=== mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel === jdub [~jdub@home.waugh.id.au] has joined #ubuntu-devel [12:48] tseng: around? [01:06] mdz: #1388 is filed somewhere in Debian too. It's still got joeyh scratching his head over the correct answer. There are some possible hacks, but he'd like to think about them. [01:06] Kamion: we should do something about it for Warty [01:06] One hack would be having the noninteractive frontend set the seen flag if and only if $ENV{DEBCONF_I_AM_RUNNING_INSIDE_DEBOOTSTRAP} or something is set [01:06] my feeling is that when the package is configured non-interactively, the questions should be marked seen [01:07] I think that makes sense regardless of whether debootstrap is part of the equation [01:07] mdz: I suggested that to Joey, but he's not entirely sure about some of the ways people are using noninteractive on their own systems [01:07] I tend to trust Joey's judgement in these matters ... [01:07] Joey has no incentive to make a determination on this issue before the Warty release, but we must [01:08] look, I'm not saying we shouldn't, OK? [01:08] ok [01:08] I'm trying to avoid having to support a fork in debconf that's too evil, by getting something supportable done before the Warty release instead [01:09] I'm not trying to put it off until post-warty [01:11] since I have the debconf author sitting ten feet from me, talking to him is worthwhile :-) [01:11] sure [01:12] does Sarge not have any debconf-using packages in base? [01:13] hmm, surely it does [01:14] yes, it does, that's why the bug is also filed in the Debian BTS [01:14] I'm just looking for it as Joey asked me to now [01:23] #238301 [01:24] thanks [01:25] I'll import it into bugzilla and mark the other one as a dupe [01:32] followed up to the new one with current brain state [01:32] will fix it when I get back [01:39] Kamion: thanks [01:39] Kamion: how is the meeting going? [01:42] very well, lots of stuff fixed [01:43] I've spent most of the meeting trying to get gtk d-i up and running, still wrestling with directfb unfortunately [01:43] but talking to Sven has made it possible to agree on stuff to remove from the powerpc CDs to make them halfway sane, and the appropriately hideous death of the root-2 floppy [01:44] oh, and I managed to boot my powerbook from a USB stick, which will speed up my d-i testing a lot in the future 'cos I won't have to burn CDs [01:57] nice [01:57] my laptop seems to quite happily boot from a USB CD-ROM [01:57] so I assume it would work with a USB stick, too, though I don't have one [01:58] I found out quite by surprise; I left a Warty install CD in the USB drive accidentally when I rebooted [01:58] my laptop (R51) just refuses to go beyond the IBM logo when a USB stick is plugged in :-/ [02:00] booting a powermac from USB requires Runes Of Power === azeem chuckles [02:01] unless your OF defines a useful devalias, which mine doesn't; I suspect it's rare to find that [02:21] Kamion: do we know of any successful installs on pre-toilet-seat powerpc/g3 laptops? [02:23] jdub: are there any such which are newworld? [02:23] if you gave a model name rather than "toilet-seat", I could check :-) [02:23] what are black g3 powerbooks? [02:23] maybe the model is "wall street"? [02:24] wallstreet is oldworld [02:24] aha [02:24] (and we don't support oldworld) [02:24] tops [02:25] the first newworld laptop was the Lombard, I believe [02:25] that was a powerbook though ... your iBook might be the Bondi? [02:25] I confess to not being able to keep track of a lot of the earlier model names [02:27] aha, fear our documentation :-) [02:27] http://archive.ubuntulinux.org/ubuntu/dists/warty/main/installer-powerpc/current/doc/manual/en/ch02s01.html#id4739961 [02:28] it's not up-to-date about a lot of things, but anything marked "powermac-OldWorld" there is unsupported === jdub_ [~jdub@home.waugh.id.au] has joined #ubuntu-devel [03:14] azeem: so [03:15] azeem: basically nothing should be listed in shlibs? === jdub is deferring the RPATH stuff, very odd ;) [03:15] well, it is my opinion that if your package does not have any regular shared libraries, but only plugin-type stuff, shlibs should not be used [03:16] but I'd be interested in the opinion of the demi-gods in here [03:16] jdub: context? [03:16] azeem: agreed [03:16] (without context ...) [03:16] shared objects used as plugins don't need shlibs entries === lamont made failure buildlogs red on http://people.ubuntu.com/~lamont/buildLogs/byDate [03:16] they should have shlibdeps though [03:17] yeah, these are all dlopened plugins [03:17] shlibdeps is all fine though [03:17] and ldd -r says they're clean? [03:17] hrm [03:17] clean being no undefined symbols? [03:18] yeah [03:18] hmm [03:19] they have undefined symbols [03:19] but those are all in the same package [03:20] the plugins should be linked with any dependent shared libraries [03:20] jdub: polypaudio-alsa does not seem to depend on libpolypaudio0 though, which I guess provides the missing symbols [03:20] if they need symbols from the binary which loads them (RTLD_GLOBAL or whatnot), that's a bit annoying, but not obviously wrong [03:21] azeem: mmm [03:21] azeem: though that's a special case, being a different package [03:21] azeem: but it does mean there's something wrong with the plugins (and why they have undefined symbols) [03:22] well, I didn't check thoroughly [03:22] this totally assumes rpath [03:23] azeem: [03:23] $ dpkg-deb -I ../polypaudio-alsa_0.5.1-1ubuntu1_i386.deb | grep Depends [03:23] Depends: libasound2 (>> 1.0.5), libc6 (>= 2.3.2.ds1-4) [03:23] [03:23] ^ wrong [03:24] you might touch the border of cdbs' applicability here [03:24] argh [03:24] cock [03:24] it's in svn :) [03:24] anyway, I'm going to bed now, *tired* [03:24] night, thanks again :) [03:24] The following NEW packages will be installed: [03:24] db4.2-util libapr0 libpcre3 libruby1.8 libsvn0 libswig1.3.21 subversion [03:24] [03:24] *bong* [03:39] jdub: what do you think about creating non-english ubuntu discussion lists? [03:39] I didn't think we'd need them so soon [03:40] there seems to at least be an -es contingent who would appreciate it [03:41] mdz: i asked mark about it, he's happy with ubuntu-users-$LANG [03:41] i'll do the -es one now [03:42] cool [03:42] i'll ask carlos to admin it ;) [03:42] and find other moderators [03:47] heh [03:48] i'm glad i set the defaults from ubuntu-users [03:48] because changing the default language makes the list hard to admin ;-) [03:58] hmm, what's up with bugzilla? No new bugs for me the whole day.. [04:12] jdub: here i am [04:15] tseng: the mono depends in the tomboy example i sent you was kinda wrong [04:15] hm [04:15] tseng: it should really depend on mono-jit | mono-mint [04:15] i think you meant mono-jit or mono-minut [04:15] yeah [04:15] ;) [04:15] mint* [04:15] i knew that but i had to go to work === tseng offers no guarantee on his own repo, as im learning on it atm [04:16] :-) [04:16] it's good [04:17] orph (alex, author of tomboy) is way pleased :) [04:17] ya works pretty well [04:17] oh neat [04:17] 20:14 #ubuntu: < whiprush> woo tseng fan club. [04:18] heh [04:18] im famous or something.. barely did anything [04:18] :P [04:18] all the credit goes to the debian devs. [04:19] mdz: is ndiswrapper a module? [04:19] mdz: so it can be in restricted? [04:19] jdub: yeah its a module + a bit of userland [04:19] there is an app to install the inf from the windows driver === jdub knows what it is, wonders what our plans are [04:21] jdub: yes, it is a module, but it's GPL [04:21] and so there's no reason to demote it to restricted [04:21] mdz: linux-image == supported, restricted == unsupported [04:22] that's a HUGE reason [04:22] restricted != unsupported [04:22] universe == unsupported [04:22] restricted is "supported the best we can given the restrictions" [04:22] restricted is unsupported [04:22] it [04:22] it's just best efforts that the packages will build and work [04:22] and we *cannot* support ndiswrapper [04:22] Mark was quite clear on this point in Oxford [04:23] i thought he was too [04:23] the userland part of ndiswrapper will print scary warnings [04:23] unsupported, best-efforts [04:23] supported, best-efforts :-) [04:23] i'll mail [04:23] because this is senseless [04:23] ok [04:24] the only clear use case for restricted is binary-only drivers [04:24] we've been over and over it [04:25] ndiswrapper is an enabler for binary-only drivers [04:26] correct [04:26] but that's a secondary issue [04:26] but it is itself an open source project [04:26] yes, it's a gross hack and it sucks [04:27] and fundamentally unsupportable === tseng notices that only some bits of asterisk are in universal [04:43] i might have to fix that next week. [05:08] evening folks [05:09] yo lamont [05:09] dieman about? [05:09] 'sup jdub? [05:09] lunchtime is up [05:09] heh [05:09] but i have not figured out what to eat [05:09] almost bed time [05:09] I hate the final week before a convention. [05:13] but nowhere near as much as I hate plumbing [05:17] mdz: having a hard time picturing 1784 as warty critical, especially since mark and I agreed to discuss auto-config of SASL/TLS post-warty [05:18] and certainly not grave for debian.. [05:18] set to important [05:18] mostly to avoid arguing about why I think it's really a wishlist bug [05:18] or dup of any one of about 6 others like it.. :-( === chris_ [~chris@chello212186066043.6.14.vie.surfer.at] has joined #ubuntu-devel [05:49] lamont: what are the postfix requests? [05:56] "grave: after I install postfix-tls, I have to actually configure SASL if I want to use it" [05:56] s/SASL/TLS/ as needed... [05:57] it's not that it does it wrong, it's that it doesn't do it... [05:57] like, add users to a sasl db, or have it use the system passwd file? [05:58] so once we have hoary opened up, that guy what knows certificates and such is gonna walk me through making postfix-tls autoconfigure sasl and/or tls.... But not before warty.. Or sarge. [06:02] jdub: hrm.. this bug came from your corner of the world... [06:10] so, what's the accepted way of turning laptop-mode off in ubuntu? because I'm fairly sure that laptop-mode is fucking up my hard drive. [06:10] will file a bugreport, if I can find some diagnostic info or something to prove it [06:13] but twice, my hard drive has made a spinup or spindown sound, and then anytime I try to access the HD, I just got an I/O error [06:13] reboot, fine... then it happened again [06:17] lamont: ahr. [06:27] lamont: that's clearly not RC [06:27] enhancement, Hoary [06:29] yeah. [06:29] I'll downgrade the debian bug further later.. :) [06:30] mdz: how are we marking hoary stuff? [06:30] lamont: target milestone: hoary hedgehog [06:30] or are we just assuming that things beyond warty will likely get revisited for hoary? [06:31] assuming mark said it should be on the list for hoary [06:31] otherwise, it doesn't need a milestone [06:31] or we need a "someday" milestone :-) [06:31] updated === doko [doko@dsl-084-057-038-142.arcor-ip.net] has joined #ubuntu-devel === jdthood [~jdthood@aglu.demon.nl] has joined #ubuntu-devel [06:47] jdub: your "spanish list" email was duplicated :-/ [06:48] but not "burning ISO image" [06:48] something weird is going on with your mail [06:48] they have different message-ids [06:48] Message-Id: <1096172902.4733.104.camel@localhost.localdomain> [06:48] Message-Id: <1096172903.4733.105.camel@localhost.localdomain> === lamont sleeps [07:26] morning [07:29] fabbione: morning [07:32] mdz: are you still around? === sivang [~pooh@CBL217-132-76-165.bb.netvision.net.il] has joined #ubuntu-devel [09:15] Where would be the right place to suggest packages for base install? pptp should be included for those on countries who must use a pptp dialer to access broadband.. [09:17] i had to go back to xp home to manually download the .deb file for pptp, and ofcourse I would have to resolve the dependecies by hand. [10:57] schweeb: remove it from power.sh [11:20] Kamion : around? [11:21] sivang: yes? [11:23] mdz: bong [11:23] sivang: pptp-linux will be on the cd [11:23] sivang: not installed by default, but on the cd [11:37] jdub : it's not currently, right? [11:43] Kamion : would it be reasonable to incorporate into base-install or post, a pptp confiugration question, just for those who lake DHCP authenticated internet broadband, who _do_ wish to update from the net? [11:44] Kamion : I'd like to try a go at it myself, however what would be the willingness to add _yet_another question to it? [11:46] Kamion : or even adding it along side the "ppp" d-i configuration scree. [11:47] well, having an easier way to configure pptp would be nice (if this is not the case already), but I doubt that the user-base is large enough to warrent an additional question anybody would see [11:48] azeem : meaning people would miss it? [11:51] azeem : moreover, we've had several problems streamlining pptp connection in israel in various distribs, debian has it going with a script that somebody wrote, and I improved, which I can adjust to all internet providers here. Would make it nice to have this built in :-) === sivang [~pooh@CBL217-132-76-165.bb.netvision.net.il] has left #ubuntu-devel [] === dand [~dand@83.103.205.136] has joined #ubuntu-devel [12:57] Is there any reason apm isn't loaded by default? === sivang [~pooh@CBL217-132-253-186.bb.netvision.net.il] has joined #ubuntu-devel [01:15] does anybody know how to disable that freakin' beep when in pure console mode? this thing's driving me crazy..:) [01:15] i disabled it for my debian sid long ago, but forgot how it goes. [01:16] : tfheen@yiwaz ~ > cat .inputrc [01:16] set bell-style visible [01:17] tnx [01:17] that would be great === sivang is still fighting to make his warty connect using pptp to the israeli broadband. === lifeless [~robertc@dsl-220.2.240.220.rns01-kent-syd.dsl.comindico.com.au] has joined #ubuntu-devel === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === elmo [~james@83.216.141.215] has joined #ubuntu-devel === sivang [~pooh@CBL217-132-86-164.bb.netvision.net.il] has joined #ubuntu-devel === sivang [~pooh@CBL217-132-86-164.bb.netvision.net.il] has left #ubuntu-devel [] === wartylog [~warthylog@port1845.ds1-khk.adsl.cybercity.dk] has joined #ubuntu-devel === Topic for #ubuntu-devel: Ubuntu development -- general discussion on #ubuntu === Topic (#ubuntu-devel): set by mdz at Mon Sep 20 07:48:40 2004 === Capri [~makolb@mnch-d9ba47c3.pool.mediaWays.net] has left #ubuntu-devel ["Bye] === T-Bone is now known as T-Gone === Capri [~makolb@mnch-d9ba47c3.pool.mediaWays.net] has joined #ubuntu-devel === Capri [~makolb@mnch-d9ba47c3.pool.mediaWays.net] has left #ubuntu-devel ["Bye] === Capri [~makolb@mnch-d9ba47c3.pool.mediaWays.net] has joined #ubuntu-devel [07:59] fabbione: here === npmccallum [~npmccallu@69-162-252-7.ironoh.adelphia.net] has joined #ubuntu-devel === vuntz [~vuntz@fennas.vuntz.net] has joined #ubuntu-devel === lamont giggles at fabbione [09:27]

stupid xfree maintainer: [09:27]

/usr/bin/make -C build-tree/xc WORLDOPTS="" IMAKE_DEFINES="-DXFree86CustomVersion='\"Ubuntu 4.3.0.dfsg.1-6ubuntu21 20040926054320 root@gropaf.esiee.fr\"' -DBuildSpecsDocs=NO -DBuildFonts=NO" World > logs/make_world.build.log 2>&1 [09:27]

make: *** [stampdir/build] Terminated [09:27]

Build killed with signal 15 after 150 minutes of inactivity [09:28] :-) [09:29] it's a good way to identify doorstop architectures [09:29] ;) [09:30] azeem: hppa is _not_ a doorstep architecture, dammit. :-) [09:32] well, my hppa box is [09:32] (with the stress on "my") [09:37] what model? === lamont contemplates filing a wishlist bug against metacity for 'focus follows users attention/eyes' [09:39] 512/40 or something? [09:40] it's lying around in the basement, haven't touched it for a year, because back then (running woody, kernel 2.2), it decided to freeze every couple of days for no apparent reason [09:40] 712/60 I could believe [09:40] pizza-box sized? [09:40] yeah, that's it === lamont has one of those kicking around here somewhere, he thinks. [09:41] my firewall is a B180 [09:41] I have to admit I got it for free because the university dumped it, a friend of mine picked it up, couldn't connect a monitor to it and then wanted to throw it away [09:41] so the hardware might be broken [09:55] nah 712's are dog slow. [09:57] Jens Schmalzing sent me a mail today that he installed potato on my father's old Mac SE/30 :) [09:57] *that's* dog slow :) === lamont one time put winblows 95 on a 386 w/16MB (back when) [10:00] bunch of folks said you couldn't do that... [10:01] heh [10:02] I remember I got royally pissed off by Microsoft when I upgraded my 386 from 4 to 8MB because the preview releases said Windows95 would run fine on it, and when it released, you needed a much better system [10:08] that machine had 15 minute boot times. Literally [10:13] Grumble. No jpilot happiness with USB sybc. [10:16] so, how do I get evo to honor the categories on my palm... [10:16] address book, that is.. [10:17] what kind of palm do you have? [10:22] tungsten C [10:24] lamont: AFAIK does the pilot-link stuff not handle the "new" PalmOS 4.x(?) features. I know it doesn't handle birthdays in contacts on my mother's Tungsten E [10:25] dunno [10:25] was justing jpilot, talking to a serial dock... [10:26] but I just gave the serial doc to my wife... [10:26] hence I need to figure out USB... === lamont upgrades his machine, just to stay current. === lamont reboots, for completeness === Mithrandir starts compiling coreutils on a microVAX running netbsd 1.4. I think it takes more time than lamont takes to reboot. [10:31] .. to run the configure script. === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [10:33] moof === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [10:41] and hotsync works again. yippee === Mithrandir notes said configure script has output about ten more lines since lamont rebooted. [10:41] LOL [10:42] USB sync is much faster than 9600 baud rs232 sync, too. :-) [10:42] god, I'm glad it caches the results. [10:42] you doing the microvax-ubuntu port, then? [10:44] so far, I'm compiling coreutils to get rid of a cronmail I've been getting for half a year. [10:44] I wonder if mark would _really_ buy the hardware to do such a port. :) [10:45] Mithrandir: he requires a _credible_ support plan. :-) [10:45] would take a double dozen machines just to _BUILD_ the beast every 6 months, I expect. :-) [10:46] we're using it in production at the computer club at the university -- that uVAX is our primary nameserver. [10:46] Mithrandir: you should steal azeem's 712... It'll scream! [10:46] we have a bunch of those as well. [10:48] apart from the fact that it's hell to log into, it's a nice box. === srbaker [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel [10:54] yo [10:55] hiya srbaker [10:55] so did anyone else start looking into the ubuntu being really slow on toshiba lappies? [10:55] it's still fast on my gf's :) [10:55] grrr === T-Gone is now known as T-Bone [11:20] lamont: ping? [11:20] yo [11:21] lamont: good news: both sbuilds are still running [11:21] yeah! [11:21] lamont: and for zsh, my mistake: sbuild wasn't killed. Looked more or like "stuck", had to kill -9 all processes === lamont is up to #420: totem [11:21] ah, ok [11:22] mrvn has reported problems with building zsh as well [11:22] T-Bone: zsh has a problem being autobuilt, we see it on amd64 as well [11:22] Mithrandir: good to know. It's now in final position in my sbuild list [11:22] T-Bone: are you are using the sbuild from db.debian.org? [11:22] lamont: #350 on ia64 [11:22] azeem: right [11:23] T-Bone: it should build fine if you SIGCONT it. [11:23] T-Bone: my xfree86 died (ia64 only here) for lack of disk space. :-( [11:24] lamont: failed on ia64 are: cyrus-sasl2, enchant, libcaca, lsb-release, mysql-dfsg, ntp, pbuilder, perl_5.8, and syck [11:24] and a bunch of gnome build-dep issues, I expect. [11:24] lamont: lol. This shouldn't happen here (20G free) [11:24] sasl2 will be a pain for us - what was the failure [11:25] [varenet@envy ~] $ grep ^Built logs/xfree86_4.3.0.dfsg.1-6ubuntu21_20040926-1716 [11:25] Built successfully [11:25] ah, that was hppa that timed out,eh? [11:25] lamont: right [11:25] fabbione: mail logs are all happy with length now - You can quit redirecting.. :-) [11:25] it's imho plain stupid to redirect all output to a file [11:26] lamont: i wonder how this would do on really _slow_ archs (m68k, arm and friends) [11:26] that was a hack for the warty buildd's of a while back. [11:26] config.status: creating man/Makefile [11:26] malloc: ../bash/dispose_cmd.c:230: assertion botched [11:26] free: called with unallocated block argument [11:26] Stopping myself..../configure: line 32173: 22342 Aborted $SHELL $CONFIG_STATUS $ac_config_status_args [11:26] make: *** [stampdir/config-stamp] Error 1 [11:26] ****************************************************************************** [11:26] Build finished at 20040926-0446 [11:26] FAILED [dpkg-buildpackage died] [11:26] log > 10MB --> bounce --> wait for lamont to manually fix it. === T-Bone goes like "Ouch, really ouch man" [11:26] t-bone: that's a bash bug [11:27] if bash is already built, install that in the chroot [11:27] then find all of the ones that died with that error, and retry them. [11:27] that is, once the run finishes... [11:27] it's not already built [11:29] lamont: #340 on hppa [11:29] 350 actually [11:29] cool [11:30] lamont: failed on hppa are: enchant, lsb-release, mozilla-thunderbird, pbuilder and xfree (because of timeout) [11:30] heh, pbuilder detects it's being built by sbuild and cops out? *g* [11:31] # Some packages may exceed the general timeout (e.g. redirecting output to [11:31] # a file) and need a different timeout. Below are some examples. [11:31] %individual_stalled_pkg_timeout = ( [11:31] xfree86 => 600, [11:31] }; [11:31] in .sbuildrc [11:32] azeem: nah - pbuilder builds pbuilder_..._all.deb, but claims to be arch: any [11:32] so it's ftbfs, but we don't care. [11:32] lamont: http://gropaf.esiee.fr/~varenet/ [11:32] ugh [11:32] azeem: yea [11:34] using a source builder from somebody who confuses 'any' and 'all' would make me uncomfortable I guess ;( [11:34] eh, :-/ or something like that [11:37] lamont: i've put some stats, updated hourly, and access to the logs folder there. Setting up the same on envy ;) [11:40] lamont: http://envy.esiee.fr/~varenet/ -- There you are. Easier than ssh ;) [11:40] heh [11:40] thanks [11:40] if you need anything else useful there, let me know :) [11:40] hmm. i'll add a reminder of the arch, too ;) [11:41] lamont: fwiw, 2.4 is definitely much more stable than 2.6 on ia64 [11:41] figures [11:41] on the bright side, warty _does_ support installing 2.4 [11:42] lol, that's a relief ;) [11:43] fuck. i wanted to chase down the toshiba problems today, but i do'nt think i'm going to get to === lamont spends a while, then notices that .gz != .py, sighs, bangs head on wall. === T-Bone spreads some nails on the wall... runs away! [11:49] does ubuntu use gksudo where debian uses gksu? [11:50] yes [11:50] ahh [11:59] the thing i like about ubuntu is exactly what Edd said. "ubuntu gives me a system that is exactly how i would have configured my debian system"