jd:/dev/blog - Free Software

Upgrading MediaWiki from PostgreSQL 8.2 to 8.3

jd — Mon, 19 May 2008 11:56:00 +0200

With my server migration I switched from PostgreSQL 8.2 to 8.3. However, I've installed MediaWiki for the awesome wiki on 8.2, but dumping from 8.2 and restoring to 8.3 just failed because of the tsearch2 extension, which is now part of PostgreSQL 8.3.

With the help of my friend Dimitri, the pgloader author, I managed to migrate the database from PostgreSQL 8.2 to 8.3. Here is how I did.

This was done with MediaWiki 1.12, and I doubt that older version supports PostgreSQL 8.3.

Dump the database schema

First, you need to dump the database schema, and then, the database data. We need to do it in two steps because we need to change the schema a bit. I assume that MediaWiki is installed in the mediawiki schema:

% pg_dump -s -n mediawiki wikidb > wikidb-schema.sql

Convert the schema

Then, edit your wikidb-schema.sql. When you installed tsearch2, you probably did it in the public schema like I did. So it created two or three data types used by MediaWiki wich are named public.datatype. These tsearch2 datatypes are now part of PostgreSQL, so they're in the pg_catalog schema. You just need to replace a few lines with public.datatype with pg_catalog.datatype. For example public.tsvector is now pg_catalog.tsvector. For your information gin_tsvector_ops is now pg_catalog.tsvector_ops.

Finished? Good, but that's not all. You need to split up the file in two parts: one with the tables creation, and another one with the constraints creation. Why? Because otherwise the data insertion will fail because of these constraints. That's not so hard, just split your files in wikidb-schema.sql with all CREATE TABLE and so one, and then you see that the SQL file begins to do some ADD CONSTRAINTS, just put this in another file, named wikidb-constraints-schema.sql. You'll also need to set the namespace before executing command in your second file, so do not forget to add a SET search_path = mediawiki, pg_catalog; line at the beginning.

Dump the database data

That's easy:

% pg_dump -a -n mediawiki wikidb > wikidb-data.sql

Create your new database

That's easy too:

% createuser -S -D -R -P -E wikiuser #(then enter the password)
% createdb -O wikiuser wikidb
% createlang plpgsql wikidb

Restore the database

That's even easier, but you need to respect this order, otherwise it will fail:

# Restore schema first
% psql wikidb < wikidb-schema.sql
# Then restore dat
% psql wikidb < wikidb-data.sql
# Then restore constraints
% psql wikidb < wikidb-constraints-schema.sql

Run update.php

Just go the the maintenance directory. Then you can try to run update.php, but it'll fail.

% php update.php
[...]
Could not open "/var/www/awesome.naquadah.org/wiki/maintenance/postgres/archives/patch-tsearch2funcs.sql".

Well it seems that MediaWiki people just forgot this file in the 1.12 archive.. You can grab it on the SVN repository here. When you've put it in the good directory, really run:

% php update.php
[...]

You should see no error. And now it works.

Give us some kudos

This is the last and hardest part of this tutorial. Since everything works, you should give kudos to Dimitri and me.

OpenArena, comment que ca m'éclate

jd — Sun, 04 May 2008 09:39:00 +0200

C'est à dire que j'avais un peu oublié Quake 3 Arena, tout ca, c'était du passé vous voyez. Et puis voilà que je tombe sur OpenArena, le jeu qu'il est forké de Quake 3 et qu'il est bien.

Alors OpenArena, c'est quoi ? C'est simple, ca ressemble à Quake 3, ca sent comme Quake 3, mais... c'est Quake 3 !@!#!

Alors évidemment, les puristes diront sûrement que non, mais en fait si. La seule différence, c'est que tous les graphismes, les joueurs, les objets, les cartes, ont été refaits à la main par les auteurs, étant donné que tout ceci n'a pas été liberé par ID Software.

Conclusion ? Ca fait juste un Quake 3 en beaucoup plus moche. Certains objects sont três bien fait (le Quad, le shotgun, le mega-health, les drapeaux, …), mais le reste est globalement vraiment pas beau. Les cartes ne sont pas sublimes non plus, ca manque de textures, de détails, etc.

Ceci dit, à part ca, ca reste du bon gros Quake, donc j'aime. Allez, je me suis même fait plaisir, j'ai jouer en tant que C4 pour défendre l'honneur, tout en mettant une claque aux gentils admins de TuxFamily qui ont leurs propres serveurs où vous pourrez peut-être me croiser quand je me détends !

Updating my .plan

jd — Fri, 18 Apr 2008 12:46:00 +0200

I've been on holidays for one week now, breaking my usual workflow. Well, that allowed me to rest and to think about what I'd like to do and things I need to handle during next weeks.

The urgent things this next days will be my primary server replacement. It is currently dying, and I already had to change its power supply twice in a month. Unfortunately, I'm now at a point where I do not have any spare piece so if things go wrong again, I'm screwed. I need to collect some money and buy a new server, or maybe get a server if someone have an old or spare one to give me, I do not know yet.

On the awesome front, I'm about to release awesome 2.3, which will be the final minor release of the major branch 2. This will lead me to work on awesome 3 at a slower rate and a cooler pace.

Then, the thing I do not have to hurry for is awesome 3. There's no big problems in awesome 2, and the xcb-util stuff are not stabilized yet. After my gentle yelling on XCB mailing list, it seems that things will move but will slowly. So I do have time to make things right and do what I want on that branch, making bugfix release on awesome 2 if needed. You can read more about futur on my last post about awesome.

All this should give me some more spare time to work on the upcoming Debian release, lenny, which I'd like to work on. Two years ago (my god, time flies), we've done good work with the french cabal squashing critical bugs and I'd like to go back on this and squash asses again.

Moving old projects to git

jd — Sat, 26 Jan 2008 10:02:00 +0100

I finally did it.

Most of my small and personal projects (telak, sysrqd and mod_defensible) were maintained in my home-svn-repository. This had the side effect that I did not have a good overview of the current status of the project, and that I never though about tagging releases.

I used git-svnimport to move this stuff to git, with great success. Now, I see that only one release on three is tagged correctly, bad me. They are now on my git server.

But with a better tool like git I'm sure I will be more precise when I will work on that source code, so it was time to switch. And now my source code is public, which is far better than before.

Generating documentation from C source code

jd — Wed, 16 Jan 2008 08:18:00 +0100

Dear Lazy Web,

I'm looking for a tool to extract and generate user documentation from C source code. No, I don't want doxygen which I already use, but something that can extract comments from C source and print them to a text file. The ultimate goal is to generate manpages (using asciidoc).

Anything that may look like that, a program, a library or a trained poney might interest me.

Fruit Fly in the box

jd — Tue, 11 Dec 2007 15:22:00 +0100

awesome 2.0 just released. Great.

Someone on #awesome pointed me to a post of Don Stewart comparing the rise of the new tiled window managers. This is quite encouraging, and I hope awesome will grow more with the next releases!

git statistics?

jd — Tue, 02 Oct 2007 15:43:00 +0200

I wonder if there are tools to do pretty git statistics, something like that:

(done on linux-2.6 git repository).

I just wrote a quick'n dirty Python script to do that, but I'm too lazy to write a full program for now…

Random news

jd — Tue, 25 Sep 2007 20:18:00 +0200

Wouhou, yesterday was quite a productive day.

First, I've released mod_defensible 1.4 which is just one line different from 1.3 but fix a huge bug (RC in Debian) with udns 0.0.9. Already uploaded to sid.

Second, I've updated greylistd in Debian and I've set myself as maintainer after 5 NMU in something like 1 year.

Then, I've pushed some updates to awesome. I may release 1.2 in the week, since it seems bug free and has some useful new features. We're now quite close to Xmonad functionalities as far as I can tell.

And the beat goes on…

Is X.org 7.3 a sucking release?

jd — Fri, 21 Sep 2007 16:37:00 +0200

I just upgraded my X server at work. Oh my god. I had a working configuration with multi-head (without Xinerama).

I was happy. Oh yes I remember yesterday everything was so fine. And now it screwed everything up.

I've got 3 LCD screens and 2 Matrox G450 PCI cards. After restarting X.Org 7.3, all I've got was two screens (connected on the first card) in clone mode. Very useful.... After 1 hour of googling I managed to use xrandr to setup a Xinerama display on this 2 screens....

I don't want Xinerama and I want my other Matrox card back. I don't know why but reading Brice Goglin post around, I understand that the old behaviour is no more available. GREAT, now I'm stuck with Xinerama that a lot of window managers don't handle correctly and I still don't know why my other Matrox card doesn't work.

Is X.Org 7.3 a piece of shit or what?

Sorry dwm

jd — Sat, 15 Sep 2007 23:37:00 +0200

This time I L O L.

Window manager sucks

jd — Fri, 17 Aug 2007 00:06:00 +0200

Now my fvwm is segfaulting almost at each startup.

Seriously, there's great software everywhere, but still no good window manager (I know, I tested almost them all).

Life sucks.

On media players, xmms and audacious

jd — Fri, 10 Aug 2007 11:59:00 +0200

Last month the new thread I was afraid about has started. Called Considerations for 'xmms' removal from Debian, the maintainers of the xmms package proposed the removal of it.

I used xmms (and, before, x11amp) for something like 9 or 10 years. I know, I love it. I can't bear other players. Amarok is heavy, rhythmbox seems like a joke to me. I like to load my 6500 files playlist in one shot, press 'j' and jump to the song I want. I don't see how it could be easier to use a music player.

Following the thread, I must admit that, yes, xmms is old, gtk1.2, unmaintained, etc. I had to switch, sooner is better.

So I use now Audacious. At least I try to. Last time I tested it, the search (jump, 'j' key) function was… not working. No it is, it works like xmms, and I'm happy. I know that upstream author of audacious dislikes being compared to xmms, but, screw you, you're like xmms, and I like it, so I like audacious for this. The plugins and the interface is like xmms, so it's great also.

However it's buggyyyyyyy. Third day of use, serious bug reported as #436557. It segfaults sometimes, or simply blocks reading a file.

The information loading 'on demand' is so crapy compared to xmms. I mean, it's fucking slow! When I put shuffle on and I lick 'next song please', then it hangs reading informations. xmms was doing what in a better way.

However, dear audacious-author-that-already-hates-me-for-comparing-xmms-and-audacious, I really think it'll be a real alternative for old xmms users like me and a lot I know (yes, I know a lot of people, damn it!). Developement seems to be very active, so I'll continue to report bug and use it.

Yes, I believe in audacious.

Evolution, you can't pass through

jd — Tue, 07 Aug 2007 08:34:00 +0200

I've switched from xmms to Audacious.

The end of 10 years of love.

At least I can use last.fm again…

Why you won't use MySQL/ndb

jd — Sun, 17 Jun 2007 11:25:00 +0200

I just failed to migrate a big project at work because of that, the D day:

Jun 15 10:46:39 hostname mysqld[20095]: Version:'5.0.32-Debian_7etch1-log'  socket: '/var/run/mysqld/mysqld.sock'  port: 3306  Debian etch distribution
Jun 15 10:46:43 hostname mysqld[20095]: mysqld got signal 11

So no, you won't use NDB, at least with MySQL 5.0.31 (included in etch). It's _so_ buggy that we get a segfault on the first DELETE statement with a 30 MB database.

Backporting version 5.0.41 from sid gives better results (at least for now).

I knew we should have used PostgreSQL from the beginning...

mod_defensible 1.2

jd — Thu, 01 Feb 2007 16:51:58 +0000

I just released mod_defensible 1.2 and uploaded it to Debian. Just have to wait that the NEW queue is processed.

In the meantime packages are available here.

mod_defensible 1.1

jd — Fri, 19 Jan 2007 16:43:00 +0000

Hopla, a new version of mod_defensible. I added a custom page in order to explain to the client why it is forbidden to access to the page.

In the mean time, my webserver load dropped to somewhat 20% less than without this module.

sysrqd 9

jd — Fri, 19 Jan 2007 15:37:55 +0000

I just released sysrqd 9th version. This version include a little page for Neil McGovern: never go in swap and allow to bind to a specific IP address.

Kicking out Web spammers with DNSBL

jd — Mon, 15 Jan 2007 11:33:00 +0000

Every project has its story. Every war has its winner, and its casualties. They were 20 millions men, fighting for their freedom.

And you'll never know their story.

Because during last week, I was looking why my Web server was so heavily loaded. And I discovered that my blog was attacked by spammers trying to post comments. They were stopped by a great plug-in named spamplemousse, which use spam keywords and DNSBL to drop spam comments.
However, this plug-in is written in PHP, like the rest of my blog, so it loads Apache and MySQL in a way that is no more acceptable: the page have still to be rendered for this !@#$ spammers.

Consequently, I decided to write a Apache 2.x module which will just drop a 403 Forbidden error page in the spammers' head using DNSBL servers. Here it is, and it is called mod_defensible.

I'm using it since 3 days now, and I got some pretty interesting result and less load on my Web server, so c'est tout bon^[1].

Notes

[1] It's all good

Easter-eggs sur Demain

jd — Sun, 17 Dec 2006 16:33:21 +0000

Durant tout le mois de décembre, la chaîne Demain diffuse dans le cadre de l'émission Label entreprise une interview du gérant d'Easter-eggs, Pierre-Yves Dillard.

Et comme c'est là que je travaille, et que c'est intéressant de découvrir le fonctionnement un peu particulier de la SSLL Easter-eggs, j'ai mis la vidéo à disposition ici.

Les Journées Moisies du Logiciels Libres

jd — Wed, 15 Nov 2006 13:49:37 +0000

Vous vous souvenez de Jimmy Pierre ? Mais si, je vous en avais parlé déjà ici.

Et bien après quelques soubresaut, le voilà de retour avec la JMLL, qu'il s'est permis ~~d'annoncer~~de spammer sur plusieurs listes de diffusions. Mais non, ce ne sont pas les Journées Mondiale du Logiciel Libre organisée tout les ans par l'excellente association ALDIL, mais la Journée Magistrale du Logiciel Libre organisé par le célèbre LUG auto-proclamé NUI.

Je souhaite un franc échec à cette évenement pathétique et grotesque, mais néanmoins amusant.