Some (not so) short news from the samba packaging team... Recent package uploads:

• Last "stable" upstream branch: 3.3.4. 3.3.4-2 is in unstable, quietly ageing before entering testing. 3.3.4-1 is in testing. In short, we have the latest upstream in squeeze
• Next upstream version: 3.4.0~pre1 is in experimental. That includes many packaging changes as upstream is now providing samba3 and samba4 altogether, which lead to source tree reorganization. Good news: there are no more "FHS" patches and upstream included needed changes for the file layout to finally be able to cope with the FHS entirely.
• We maintain many backports:
  • 3.3.4 was just uploaded to lenny-backports on backports.org. Maybe soon to come a 3.3.4 upload for etch-backports
  • unofficial backports on http://pkg-samba.alioth.debian.org: here we will maintain "the latest upstream version of the same upstream branch than what we have in Debian stable suites". That seems complicated to say and an example says it better: lenny has 3.2.5, so this repository has upstream 3.2.11 with the same packaging than 3.2.5. We made an exception for etch: instead of backporting the latest 3.0 upstream (etch has 3.0.24, upstream has 3.0.34), we backported the latest 3.2: we consider that nobody is now seriously considering to run 3.0, except those people who are very happy with the official packages in etch.
  • All these backports will be useful when tracking down bugs and ask our users if they reproduce them with the latest upstream versions, as samba setups are often very tricky to reproduce in our own environments.

Other things are coming, such as the work on having samba3 and samba4 packages coexist. Jelmer Vernooij prepared everything after our discussions at the SambaXP conference and it should be uploaded "soon". A few patches we have should be integrated upstream as well, therefore making the Debian diff smaller and smaller. Also, Luk Claes joined the maintenance team and helped setting up the unofficial repository for backports. The bug count is going up again these days as I don't spend much time triaging the bugs and some are very tricky to reproduce. As usual, help would be welcomed. I'm fairly sure that several of these are user errors but I often lack time to prove this enough for closing the bug. We should go to to 50-60 bugs or so.

Robert McQueen reported bug #530000 on Friday May 22th, against the dbus package. Bug #520000 was reported as of March 16th 2009. We are now close again to our "10,000 bugs in 2 months" pace. So, bug #540000 should be reported during Debconf 9, as it seems...

Thanks, CRM114. From yesterday morning (where I ran my daily dist-upgrade) to now, you just *trashed* all my legitimate mail. Oh, sure, I got warned in some way by NEWS.Debian entries, but I didn'tfigure out that it was URGENT that I change my *.css files. What was happening is, in short, that all non whitelisted mail was just...reduced to a blank line. And I didn't figure out immediately because I have several whitelist entries, so I was still receiving *some* mails. So, it is not only "stop working"....it is doing much harm. If you're a CRM114 user and haven't upgraded to 20090423-1, you've been warned. I'm currently considering what to do and if a bug report is warranted (RC, of course). Oh, for sure, there *is* a warning through NEWS.Debian but is breaking existing behaviour so badly really something that we want in Debian? Losing legitimate mail is the worse thing a program can do, isn't it?

Now, I'm registered for Cologne Marathon. So is No l and probably Ralf Treinen. Any other DD or Debian contributors wanting to join in? I know that Dirk Eddelbuettel will probably not (btw, nice run at Boston, Dirk)....but others certainly can...and should ! October 4th, 2009 is still far enough for you guys to plan this..:-)

Just found by my son who's as crazy as I am abut SpongeBob Square Pants:

In case you last tapping with your touchpad with the latest X.org update in sid, you might want to read bug #497523 or just add the following to your xorg.conf file:

        Option          "TapButton1"            "1"
        Option          "TapButton2"            "2"
        Option          "TapButton3"            "3"

in the InputDevice section that's relevant for your touchpad. ...and good luck exploring the joy of X.org updates (I am currently back in time fighting again with a jerky and unreliable Suspend-to-RAM which sometimes freezes my display...or just exits the KDE session...or sometimes just works....).

Just like Wouter, I had no real preference among the two DPL candidates. I have respect and good opinion of both Zack and Sledge and I had no real way to make a difference between both of them. Still, I voted (112 or 11-, which makes no real differences). That goes with my personal feeling about right to vote, which I always used in my life as I consider it to be one of the most important rights we have, overall. No intent to put shame on Wouter, of course. The right to not vote is also very important and he even doesn't have it in his country. Congratulations to Steve and many thanks to Zack for running. That DPL election was a quiet one (too quiet?) but maybe Debian has reached its maturity by being close to 16 now? :-)

So, following the recent announcement that kde4 was uploaded to unstable, and following the positive feedback from Christian Perrier, I decided to give kde 4 a try.. Ironically, I am exactly in the same context as when I opened this blog, to complain about the state of kde 4 back at that time... So, I am quite satisfied by the current state. Clearly, it is usable, and most of the criticisms I had in the previous try are resolved now. Only the IRC plugin is still missing for Kopete. To me (...) - Linux / debian

I finally more or less completed my switch to KDE4. Thanks the great great KDE maintenance team we have. Simultaneously, X.org packages were updated so possible problems may have ben triggerred by it more than KDE, though. At first glance, this new KDE is nice and has great improvements but generally speaking painfully slow..:-) So, I began deactivating again all visual effects so that my poor old Dell X1 survives the upgrade. I was also greatly annoyed by losing the "Terminal sessions" which I was very actively using with Konsole. No more "Terminal Sessions" menu in my task bar to connect to various hosts over the network. Bleh. I have to redo everything with Konsole's "profiles". Also sucking a lot is the default setting for fonts in Konsole (antialiased fonts). It makes scrolling painfully slooooooow. I just deactivated this and it's much much much better. I also lost my old mail notification utility (Korn). It is still working but having it running leaves a kio_mbox process eating all my CPU. Actually, korn is still something from KDE3, so I might expect it to be rebuilt for KDE4, hopefully. I never found anything as good as it in the gazillion of mail notification utilities we have. Maybe my dear Lazyweb will help. I'm still left with weird keyboard behaviour which is hard to identify. Apparently the delay before key repetition starts when one leaves a finger on a key has changed in some way and despite my attempts to restore this in the adequate KDE thing, it doesn't change that much. I suspect some weird interaction with X.org stuff. Such "minor" details are very annoying as they interfere with year of (bad) typing habits. In general, as a very old KDE user (I adopted it back in 2001 or so because of Kbabel), I'm somewhat happy with that upgrade but it will take time to be used to all the new concepts (ah, these "Plasmoids" thing...) and also restore the overall speed of my machine...:-)

Done. I'm right here now sitting on my couch and trying to recover my legs after that first marathon ever. As planned, I met Ralf Treinen and No l K the at the start of the Paris Marathon. As No l was targetting 3h30m while Ralf and I were targetting 4h, we left him go with the faster runners. As No l doesn't blog that much, I think I can say it now: he managed to run in 3h28m, which is awesome. He is incredibly happy with that (we unfortunately didn't manage to meet again after the race as the crowd was too huge and the cellphones were of course miserably failing). Ralf and I ran together the first 12 kilometers. Then we split because he had to stop for a few seconds. Up to the 21st kilometer, I kept the pace for 4h and I was even running a little bit faster. However, this advance vanished slowly from km 25, up to km 35 where the virtual runner that was running 4h passed me..:-). Ralf also passed me around km 38 which is indeed the moment where I was blowing up, running slower and slower. The last 3 kilometers were a nightmare but, still, I managed to finish and my GPS watch tells me I ran for 4h10m. The official time is 4h12m56s, because of a huge "runner jam" at Bastille where we had to wait for 3 minutes before being able to enter the Faubourg Saint-Antoine street. Running through Paris is really magic, particularly with the nice sun we had ( not *too* hot, though) and I am really proud and happy to have my first marathon finished *without walking*. Yay.

Yipee! Just the day before my first marathon, with No l K the (who's targetting 3h30, crazy No l) and Ralf Treinen (who's targetting 4h), I got a Garmin Forerunner 405 Sport GPS device as birthday gift. So, I'll apparently be able to enter the modern world and adapt my pace more easily than earlier (with my mechanical podometer and the stopwatch function of my cell phone). And that thing has even a Debian package, maintained by...Ralf and No l..:-) ...though I have no idea whether it works with a Forerunner 405 as the original author apparently owns a 305. We'll see. If you somehow happen to see blog posts of mine with the GPS track of the Paris Marathon, I made it work (or I cheated and used a Windows machine...).

April 2nd: time for me to turn 30. Yay, I'm grown up.

Mark Purcell reported bug #520000 on Monday March 16th, against the gsm-utils package. Bug #510000 was reported as of Dec 28th 2008. So, it took us 2.5 months for 10,000 bugs. This is still slower than our rate of 10,000 bugs/2 months but more than the pace we had for 500000-510000. Of course, in the meantime, we released lenny on Feb 14th and, therefore, the development started again and, with development, new bugs come in. To illustrate this, one will notice that bug #515000 was reported on Feb 12th. So, it took 46 days for the first 5,000 bugs and then only 32 for the remainder 5,000. In short, we're probably back to our pace of 10,000 bugs in two months. In case someone cares, the first bug reported after the release of Lenny is #515247, reported by Robin Kluth against the gvfs package. It followed the release of lenny by 25 minutes and 14 seconds. That bug isn't fixed yet but is already reported upstream.

The first version of samba packages (2:3.3.1-1) that is different from the version in lenny (2:3.2.5-4) entered testing this week-end. The first update of samba packages that's aimed for lenny has also been accepted for the next point release of Lenny, through stabe-proposed-updates. This is version 2:3.2.5-4lenny1. It fixes two important bugs discovered after the release of lenny that potentially affect many users and were fixed upstream in post-3.2.5 releases In the constant tracking of a very active upstream, I also uploaded packages for 3.3.2 in unstable. That was quite a straightforward process, of course helped by Steve Langasek, who is a constant active support in the maintenance of samba packages. I really like the team we're having now, with Steve being very picky and motivated by quality and /me doing most of the nasty repetitive work to try saving his time for more important tasks..:-) The next plan is updating packages on backports.org. Having 3.3.1 packages in testing now allows us to upload them in lenny-backports which should be good for lenny users who want to track upstream. Finally, my proposal for a talk about "Packaging Samba for Debian and derivatives" has been accepted at Samba XP, the annual conference of Samba developers and users in G ttingen, Germany. For the first time, I'll be a speaker at this conference which I attend yearly since...it exists. My "german pilgrimage" of the year in Northern/Central Germany... Ah, and I'm still preparing for the Paris Marathon, in 3 weeks now, which I will run with Ralf Treinen and No l K the. I feel in good shape for this as, this week-end, while I promised myself to "not run for too long", I just ran 19km and felt this was a quite "short" run. Still not in the performances of Dirk who mentions that 1h36 for a semi is "good for someone who didn't train that much during winter"...but I think I'll never reach that level of performance. I should have started running long distance 20 years ago for this, I'm afraid.

Apparently, the sync magic I'm using between my laptop (where I usually write my blog entries) and my home server (which serves them with Blosxom) was broken since about 3 weeks. So, dear readers, sorry but you'll have to see old entries which never made it to my blog. Apologies.....old rants are not very useful, but I decided to leave them..:-)

Yesterday evening, I registered for Debconf9 in C ceres. I also registered my first proposed event: the now traditional cheese and wine party. Being in Spain with no regulations on food import for the traditional cheese providers in the Debconf crowd, I promise that this cheese party will be E.V.I.L. And Debconf 9 will be a great success, of course. See you in C ceres!

I was not in a work mood tonight and then decided to lose my time reading about timezones (thanks to a Chinese user who wants to add Beijing to the list of timezones for China...which has only one, including the colonies of Tibet and Xinjiang). The list given on Wikipedia is a completelt hilarious time waster, indeed. My favourite one is probably the mention then some parts of British Columbia (Canada) are in UTC-7, namely Most of Peace River Regional District except Fort Ware, Pink Mountain as well as Regional District of East Kootenay and finally (take a breath!) Regional District of Central Kootenay east of the Kootenay River and parts east of Kootenay Lake that are south of and including Riondel (Creston doesn't observe DST), and Columbia-Shuswap Regional District east of the Selkirk Mountains. Canada timezones are really a gem, where one learns that Northwest Territories (one of the largest states) follows UTC-7 except a town named Tungsten and.....a tungten mine. Some other gems in that page are the places that leave in the future at UTC+13 and UTC+14, in Tonga and Kiribati. A other one? Arizona is also funny with the entire state being UTF-7, but the Navajo nation not observing DST, but the Hopi reservation (that's an enclave of the Navajo nation) observing it. That gives an interesting nightmare when travelling over there in summer. And finally, living in Florida probably requires a very high intelligence to understand what time it can be when "east of the Apalachicola River, plus the portions of Franklin County and Gulf County south of the Intracoastal Waterway, west of the Apalachicola River" uses UTC-5 while "west of the Apalachicola River, except for the portions of Franklin County and Gulf County south of the Intracoastal Waterway" use UTC-6. There are other such hilarious thingies on that page. I hope you'll enjoy it as much as I did.

37 years since I first heard Peter Gabriel's voice (Genesis' Nursey Crime, one of the first LP I ever bought) and about 34 since I first saw him on stage in what turned out to be Genesis' last show in France (St-Etienne, my home city, may 1975). Since then, I tried to keep up with PG's appearances here and there and he has always been my favourite artist. So, if any of this blog's readers happen to be in Latin America and enjoy PG's music, just try to go there:

• March 20 Lima, Peru / Esplanada Del Monumental
• March 22 Buenos Aires, Argentina / Estadio Velez Sarsfield
• March 24 Santiago, Chile / Arena Santiago
• March 27 Mexico City / Distrito Federal
• March 29 Guadalajara, Mexico / Jalisco
• March 31 Monterrey, Mexico / Nuevo Leon

No, Debian folks, not our former DPL...:-) My, by far, favourite skipper in the greatest race around the world (alone on a 60-feet boat without assistance), the Vend e Globe, arrived last night. I mean Samantha Davies, 34 y.o. female british skipper, who finished 3rd (or 4th, depending on Marc Guillemot's arrival time), who illuminated these weeks with her videos and messages always full of positive attitude and happiness....which is certainly harder to do when surfing in the southern latitudes over 20 knots and thousands of miles away from any land. I recommend those of you who never went on the VG web site to just navigate around the videos and spot those sent by Sam, to understand. She made it with a 9-year old boat (that completes its third VG), keeping up with performance while preserving the boat (probably the one in the best condition after arrival). and all Vend e Globe fans will always remember her constant smile. All this explains why I have even more admiration for Sam's performance than I have for Michel Desjoyaux (the winner) or Armel Le Cl ac'h (2nd), both among the world's top skippers (along with Lo ck Peyron, imho). The race is not over, far from this. Eight skippers still have to arrive. Three of them (Marc Guillemot, Brian Thompson and Dee Caffari) will arrive on Monday or so, while Arnaud Boissi res and Steve White wtill need one or two weeks and Rich Wilson, Rapha l Dinelli and Norbert Sedlacek still are in the southern Atlantic ocean. And kudos also have to go to the 21 skippers who couldn't complete the race, certainly the most difficult Vend e Globe ever. I wish I would live close to Sables d'Olonne to have a chance to see a VG arrival some day. Great race, great adventure. Sad that it happes only every 4 years..:-)

Suspense..... Will we release Lenny today? :-) Whatever happens, I like the way these last two weeks were handled. The clear communication from the release team made it clear to everybody what the deadlines were. That helped a lot in cooking up good Release Notes and the great Release Annoucencement, as well as their translations. Of course, one can regret that all had to happen in the very last weeks (the RN could have been prepared earlier but nobody was in that mood except certainly Martin Borgert who is trying to shake up the tree for months...). Again, great work, everybody and I wish the best to the folks who will burn their week-end up to polish the last bits of that release.