jd:/dev/blog - debian - Commentaires

ATL1E support in 2.6.26-1 - elbenditu

elbenditu — Thu, 09 Oct 2008 11:56:30 +0200

it really is, thank you.

Save power, aka easy CPU frequency scaling - Matthew W. S. Bell

Matthew W. S. Bell — Thu, 14 Aug 2008 02:44:31 +0200

http://mjg59.livejournal.com/88608....

Save power, aka easy CPU frequency scaling - MJ Ray

MJ Ray — Wed, 13 Aug 2008 20:17:29 +0200

How do you think cpufreqd compares to cpudyn?

Save power, aka easy CPU frequency scaling - michael schurter

michael schurter — Wed, 13 Aug 2008 17:59:10 +0200

cpufreqd on Lenny/Sid works great on my new AMD X2 desktop. I'm at 1 GHz (the lowest speed) most of the time unless I'm using Flash or playing video.

xulrunner hacking - glandium

glandium — Tue, 12 Aug 2008 11:24:01 +0200

Note that you didn't need to actually build the thing to patch the file you patched: you can just directly modify the file in /usr/lib/xulrunner-1.9/components.

[ping] One step further for XCB into Debian - New

New — Mon, 30 Jun 2008 16:59:35 +0200

New

...

New laptop: EeePC! - jd

jd — Mon, 25 Feb 2008 17:27:24 +0100

I still use it, not every day because I prefer to type on my big dual-screen workstation, but i usually use it in train, travels and on holidays.

New laptop: EeePC! - afi

afi — Thu, 14 Feb 2008 15:16:58 +0100

but alors you are french?

A dummy user.

New laptop: EeePC! - ptecza

ptecza — Thu, 14 Feb 2008 14:59:25 +0100

Hi Julien! Is it only next high-tech toy for kids or it can be really useful thing for such developers like you? How do you use EeePC? Do you still use your laptop? Any your opinions are very welcomed

New laptop: EeePC! - jd

jd — Sat, 26 Jan 2008 09:57:03 +0100

Wolf: no because I did not have eeepc-acpi. No I have and I can see it in the USB device list. I have not yet try to use it.

Booting Debian on it with gdm and then awesome take something like 30 sec, it's ok for me

New laptop: EeePC! - Jon

Jon — Tue, 22 Jan 2008 12:44:38 +0100

Enjoy! The keyboard imho is much more of an issue than the screen (which is often the first thing people complain about)

New laptop: EeePC! - sbdep

sbdep — Tue, 22 Jan 2008 01:51:30 +0100

You should be able to decrease the boot time. The Xandros system provided was targeted to boot in 15 seconds. Most of the changes made to achieve that were to ensure that you only start the services necessary for the basic desktop. I think the boot time is longer if you use the optional "full desktop" since it needed to start more KDE services.

New laptop: EeePC! - FrIkI

FrIkI — Mon, 21 Jan 2008 19:26:07 +0100

I'm still waiting for mine. Good luck ^_^

New laptop: EeePC! - wjl (Wolfgang Lonien)

wjl (Wolfgang Lonien) — Mon, 21 Jan 2008 18:36:54 +0100

Did you get the Webcam working, and how?

cheers,
Wolfgang

New laptop: EeePC! - M.Pomme

M.Pomme — Mon, 21 Jan 2008 17:42:14 +0100

... "Dummy users" ... ?

http://www.istockphoto.com/file_clo...

New laptop: EeePC! - krazykit

krazykit — Mon, 21 Jan 2008 16:27:20 +0100

Congratulations on the new toy! You'll get used to the keyboard soon enough. As far as tuning stuff goes, ext3 with an ordered journal may not be ideal for the life of the solid-state drive. writeback mode, while slightly more dangerous with data, should help extend the life of the drive, as the data itself isn't being written twice.

Testing out rebuildd 0.3.0 - jd

jd — Mon, 03 Dec 2007 15:14:31 +0100

Easily doable, you just have to modify your pbuilder instance to run wrap dpkg-buildpackage calls.

Testing out rebuildd 0.3.0 - Seo Sanghyeon

Seo Sanghyeon — Sun, 02 Dec 2007 19:20:10 +0100

How about parallel building with dpkg-buildpackage -j2?

Time < work - jd

jd — Fri, 16 Nov 2007 10:24:54 +0100

it will come with awesome 2.0-rc1.

Time < work - Dave

Dave — Fri, 16 Nov 2007 03:50:15 +0100

I'm looking forward to awesome in Debian!

Laptop problem - sae

sae — Mon, 29 Oct 2007 09:36:10 +0100

Pour Helllo -> vous bloguez à une heure bien tardivbe

Ubuntu: not contributing to Debian, and even doing bad work - bytecoders

bytecoders — Thu, 04 Oct 2007 14:05:29 +0200

Ubuntu no contribuye a Debian y encima hace el trabajo mal
Trackback:
http://bytecoders.homelinux.com/con...

Ubuntu: not contributing to Debian, and even doing bad work - RAOF

RAOF — Wed, 03 Oct 2007 09:35:35 +0200

All our buildds have /bin/sh -> dash, this is our default sh, and I (mistakenly) thought that \n -> newline in echo was POSIX (it's implementation defined, apparently). Stupid man sh, stupid me!

Your fix is really a fix, being actually portable. The Ubuntu bug has been re-opened.

Ubuntu: not contributing to Debian, and even doing bad work - MadCoder

MadCoder — Wed, 03 Oct 2007 09:26:38 +0200

FWIW the proper fix is not to use printf at all but use:

python -c "statement1; statement2; statement3"

@Chris: I thought Ubuntu had a CoC ? or does it apply only in teletubyland (_that_ would be a quite convenient CoC then) ?

Ubuntu: not contributing to Debian, and even doing bad work - jldugger

jldugger — Wed, 03 Oct 2007 09:25:23 +0200

It can be frustrating to send patches upstream and be shot down, but I don't think that Ubuntu developers should allow the letdowns to discourage them, even if they comprise the majority of interactions with Debian.

In this particular case, the developer seems to have decided Debian upstream wasn't interested in fixing bash-isms. I'm glad to hear you are. Indeed, I wish there was a better way to handle. Partly I think the direct antagonism directed at Ubuntu makes people care less about working with Debian, resulting in less than optimal systems like patches.ubuntu.com with little or no documentation. I realize that the amateur efforts that sometimes represent Ubuntu Universe are less than perfect. Still, I like Ubuntu and Debian, and I hope the future finds some better policies, algorithms and communications to make this work with less friction on both sides. A posting on a blog indexed by Planet Debian should not be the best avenue to pursue such things, but it seems for the moment it is =(

Ubuntu: not contributing to Debian, and even doing bad work - jd

jd — Wed, 03 Oct 2007 09:01:34 +0200

We assume that /bin/sh can be dash as well.

If the fix is correct in Ubuntu, you assume that /bin/sh will never be bash?

Ubuntu: not contributing to Debian, and even doing bad work - RAOF

RAOF — Wed, 03 Oct 2007 03:37:29 +0200

I filed & fixed the Ubuntu bug (the fix is correct in Ubuntu, but /usr/bin/printf is indeed more correct, thanks).

I can't remember exactly why I didn't file a Debian bug as well, but it was probably because: (1) I didn't have a Debian system at the time to test it on, and (2) It shouldn't have adversely affected the Debian package at all, since you use bash for /bin/sh.

Ubuntu: not contributing to Debian, and even doing bad work - Chris Cunningham

Chris Cunningham — Wed, 03 Oct 2007 00:54:09 +0200

All things considered, considering that I haven't the patience of saints, I probably wouldn't make much of an effort to submit my patches to sociopathic dicks (that's debian-fr, by the way) if I could, either. Unsurprisingly, this is self-perpetuating, which is why it's 500% more likely that a random debian-fr user is going to troll Planet Debian about uncooperative upstreams/downstreams comparing to normal DDs.

Apologies to the few French DDs tarred with this brush who don't spend most of their free time making other free software developers less inclined to cooperate with debian. But then, I imagine they stopped reading this blog a while ago.

- Chris

[ping] Ubuntu: not contributing to Debian, and even doing bad work - cpunk - the cypherblog

cpunk - the cypherblog — Tue, 02 Oct 2007 22:35:04 +0200

Ubuntu Debian

Noch gut erinnere ich mich an den abschätzigen Kommentar eines Freundes, Ubuntu sei ein Betriebssystem für Hippies... Dazu mag allerdings so gar nicht der Vorwurf der Debianer passen, Ubuntu profitiere zwar eine Menge vom Debian-Projekt, gebe aber......

Ubuntu: not contributing to Debian, and even doing bad work - jd

jd — Tue, 02 Oct 2007 22:19:54 +0200

Already commited with printf, yes.

Ubuntu: not contributing to Debian, and even doing bad work - Daniel Jacobowitz

Daniel Jacobowitz — Tue, 02 Oct 2007 22:15:27 +0200

If you still want to fix the bug, I recommend /usr/bin/printf for this sort of thing.

Is X.org 7.3 a sucking release? - Vincent Bernat

Vincent Bernat — Tue, 25 Sep 2007 10:10:14 +0200

On three machines that were running dual head setup, none of them is able to use dual head with Xorg 7.3. One Intel, two ATI. This release is really a pain...

Is X.org 7.3 a sucking release? - jd

jd — Sat, 22 Sep 2007 13:24:47 +0200

Grow up? hm, lol ?

I've got 2 computers, 2 non-functionning computer, so yes X.Org 7.3 is worse than ever.

Seriously, try to be objective a little and grow up too: don't take everything and every word literraly.

Is X.org 7.3 a sucking release? - daniels

daniels — Fri, 21 Sep 2007 23:27:24 +0200

Yes, it's a complete piece of shit, and we love nothing but to make software worse and worse, to annoy people who don't appear to have ever contributed anything to it (including bug reports), but still feel a sense of entitlement.

But seriously though, grow up.

Is X.org 7.3 a sucking release? - garaged

garaged — Fri, 21 Sep 2007 18:56:16 +0200

I can tell that 7.3-2 is not that good because just googleearth crashes X, the acceleration improvement is notizable, but is a "little bit" more unstable

[ping] DeFuBu contest #3 - Valium.

Valium. — Fri, 21 Sep 2007 02:17:18 +0200

Valium.

Valium side effects. Valium....

i386--, amd64++ - jd

jd — Thu, 16 Aug 2007 19:21:13 +0200

Good to know, I did not knew that the CPU could scale down more.

I was happy with my old Asus P4P800...

Debian archive rebuild for nostrip finished - klokop

klokop — Thu, 16 Aug 2007 17:36:08 +0200

About pr0n in software: libPr0n (http://www.stukjewebgebeuren.nl/nod...)

i386--, amd64++ - directhex

directhex — Wed, 15 Aug 2007 20:27:10 +0200

Asus boards seem to be getting worse and worse - I have a home server with a top-end Asus inside it, which suffers from inadequate CPU frequency scaling (you ought to be seeing 1GHz or less on your Core 2, it's not going that low because Asus' ACPI implementation is busted)

It also suffers from major issues with networking - see https://bugs.launchpad.net/ubuntu/+... and a whole host of similar bugs for example. Fishing furhter into the problem suggests Asus have written a completely broken firmware for their onboard networking.

Unimpressed with the company.

On media players, xmms and audacious - ttm

ttm — Sat, 11 Aug 2007 02:06:44 +0200

MPD ? (Music Player Daemon)

large choice of clients...

http://musicpd.org

On media players, xmms and audacious - romain

romain — Fri, 10 Aug 2007 17:38:08 +0200

beep-media-player ?

On media players, xmms and audacious - Anonymous

Anonymous — Fri, 10 Aug 2007 17:31:55 +0200

Random suggestion: Quod Libet works very well with huge music libraries. It also has excellent search, which would address your desire to just load your entire playlist and jump to a song you want.

On media players, xmms and audacious - samvais

samvais — Fri, 10 Aug 2007 15:53:54 +0200

too bad if xmms is going away. it's my only option when I use my Logitech wireless usb-soundcard ( Bus 003 Device 006: ID 0b7a:0100 Zeevo, Inc.) since so far I haven't found another music player that can chooce the output device without running alsaconf or changing the default output device for all sound. I want my music player to stream the music to my stereos and iceweasel et al to play their sound on my laptop. I admit that exaile and rhythmbox are much more versatile and modern than xmms, I wish there was a way to chooce the sound output in every player.

Implementing Debian policy check with rebuildd - glandium

glandium — Tue, 07 Aug 2007 11:37:01 +0200

The problem I have is that even when not building with -g, there are debugging information coming from the libc in the binaries So, for the moment, I don't know how to tell if there are lines debugging info in a binary or not (.debug_line section exists and is not empty), though it seems possible to say if -g was used by looking if .debug_loc and .debug_frame sections are present.

Implementing Debian policy check with rebuildd - jd

jd — Tue, 07 Aug 2007 11:11:45 +0200

Suggestions are welcome.

Implementing Debian policy check with rebuildd - glandium

glandium — Tue, 07 Aug 2007 07:42:31 +0200

Next step will be to determine what kind of symbols are provided in the files. A file could actually be not stripped, and still have no debugging information. I was trying to see with objdump and readelf the kind of tests that could be made to detect this.

About unit tests - Marius Gedminas

Marius Gedminas — Fri, 13 Jul 2007 18:59:41 +0200

I cannot count the cases where I wrote unit tests that found bugs in trivial functions I was absolutely certain were error-free.

Unit tests are very useful.

About unit tests - Steve

Steve — Fri, 13 Jul 2007 15:56:17 +0200

Probably a little of both

Actually I've found unit tests to be extremely useful for coding, and I try to add them to all new code I touch.

What a real RC bug squasher must have - Bernd

Bernd — Thu, 05 Jul 2007 19:16:52 +0200

Is it also possible to use a complete directory instead of a single image?

Would be nice to have small random pictures on the desktop.

Why you won't use MySQL/ndb - jd

jd — Mon, 18 Jun 2007 21:25:59 +0200

For now I have not been able to reproduce the bug, so I keep my fingers crossed

Why you won't use MySQL/ndb - Stewart Smith

Stewart Smith — Mon, 18 Jun 2007 02:42:57 +0200

There have been a good number of bug fixes in the gap between .31 and .41 for NDB, so .41 will no doubt end up being a lot better.

I haven't seen the bug you describe before... so perhaps it's something related to the debian package. (certainly report it to the debian BTS, debian packaging of MySQL is fairly good..).

Certainly if you can repeat it with .41, we'll get right on it... as it would certainly be a new bug.

News from the battlefront - Sandrine

Sandrine — Wed, 04 Apr 2007 17:58:32 +0000

Fight!

The new DeFuBu contest - francois matias

francois matias — Wed, 21 Mar 2007 20:28:25 +0000

The same contest with windows would be a lots of more funny i guess .. lol..

The Etch song - Manu

Manu — Fri, 09 Mar 2007 13:33:14 +0000

programming make me crazy too

Debian Developers, you all suck - julien

julien — Wed, 07 Mar 2007 14:12:05 +0000

French people Rocks!

My way to have etch released on time - Stéphane

Stéphane — Fri, 02 Mar 2007 02:41:19 +0000

Don't be sorry

DPL 2007: Raphael Hertzog - jd

jd — Wed, 28 Feb 2007 17:21:31 +0000

Ca s'appele de l'humour, mais apparement, tu ne connais pas.

DPL 2007: Raphael Hertzog - Suomynona

Suomynona — Wed, 28 Feb 2007 17:17:24 +0000

Ca fait chic de démolir les gens publiquement ! C'est ça le nouveau Debian.

Bonjour, je suis anarchiste - bart

bart — Tue, 27 Feb 2007 09:48:18 +0000

moi tant qu'un soft marche je men fou si c libre, gratos ou quoi

My way to have etch released on time - greg

greg — Tue, 13 Feb 2007 19:37:27 +0000

Nice

Article about apt-build - MAxime

MAxime — Mon, 12 Feb 2007 01:44:12 +0000

THANKS DUDE FOR SUPPORTING ME... BY THE WAY U OWe ME 10000000000 million dollars for the corrections.... By the waymy english powns yours.

DPL election begin - Des

Des — Tue, 06 Feb 2007 00:04:37 +0000

So, Are you going to be a candidate this year?

DeFuBu contest #6 - Nelson

Nelson — Tue, 23 Jan 2007 04:19:05 +0000

Nice to see how people are polite... (#270877)

DeFuBu contest #6 - meuh

meuh — Mon, 22 Jan 2007 17:27:21 +0000

C'est pas chez Debian, mais c'est sympa aussi:

http://trac.gajim.org/ticket/2918

I KID YOU NOT! - eclipse

eclipse — Wed, 17 Jan 2007 00:47:29 +0000

Tres parlante cette image !

Bonjour, je suis anarchiste - zero

zero — Sun, 07 Jan 2007 15:10:15 +0000

Héee béeh! Le logiciel libre pour la competitivité, ou comment gagner plus d'argent avec des chose "gratuites"

apt-get install anarchisme

Where is my Etch - Quelqu'un d'autre

Quelqu'un d'autre — Wed, 13 Dec 2006 13:10:41 +0000

Au fond on est tous un peu artiste!

Where is my Etch - Frederic de Villamil

Frederic de Villamil — Tue, 05 Dec 2006 21:34:48 +0000

Énorme...

Et en plus je vois que tu aimes le go ? Double combo dans le bou goût dis moi

DeFuBu contest #5 - OdyX

OdyX — Fri, 01 Dec 2006 15:46:21 +0000

http://bugs.debian.org/397749 seems good candidate for next contest...

What a real RC bug squasher must have - Alex Medina

Alex Medina — Sat, 11 Nov 2006 19:19:44 +0000

Thanks, you save me a lot of time.

The man who did not know he had an amd64 - Julien

Julien — Thu, 09 Nov 2006 06:51:51 +0000

Hi,

Thanks for your detailed explanations!

It is a really strange feeling...

Cheers, Julien

The man who did not know he had an amd64 - jd

jd — Wed, 08 Nov 2006 21:38:58 +0000

Hello Julien !

You're right, lm means 'long mode', so you're machine is an amd64 too! Welcome to my life!

IA64 are Itanium processors, not Pentium, and are real 64 bits processor AFAIK.

If you have 'lm' flag, it means that you have support fort 64 bits instructions, like AMD processors have.

"amd64" arch is the name of the arch, but it also support em64t processor like we have. Take a look on: http://www.debian.org/ports/amd64/

I am not sure which improvement it can gives to you on a desktop machine, especially that you may have problem with something like proprietary Flash player (it does not exists for other arch than i386). Anyway such limitation can be bypassed by using a i386 chroot (32 bit code (i386) is executable with amd64 architecture).

If you want to know why it's better take a look : http://en.wikipedia.org/wiki/X86-64 or http://fr.wikipedia.org/wiki/AMD64 (french)

Cheers.

The man who did not know he had an amd64 - Julien

Julien — Wed, 08 Nov 2006 20:39:55 +0000

Hi,

Just wondering how what you think is a Pentium 4 can be and amd64? I have just checked my cpuinfo and discovered the same thing on my main machine!

lm means "long mode", ie. 64 bits instructions support, am I right? But how do you know it is amd64 (which I would have considered to be only for AMD processors), and not IA64?

I think I have understood that amd64 is also for EM64T processors. EM64T processors are the newer 64bits architecture from Intel, thus I can guess that any recent processor is amd64 and not EM64T...

I am quite lost with that.

What are the improvements in such cases? Do you think it is really worth "reinstalling" a desktop machine?

Cheers, Julien

[ping] What a real RC bug squasher must have - tempurpedic

tempurpedic — Wed, 01 Nov 2006 11:08:05 +0000

tempurpedic

Wie du tausend "real RC bug" fur eine alte Version von Xen 3 erhalten konntest behauptend, von Firefox zu sein, und behauptend, von einer BBC RSS Zufuhr verwiesen zu werden...

Open Picture - abi

abi — Tue, 31 Oct 2006 14:35:20 +0000

mahahahaha, excellent!

Knud can't - Zorglub

Zorglub — Fri, 27 Oct 2006 14:27:23 +0000

Eh, you missed your chance... AJ was actually sitting on top of a dunk tank in January 2004 in Adelaide during Linux Conf Australia... see for example http://lwn.net/Articles/66669/ for a photo.

You could have paid to throw balls and get him into the tank (I feel obligated to mention that the money was going to a charitable organisation, and it was not Debian -- not even computer-related).

Maybe that's where the name comes from, who knows.

Zorglub

Bonjour, je suis anarchiste - riek

riek — Sat, 21 Oct 2006 19:41:23 +0000

En effet, ne nous égarons pas. La lutte pour le libre se gagne sur le terrain de la compétitivité. Richard Stallman est un visonnaire. La lutte technologique passe par la force de collaboration. La hiérarchie structurelle de l'entreprise est un frein à l'inovation personnel. L'open-souce a de beaux jours devant lui. Il s'agit d'un art du travail en collectivité au delà des clivages politiques. bon week-end à tous

News from the battlefront - Baptiste

Baptiste — Thu, 19 Oct 2006 19:12:46 +0000

It is strange that you never cared about those bugs for so long, and now they become your first priority just to annoy your adversaries. I guess internal politics is a powerfull motivation That being said, I hope the Dunk Bank project keep their scripts around. They can later be made into an automated test suite for Debian. Cheers, BC

News from the battlefront - sukria

sukria — Thu, 19 Oct 2006 14:51:13 +0000

thebluesgnr: did you ever read the Debian Social Contract?

I cannot read on the contract something like "our priority is releasing quickly stable releases" but I can read the following: "Our priorities are our users and free software".

The overall quality of the Debian operating system matters more than getting the release out "on time". I am for quality, even if that's not good for people who are getting paid for quite another target.

The Dunk Bank project is a reminder, in a way.

News from the battlefront - Sam Hocevar

Sam Hocevar — Thu, 19 Oct 2006 14:46:33 +0000

And whoever started the "Let’s release Etch, even with bugs and non-free software" project should be praised, I guess.

News from the battlefront - thebluesgnr

thebluesgnr — Thu, 19 Oct 2006 14:38:08 +0000

Anyone that starts a "Let's delay Etch" project should be kicked from Debian.

Bonjour, je suis anarchiste - Toutoune25

Toutoune25 — Wed, 18 Oct 2006 20:47:16 +0000

Certains utilisent le logiciel libre pour sa performance, d'autres pour sa liberté... « Free software, free society ».

Olivier;

Bonjour, je suis anarchiste - OdyX

OdyX — Wed, 18 Oct 2006 08:03:53 +0000

Glazman dit: «le Logiciel libre, c'est d'abord du logiciel», je pense «le Logiciel Libre, c'est d'abord Libre». On n'est «juste» pas d'accord. Ce sont deux visions (incompatibles ?) différentes du même concept, pas de quoi fouetter un chat...

Bonjour, je suis anarchiste - riek

riek — Wed, 18 Oct 2006 07:38:37 +0000

bonjour, je suis libéral j'adore la souplesse de debian. Mais le fatra politique autour du système me rebute. Je recherche la stabilité et la performance du système. Debian est une preuve que le code ouvert est plus fort car travaillé plus facilement. Aussi le Logiciel Libre est une aubaine pour nos entreprises. Les libéraux ne sont pas tous des vampires. Baissez vos faucilles et vos marteaux. Et chaque jour devenons un peu meilleur comme nos chers Logiciels Libres.

Sparc buildd and howto - jd

jd — Sat, 14 Oct 2006 18:56:55 +0000

Just do it

Sparc buildd and howto - abi

abi — Sat, 14 Oct 2006 18:32:39 +0000

hi,

what about moving this to wiki.debian.org?

Final score for the French BSP - lees

lees — Sat, 14 Oct 2006 14:20:16 +0000

No? Really ...?

Sparc buildd and howto - jd

jd — Sat, 14 Oct 2006 00:45:27 +0000

In fact, that's just a matter of tweaking some configuration files, add others wanna-build databases and add others chroot.

Not too complicated, but do not hesitate to contribute.

Sparc buildd and howto - Alex

Alex — Fri, 13 Oct 2006 23:04:24 +0000

Thanks for the how-to. It is useful and well-written. I wonder, though, if you could expand it to the setup that could build packages for several distributions, not just unstable.

Sparc buildd and howto - Steve Kemp

Steve Kemp — Fri, 13 Oct 2006 15:23:44 +0000

That looks like a great guide. I've tried multiple times to setup a buildd and always failed.

I will try following your recipe this evening and seeing how it works out. If it works I think I'll post a guide too.

Thanks a lot for sharing, more documentation on these things is always a good thing.

My way to have etch released on time - er:k

er:k — Wed, 20 Sep 2006 22:10:11 +0000

Pas mal...

kexec support in initscripts - Chris

Chris — Fri, 15 Sep 2006 18:02:23 +0000

I wrote a little python script called kexec-chooser to make rebooting with kexec a bit easier (on Debian anyway). Check it out here: http://atlee.ca/blog/software/kexec-chooser/

kexec support in initscripts - Nikolaus

Nikolaus — Fri, 15 Sep 2006 15:03:30 +0000

This sounds like a great idea. I must admit that I haven't tested kexec yet since I'm running Stable exclusively. But if it does what it says, this would be really super-cool to have in Etch!

DeFuBu contest #2 - nion

nion — Fri, 25 Aug 2006 23:14:11 +0000

great! Maybe you can bring this as joke part into dwn. Keep on reporting them

The Etch song - Anony Mouse

Anony Mouse — Wed, 16 Aug 2006 18:02:44 +0000

Novell and RedHat fighting about Xen - thebluesgnr

thebluesgnr — Sat, 12 Aug 2006 01:39:01 +0000

According to Red Hat Xen will be stable in 3 or 4 months, when they ship RHEL 5.

Novell and RedHat fighting about Xen - Jon

Jon — Fri, 11 Aug 2006 21:11:33 +0000

I'm glad that Redhat believe it isn't ready yet: Just look at the xen "support" in Fedora Core (4, specifically: maybe they fixed it in 5).

Novell and RedHat fighting about Xen - jd

jd — Fri, 11 Aug 2006 18:44:53 +0000

But there's a difference between claiming that Xen hypervizors from the 3 series will work any Xen patched kernel, and telling that the kernel API/ABI will never be stable.

Because the first statement is wrong in the facts.

Novell and RedHat fighting about Xen - Béranger

Béranger — Fri, 11 Aug 2006 18:40:00 +0000

« Xen can not pretend to be stable, mature, or whatever you want, because it is still not integrated into the Linux kernel tree, and thus far for a good reason: the API/ABI is not yet stable. »

Désolé, Julien, but the kernel API will _never_ be stable, Greg KH dixit. It's stupid to be this way, mais les jeux sont faits, rien ne va plus !

As for Xen, I agree with Red Hat and with you. With a _server_ distro, amateurism is not acceptable, unstability being the worst part of it.

The Etch song - vorador

vorador — Fri, 04 Aug 2006 13:21:57 +0000

wimp !

The Etch song - jd

jd — Wed, 02 Aug 2006 08:05:15 +0000

The Etch song - mike

mike — Wed, 02 Aug 2006 04:18:38 +0000

Can you provide a voice recorded song? :P

Debian Developers, you all suck - jd

jd — Sat, 29 Jul 2006 09:35:37 +0000

It's still possible to downgrade RC bugs which are not really ones.

Debian Developers, you all suck - mike

mike — Sat, 29 Jul 2006 02:09:16 +0000

http://mykey57.free.fr/pub/picture/yousuckroom.jpg

Debian Developers, you all suck - Simon

Simon — Sat, 29 Jul 2006 01:05:42 +0000

I appreciate it is slightly tongue in cheek, but some of the release critical bugs still appear to be not release critical. Some even only occuring on architectures that Etch won't release on.

Perhaps we need /. style moderation, where if your previous bug reports are categorized too highly you lose the karma to place new release critical bugs.

Did I just read Windows Vista will release "when it is ready", those guys at Redmond are learning quick

Debian Developers, you all suck - jd

jd — Fri, 28 Jul 2006 23:45:17 +0000

I said "people like Steinar". Not only Steinar.

I meant, people working on fixing RC bugs.

There's still too many DD working in their corner and increasing the RC count.

Debian Developers, you all suck - Jon

Jon — Fri, 28 Jul 2006 23:40:29 +0000

I think Steinar's work is brilliant and a true inspiration: However, he's not the only person working hard. If you read about the fixes he's done, often it's "fix by NMU using patch in the BTS". Great. Someone else has written the patch. If you are a DD, you can NMU and upload the fix. If you are not, what motivation do you have to work on the bug count, if credit goes to whoever takes your work and uploads it on your behalf?

DeDuBu contest #1 - jd

jd — Tue, 25 Jul 2006 17:35:18 +0000

Thanks, fixed.

DeDuBu contest #1 - alex

alex — Tue, 25 Jul 2006 17:28:29 +0000

Simon didn't report the winner, kingsley did.

The new DeFuBu contest - benj

benj — Sat, 15 Jul 2006 22:51:59 +0000

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=177128 looks like a good candidate to me

Xen.. ACCEPTED - glandium

glandium — Mon, 24 Apr 2006 23:41:23 +0000

Is it normal that the maintainer for xen-3.0 in the new queue is "Debian amd64 Build Daemon" ?

Xen.. ACCEPTED - jd

jd — Mon, 24 Apr 2006 18:17:45 +0000

We will start working on backport now, yes.

For libc6, I'll try to see if this possible. Don't know yet.

Xen.. ACCEPTED - Sven Hartge

Sven Hartge — Mon, 24 Apr 2006 18:14:25 +0000

Great.

Let the backporting begin!

Seriously: Will there be backported Xen packages and libc6-xen-packages available on backports.org?

Xen howto update - soda

soda — Fri, 14 Apr 2006 18:18:38 +0000

Merci beaucoup pour ce tuto. J'avais déjà essayé xen à partir des sources et j'avoue que c'était vraiment pas sexe. Avoir une bonne intégration avec Debian permet ainsi d'avoir quelque chose de simple et maintenu.

\o/ jd

Status of Xen in Debian - Anon

Anon — Wed, 12 Apr 2006 02:57:33 +0000

Thanks for you and your team's hard work with Xen.

php-sqlite3 - jd

jd — Sun, 02 Apr 2006 23:16:55 +0000

It may work, but packages were compiled or sid. I'll probably try to upload backports to backports.org.

php-sqlite3 - Pete again

Pete again — Sun, 02 Apr 2006 23:05:19 +0000

For what it's worth, the binary doesn't need such strict dependencies as listed in the Deb package you've made. The binary's just fine for Sarge. I extracted the /usr/lib/php4/20050606/sqlite3.so from the package with alien --to-tgz, copied sqlite3.so to /usr/lib/php4/20020429, made the necessary changes to /etc/php/apache2/php.ini, and it's all worked a treat :-).

php-sqlite3 - Pete

Pete — Sun, 02 Apr 2006 22:31:30 +0000

A good idea. The install didn't work for me though. Using Stable (Sarge):

dpkg --install php4-sqlite3_0.3-1_i386.deb

Selecting previously deselected package php4-sqlite3. (Reading database ... 34078 files and directories currently installed.) Unpacking php4-sqlite3 (from php4-sqlite3_0.3-1_i386.deb) ... dpkg: dependency problems prevent configuration of php4-sqlite3:

php4-sqlite3 depends on libc6 (>= 2.3.5-1); however:
 Version of libc6 on system is 2.3.2.ds1-22.
php4-sqlite3 depends on libsqlite3-0 (>= 3.2.8); however:
 Version of libsqlite3-0 on system is 3.2.1-1.
php4-sqlite3 depends on phpapi-20051025; however:
 Package phpapi-20051025 is not installed.

dpkg: error processing php4-sqlite3 (--install):

dependency problems - leaving unconfigured

Errors were encountered while processing:

php4-sqlite3

About Xen in Debian - Spike!

Spike! — Tue, 28 Feb 2006 04:21:28 +0000

Wow, that was a long thread. It looks like you guys are ironing out the issues in that thread though. I sort of agree w/ Bastian though...in order for Xen to get its speed, it needs invasive changes in the kernel (not a big surprise). Ideally, Linus will integrate it into the mainline Linux kernel but that won't happen until it's stable enough. In the meantime, there's a big chunk of it that has to go into the kernel and another big chunk that does not (the UI and userspace stuff). The big chunk that has to be integrated into the kernel should be handled by the Debian kernel maintainers.

Thanks for all the work you guys are putting into it. I'm itching to try out the Intel VT and AMD Pacifica support for running Windows on Xen at high speed

lighttpd backports - StefanB

StefanB — Wed, 08 Feb 2006 08:24:32 +0000

Thank you very much! This is exactly what I was waiting for. My small sarge server (a laptop with only 32MiB RAM) also seems to smile because of this.

StefanB

checksecurity - jd

jd — Sun, 22 Jan 2006 15:53:32 +0000

Ouais, mais je trouve que c'est encore un peu trop usine a gaz. Je ne sais plus lequel j'avais tester, celui la ou un autre dans le genre. Mais je verrais, je rejetterais peut etre un coup d'oeil.

checksecurity - neo

neo — Sat, 21 Jan 2006 09:42:42 +0000

Je te conseille Tiger, très bon logiciel.

Au fait, félicitations pour la fin de tes études.

Bad maintainer - yooseong

yooseong — Fri, 09 Dec 2005 06:24:03 +0000

I got shy with your post. I feel very shameful on myself with my package maintainance.

Debian bag - matth

matth — Mon, 04 Jul 2005 00:23:11 +0000

Hi, You'll notice that the logo debian turn the other way

Laptop problem - hello

hello — Sun, 22 May 2005 02:08:58 +0000

hello

my name is ebrahim i have problem with my laptop the screen thing and i dono wut i have to do ?

thanks for helping

Debian bag - Julien Danjou

Julien Danjou — Sat, 07 May 2005 15:32:38 +0000

It was in E. Leclerc in Arconnay (72), France

Debian bag - Florent Bayle

Florent Bayle — Mon, 02 May 2005 21:04:32 +0000

Very nice bag, but... What is the name of you supermarket ?

In the core of apt - Julien Danjou

Julien Danjou — Thu, 17 Feb 2005 17:10:16 +0000

Anonymous coward:

When I worked at Mandrakesoft, I had to use URPMI (the Perl lib), and fortunately the original author Francois Pons explained me how it works, because there is no many comments in the source.

So I can compare, and yes apt rox

In the core of apt - Anonymous coward

Anonymous coward — Thu, 17 Feb 2005 17:04:00 +0000

> APT IS A MAZE !

Ahah. Just for fun, once, take a look at urpmi (Mandrake package (mis)handling tool), and try not to die from a heart attack: returns 0 for some errors, or sometimes returns the same code for different errors, without even talking about Mandrake package naming scheme, which is a terrible mess. So, you can't complain, you're a lucky man.

Laptop problem - Alberto Reyes

Alberto Reyes — Sat, 15 Jan 2005 17:51:17 +0000

Try using another hub/switch you are using...

apt-build 0.10.0 - sod`

sod` — Thu, 06 Jan 2005 20:28:40 +0000

Salut,

Peut-on désormais utiliser apt-build pour builder tout son système ? Par exemple, c'était plus ou moins foireux pour certains paquets critiques auparavant, comme la libc6 et cie ... as tu essayé ? ça fonctionne bien maintenant ? As tu eu des retours d'expérience de personnes utilisant le cvs ?

In the core of apt - sod`

sod` — Sun, 02 Jan 2005 14:13:08 +0000

Tu as l'air de te réinvestir dans le projet apt-build, et à vrai dire, je t'en félicite. Cependant, juste une question et une remarque (qui seront peut-être superflue, vu que je n'ai pas lu le code du programme) : pourquoi n'utilises-tu pas la librairie perl libapt-pkg-perl ? (elle n'est peut etre pas adaptée à ce que tu souhaites faire). Et dans le cas ou elle n'est pas utile, as tu l'intention de faire une librairie perl pour controler la gestion des paquets ? Je pense qu'une telle librairie serait ***vraiment*** utilisée et très utile ...

Sur ces quelques remarques, bon courage (et bonne année)

apt-build revolutions - Lunar

Lunar — Fri, 31 Dec 2004 18:45:35 +0000

More than the optimisation thing, what I find really interesting in apt-build is the patch feature.

Mailman is a good example of a package in which you happen to add a lot of patch. It would be great if apt-build could track patches applied and reapply them during apt-build upgrade...

apt-build revolutions - Steve Kemp

Steve Kemp — Fri, 31 Dec 2004 16:35:09 +0000

PS.

http://www.debtoo.org/

apt-build revolutions - Steve

Steve — Fri, 31 Dec 2004 16:33:32 +0000

Looks like an interesting tool, my solution was "debian-builder" and "wrap-gcc" which is written in perl.

I started so that I could use the SSP protection offered by GCC, but it's very similar to your wrapper and pentium-builder which I think is still in the archive.

Steve

apt-build revolutions - :emaN

:emaN — Fri, 31 Dec 2004 16:02:57 +0000

Yeah, kill Gentoo #_#

Debian stuff - neo

neo — Wed, 15 Dec 2004 08:18:01 +0000

You don't really have to clear your mind, but you have to clear your stomach of all different alcohols you drunk this last days :p

Article about apt-build - artime

artime — Tue, 14 Dec 2004 14:11:26 +0000

Not yet read it. Just posted to indicate that the link to GNU stow's article is broken.

And some comments comparing apt-build with apt-src and apt-fu would be great!

Au relire!

Article about apt-build - Chris Daniel

Chris Daniel — Tue, 14 Dec 2004 06:50:51 +0000

I am liking apt-build, but it's throwing errors about a trailing "-" in the directory name of the pkg. I'll submit a real bug report, I promise... Oh, wait, it's a nonexistent source package. nevermind and thanks!

pure-ftpd with virtual chroot - Swiip

Swiip — Thu, 09 Sep 2004 17:32:03 +0000

Great, it woks perfectly !!!!!! Could you give some help about how rebuild a deb with diff option ? Thank you.

Native Debian packages - Matt Brubeck

Matt Brubeck — Sun, 15 Aug 2004 18:20:27 +0000

The wxwindows2.4 maintainer adds a "nano" version when he makes packaging revisions (e.g. "2.4.2.6" for the sixth package based on upstream 2.4.2), but this is not ideal for many reasons -- among other things, it's hard to tell which part is the upstream version.

Native Debian packages - algernon

algernon — Sat, 14 Aug 2004 21:15:06 +0000

You can fix it without an epoch if no .orig.tar.gz was ever uploaded (for that version, of course). Simply grab the upstream tarball, create a version that is not debian native, and upload it. Since native packages have foo_1.0-1.tar.gz, and non-native ones have foo_1.0.orig.tar.gz, there will be no filename clash.

If there ever was an orig.tar.gz for that particular version, that's a harder case.

pure-ftpd with virtual chroot - neo

neo — Fri, 25 Jun 2004 14:26:01 +0000

Hi acid,
Thanks a lot for your package, it is very useful.

I hope the maintainer will understand he made a mistake and correct this soon ...

Laptop and roaming - Josh Triplett

Josh Triplett — Thu, 24 Jun 2004 17:45:36 +0000

Personally, I use waproamd and ifplugd, to manage the wireless and wired interfaces respectively. With that, the wireless interface automatically finds a usable network (using a WEP key if I've configured one) and the wired interface comes up as soon as I plug in the cable. In my case, all of my networks use DHCP, so I don't need anything more than that. If you do need different configurations for different networks, use ifupdown-roam, available at panopticon.csustan.edu/th... .

Laptop and roaming - dam

dam — Wed, 23 Jun 2004 13:39:02 +0000

I looked at that recently, the mos usefull configuration in debian is ifupdown + ifplugd + resolvconf + ifuproam

ifuproam can be found at panopticon.csustan.edu/th... and is created specificaly for debian

Laptop and roaming - Wouter Verhelst

Wouter Verhelst — Wed, 23 Jun 2004 09:40:47 +0000

I don't do roaming; I hardly ever use my laptop on networks where I can't change firewall configurations. Thus, what I do is make sure the network fits right in, rather than try to change the laptop. That's easy:

* I hijack SMTP connections on smarthosts, so that the laptop can think it connects to remote hosts, although it really connects to the local smarthost, whichever that one is.
* Same goes for transparent proxies, where proxies are required.
* Stuff that cannot go to random servers (NNTP, IMAP) is done through some SSH tunnels which I configured in ~/.ssh/config; my mailclient and leafnode setup think their "servers" are localhost:1143 and localhost:1119, respectively.

Laptop and roaming - Ross

Ross — Tue, 22 Jun 2004 20:54:27 +0000

hotplug + ifplugd + guessnet + resolvconf + masqmail gives me automatic interfaces and correct SMTP routing wherever I am, whenever I insert my wifi card or plug in an ethernet cable.

It even uses /etc/network/interfaces as the configuration file, so everything is transparent.

[ping] Explanation for my choice about the latest general resolution - slef-reflection

slef-reflection — Tue, 22 Jun 2004 20:09:35 +0000

Debian GR voting

A few people have expressed interested in how people vote and...

[ping] Laptop and roaming - slef-reflection

slef-reflection — Tue, 22 Jun 2004 20:09:23 +0000

Laptop and roaming

Julien wonders what to config network on a laptop....

[ping] Explanation for my choice about the latest general resolution - slef-reflection

slef-reflection — Tue, 22 Jun 2004 20:06:59 +0000

Laptop and roaming

Julien wonders what to config network on a laptop....

Laptop and roaming - jorge

jorge — Tue, 22 Jun 2004 15:49:12 +0000

apt-get install laptop-net

Laptop and roaming - fast

fast — Tue, 22 Jun 2004 14:07:56 +0000

use laptop-net.
This nice tool uses arp-requests to look where you are...

Explanation for my choice about the latest general resolution - Antti-Juhani Kaijanaho

Antti-Juhani Kaijanaho — Sat, 19 Jun 2004 13:41:40 +0000

Thank you

For some reason, the ballot is not reproduced by Planet Debian.

(I hope I don't mess anything up here. I don't understand the French language instructions in this comment form.

libparportled support for AMD 64 - soda

soda — Mon, 07 Jun 2004 12:23:58 +0000

Je tiens tout de même à souligner le caractère important de ce portage ! En effet, quoi de plus utile que de faire ressembler sa machine à K2000 ? Ou encore, de la décorer pour en faire un arbre de Noel ?
Bref, merci Julien, grâce à toi, les richissimes possesseur d'AMD64 ne seront pas mis de coté. Comme on nous le démontre tous les jours, c'est ça le libre : l'ouverture à tout le monde de logiciels aussi stupides qu'inutiles

NB : les smileys automatiques, ça fait tout pourri quand même