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

more bugs – bootloader and mksh

01.02.2009 by tg@
Tags: bug event grml mksh snapshot

Again. I almost have a MirGRML+bsd4grml ISO ready (exactly 72 MiB), just for the fun of it, and so that people can toy with it – and test the integration. But nooo, I even hacked a disklabel, yet it wouldn’t access the filesystem. The “machine label” command shows what’s wrong (and hints how to fix it), and I missed a corner case in disc I/O due to adding two not-so-independent-from-each-other scenarios during the El Torito merge.

mksh has funny behaviour with ${foo/@(%)/x} failing – only in Unicode mode though.

ext2fs damage

01.02.2009 by tg@
Tags: bug

Fun: MirBSD #8, #10 and #10-current can’t edit or create symlink(7)s on ext2fs, MirBSD #7quater could. They hang the filesystem, as well as the parent of the mount point, on access, and damage the filesystem. I have no idea… (msdos is damaging long filenames, btw.)

4.4OpenBSD can… oO

Benny has done more work on ext2fs, maybe he’ll fix it.

more bootloaders *grml*

31.01.2009 by tg@
Tags: bug event mksh snapshot unicode

To counter the bug in Parallels Desktop, I rewrote all of our Master Boot Record (mbrldr, mbrmgr), Partition Boot Record (bootxx), and BIOS disc detection code in boot(8/i386). I've also changed the magics, API etc. a little between these, rewriting or removing quite some parts of both installboot(8/i386) and a.k.a. self-installing bootblocks. Oh, and MBR and PBR are smaller, or rather, have more room for informatory texts and data block storage now.

Lucas “laffer1” Holt from that cat's BSD helped me testing, qemu and bare metal IBM X40 works, so I suppose it's good. Can't test on VMware Server 2 right now.

On the other front, I'm writing this wlog entry on MirGRML, which is finally sort-of finished (which is why it has much less Unicode than a regular posting from my laptop). lynx-cur in sid is broken though, due to it using GnuTLS. Gah! Anyway, I see a sid “wtf” repo coming, and it was workarounded; next time I'll just do things differently.

Lukas “smultron” , the graphician of our partner project mnbsd, helped a little with the label (while I'm not versed in that field, if I had gotten the right material to work with I could've managed it all alone this time), and all that's left now is the Live-CD part and baking (and testing! even on sparc...) the ISO 9660 image.

There will be 297 MirBSD Triforce™ DuaLive™-CDs (two will go to Kiwi land to swishy, one is already reserved for XTaran) and 50 MidnightBSD Live DVDs at FOSDEM.

gecko2@ and bsiegert@ were not too helpful today, although I have to admit that real life often has precedence. I merely read some and went to sleep early yesterday too (and disabled both POTS and cell phone so that I couldn't be waked, haha!) instead of hacking late. Luckily, the deadline is not that dead, and I'll almost certainly make it tomorrow, in time. Thanks again, Daniel Seuffert from, for all support you and others give to the smaller BSDs. Heh, and laffer1 is not quite done yet with his ISO (or UDF?) either.

2009-01-24 snapshot uploaded, pending validation

25.01.2009 by tg@
Tags: grml snapshot

Yesternight, I built an i386 snapshot (including a bsd4grml). This is already uploaded on /MirOS/current/i386/ but beware, it is pending validation, and some things (like the serial console or combined i386+sparc ISOs) are missing. Note that there will be, due to time constraints, no new sparc snapshot, we’ll just use the last one I built.

If people who actually use MirOS/sparc were to talk to us, this situation might improve.

It’s still a couple of days until the Triforce™ ISO and CD label are due, but we’d appreciate independent test results, as usual.

Plans for the FOSDEM Edition Triforce™ (MirOS DuaLive™ + MirGRML) CD-ROM are to add a few packages to the usual baseLive image: IceWM, screen, rsync(GPLv2), mc; possibly ent, lzmadec

Watch the news for announcements about usability of the 2009-01-24 snapshot.

Update: the three serial console floppies and the five ISOs are there and should arrive at the public mirrors tonight. They are in fact bootable and quite probably usable, but bsiegert@ still has to do the validation by doing a full install and some port builds.

Das Wetter bessert sich

25.01.2009 by tg@
Tags: geocache

Heute mal wieder das Vélo aus- und mich gut eingepackt:
(Update: images moved here)
        inc     dword ptr ds:[caches]

Diesen hatte ich auch schon länger auf’m Kieker, allerdings nie die Muße, dort hinzufahren, bei dem Wetter, und der FTF lockte auch nicht, weil er – mal wieder – auf der Kommerzseite Tage bevor er auf der freien Seite zu finden war geloggt wurde. *grml*…

Liebe Plänet Sümlink Leser, weil so viele von euch zu <zensiert /> sind, sich das Root CA Certificate zu installieren, und die Bloatzilla-Leute ihren A^H^H^H^H Hinterteil auch nicht bewegt kriegen, editiert der XTaran das https in ein http. Zu dumm, daß ich auf vielen Seiten nur noch SSL erlaube. Nutzt also den „Mehr...“-Hyperlink, um die Statistikbildchen zu sehen, und hofft, daß die neue CAcert-Root bald live und in die Brauser geht. MirPorts Framework Nutzer habens natürlich einfacher…

Gleich geht’s erst mal zu meinem Stammrestaurant, einem Jugoslawen, ääh Kroaten (Primošten oder auch Dalmatien Grill) mit bogus…


24.01.2009 by tg@

My pcc wchar_t patch thread gets attention again. Yay!

Good I just installed MirOS on soon-to-be-tear (althouh on odem’s old 2.5″ HDD. I cou^H^Haught an error due to /dev/MAKEDEV using the installer’s user and group databases, which were missing some. And good I compiled a kernel beforehand… bzero(3) is optimised in locore, memset(3) isn’t. Lession learned? Maybe. I bet I’ll do quite some of these “Flüchtigkeitsfehler” again in the future, I always used to do so, even at school.

Damn. SCSI is expensive, and SSDs are available only for SATAn.

Whew, finally weekend. I slept what appears to be 10 hours and dazed for some more. Now I’m unwilling to do any work, but I’ll do what I’m owing y’all. (Yes, normally I try to write British English.) I estimate we’ll have a new i386 snapshot, and mika his special edition, by tomorrow evening. Too bad I delayed testing the MirGRML due to the dayjob workload and general… unwillingness. Anyway, FOSDEM is coming, we’ll shine.

Lucas “laffer1” Holt from our other ally MidnightBSD will also do a special ISO for FOSDEM, which we’ll distribute among people. I wonder if Debian GNU/kFreeBSD does some, maybe an updated GING CD, but I guess that’s up to the Debian people, not for us BSDers.

Since I started working, I can’t bring myself to eat anything in the morning, nor really in the evening. I wonder why that is… I still have a package of black bread here I bought when gecko2@ visited me, that would be 31 Dec I guess. I probably have to throw it away by now, but that sucks. I hope I don’t get even more ill… my colleagues and I have not been feeling all that well for days.

On building a cross-compiler package: I wish I could bring myself to get it done by FOSDEM, maybe Robert Schuster would help baking the JDK.

Anyway, I’d better stop concocting some text here, and instead bring the mock-tear installation into working state and get it to start rsynching, cvs(1) co(1)ing, and compiling…

Sponsor for FOSDEM

21.01.2009 by tg@
Tags: event

Sponsors make it possible for us to attend popular events, paying for travel and accomodation (we do not necessarily require a hotel, though, but it should be a clean place to sleep) cost. We are still looking for more sponsors making it possible to travel to even more events (Linuxtage Chemnitz, Essen, Graz, Luxembourg, for example) – flyers and CDs are already covered by or ourselves, so it would just be that and, if you have money to spare, food and drink. Come and talk if you’re interested (in return, like here, we will publicly thank you or provide custom MirOS editions, possibly Live DVDs with binary packages, etc).

Heute mit einem Kollegen in der Mittagspause aufm Weg zum Chinesen:
(Update: images moved here)

FOSDEM, the Free and Open Source Software Developers' European Meeting

Who’s not? ☺

My boss approached me just this morning… looks as if gecko2@ and I can use the bigger company car from here, expense paid, and take the other two who are going to Bruxelles with us. I’m in the process of organising them a hotel now though – maybe bsiegert@ the francophile can help with that. gecko2@ is going for a youth hostel though, as usual, as he has a membership anyway. Benny and I are already hotel’d.

Now if I were on Plänet Debilian…

SCSI-induzierter Rückschlag

20.01.2009 by tg@
Tags: bug event grml hardware snapshot

Heute hat mein Rückschlag, wenn man von den leichten Kopfschmerzen und der arbeitsinduzierten Müdigkeit absieht, einen Namen: SCSI. Ich wollte, damit wir mit dem neuen Snapshot endlich mal weiterkommen, tear aufsetzen (hey, ich schreib ja immer noch im Blocksatz im HTML-Quelltext ohne Mühe oder es zu wollen oder zu schummeln, wow) und dann kompilieren (quasi als Abnahmemessung). Dummerweise habe ich, auch dank gecko2@, nun eine Idee, woher die SCSI-Fehler kommen könnten, denn die Plattenprüfung im BIOS zeigt keine Fehler an. Ich fürchte, daß das LVD-Kabel kaputt ist (i386-Hardware ist eh doof, ein Kabel, ein Terminator und fünf Konverter von SCA auf UW-SCSI sind nötig, wo meine SPARCstation 20 das bereits als Backplane hat).

Es läuft vermutlich darauf hinaus, daß ich odems 2.5″ 80G IDE Platte nehme. Bonus: ältere cvs und Distfiles sind schon drauf, sodaß es mit einem schnellen rsync sowie einem erneuten Checkout (da ich / ja eh’ plattmache mit dem letzten Zwischensnapshot) gehen sollte, und ich sogar ein paar Pakete (zumindest IceWM) für den LiveCD-Teil backen könnte.

Toll. Ich habe sogar schon ein vorläufiges ISO zum MirGRML testen, bin aber irgendwie platt und auch von Technik mal wieder angenervt. So komme ich nie zu was… ich glaube, ich fang morgen mal so früh wie geht an, und bin dann ausnahmsweise schon wo’s hell ist zu Hause.

Also: Am Snapshot wird gearbeitet; mbsd4grml kommt direkt im Anschluß; grml-mir muß noch getuned (Konfigs), aktualisiert (lynx-cur UTF-8 Bug wo ich vorgestern drüber schrieb) und getestet werden; den sparc-Teil nehme ich vom letzten Snapshot; für DuaLive™ haben wir somit alles, Extrazeugs mache ich nur wenig und nur für i386, und für Triforce™ siehts gut aus – eine offizielle Ankündigung kommt auch irgendwann, wenn für beide Seiten alles funktioniert und integriert ist. Die tear-Migration ist – leider – hingegen auf unbestimmte Zeit verschoben. (Damn!)

Bißchen malen müßte ich noch, damit die FOSDEM Edition CDs auch sowohl das grml-Logo (habe ein OK von mika), den Shilouetten-Dæmon (altbekannt, mit OK von rcollette und mckusick), und ein Triforce haben. Es paßt mir, ausnahmsweise, auf Arbeit einen KDE-Desktop mit gimp und Inkscape nutzen zu „dürfen“ (peu à peu krieg ich den weniger nervig konfiguriert, und ㈠ kenn ich KDE 1 schon auf ecce!GNU/Linux 1.0 und ㈡ mag ich kmahjongg eh’ und konqueror besser als M*zilla Schrott, also ruhe da!, auch wenn ciruZ mir GNOME verkaufen will). Mal sehen, was ich da so hinbekomme. Wobei, à propos Bildchen: wann lernen die Leute eigentlich, daß man keine Formate wie JPEG ob ihrer verlustbehafteten Kompressionsverfahren herumschickt – zumindest zum Bearbeiten? *seufz*… Zumindest denke ich, daß ich das ganz gut hinkriegen werde, hab da was im Kopf, sogar ohne smultron, lediglich die Schrift könnte knapp werden, kA ob der schwarze Hintergrund in dünne Linien reinläuft. Mal Daniel fragen…

Ha! Laufend mehr Leute krieg ich vom RANDEX-Protokoll und den Vorzügen des Entropieaustauschs, Zufall allgemein, usw. überzeugt. Way to go! Nun bin ich gespannt, ob Vutral mal was zu APS findet und was man mit Mumble und randomness so machen kann.

So much for a weekend

18.01.2009 by tg@
Tags: grml snapshot

I seem to have written about it already. Damn. Cost me nearly the entire of this weekend day (yesterday it was more lashing the alliance with mika), and then I still had to fix it – turns out makefs(8) was at fault again. Now who’d have expected this?

Owkay. We have a working bootloader (sans the scheduled rewrite of the cd9660 filesystem, see my last wlog entry), a cleaned up system for both creating the baselive (and DuaLive™ and Triforce™) CDs, stuff for making the bsd.rd for grml, a working ports/sysutils/pxegrub, the legal issues cleaned up, the makefs(8) bug fixed, lynx(1) updated as well as bugfixed (UTF-8 was broken), the MBR source code portable to not include <machine/asm.h>, and the worst problem with the OpenLDAP’s port MESSAGE fixed (discovered at work).

Now it’s later than intended again, but I’ll regenerate the port index later tonight, head to bed then work, and will set up tear tomorrow, for building the snapshot-to-be-uploaded, while writing a new bootloader man page. That should about be it.

I wonder if we should rename MirOS #11 to #12 and issue a release from what we have in -current now (pretty stable it is, plus #10 isn’t really maintained) like I did in the early years, when I start breaking HEAD to merge OpenBSD changes into it. Mh, talk with Benny and replaced.

This 40 work hours per week scheme is annoying. But I guess if all had 34 work hours per week, they’d need additional personnell which has high cost overhead. Damn! If I divided my salary by 40/34, I could still make a living (not much worse than now; still can’t afford much).

Woohoo. replaced commits, n0-1 (FreeWRT) joins the RANDEX contributors club and joins ##/dev/arandom. Just Benny was on vacation, and, again, I didn’t know of it ☺

I seem to recall that IBM Thinkpads have a movement sensor. We may use it as entropy source – not for normal on-the-desk use cases, but still – every bit counts. This is probably a Vutral-worthy idea. Someone should, really, look into that. Kabelaffe says it has few states, but both level and edge (timing) values.

Another thing Mumble and Murmur, which he showed me, could do is to collect entropy too – Mumble (the client) continuously records anyway (to intelligently try to find out when we talk loudly – it works surprisingly well), which hashed can contribute entropy (even with hardware mute it’ll get electric noise or somesuch); it could also hash the current channel conversation a.k.a. input from the network. Now I need to persuade gecko2@ to set up Murmur, the server, on thetis, and get bsiegert@ to port qt4-x11, as well as all the other prerequisites of Mumble, the client, to MirBSD. May as well be he ports both Mumble and Murmur, that latter I could set up on eurynome. At least it has more bandwidth than Kabelaffe’s home ADSL.

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