.pl (Mariusz Mazur)
Date: Mon, 21 Oct 2002 17:46:10 +0200
Subject: problem with rpm 4.1
In-Reply-To: <20021021154131.GA30345@roke.freak>
References: <20021021154131.GA30345@roke.freak>
Message-ID: <200210211746.10678.mmazur@kernel.pl>

On Monday 21 October 2002 17:41, Michal Moskal wrote:
> Of course php-common-4.2.3-1 also provides php-common = 4.2.3 so rpm is
> wrong. I've seen few similar situations.

And so have I.


-- 
Ka?dy cz?owiek, kt?ry naprawd? ?yje, nie ma charakteru, nie mo?e go mie?.
Charakter jest zawsze martwy, otacza ci? zgni?a struktura przeniesiona z 
przesz?o?ci. Je?eli dzia?asz zgodnie z charakterem wtedy nie dzia?asz w og?le
- jedynie mechanicznie reagujesz.                 { Osho }


From malekith at pld-linux.org  Mon Oct 21 18:42:53 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Mon, 21 Oct 2002 18:42:53 +0200
Subject: problem with rpm 4.1
In-Reply-To: <200210211746.10678.mmazur@kernel.pl>
References: <20021021154131.GA30345@roke.freak> <200210211746.10678.mmazur@kernel.pl>
Message-ID: <20021021164253.GA29420@roke.freak>

On Mon, Oct 21, 2002 at 05:46:10PM +0200, Mariusz Mazur wrote:
> On Monday 21 October 2002 17:41, Michal Moskal wrote:
> > Of course php-common-4.2.3-1 also provides php-common = 4.2.3 so rpm is
> > wrong. I've seen few similar situations.
> 
> And so have I.

Forwarded to rpm-list...

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From jajcus at pld.org.pl  Mon Oct 21 18:45:15 2002
From: jajcus at pld.org.pl (Jacek Konieczny)
Date: Mon, 21 Oct 2002 18:45:15 +0200
Subject: [MBT] new entry in pkg oidentd "Buffer overflow..."
In-Reply-To: <Pine.LNX.4.44.0210211648030.29788-100000@sigma.wsb-nlu.edu.pl>
References: <Pine.LNX.4.44.0210211648030.29788-100000@sigma.wsb-nlu.edu.pl>
Message-ID: <20021021164515.GB21320@nic.nigdzie>

On Mon, Oct 21, 2002 at 04:51:20PM +0200, Marcin Krol wrote:
> hangs due to "out of memory" problem. This is happening only on oidentd
> 2.x. I've never noticed that kind of behaviour on any other identd
> software, including oidentd 1.x. And after upgrading to 2.x, this happens
> at least one  time per week. Its irritating because if I'm away from
> server when oidentd starts eating up memory, machine will hang and will
> not respond until I'll reboot it. BTW: oidentd ping-pong will not eat
> memory with speed >= 50 megabytes per second.

What is the output of the following command?
wc -l /proc/net/ip_conntrack

Greets,
        Jacek


From kloczek at rudy.mif.pg.gda.pl  Tue Oct 22 09:48:16 2002
From: kloczek at rudy.mif.pg.gda.pl (=?ISO-8859-2?Q?Tomasz_K=B3oczko?=)
Date: Tue, 22 Oct 2002 09:48:16 +0200 (CEST)
Subject: [MBT] new entry in pkg oidentd "Buffer overflow..."
In-Reply-To: <Pine.LNX.4.44.0210211648030.29788-100000@sigma.wsb-nlu.edu.pl>
References: <Pine.LNX.4.44.0210211648030.29788-100000@sigma.wsb-nlu.edu.pl>
Message-ID: <Pine.LNX.4.44L.0210220945100.11246-100000@rudy.mif.pg.gda.pl>

On Mon, 21 Oct 2002, Marcin Krol wrote:
[..]
> > it will cause next indent from <host> .. classic indent ping-pong.
> > How to avoid this ? Simple for do not use indent specyfic ruler for indent rntrues in hosts.{allow,deny}.
> > Summarize: this not bug .. this feacture :)
> >
> > *** State changed to 'ignored'
> 
> OK. I know this. But... If there are no rules for ident in
> hosts.{allow,deny}, and single ident request from some host/service
> causes, lets say 7000 oidentd processes, this is not a classic ident
> ping-pong.

Try run spare 7000 httpd waiting for incoming connection and you will have
similar stuaction. Is this in this case bug in httpd ? No because this 
configuration bug as same as not inteligent using indent lookup rules in  
hosts.{allow,deny}.

kloczek
-- 
-----------------------------------------------------------
*Ludzie nie maj? problem?w, tylko sobie sami je stwarzaj?*
-----------------------------------------------------------
Tomasz K?oczko, sys adm @zie.pg.gda.pl|*e-mail: kloczek at rudy.mif.pg.gda.pl*


From wiget at pld.org.pl  Tue Oct 22 11:33:06 2002
From: wiget at pld.org.pl (Artur Frysiak)
Date: Tue, 22 Oct 2002 11:33:06 +0200
Subject: problem with rpm 4.1
In-Reply-To: <20021021154131.GA30345@roke.freak>
References: <20021021154131.GA30345@roke.freak>
Message-ID: <20021022093306.GC14515@localhost.localdomain>

On Mon, Oct 21, 2002 at 05:41:31PM +0200, Michal Moskal wrote:
> poldek> install php-common-4.2.3-1 --force     
> Processing dependencies...
> There are 1 package to install:
> I php-common-4.2.3-1
> Executing sudo /bin/rpm --upgrade -vh --force --root / --noorder...
> error: Failed dependencies:
> 	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
> There were errors
> poldek>
> 
> Of course php-common-4.2.3-1 also provides php-common = 4.2.3 so rpm is
> wrong. I've seen few similar situations.

Show me rpm -qp --qf '%{NAME}\n' php-common-new-version.arch.rpm
And rpm -qp --provides php-common-new-version.arch.rpm

Regards
-- 
Artur Frysiak
http://www.pld-linux.org/


From malekith at pld-linux.org  Tue Oct 22 13:14:23 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Tue, 22 Oct 2002 13:14:23 +0200
Subject: problem with rpm 4.1
In-Reply-To: <20021022093306.GC14515@localhost.localdomain>
References: <20021021154131.GA30345@roke.freak> <20021022093306.GC14515@localhost.localdomain>
Message-ID: <20021022111423.GB10633@roke.freak>

On Tue, Oct 22, 2002 at 11:33:06AM +0200, Artur Frysiak wrote:
> On Mon, Oct 21, 2002 at 05:41:31PM +0200, Michal Moskal wrote:
> > poldek> install php-common-4.2.3-1 --force     
> > Processing dependencies...
> > There are 1 package to install:
> > I php-common-4.2.3-1
> > Executing sudo /bin/rpm --upgrade -vh --force --root / --noorder...
> > error: Failed dependencies:
> > 	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
> > There were errors
> > poldek>
> > 
> > Of course php-common-4.2.3-1 also provides php-common = 4.2.3 so rpm is
> > wrong. I've seen few similar situations.
> 
> Show me rpm -qp --qf '%{NAME}\n' php-common-new-version.arch.rpm
> And rpm -qp --provides php-common-new-version.arch.rpm

Hmm.. Strange I posted detailed bug report to rpm-list at redhat.com
yestarday (IIRC it does not have only-subscribers-can-post policy), 
and it does not appear in google group:linux.redhat.rpm.

So again:

[malekith at roke rpm-test]$ sudo rpm -Uvh php-common-4.2.3-5.i686.rpm 
warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
error: Failed dependencies:
	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
[malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-1.i686.rpm|grep php-common
php-common = 3:4.2.3-1
[malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-5.i686.rpm|grep php-common
warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
php-common = 3:4.2.3-5
[malekith at roke rpm-test]$ rpm -qp --qf "%{NAME}\n" php-common-4.2.3-*
php-common
warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
php-common
[malekith at roke rpm-test]$ 

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From wiget at pld.org.pl  Tue Oct 22 15:28:50 2002
From: wiget at pld.org.pl (Artur Frysiak)
Date: Tue, 22 Oct 2002 15:28:50 +0200
Subject: problem with rpm 4.1
In-Reply-To: <20021022111423.GB10633@roke.freak>
References: <20021021154131.GA30345@roke.freak> <20021022093306.GC14515@localhost.localdomain> <20021022111423.GB10633@roke.freak>
Message-ID: <20021022132850.GD9147@localhost.localdomain>

On Tue, Oct 22, 2002 at 01:14:23PM +0200, Michal Moskal wrote:
> 
> So again:
> 
> [malekith at roke rpm-test]$ sudo rpm -Uvh php-common-4.2.3-5.i686.rpm 
> warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> error: Failed dependencies:
> 	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
> [malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-1.i686.rpm|grep php-common
> php-common = 3:4.2.3-1
> [malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-5.i686.rpm|grep php-common
> warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> php-common = 3:4.2.3-5
> [malekith at roke rpm-test]$ rpm -qp --qf "%{NAME}\n" php-common-4.2.3-*
> php-common
> warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> php-common
> [malekith at roke rpm-test]$ 

And again with --nosignature --nodigest ?
Or before 
$ rpm --import ftp://ftp.pld.org.pl/dists/ra/PLD/i386/PLD/PLD.gpg

Regards
-- 
Artur Frysiak
http://www.pld-linux.org/


From malekith at pld-linux.org  Tue Oct 22 15:34:10 2002
From: malekith at pld-linux.org (Michal Moskal)
Date: Tue, 22 Oct 2002 15:34:10 +0200
Subject: problem with rpm 4.1
In-Reply-To: <20021022132850.GD9147@localhost.localdomain>
References: <20021021154131.GA30345@roke.freak> <20021022093306.GC14515@localhost.localdomain> <20021022111423.GB10633@roke.freak> <20021022132850.GD9147@localhost.localdomain>
Message-ID: <20021022133410.GA17541@roke.freak>

On Tue, Oct 22, 2002 at 03:28:50PM +0200, Artur Frysiak wrote:
> On Tue, Oct 22, 2002 at 01:14:23PM +0200, Michal Moskal wrote:
> > 
> > So again:
> > 
> > [malekith at roke rpm-test]$ sudo rpm -Uvh php-common-4.2.3-5.i686.rpm 
> > warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> > error: Failed dependencies:
> > 	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
> > [malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-1.i686.rpm|grep php-common
> > php-common = 3:4.2.3-1
> > [malekith at roke rpm-test]$ rpm -qp --provides php-common-4.2.3-5.i686.rpm|grep php-common
> > warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> > php-common = 3:4.2.3-5
> > [malekith at roke rpm-test]$ rpm -qp --qf "%{NAME}\n" php-common-4.2.3-*
> > php-common
> > warning: php-common-4.2.3-5.i686.rpm: V3 DSA signature: NOKEY, key ID f9651d5a
> > php-common
> > [malekith at roke rpm-test]$ 
> 
> And again with --nosignature --nodigest ?
> Or before 
> $ rpm --import ftp://ftp.pld.org.pl/dists/ra/PLD/i386/PLD/PLD.gpg

That's not the problem (it's only a warning):

[malekith at roke rpm-test]$ sudo rpm --import ftp://ftp.pld.org.pl/dists/ra/PLD/i386/PLD/PLD.gpg
[malekith at roke rpm-test]$ sudo rpm -Uvh php-common-4.2.3-5.i686.rpm 
error: Failed dependencies:
	php-common = 4.2.3 is needed by (installed) php-4.2.3-1
[malekith at roke rpm-test]$ 
	

-- 
: Michal Moskal ::::: malekith/at/pld-linux.org :  GCS {C,UL}++++$ a? !tv
: PLD Linux ::::::: Wroclaw University, CS Dept :  {E-,w}-- {b++,e}>+++ h


From orzech at pld.org.pl  Mon Oct 28 11:16:51 2002
From: orzech at pld.org.pl (Tomek Orzechowski)
Date: Mon, 28 Oct 2002 11:16:51 +0100
Subject: SPECS: nut.spec (HEAD)
In-Reply-To: <nut.spec.1.43@cvs.pld.org.pl>
References: <nut.spec.1.42@cvs.pld.org.pl> <nut.spec.1.43@cvs.pld.org.pl>
Message-ID: <20021028101651.GA20834@alpha.hq.cs.net.pl>

[28 pa?dziernik, 2002] orzech wrote:

>Module name:	SPECS
>Changes by:	orzech	02/10/28 11:10:23
>
>Modified files:
>	nut.spec
>
>Log message:
>- fixed broken init
>- rel. 5, STBR
[...]
>-Release:	3
>+Release:	4

Ignore my mistake. STBR, please.

-- 
  _  _._. _  .|_
 (_)|  /_(/.(.| )
Tomek  Orzechowski


From dobrek at itp.uni-hannover.de  Tue Oct 29 11:30:39 2002
From: dobrek at itp.uni-hannover.de (Lukas Dobrek)
Date: Tue, 29 Oct 2002 11:30:39 +0100
Subject: [kreutzm@itp.uni-hannover.de: Please package texdoctk]
Message-ID: <20021029103039.GA23530@zibal.itp.uni-hannover.de>

----- Forwarded message from Helge Kreutzmann <kreutzm at itp.uni-hannover.de> -----

> Date: Tue, 29 Oct 2002 09:33:43 +0100
> From: Helge Kreutzmann <kreutzm at itp.uni-hannover.de>
> Subject: Please package texdoctk
> To: ?ukasz Dobrek <Lukasz.Dobrek at itp.uni-hannover..de>@itp.uni-hannover.de
> 
> Hello Lukasz !
> I did not find a way to submit packaging requests on bugs.pld.org, so
> I ask you directly. It is a very small programm. I went throught this
> on suker and it almost works and I think "almost" because there seems
> to be some problems in the underlaying tex/kapthesa-layer (see below).
> The recepie:
> 
> # create directory
> # get
> # http://www.geophysik.uni-frankfurt.de/~ruedas/download/texdoctk-0.5.1.tar.gz
> #                         (or from CTAN)
> # unpack (no new subdirectory is created)
> # probably get newer texdoc-10x.dat file (don't now where from, just
> #                                         saw references)
> # cp -ivp textdoctk $PREFIX/bin
> # chown root:root $PREFIX/bin/texdoctk
> # mkdir $PREFIX/share/texmf/texdoctk  (must be main tex-tree)
> # cp -ivp texdoc-102.dat texdocrc.defaults
> # $prefix/share/texmf/texdoctk
> # chown root:root $PREFIX/share/texmf/texdoctk/*
> # Customization of data file is possible:
> #             cp texdoc-10x.dat texdoc-local.dat and adjust
> # texdocrc.defaults can be updated as well
> # Run texhash
> 
> # Needs: Perl, perl-Tk
> 
> 
> On suker I unfortunately get when trying to build documents:
> 
> mkdir: cannot create directory `/var/cache/fonts/pk/modeless': No such
> file or directory
> kpsestat: /var/cache/fonts/pk/modeless/..: No such file or directory
> chmod: too few arguments
> Try `chmod --help' for more information.
> mkdir: cannot create directory `/var/cache/fonts/pk/modeless/public':
> No such file or directory
> kpsestat: /var/cache/fonts/pk/modeless/public/..: No such file or
> directory
> chmod: too few arguments
> Try `chmod --help' for more information.
> 
> Thanks a lot
> 
>             Helge
> -- 
> Helge Kreutzmann, Dipl.-Phys.               Helge.Kreutzmann at itp.uni-hannover.de
>   gpg signed mail preferred    gpg-key: finger kreutzm at rigel.itp.uni-hannover.de
>     64bit GNU powered                  http://www.itp.uni-hannover.de/~kreutzm
>        Help keep free software "libre": http://www.freepatents.org/

----- End forwarded message -----

-- 
?ukasz Dobrek
   An optimist believes that we live in the best of all possible worlds.
   A pessimist is sure that this must be so.
http://www.pld-linux.org


From wrobell at ite.pl  Tue Oct 29 11:55:10 2002
From: wrobell at ite.pl (wrobell)
Date: Tue, 29 Oct 2002 11:55:10 +0100
Subject: [kreutzm@itp.uni-hannover.de: Please package texdoctk]
In-Reply-To: <20021029103039.GA23530@zibal.itp.uni-hannover.de>
References: <20021029103039.GA23530@zibal.itp.uni-hannover.de>
Message-ID: <20021029105510.GD16781@pred.posexperts.com.pl>

On Tue, Oct 29, 2002 at 11:30:39AM +0100, Lukas Dobrek wrote:
> ----- Forwarded message from Helge Kreutzmann <kreutzm at itp.uni-hannover.de> -----
> 
> > Date: Tue, 29 Oct 2002 09:33:43 +0100
> > From: Helge Kreutzmann <kreutzm at itp.uni-hannover.de>
> > Subject: Please package texdoctk
> > To: ?ukasz Dobrek <Lukasz.Dobrek at itp.uni-hannover..de>@itp.uni-hannover.de
> > 
> > Hello Lukasz !
> > I did not find a way to submit packaging requests on bugs.pld.org, so
> > I ask you directly. It is a very small programm. I went throught this
> > on suker and it almost works and I think "almost" because there seems
> > to be some problems in the underlaying tex/kapthesa-layer (see below).
> > The recepie:
The latest tetex snapshots contain the texdoctk, just:

$ ./builder tetex.spec

It creates over 170 packages and is not widely tested, so you are warned.

    wrobell <wrobell at ite.pl>