Developers’ Weblog

Sponsored by
HostEurope Logo

Developers’ Weblog

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

catching up the TODO

01.01.2010 by tg@

Wow. Cleaned up mksh, did NMUs for T&S 1, tested nwcc, validated mksh-current (what will be the base of R39b) on HP-UX, but there’s still a lot of things I need to do. A new MirBSD snapshot awaits, more ports fixes, new LLVM, I’ve got to fix the typos I found, and much more. And it’s already this late. At least I feel like I’ve achieved somewhat much ☺ I somehow need to put an order or priorities to tasks.

The 26C3 gang has also returned, with delays that seem funny to those not involved… and tonight snow came back as well. Benny still works on the pkgtools configuration stuff. I wonder how much effort it would be to patch both pkgtools and apt to play together optionally… we wouldn’t lose that what pkgtools already have as high-level functionality, gain some of dpkg’s low-level functionality (must be reimplemented though), and for these who want, a (GPL’d…) high-level tool for binary package management.

FreeWRT is also slowly gaining a little more traction, even from its founder (who asserts that FreeWRT 1.0-stable is the only OS for WRT-like devices that does WDS correctly, and feeds us patches). I still think we need MirWRT ☺ with a ports-like instead of buildroot-based system, but will never have enough time and interest to do that…

And: our website source code is in fact superiour to many others ☻ and quite extensible and flexible…


01.01.2010 by tg@
Tags: debian

My first RC Bug-squashing Day (or rather night):

  • bug #552791 – acorn-fdisk – Copyright file does not contain verbatim copy of the license or a pointer to one
  • bug #562647 – gidentd – Does not work with ipv4 after recent change in netbase
  • bug #558812 – dietlibc – incorrect license in debian/copyright
  • bug #531937 – autossh – FTBFS on mipsel due to missing -fPIC

I picked all of them mostly randomly from the list Zack gave me, and except dietlibc they are packages I had not even heard of before. The first one begun easily (track down licencing information, pimp debian/copyright, but I ended up fixing compiler and lintian warnings and even wrote a manpage for it while there (but for this one, I didn’t bump the Standards-Version). The second one was only the second one to complete because the others took longer; it’s basically a change of a dæmon to use two instead of one listening socket, to work with a “doble stack” OS instead of just a “dual stack” OS by not using v4-mapped IPv6 addresses (I considered if to use select(2) or poll(2), or to just fork and have two dæmons running, but that seemed too ressource-consuming to me so I chose the less-complicated poll(2) method, looking at popa3d(8)’s source code (inherited from OpenBSD) in the MirBSD tree since I could not find my network programming book. The third one was basically communicating with upstream; the bug can be blosed with no change to the package. The fourth one took me a while; luckily I have qemu 0.11.0 on MirBSD, and aurel32’s mipsel qemu images helped a great deal; however, cowbuilder --create failed for me, so I ended up waiting almost the entire night for a-g d-u to finish; in the end, it was simply a bug in upstream’s which is only exposed due to a toolchain bug on mips(el).

To do: my AM Zack needs to upload the NMUs (after checking, of course); I need to communicate some of the fixes upstream (the gidentd upstream is NXDOMAIN ☹), produce a very small testcase for the mipsel toolchain bug, maybe fix some more mipsel FTBFSen as I have a working qemu instance now, but maybe I’ll do that at the BSP when I can’t find IPv6 bugs or so that I feel I can fix (I also want to do an mksh release which has to be prepared first RSN, and there’s still the need to formally publish the MirBSD-current bi-arch snapshot and make another ISO out of it for BT and prepare the multi-BSD USB stick for…).

Annoyances: a-g d-u could ask me things at the start before working for some four+ hours instead of in the middle, and the same questions several times (PAM restart). The sid kernel doesn’t boot today but did yesterday, the lenny kernel produces this:
Starting the hotplug events dispatcher: udevdudevd[320]: udev: missing sysfs features; please update the kernel or disable the kernel's CONFIG_SYSFS_DEPRECATED option; udev may fail to work correctly
(I hate udev.) And, worst of all, these annoying fireworks (some sounded like originating from inside our staircase, I pity the neighbour’s cats) when one wants to hack… Finally, I *loathe* CDBS. Debhelper v5 rules!

Oh, and I also was under the impression that “Firstname LASTNAME” was a French thing, and to a much lesser extent Asian. (@bubulle)

new snapshot upcoming

29.12.2009 by tg@
Tags: snapshot

I’m almost there: i386 and sparc have compiled complete. Now I “just” need to do some release engineering (consider whether to re-roll the ports10.ngz dist set; creating the combined manifold Midi-ISO, various things with checksums and signatures, announce and upload) then we’ll have a new snapshot. Finally, I also plan to create a big ISO to replace the “wtf ist hallowe’en” edition. No new MirGRML but nevertheless bugfixes for grub.cfg (memtest and vga modes).

The other side is that I didn’t have to do much for this, luckily; I was still suffering from some kind of cold+stomach ache. I hate that. I bought vitamines etc. for $alot € yesterday to fight it better though.

OpenBSD can build the kernel with pcc now, nice. I thought ÆrieBSD was the one with switching to a GNU-free toolchain goal though…


27.12.2009 by tg@
Tags: debian

With the Lintian 2.3.0 saturday-after-christmas release (by the way, over here if it’s done twice it’ll really become tradition) I’ve run its spelling tests over all of MirOS CVS repository. The result: 293 kinds of typos in 35857 souce files. (Although there are the case things too. Without them, I have 51 typos in 7206 files. Asides from false positives (I used fgrep -rwl[i], and -i and -w don’t play well together, and -w mis-catches “GTK+” as “GTK”) I probably can’t (API, source code) or won’t fix all of them though.)

However, I have some rather hot asia-style food to eat right now, and will need to get up early tomorrow for work, so I am not applying/fixing them right now. (bsiegert@ and gecko2@ however are enjoying themselves at 26C3, see their wlog entries.)

Note that all of today’s fixes will not make it into the next MirBSD snapshot already, since it’s built (i386) and building X11 already (sparc). On the other hand, the next bunch of WTF *.deb files will have them. I also need to fix makefs upstream for Hurd… and continue the T&S questionnaire… *sigh*

Update: I suppose this is my “Hello, Plänet Debian!” posting (thanks aptituz!)… well, my packages in the archive were already lintian clean, in case someone wonders (I did recheck with 2.3.0 though). My point was, why not use checking tools from one “universe” for another one, viceque versa? (Similar to synergy effects from knowledge.)

26C3, day 0 and 1

27.12.2009 by bsiegert@
Tags: event

I finally have a wristband for 26C3. The wristband is what is controlled on the entrance. I had preordered the ticket and paid by bank transfer, and the receipt with the barcode only arrived on december 25. In earlier years, you came on the 26th, crashed in the hackcenter, and at some point, when there were only ten people in the queue at the cash desk, you got your ticket easily.

Well, this year, things were a little different. First, there was a semi-public ticket presale. What's more, rumors about a very limited number of tickets were floating around. And finally, while the cash desk was said to open at 2000, the software was not ready by then (!) and the opening was deferred to 2130. The "fast line" cash desk for people with presale tickets was only opened around midnight, when I had already given up hope and joined the huge queue of several hundred desperate hackers. (Thanks, btw, to ScottyTM for pointing out that the "fast line" had been opened.) Finally, I was told that my ticket "did not exist in the system", that I was an evil cracker for trying to enter with a fake ticket, and that I should check back on the 27th. Great.

The end was happy and somewhat anticlimactic: I checked back this morning, there was no queue, and my presale ticket was accepted without any problem. Most of the talks I saw today were about network neutrality, censorship and related topics. I found the talk by Jérémy Zimmermann from La Quadrature du Net especially interesting. Part of the talk was an introduction to lobbying: calling your EuroParl representative, sending e-mails and insisting on your point "raises the political cost of certain decisions".

In addition to OpenOffice 2.0 (Linux-i386 on MirBSD) and the Lotus SmartSuite Millennium, I downloaded SoftMaker Office 2008 for Linux-i386 (on MirBSD) and Windows® (free serial numbers until end of 2009, since SoftMaker Office 2010 (for Windows®) is out). This was pretty easy: extract tarball, throw away installer, extract the second tarball in /opt/SMo (along with /opt/OOo), and it Just Works™ (with ports/emulators/fedora installed).

So now I have three choices for creating documents (although I won’t use OOo for it, I think it’s unreadable), and three for reading files in MS Office format, and two (OOo and SMo) for reading documents in some weird .ZIP-like format (OASIS or something), if they get sent to me. I’ll toy around with SoftMaker Office 2008 on both platforms and will see how it compares to Lotus WordPro.

So this is basically positive press to show another Linux-i386 blob that works as-is on MirBSD/i386 with compat_linux(8) – thanks! Others make this much more difficult.


26.12.2009 by tg@
Tags: snapshot

I somewhat had hoped the TLS extension would be finished this month. Too bad. I will still wait for an OpenSSL patch, and maybe options for clients to control behaviour (on connections to servers that don’t have it implemented yet), such as lynx(1).

Alas, we need a snapshot by January, to put on USB sticks, so I’m compiling one. Other than that, recovering a little over the weekend, will need to work next week as well. Taking things slower. I think I caught a slight cold+stomach ache last weekend, when it was freezing pretty hard.

Benny is at 26C3, but there’s probably no official MirOS presence. You’re still cordially invited to talk to him and gecko2@ who’s there as well.

Stehen, Warten, Beten

17.12.2009 by tg@
Tags: rant

SWB. Stehen, Warten, Beten.

Am Dienstag Abend waren wir mit Roland in der Bonner Innenstadt. Hiernach wollte ich gerne nach Hause fahren, aber die StadtWerke Bonn machten mir, wie so oft, einen Strich durch die Rechnung.

Das ist eigentlich gar nicht so schlimm, denn wäre ich zu Fuß gegangen hätte ich sicherlich nicht mehr als eine Dreiviertelstunde benötigt. Aber wir haben ja das tolle Informationssystem, das uns allerdings nur anzeigte: „Betriebsstörung. Unregelmäßigkeiten im Fahrplan möglich.“

Bei „möglich“ wäre ich hellhörig geworden, aber das stand schon den ganzen Tag da, und morgens auf dem Weg zur Arbeit war eigentlich nichts los und wurde nichts durchgesagt. Also warte ich auf die nächste Bahn. Eine ganze Viertelstunde lang – anscheinend ist (gestern Abend nach der Arbeit übrigens auch) mal wieder eine Bahn ausgefallen, ohne daß dies angezeigt oder durchgesagt wurde, trotz moderner Technik. (Die Deutsch Bahn schafft dies zumindest auf ihrer Webseite, und Durchsagen gibts auch öfters, Anzeigetafeln sind dort ja etwas unflexibler.)

Dann kam die Bahn, ich frag’ was denn los sei. „Ja, ich fahre auch nur (1 Haltestelle), von da aus müssen Sie zu Fuß (2 Haltestellen) laufen, da fahren wieder Bahnen, wir haben da ein Riesenloch in der Straße.“ (Kurzum ein ganzer Streckenabschnitt abgeschlossen, ohne eine Möglichkeit, dort weitere Verkehrsmittel reinzubringen.)

Ich laufe also, frustriert, und stehe mir an der Zielhaltestelle – wohlgemerkt bei Minusgraden – wieder die Füße in den Bauch. Die gerade gewendet habende Bahn eine knappe Viertelstunde lang gut sichtbar für uns auf dem Stumpfgleis, Lichter an, tätigkeitslos.

Alles in allem habe ich annähernd eineinviertel Stunden nach Hause benötigt, und das ist bei einem monatlichen reduzierten Abonnementpreis von 66.50 € einfach inakzeptabel, zudem die tollen neuen Anzeigetafeln intakt waren. Hätte ich das gewußt wäre ich gelaufen oder hätte mir ein Taxi genommen, aber nein, die StadtWerke Bonn machen ihrem Namen „Stehen, Warten, Beten“ alle Ehre. Wenn ich nicht mein Abonnement eh’ schon bei der Deutschen Bahn hätte würde ich es sofort kündigen.

To add insult to injury, die Anzeigetafeln waren mit einem teils statischen teils durchlaufenden Text so dermaßen „voll“ belegt, daß man es nicht für nötig hielt, die normalerweise dort verfügbare Uhr (man sieht schön, daß alle Bahnen an den Endhaltestellen immer eine Minute zu spät abfahren – aber von einem Ex-1€-Mitarbeiter weiß ich ja, daß die SWB drei(!) unterschiedliche Fahrpläne haben…) anzuzeigen. Dann hätte man ja entscheiden können zu laufen – insbesondere wenn man informiert gewesen wäre. (Oder zumindest dann drei statt zwei Haltestellen laufen und eine Bahn früher nehmen. Oder zwei, je nachdem.)
Ich meine, für den Wasserrohrbruch zwischen Straßen- und U-Bahn können die ja nix, aber informieren können hätte man mich wenigstens, denn am nächsten Tage (gestern) mußte ich sehr früh ’raus.

Yesterday, I posted the following proposal to the mailing list. I am putting it here, too, for future reference.

I have been thinking about how we can improve the pkgtools with some new and useful features. There are two that would be easy to implement and would give us some kind of "poor man's apt":

Replace PKG_PATH, which at the moment is either a single URL or a colon-separated list of local paths, by a simple config file. It could also include a few other directives, like the default pager for MESSAGE files. I am thinking of the following:


The "source" lines are just local paths or URLs with a priority value, where 0 is the highest one.

For http, we have the problem of getting the file names. I propose to put an index.txt file, with just one filename per line, into each "repository". The indices might be cached so that we know what files are where. For http, we could even automatically update by using an "If-Changed-Since" header, or have something akin to "apt-get update".

It would be very easy to just search the repos in order of priority and take the first match for the pkgspec given. Or we could use findbestmatchingname() to decide which package to take.

These changes can be implemented in a reasonable amount of time, I think. Maybe during the CCC?

I smell an antitrust case coming up

12.12.2009 by tg@
Tags: mksh rant security

Oh the joy…

20:54⎜«smultron» mira: i just upgraded the big server from 10.5 to 10.6... and apparently the upgrade script removed /bin/mksh... now I don't have a shell and quits immediately... any ideas?

My suggestion – ssh -t servername /bin/bash --login – doesn’t seem to help:

20:59⎜«smultron» oh great
20:59⎜«smultron» ssh just keeps asking for the password
20:59⎜«smultron» then gives this:
20:59⎜«smultron» Permission denied (publickey,gssapi-keyex,gssapi-with-mic,keyboard-interactive).

Maybe gecko2@ can help… or bsiegert@… who knows.

In other news, no reaction at all on the Android front on my proposal to get mksh either to replace NetBSD® 0.x ash, or, at least, add it for developers.

Bei VIAG Interkom einen Händivertrag (Händi, schwäbisches Fremdwort, von "Jo hän di denn do koa Kabel dran?") kündigt, geht das relativ gut, man bekommt nur zwei Monate später eine Rechnung über 0,00 €. Den T-D1 Mitarbeitervertrag (Azubi...) wird man auch gut wieder los, obschon er über die Laufzeit hinweg funktioniert (wohl ein Abschiedsgeschenk des Ausbildungskonzerns). Bei Debitel hingegen wird der postalische Eingang der schriftlichen Kündigung um ein paar Tage verzögert, und dann hat man den Vertrag noch knapp 15 Monate lang am Hals. Die Frechheit ist aber BASE: die Karte funktioniert, nach meiner Kündigung gegen Anfang des 2. Quartales 2008(!), immer noch. Auf Anfrage teilt man mir mit, daß "leider" in ihrem System keine Kündigung ersichtlich sei, und ich noch bis zum 5. Dezember 2010(!) damit leben müsse.

Hoffen wir, daß Netcologne, falls ich denn den dortigen Mobilfunkteil meines Kombipaketes mal loswerden will, angenehmer ist. Immerhin ist das die einzige Telko, die auch mal Bestandskundenaktionen macht, mit denen ich seit 2001 ununterbrochen extrem zufrieden bin, und die nur ein Mal größere Störungen hatten (November 2001, wohl noch neu) und einem für einen halben Monat öfters gestörten DSLs einen ganzen Monat Telephon- und DSL-Grundpreis erstattet haben; die defekte Splitter und Leihmodems (NTBBA) innert Stunden ohne große Fragen austauschen; die einem alle paar Jahre mal sowohl den Grundpreis reduzieren als auch der DSL Geschwindigkeit erhöhen...

Also, Leute, kauft nicht bei BASE! (ciruZ ist mit zufrieden, die sind aber Prepaid, das heißt zwar keine Knebelverträge, aber auch keine unlimitierten Telephonate im Ausland, z.B. um gecko2 auf belgischen Autobahnen zu helfen.) Daß es auch anders geht zeigt die Deutsche Bahn in einer anderen Kaufempfehlung (diese hier ist allerdings eher eine leicht frustrierte Verfehlung). Mein Brüderchen fand, ich sollte das hier dokumentieren/ablassen; keine Bange, das hier wird nicht wie bei Fefe eine Produktbashingseite werden (dafür sind einfach alle Sachen in Existenz zu... kaputt, das ist mir meine Zeit nicht wert).

Naja, mal die Kündigungsbestätigung aus den Altunterlagen raussuchen, um deren Zusendung ich letztes Jahr gebeten hatte. Warum wohl?

Happy Benzday!

07.12.2009 by tg@

Since I don't reach you via IRC, phone or Natel™ let me wish you a happy benzday here. ☺☻

I think it's perfectly okay for libobjc to depend on libgcc_s like libstdc++ does. So let's not disable gobjc from base. Rather, make everything except C version-dependent. (Bump base vsn afterwards, the gcc vsn has been already.) I think we'd also best rename the clang executable and make a wrapper using -L, -rpath and -I flags from its CCLD instead of always using these from the system compiler. (I wonder if it's worth the effort to make the C++ header files version dependent as well. Probably.)

We somehow need a way to differentiate ABIs in MirPorts, as there will be several compilers. On MirBSD: base-gcc3.4.6, port-gcc4.4.2, port-llvm-gcc4.2.1, port-pcc, port-nwcc (this list is for i386).

Benny, may I encourage you to hack on pkgtools the next Muttenzday? Especially I'd like to have that file extension retaining/cycling for auto-dependent packages in, it's basically a showstopper for LZMA compressed binary packages. (Will still be LZMA1 for some time, as there is not yet a stable xz release.)

The hardy at MirDebian "WTF" Repository section contains a KDE 3 source and binary package of KWalletCLI 2.00 (built for Debian Lenny and K?buntu Hardy). A KDE 4 package (for Debian squeeze/sid, newer *bunti and ones with KDE4-backport) will be worked upon as soon as I can get into the pkg-kde Alioth group.

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

MirOS Logo