[12:02] momentarily [12:04] given back [12:06] thanks === mdz beats dpkg-reconfigure into submission [12:14] Setting up cupsys (1.1.23-1ubuntu2) ... [12:14] /var/lib/dpkg/info/cupsys.postinst: line 31: ps: command not found [12:14] /var/lib/dpkg/info/cupsys.postinst: line 39: ps: command not found [12:14] chown: cannot access `/usr/bin/lppasswd': No such file or directory [12:14] lamont: evolution-webcal build :( [12:16] mdz: here now [12:16] mdz: giving you X_SETBACKTITLE love in cdebconf [12:17] although using it is a bit fiddly because you have to call it after the main menu starts up but before localechooser appears *shrug* [12:20] mdz: just noticed my resync failed when I fire it up (I didn't notice so no download) @ERROR: Unknown module 'daily-live' [12:21] seb128: there's already a bug in the bts about that [12:21] sivang: put cdimage/ before daily-live/ [12:21] i think this just changed my life: http://conkeror.mozdev.org/ [12:21] seb128: hrm.. then again, I think that could be my upload... [12:21] sivang: rsync is laid out slightly differently because rsync doesn't do virtual hosting, unlike http [12:21] Kamion,mdz : noted, thanks :) === sivang downlaoding [12:22] mako: I love the bottom item on their features list [12:23] sivang: note rsync isn't all that useful for the live CD yet [12:23] Kamion: I know, I just use it as means of letting me continue a broken download, should I loose power or net acess. [12:24] mako: there's a package in universe or something? [12:25] sivang: it's an XPI for mozilla so you can just click a link and install it [12:26] mako: ah nice [12:26] Kamion: i don't have a mouse either! this guy ROCKS [12:28] it's slightly a mindfuck but i think i'm loving it [12:28] "This means all the keybindings and to-die-for features of Emacs that can be imitated by a javascript/XUL web browser Just Work." [12:29] it does nxml-mode and psychoanalyze-pinhead? :p [12:29] Keybuk: hehe [12:30] there are 2-3 mouseless mozilla projects [12:30] this one seems great but kind of buggy maybe === elmo [~james@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel [12:33] evening elmo [12:33] hey [12:34] Kamion: so I was working on the debconf/cdebconf situation today [12:34] Kamion: and the cleanest solution I came up with was to modify dpkg-reconfigure to allow it to use an existing frontend [12:34] Kamion: and then run it in a chroot before pivoting [12:34] Kamion: does this sound sane to you? === lamont notes idly that the phrase "does that sound sane" generally indicates a lack of belief on the part of the speaker... [12:35] Kamion: I tried to get joeyh's opinion, but he didn't seem to be around [12:35] Kamion: I did a proof of concept using the current live CD, and this method works [12:36] mdz: what does the modification look like [12:36] ? [12:36] mdz: joeyh is around now, he was speaking on #debian-boot just a minute ago [12:36] Kamion: it basically just conditionalizes a bunch of code in dpkg-reconfigure [12:36] IME he doesn't tend to read scrollback much [12:36] I mean is it optional? [12:36] emailed you a copy of the patch [12:37] there's some stuff in base-config I think you might need to be careful about not breaking [12:37] Kamion: it automatically changes behaviour if DEBIAN_HAS_FRONTEND is set [12:37] ugh [12:37] think I'd prefer it to be explicit, but ... [12:37] dunno [12:37] well, the current behaviour just crashes and burns if there's a frontend running [12:37] so I didn't see a reason to preserve that [12:38] hm [12:38] but it's simple enough to make a flag for it [12:38] maisde you a copy of the diff [12:38] mailed, even [12:38] mdz: he said he didn't catch your earlier question [12:39] mdz: presumably you have to load the templates into cdebconf, as we discussed? [12:39] Kamion: yeah, did that [12:39] ok [12:39] cool [12:39] debconf-loadtemplate d-i /target/var/lib/dpkg/info/xserver-xorg.templates [12:39] XORGFORCEPROBE=yes chroot /target dpkg-reconfigure xserver-xorg [12:39] that's it, with the patch [12:40] Kamion: this could be useful for rescue mode as well [12:40] how so? [12:40] guided repairs [12:40] rescue mode tends to assume that any installed system that might exist is arbitrarily broken [12:41] hoping that debconf miraculously works in the target is too ambitious, usually === doko [doko@dsl-082-082-189-242.arcor-ip.net] has joined #ubuntu-devel === lamont grumbles at his machine === ogra_ [~ogra@p508EBE24.dip.t-dialin.net] has joined #ubuntu-devel [01:00] Kamion: it's not the sort of thing I would do unconditionally, but it seems like an interesting idea to be able to have menu items which could talk to debconf in the target to do reconfigure/reinstall type things [01:01] yes === lexhider [~lexhider@dip-220-235-84-93.vic.westnet.com.au] has joined #ubuntu-devel [01:02] though I think it's slightly more useful in the Debian context, to allow debootstrap-installed packages to ask questions at lower priorities === usual [~colin@alb-69-202-45-16.nycap.rr.com] has joined #ubuntu-devel === usual [~colin@alb-69-202-45-16.nycap.rr.com] has joined #ubuntu-devel === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel [01:27] /sleep === Kamion -> bed [01:34] so, -4 is busted, yes? [01:34] linux-image-2.6.10, that is [01:37] lamont: yes [01:38] then I won't upgrade from that snapshot of the archive [01:40] Mithrandir: ? [01:51] Setting up mozilla-firefox (1.0-2ubuntu3) ... [01:51] Updating mozilla-firefox chrome registry...E: Registration process existed with status: 1 [01:51] E: /usr/lib/mozilla-firefox/extensions/installed-extensions.txt still present. Registration might have gone wrong. [01:51] mv: cannot stat `/usr/lib/mozilla-firefox/defaults.ini': No such file or directory [01:51] hrm... that's not right [01:51] thom??? [01:51] yeah, I know. sleeping [01:54] hrm, i didn't get that on my upgrade [01:56] jdub: it could be that rebuilding the chroot might help [01:56] things have been installed/removed over and over and over and over in that chroot === lamont heads off to class [02:02] jdub: you have a current gnome install? [02:02] panel -> properties -> autohide... does it? [02:03] (doesn't here, but not 100% current...) === lamont really heads to class. back in 3.5-4 hours. [02:05] mdz: ... [02:05] lamont: gnome-session -> libxau-dev === sivang burns the live iso [02:17] nice to see that nautilus showing the content of the RW cd doesn't block the burner from working. === chrisa [~chris@nullcode.org] has joined #ubuntu-devel [02:22] daniels: how is the X configuration stuff coming? [02:24] mdz: good thanks, it seems to be pretty solid at the moment; the bandwidth is a bit flogged, so live CD hasn't come through yet, but I think this X upload is pretty much ready to go, and I've also dealt with two merges (one of them being glibc) [02:24] mdz: if I gave you new debs, would that be useful? [02:25] brr [02:25] daniels: yes [02:25] well [02:25] me, or the archive? [02:25] me -> no, archive -> yes [02:29] mdz: k [02:30] daniels: how does it stand? [02:30] did you implement the bit to ignore the previous values if a probe is forced? [02:30] mdz: yeah [02:30] XORG_FORCE_PROBE now throws away everything it's got earlier and does a reprobe [02:30] note the change in variable name [02:31] as well as fairly large backports of i8xx and via drivers, disabling horizontal scrolling per default, fixing the altgr-on-macs bug, dpkg-architecture, fixing the original-imacs-won't-show-anything bug, a couple of xkb stuff, this change, and a semi-related script cleanup [02:40] rock [02:47] Kamion: around? could use some cdebconf enlightenment === ogra [~ogra@p508EBE24.dip.t-dialin.net] has joined #ubuntu-devel [03:02] daniels: Thom sees the same problem with suspend/resume after playing with vbetool [03:02] Have you got nifty keen DRI save/restore functionality yet? [03:11] mjg59: wack; not yet, no [03:11] mjg59: i'm dragging back mesa 6.2 for my next xorg revision === BlackHussar [~chatzilla@corp.stamps.com] has joined #ubuntu-devel === usual [~colin@alb-69-202-45-16.nycap.rr.com] has joined #ubuntu-devel [03:41] mdz: we should sync tla 1.3 if we haven't already [03:47] is amaya supported or unsupported ? [03:47] Cause its fucked, and does anyone care ? [03:58] use apt-cache show to see whether it's in main or universe [03:59] ah universe [03:59] ok, so does anyone care;) [03:59] if you want to fix it ... [04:00] holy shit, 60.5% [04:00] the rsync may actually finish this week [04:01] :) [04:11] lifeless: patches gratefully received [04:14] mdz: for amaya or tla 1.3? [04:15] lifeless: amaya [04:15] I may have a peek if I get some time. [04:15] tla 1.3, we're in freeze now [04:15] mdz: it was released before christmas, but asuffield sucked. [04:16] mdz: its quite important to have tla 1.3, because its got the same revlib format as baz, which tla 1.2 doesn't. [04:16] mdz: the muppet who did the patch originally didn't do it backwards compatibly. [04:18] Exceptions requiring confirmation: [04:18] * Packages in or relating to FeatureGoals? [04:18] * Minor fixes, if the upstream change is a micro-increment (or equivalent) [04:18] * Major fixes, particularly blockers, if the upstream change is a minor-increment (or equivalent) [04:18] * Exceptional circumstances === lifeless claims exceptional circumstances for tla 1.3 and baz 1.1 [04:19] baz 1.1 is an internal project, so I think it deserves it [04:20] but why should we track tla releases through the freeze? [04:20] it sounds like they're actively doing feature development [04:20] because having tla 1.2 there will make it harder for users to use baz. [04:20] we don't need to go for tla 1.4, only 1.3 === `anthony [~anthony@ekorp-203-63-137-225.eoff.ekorp.com] has joined #ubuntu-devel [04:46] bazaar 1.1 didn't make the upstream freeze either, did it? [04:47] no, but that was pre-discussed [04:56] Thanks to vbetool, I've now got this Dell8600 laptop suspending to RAM, except that when it wakes the external USB mouse is no longer working. Can someone point me to a tool that might let me reset that stuff? [05:01] stub: try unloaded and reloaing the usb)uhci or whatever driver === zenrox [~zenrox@wbar7.sea1-4-10-181-189.sea1.dsl-verizon.net] has joined #ubuntu-devel [05:56] morning === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel === OddAbe19 [~OddAbe19@pcp02542642pcs.lncstr01.pa.comcast.net] has joined #ubuntu-devel [06:41] morning fabbione [06:42] hey lamont === lamont cries [06:43] I fear my mirror will never catch up [06:43] what's up? [06:43] ahhaha [06:43] i still have to upload -6 [06:43] ain't my fault... [06:43] not this time at least [06:44] nah -evo data server and mozilla-* are . [06:44] but before they get there, you will be [06:49] mdz: wvstreams uploaded. === ironwolf [~ironwolf@c-24-6-169-124.client.comcast.net] has joined #ubuntu-devel === lamont plays with a new keyboard. === ultrafunk [~pd@eth779.vic.adsl.internode.on.net] has joined #ubuntu-devel [07:07] lamont: f11 still has I/O problems [07:07] i did try to make some working configs but it takes just too much time... sorry [07:08] np. does -6 have where you're at then? [07:09] lamont: -6 has -pa10 but not valid configs [07:09] ok. [07:09] I'll work with it over the weekend or something [07:09] fabbione: do you want latest inotify patches? [07:09] right now is sleep time, thouhg [07:09] lamont: so basically what you have to do manually is to dpkg-buildpackage and while it steps in the configs try to answer in the appropriate way until it builds ;) [07:10] fabbione: http://lwn.net/Articles/118107/ [07:10] fabbione: and capture the config. [07:10] fabbione: just checking to see if that's the very latest [07:10] lamont: than copy them from debian/build/build-$flavour/.config to debian/config/hppa/$flavour [07:10] lamont: and send them to me [07:10] jdub: hold on... [07:11] jdub: it'a all documented in the source btw.. you can check yourself... [07:11] fabbione: ok. g'night then [07:11] debian/external-drivers [07:11] lamont: g'night [07:11] fabbione: our kernel source won't tell us if you've used the latest patch aailable :) [07:12] jdub: yes it does [07:12] fabbione: it can't, you need extra information, ie. what the latest inotify patch is :-) [07:12] jdub: apt-get source linux-source-2.6.10 and read debian/external-drivers [07:12] jdub: that one will tell you what version is in [07:12] and you can compare with what is available [07:12] fabbione: yes, that will tell me the version in the kernel, not which versions are available [07:12] meanwhile, i'm not downloading the kernel source [07:13] jdub: you are just lazy and i am still waiting info on your DVB bug report [07:13] fabbione: bandwidth issues [07:13] i have to get the card back to find out [07:14] fabbione: which version of inotify is in our kernel? [07:14] i'm checking what the latest 'stable' version is atm [07:15] inotify | 0.17-wip-rml-2.6.10-rc2-12 | ok | 15/12/2004 | http://www.kernel.org/pub/linux/kernel/people/rml/inotify/v2.6/ [07:15] ok [07:15] http://www.kernel.org/pub/linux/kernel/people/rml/inotify/v2.6/0.18/inotify-0.18-rml-2.6.10-4.patch [07:15] so that's a start [07:16] jdub: i have other things to deal with right now.. like a few tons of security patches [07:16] also we are in UVF [07:17] that's a new upstream version [07:18] as yet, we have no policy about kernel patch relationship with UVF [07:18] i will file a bug for you [07:19] dude.. [07:19] read the changelog [07:19] The primary change since the last post is a reworking of the locking and [07:19] refcounting. A bit more such reworking is expected. [07:20] meanign that the new version is not completed [07:20] i know [07:21] is inotify fucked? [07:21] mdz: no. jdub is asking for a new upstream version of it [07:21] but I have heard talk about it being buggy [07:21] mdz: you'll get bug cc in a moment [07:21] mdz: i believe that's mostly gamin's support of inotify, which i am considering disabling [07:22] and so what is the point of having inotify if there are only 2 thing that actually uses it? [07:23] fabbione: because everything in gnome uses gamin (via the fam abi) [07:23] fabbione: so if you're using inotify for monitoring, you don't block disks from unmounting and so on [07:23] which is preferable to dnotify [07:24] mdz: i believe that's mostly gamin's support of inotify, which i am [07:24] considering disabling [07:24] than i miss something [07:24] you need inotify for gamin [07:25] is gamin not going to be fixed for hoary? [07:25] and you want to disable inotify support for gamin [07:25] jdub: bah, LWN went and published that hack-of-the-day live CD rather than the subsequent (much nicer) daily builds [07:25] I deliberately sent that to -devel and not to -announce because it wasn't ready for that kind of publicity [07:25] fabbione: gamin supports inotify as well as dnotify, but its support for inotify is not great [07:25] it seems [07:26] mdz: not sure if gamin/inotify will be fixed, thus considering disabling the inotify support [07:26] mdz: i will track it beyond UVF to see if there are related bugfixes [07:27] jdub: have you tried the latest round of hoary live CDs? [07:27] no, haven't committed to downloading them yet ;) [07:27] 3 architectures, fewer calories, better tasting [07:27] dc-built? [07:27] daily [07:27] fresh and hot [07:27] cool [07:27] i'll pull one down during off-peak tonight [07:27] no X yet, but daniels is working on it AHEM [07:27] i've been watching the casper changes with interest :) [07:28] it's such a good idea [07:28] I must admit, it is [07:28] :) [07:28] casper is like 100 lines of shell on top of one fucking good idea [07:28] proven by beauty? :) [07:29] mdz: how do I mount the cloop image from the live CD? [07:29] mdz: or otherwise test X [07:29] daniels: boot it? [07:29] mdz: lwn love us to omuch [07:29] mdz: and, heh, yeah, they do love us === daniels stares at mdz. [07:29] modprobe cloop && losetup /dev/cloop0 /path/to/filesystem.cloop && mount /dev/cloop0 /mnt [07:29] mdz: can I netboot the live CD? === ultrafunk [~pd@eth779.vic.adsl.internode.on.net] has joined #ubuntu-devel [07:29] daniels: dude, yes you can ;-) [07:30] tftp get filesystem.cloop [07:30] ... [07:30] ;-) [07:30] jdub: d-i netinst + NFS mount [07:30] heh, nice [07:30] my laptop sort of lacks a cd-rom [07:30] ls [07:30] it should work with zero or few changes [07:30] daniels: no docking station or USB? [07:30] mdz: yum (such a good idea) [07:30] mdz: i have usb [07:30] but no docking station yet [07:30] agh, hm [07:31] daniels: I use a USB drive for this stuff pretty much exclusively; it's great [07:31] surely you could copy the cloop to your hard drive? [07:31] ran me about USD 100 for DVD+/-/RW/RAM and USB enclosure [07:31] daniels: I don't really see how the cloop image is necessary for testing, tbh [07:32] but you can chroot into it easily enough === BlackHussar_ [~BlackHuss@dsl081-081-225.lax1.dsl.speakeasy.net] has joined #ubuntu-devel [07:34] mdz: yeah, I should probably grab one [07:35] mdz: well, failing being able to boot the thing ... but yeah, I'll see if I can't go grab one [07:35] daniels: you can also boot it in qemu [07:35] sluggish, but works [07:35] qemu -cdrom hoary-i386-live.iso [07:35] i should try the powerpc one under pearpc :) [07:36] er [07:36] s/^i/someone with more bandwidth/ [07:36] someone just renamed the UserPages page in the wiki to their personal page [07:36] haha [07:36] ??changed: [07:36] -{{{This is a list of all userpages}}} [07:36] - [07:36] Una bienvenida desde el Peru [07:36] yay wiki [07:39] that's it, i'm switching to the other theme [07:40] my login is never remembered :| [07:40] you can login?? [07:40] with https, yeah [07:41] oh, the zwiki theme doesn't have a recent changes link [07:41] boh [07:43] oh man [07:43] someone sent a 16K please unsubscribe me repluy [07:43] yeah [07:44] imminent death of usenet predicted === jdub replies off-list === mdz changes the live CD to _not_ change the hardware clock [07:54] mdz: 5472.. do you want to comment on it? it can still make it for -6 but i want to know before (possibly) breaking the ABI again [07:56] if it will fix some of the problems with inotify, great [07:56] if not, I'm not sure it's worthwhile [07:56] jdub: ? [07:57] mdz: hm [07:57] leave it for now [07:58] mdz: is there an easy way to transfer files from the host OS to the live CD? [07:58] rml's next release will be more useful it seems, and it won't fix the gamin/inotify stuff [07:58] daniels: yeah, boot it :-P [07:58] heh [07:59] not that I have a working CD burner right now, anyway [08:00] jdub: ok [08:05] daniels: that's a problem [08:06] mdz: what was the rationale to use the e100 driver in place of the eepro100 ? [08:06] if there is any... [08:07] mdz: yeah, I have two dual-layer DVD burners on order [08:07] mdz: i'll be in much better shape when my shiny amd64 turns up [08:07] right now, all I have is the motherboard [08:07] fabbione: herbert said eepro100 was deprecated [08:07] iirc [08:08] ah [08:08] https://bugzilla.ubuntu.com/show_bug.cgi?id=2156 [08:08] for reference [08:08] because both Debian and us are having problems with the e100 but not with the eepro100 [08:09] maybe ask dilinger or someone, get a second opinion [08:09] T-Bone is alive ;) [08:09] we already discussed it [08:10] fabbione: I meant ask about e100 vs. eepro100 [08:11] daniels: I bought my single-layer *just* before the dual-layers came out for the same price [08:11] it can apparently read (but not write) them with a firmware update [08:11] mdz: i bought my x40 about 5 days before they brought out faster models and dropped all the prices [08:11] mdz: yeps... that's what we discussed about [08:11] mdz: whoohoo! === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel === d3vic3 [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-devel === abelli [~abelli@84.222.38.245] has joined #ubuntu-devel [08:45] smurfix: ping [08:45] hi everybody [08:45] hi abelli === sid77 [~sid77@151.11.187.66] has joined #ubuntu-devel [08:45] was it you asking for the it-devel list? [08:45] jdub: yes [08:46] jdub: but its ok now... [08:46] hi [08:46] abelli: i spoke to mako about it; it would be more appropriate to keep all devel discussion on one list [08:46] abelli: and country-team oriented stuff on one list (ubuntu-it) [08:47] jdub: i agree with that, but since we are dealing with some teachers [for projects about Ubuntu in the Italian School System] , and [08:47] many of them do not speak english, i needed an italian mailing list. [08:47] sounds like a good project for ubuntu-it [08:47] so now we have ubuntu-it@lists.ubuntu.com, EXCLUSIVELY for SUPPORT [08:48] and sviluppo@ubuntuitalia.org for administrating our projects [08:48] jdub: you can find some of them here... http://www.ubuntulinux.org/wiki/AndreaAbelli [08:49] mdz: I AM THE SHIZZLE [08:50] Word! === Treenaks pokes SuSE into accepting his jhbuild-built gnome CVS HEAD goodness [08:51] mdz: ok, I have something that works and generates an entirely new configuration with XORG_FORCE_PROBE=yes dpkg-reconfigure -phigh xserver-xorg [08:52] daniels: awesome, upload it in time for me to use it for live CD testing tomorrow? [08:52] i was actually looking for smurfix: during the CC he said he had told us to get to the meeting... but i can't remember/find his advices /in the logs .. and i dont want to appear a deserter [08:52] mdz: fo'sho [08:53] mdz: number of hours I have to upload? [08:53] mdz: i'd like to do the usual laps-of-the-data-centre thing with test builds on every architecture [08:54] jdub: what about ubuntu-it's rootly powers? i receive a cc: by mako about this. [08:54] daniels: does the change require to ship new files? [08:54] abelli: that was waiting on appointment of loco team leaders [08:54] abelli: has that happened? [08:54] fabbione: i don't think we've touched the manifests ... but I did fix an ia64 FTBFS that lamont didn't notice :) [08:54] CC meeting is too late for me to sanely get to [08:54] daniels: ehe ok [08:54] daniels: I'll be awake again at about 1800 UTC probably [08:55] fabbione: even so, there are some fairly significant code changes (i8xx from HEAD to get rid of all the horrid widescreen panel video BIOS changes, via from unichrome.sf.net), so I'd feel a lot safer :) [08:55] mdz: when does the daily build run? [08:55] daniels: now [08:55] (0800 I think) [08:55] I can run a build by hand in the morning, though, if X is ready [08:56] daniels: last time we checked the unichrome was utterly broken... [08:56] daniels: did they fix stuff around? [08:56] fabbione: yeah, and it should finally run on the craptop [08:57] mdz: ok, so let's assume 'now' is a missed target ... [08:57] mdz: i'll upload after my test builds are done [08:57] daniels: last time Treenaks was complaining about tvout being broken [08:57] mdz: say ... upload by 1000 UTC? [08:57] jdub: sorry it was batty time...yes [08:57] daniels: don't they take 4 hours to build or something? [08:57] fabbione: hm. well, the unichrome guys claim it will give us basic 2D functionality on everything they know of, so that's a huge win over what we have already. [08:57] I want to get X working in the live CD by end of day tomorrow if possible [08:57] mdz: 32min on amd64, 74min on powerpc, ~90 on i386. [08:57] fabbione: that was a long time ago [08:57] fabbione: if you have sid-installable xorg packages I can try again.. [08:58] mdz: so our worst case has it being in the archive by about 1400 [08:58] mdz: at which time you're still asleep [08:58] brb [08:58] that's fine [08:58] daniels: I thought you meant the 1000 UTC after that one [08:58] fabbione: people have been tweaking tvout on unichrome.sf.net afaik [08:58] you and your international date line [08:59] mdz: nope, the 1000 UTC in ~2h [08:59] mdz: er, it's 0800 UTC now [08:59] i'm not talking about my timezone === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel [08:59] anyway, back in a bit [09:04] mdz: http://www.novell.com/documentation/suse91/suselinux-adminguide/html/apas02.html [09:05] mdz: published by novell (haw haw) [09:05] jdub: let me guess: they like reiser [09:07] nup [09:07] it's very balanced (ie. ext3 naturally comes out as safest) [09:07] hmm [09:07] it doesn't talk about weaknesses of any of them [09:08] yeah [09:08] probably don't want to get sued [09:08] ;-) [09:08] jdub: by Hans Reiser? [09:08]