[00:00] slangasek, lamont: pong: do we care about db4.5 anymore? [00:01] doko: it's more about the divergence from debian wrt shifting to NPTL [00:01] and I expect that at least a little bit of universe will take a while to shift from db4.5 [00:03] * slangasek nods [00:03] lamont: I didn't activey change it [00:03] does openoffice.org-core have any special upgrade requirements for bdb? [00:03] no, it's just used for the help [00:03] that's the only reason I see not to kick db4.5 out of main; but then it should still get a sync or merge or whatnot [00:03] soo tightened deps should work [00:04] anyway, if we don't need divergence on the mutex configure option, db4.5 can be a sync [00:11] lamont, doko: ok, sorry for the bother; it turns out --enable-pthreadsmutexes isn't supported at all by db4.5 so was a total no-op, and Debian already DTRT with --enable-posixmutexes [00:13] slangasek: heh [00:27] * StevenK sighs at helix -0ubuntu3 failing on all arches for two different reasons === `23meg is now known as mgunes [00:44] * slangasek blahs, botching the subscription to bug #164227. Can someone take ubuntu-main-sponsors off of there, since it looks like that wasn't required? [00:44] Launchpad bug 164227 in db4.5 "Please sync db4.5 4.5.20-11(main) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/164227 [00:45] slangasek: you need someone from the team to do that [00:48] slangasek: Hold on, I can do it [00:49] elmo: sure; I assumed there'd be someone from the team on this channel :) [00:49] slangasek: Done. [00:49] You know what happens when you assume... [00:50] ... StevenK executes! [00:50] bddebian: In this case, that's crap - u-m-s is a large team and this is the main development channel [00:50] It was a joke, sheesh :-) [00:53] * Hobbsee waves [00:57] hey Hobbsee [00:58] Heya again Hobbsee :) [00:58] :) === dendro-away is now known as dendrobates [01:32] huh, http://merges.ubuntu.com/main.html looks at experimental? [01:32] slangasek: for some packages [01:33] TheMuso: ping, please deal with gnome-speech. [01:34] Hobbsee: I have done, and its in the sponsors queue. [01:34] TheMuso: ahhh, neat. bug #? [01:34] Hobbsee: how is that determined? [01:34] slangasek: i think by bugging keybuk. [01:34] Hobbsee: e.g., diffutils experimental version is listed on there, but that's not what hardy is currently merged with... [01:34] bug 162127 [01:34] Launchpad bug 162127 in gnome-speech "Merge gnome-speech 0.4.16-2ubuntu1." [Undecided,New] https://launchpad.net/bugs/162127 [01:37] right, libsvg done. [01:44] Hobbsee: you have to be tired. [01:45] dendrobates: why in particular? [01:45] * Hobbsee only got up around an hour and a half ago [01:46] Hobbsee: you were up sooo late. [01:46] i know. [01:46] work's turned me somewhat of an insomniac, i'm afraid. [01:46] Hobbsee: it must be nice to be young. That would kill me. [01:46] heh :) [01:46] Young, who's young? :-) [01:47] * Hobbsee was asleep for....7.5 hours or so... [01:47] Hobbsee: not too bad, it seems like you were on irc just a few hours ago. [01:48] dendrobates: yeah,k i know [01:48] * TheMuso applauds Hobbsee. [01:48] Thats a good sleeping time length. :p [01:48] hehe [01:48] dendrobates: i dont deal well on <6. <4 is particularly bad. [01:48] pity :) [01:48] * RAOF prefers an number around 10. [02:06] im suspecting that Hobbsee is actually a zombie [02:06] dendrobates: besides, didn't you know that green aliens don't sleep? [02:06] ha [02:07] except when they turn purple. [02:11] dendrobates: this is a more acurrate picture of Hobbsee http://www.rightblueeye.com/blog/wp-content/uploads/2007/06/zombie_bunny.jpg [02:12] *snort* [02:12] arggh [02:13] heh [02:14] * LaserJock locks that away in the "Hobbsee" files [02:15] * Hobbsee beats LaserJock [02:15] LaserJock: PONIEZ PLZ [02:15] looks more like mneptok to me, though [02:28] Hobbsee: mneptok looks far healthier than that, I'm sure [02:28] ajmitch: i wouldn't bet on it... [02:29] true [02:29] the amount of time he spends on irc, can't be good for anyone [02:29] mhr? [02:29] * ajmitch joins in the request for more ponies [02:29] Ponies! [02:30] Perhaps ShipIt could bundle a pony with each shipment or CDs. [02:30] ajmitch: the time he spends doing support cant be good for anyone :) [02:30] customer service for long periods of time == bad. [02:32] he did have the advantage of a warm, fluffy shell of insanity to protect him [02:32] hehe [02:32] only works for so long, though. [02:32] * Hobbsee wonders if banana lady as whinged to corporate yet. [02:32] Hah [02:33] * StevenK holds you close [02:33] she probably has [02:33] i'll probalby hear about it sooner or later [02:33] Hum [02:33] utterly fabricated, of course [02:33] and you won't care at all [02:34] & you can go in there & laugh [02:34] nah, they won't call me in there. they'll likely just tell my manager. [02:34] banana... lady? [02:34] bonus points to her if she fabricates it into a 3 page written complaint, though [02:34] slangasek: i had a nutter of a lady who was unhappy as she refused to read the signage, for the prices of bananas. [02:34] so called up to complain [02:35] apparently she couldn't understand the difference between which price referred to the prepack (hint, the one that says "each"), and the loose. [02:35] that's pretty special :) [02:35] apparently having the branding on the prepack, and also on the sign, was not sufficient documentation. [02:36] sorry, i'm not giving her a refund because she doesn't like the way the world works :) [02:36] well, it to her for free. [02:36] what exactly is a prepacked bananna? [02:37] pwnguin: that's where you pack the bananna beforehand [02:37] and so the bundle of them is sealed in plastic [02:37] ajmitch: yes, i thought so. [02:37] lamont: Yeah, to the yellow material called “banana peel”. [02:37] lol [02:38] so... how do I get upstart to use a new version of /etc/event.d/ttyS1? [02:38] on gutsy [02:39] lamont: It should just deal, upstart uses inotify [02:39] haven't seen the -n 3 that I added to the mgetty invocation yet... [02:40] it's been 2 days, how much longer should it need?? :-) [02:40] lamont: Kill mgetty and have upstart respawn it? [02:40] did that too [02:40] Wave your arms and beg Keybuk for help? :-) [02:41] I expect that he's sleeping [02:41] Me too [02:42] phones are useful. [02:56] StevenK: I expect that I could use the redmond solution [02:56] Hobbsee: I'm _not_ going to call keybuk at 3AM [02:56] lamont: Yeah, that might fix it. :-) [02:57] lamont: surely you'd only do it when you were very well hidden, so you didnt' die. :) [02:57] Hobbsee: I only call people at 2AM/3AM when I can justify it. [02:57] haha [02:58] * Hobbsee turns her phone off at night, but apart from that, has a "no particular hours are bad to call" policy [02:58] we knew someone who answered the phone after 10PM with "This better be good." [02:58] haha [02:58] * Hobbsee gets the odd call at 1am or so [02:58] or midnight [02:59] my call at 2AM one night went: "This better be good." "This is LaMont, I need to speak with $person." "do you know what time it is???" "Yes." "Let me get her." [02:59] haha [03:00] which is how I know that he really does answer the phone that way after 10PM [03:02] I have gotten a call at 3am local, and told the person calling, "This better be a real problem, or I will be eating your heart later today." [03:03] (It turns out it was, so they were lucky) [03:03] yes...i dont give out my home phone number precisely for this reason. [03:04] I was a sysadmin at the time, and I told people if $machine or $service call me, I don't care what time it is [03:04] hrm... cdrom jacked up. I guess I do get to reboot [03:05] StevenK: yeah. same here, pretty much [03:05] StevenK: if you're getting paid, it's different. [03:06] My father had the same first and last name as another man in the same city when I was growing up. This man had a LOT of female friends. Some of which just had to talk to him in the middle of the night. His number was unlisted, so my Dad got the calls. My Mother did not appreciate cute young things calling for Gary in the middle of the night. [03:06] actually, i had a prank caller for a while. so i started them calling back while i was interstate, which was aroudn 1am local. muhahaha :) [03:06] ScottK: Maybe your father did. :-P [03:07] StevenK: He thought it was kind of funny. She didn't see the humor. [03:07] * ScottK used to get called in the middle of the night to make decisions about stuff. I could never remember what I decided, so I always had to be sure to read the logs first thing in the morning to see if I'd been called and what I said. [03:08] keescook: you still online? [03:09] ScottK: I used to tell people if they called me to fix stuff in the middle night, *they* could e-mail the systems team to say what the problem was, and what I did, because I usually wouldn't remember [03:09] StevenK: This was way back in ancient history when there was no e-mail at the facility involved. It was only paper logbooks. [03:10] ah. paper. [03:10] and console hardcopy [03:10] those were the day [03:10] s [03:10] * lamont reboots to get the CD out of his machine [03:10] brb [03:14] StevenK: yep. redmond solution was a winner for upstart [03:15] Heh [03:21] When I was in the Navy, turning a system off/on again was often referred to as a "Raytheon reset" - Raytheon being a large, famous Defense contractor here. [03:22] I recognize the name Raytheon. [03:22] I have no idea why [03:22] they make some pretty good misiles [03:23] (they also sponsored my state Science Olympiad) [03:30] StevenK: ever use shrink tubing? [03:30] poly switches? [03:30] lamont: I've used heatshrink [03:30] That's kind of fun [03:30] oh wait. that's RayChem [03:31] /dev/md1 679749632 596355428 69586328 90% /home [03:31] sigh [03:32] I guess it might be time to do a little house cleaning [03:32] Shoulda used LVM [03:32] * StevenK ducks [03:32] nah - I just need to buy another sata controller and a couple drives, then reshape the array. :-) [03:32] Hrm. Is that gluck? [03:32] that's my workstation [03:32] Cool. ~ 650G [03:33] /dev/cciss/c0d0p8 625344192 575931328 17647140 98% /home [03:33] that's gluck [03:33] Heh, smaller, and fuller [03:33] more users, too [03:33] And a Compaq hardware RAID controller [03:33] yeah. swraid has some advantages, and some disadvantages [03:34] my machine claims to have 2 raid1 controllers [04:04] * lamont -> bed [04:15] I downloaded Ubuntu Gutsy Beta, which worked pretty well. [04:15] However, I'm now in a total bugfest. [04:16] Are there any known bugs with not reinstalling when the release came out? [04:23] spasticteapot: Are you asking if upgrading from the Beta to the Final is known to cause problems? [04:23] Er, yes. [04:25] spasticteapot: I'm not aware of any significant regressions. [04:25] It's quite normal to not re-install, but just upgrade. [04:25] I've done it on more than one box. [04:25] No troubles here. [04:27] My keyboard brightness adjustment buttons don't work....the laptop regularly refuses to go into sleep mode.... laptop_mode does not engage automatically....and power use is abysmal. [04:27] I have an X40. This is a small laptop. [04:27] do all the archive managers handle SRU proposed->updates copying or is that just pitti? [04:27] How the fnord am I pulling 13 watts running Abiword? [04:27] spasticteapot: sleep mode, known bug unknown cause on Intel graphics + core*duo [04:28] Gronk. [04:28] spasticteapot: laptop_mode does not engage because it sets aggressive options that may wear your hard drive prematurely [04:28] Er.... [04:28] How, exactly, am I supposed to use my laptop as a laptop? [04:28] I get two hours of battery life! Maximum! [04:28] spasticteapot: power usage.... hmm, good question, I have a macbook myself and am looking answers to that [04:29] spasticteapot: there was a fun internet war over laptop_mode parking the drive too much, i gather [04:29] I'm debating just selling the stupid thing, repairing my old ULV X40, and throwing in a CompactFlash hard drive. [04:29] spasticteapot: what wattage were you expecting? [04:29] out of curiousity [04:29] 13W sounds typical for non-ULV core 2 duo type hardware [04:29] I know a fellow who managed to get his T41 down to 10w. [04:29] spasticteapot: Tried powertop? [04:29] Yes. [04:29] if you spin down the hard drive that should get you 2W [04:30] if you turn down wireless that should give you 1W [04:30] With wireless disabled AND laptop_mode enabled, I get 12w. [04:30] Removing tracker/strigii for Ubuntu/Kubuntu ought to help with battery life. [04:30] More like 14w without. [04:30] Tracker? [04:30] a search indexer [04:30] spasticteapot: enabling laptop_mode doesn't really save much power [04:31] And eater of CPU cycles and disk I/O [04:31] like beagle / updatedb [04:31] I'd say 11W is an extremely optimistic idle wattage for non-ULV core 2 duos [04:31] * pwnguin pulls around 24 watts [04:31] I can only reproduce that on my macbook on an idle kernel with no init process and lowest frequency state [04:32] all modules unloaded except required to bring up an initramfs [04:32] C3 99.9% 300-500ms residency [04:32] that's what I'd consider to be an optimistic idle :) [04:32] * RAOF also pulls in ~24W [04:32] my macbook during normal use is 14.5-15W compiz running GNOME [04:32] almost stock Ubuntu setup [04:33] Anyway to undervolt my CPU? [04:33] I know you can do it through Windows. [04:33] I think there are kernel patches to do so [04:33] Gentoo had the best documentation on that, you might wanna look at their wiki [04:33] they have some really nice info about these kinds of power hacks [04:33] note that gentoo wiki is not regulated by gentoo itself [04:33] so please santiy check what you find there ;) [04:34] right it's user contributed documentation [04:34] more than that [04:34] pwnguin: sanity check and undervolting don't go together [04:34] its user contributed and not developer reviewed [04:34] basically, gentoo wiki is its own thing with its own administration [04:34] I shoulda spent the extra few hundred bucks for an X60s [04:35] They said I'd get 3 hours for the X40 in the review...3.5 for the X60s. [04:35] With what battery? [04:35] One fellow I talked to was only getting four hours on an X60s with an eight-cell battery. [04:35] StevenK: Technically, six hours with the eight-cell. [04:35] the x40s are quite old [04:35] I get over six with the eight cell on my X40 [04:36] hmm [04:36] this graph is wierd [04:36] according to gnome-power-monitor, i use between 21 an 6 watts [04:36] I had an X40. [04:37] Someone killed it. [04:37] Now I have this wretched X60, which I hate. [04:37] A lot. [04:39] interesting [04:39] 7.7% ( 10.0) totem : do_nanosleep (hrtimer_wakeup) [04:40] totem is not playing anything [04:40] 10 times a second [04:40] pwnguin: VLC does similar things on OS X for me, havne't tried in Linux [04:41] * RAOF wishes 10 wakups/sec could be 7% of his total. [04:41] well dont use nvidia ;) === tritium_ is now known as tritium [04:42] Yeah. Or x86-64. [04:42] oh yea [04:42] that'll be something to look forward to in the new kernel ;) [04:43] So it seems my X40 pulls 10W [04:43] (with wireless on) [04:43] How should I disable Tracker? [04:43] StevenK: I'm very tempted to trade you. [04:43] apt-gte remove tracker ;) [04:43] err i think its trackerd or soemthing [04:43] spasticteapot: It's my precious and not yours. :-) [04:44] I miss my X40. [04:44] Admittedly, the ability to run Compiz will be nice. [04:44] tracker is it i think [04:44] If an when the stupid bums at Intel get off their asses and write me a wifi card driver. [04:44] Or, at least, one that works. [04:44] Tracker shouldn't kill battery life, at least in gutsy. [04:45] Or rather, not too hard. It doesn't (shouldn't) index while on battery. [04:45] Well...lemme see. [04:45] Wow. 15.3 watts. [04:45] I hate this thing. [04:46] I wonder how much it will sell for on eBay.... [04:46] man [04:46] im sitting at 16.0, and its better than ever :P [04:47] Yes, well...you have a MacBook. [04:47] Because selling a laptop because it draws too much power is sensibel [04:47] no i dont [04:47] sensible [04:47] i have a toshiba tablet [04:47] StevenK: Let me rephrase that: Selling a laptop because it won't get you through more than two consecutive classes is sensible. [04:48] RAOF: tracker has an extremely negligble impact on battery life here [04:48] Maybe it's just the initial indexing. [04:48] is the idea to replace updatedb with tracker? [04:49] cuz honestly, i dont search my files that often, and updatedb is a pita [04:49] I think I'm gonna build a ram distro of Gutsy for when I'm on the road [04:49] spinning down the hard disk gets me from 15->13.5W [04:49] the best improvement I've seen [04:49] no, the best improvement is removing nvidia modules ;) [04:50] ? [04:50] pwnguin: intel baby :) [04:50] well [04:50] that means nothing to me [04:50] 24.1% ( 11.8) : iwl3945 [04:51] Hmm. [04:51] 24.1% ( 11.8) : iwl3945 [04:51] doh [04:51] 66.6% (214.8) : PS/2 keyboard/mouse/touchpad [04:52] 34.4% (116.8) firefox-bin : schedule_timeout (process_timeout) [04:52] 17.6% ( 59.8) : ohci1394, uhci_hcd:usb6, HDA Intel, iwl4965 [04:52] 14.7% ( 49.8) : extra timer interrupt [04:52] wow [04:52] what plugin is that? [04:52] pwnguin: I bet flash video based ad [04:54] hmm. even with adblock off on tomshardware its not that high, but i can see how that happens [04:55] pwnguin: and firefox is noisy anyway :) [04:59] Hello [04:59] hi DaSkreech [04:59] I would just like to know if the Live CD installer has any more steps to complete after it installs grub [04:59] hello [04:59] I think it just asks to reboot [05:00] So if that step fails and I reinstall grub from the Live Cd I should be ok? [05:00] yeah [05:00] i've had that happen several times [05:00] thank you :) [05:00] sure thing :) [05:00] long time no see, man :) I've drifted far from the Kubuntu world it seems [05:01] Kome to the Blue side! [05:01] I am your father [05:01] :) [05:01] Well dark Blue [05:01] Purple sometimes.... [05:01] I'll stay with managing that torrent klient from your side and that's enough KDE for me to take in ;-) [05:01] :-) [05:02] I see you have the Hobbsee_ one sampling short trolls as well :) [05:02] stop stealing our citizens! [05:02] haha you made her leave :) [05:02] heh [05:03] oh the other one is still here [05:03] Ha ha :) [05:03] There is no escaping the longpointystick [05:03] LongPointyStick: Would you agree? [05:03] DaSkreech: tell ya what, get your 3D right prism desktop by default and I'll use KDE for LTS :) [05:04] indeed. DOOOOOM!!!!!!!!!!!!!!!!!! [05:04] jdong: Isn't that 3D cubiod? :-P [05:04] jdong: You know we were told not to use KDE4 for LTS :-P [05:04] but if you wanna You can help us test it for hardy :) [05:04] StevenK: it's not cuboid if you don't have 4 desktops :) [05:04] jdong: Hmph [05:04] whee geometry! [05:04] People keep complaining it's not a cube cause you have two desktops by default [05:05] DaSkreech: it's not even a cube with 4 [05:05] it isn't? [05:05] DaSkreech: nah length != width [05:05] silly widescreen laptops [05:05] DaSkreech: it has 3 unique dimensions, making it a cuboid or rectangular prism [05:05] DaSkreech: when desktops != 4, it can be generalized as a right prism [05:06] I got it to balance on the tip as a Pyramid which was fun [05:06] *cough* triangular prism [05:06] but KDE4 ships in a Month! :) [05:06] haha ok I'll stop being the geometry nazi tonight ;-) [05:06] jdong: I couldn't see through it so it was a pyramid [05:06] :-P [05:06] DaSkreech: LOL [05:08] jdong: interested in running a KDE4 session ? For side testing? [05:09] not currently, but definitely in the future :) [05:09] Like in a month? :) [05:09] while we're talking about KDE, can someone triage the gutsy side of bug #163417, I really strongly think it should be handled as a -security matter [05:09] Launchpad bug 163417 in kdesudo "kdesudo+dolphin leads to command execution vulnerability" [Undecided,Fix released] https://launchpad.net/bugs/163417 [05:10] I thought that was fixed earlier? [05:10] Could be wrong [05:10] it was uploaded to Hardy [05:11] Whoot Obama wants ODF :) [05:11] I have no idea what that means === greeneggsnospam is now known as jsgotangco [07:22] Hi everybody, I would like to show some new specs about the high priority bug 53387 that has great chances to get it fixed, can-I talk to somebody about this? [07:22] Launchpad bug 53387 in wpasupplicant "Manual WPA networks doesn't connect at boot" [High,Confirmed] https://launchpad.net/bugs/53387 [07:24] The facts is that removing the file /etc/udev/rules.d/85-ifupdown.rules solve this problem. Since this file starts the network before init.d, this gives great clues [07:24] saivann: Then don't mark the device as auto in /etc/network/interfaces ? [07:25] StevenK : The device is marked as auto in /etc/network/interfaces, if it's not, it will not automatically connect at boot [07:26] saivann: If it's not marked as auto then /etc/udev/rules.d/85-ifupdown.rules will leave it alone [07:26] StevenK : But the final result will be the same, the network won't automatically connect at boot [07:28] StevenK : Is there a way to add a exception in the /etc/udev/rules.d/85-ifupdown-rules to make it ignore all wireless devices? [07:29] I daresay there is, but I don't know it. [07:31] Hum.. I want to make sure that this bug get fixed quickly since it affects a lot of ubuntu users and since we have pertinent informations on it [07:32] StevenK : Do you know how and when WPA drivers are started by netbase? [07:33] saivann: Nope, I don't, sorry [07:33] StevenK : Thanks for your help :) [07:35] asac: could probably help there [07:37] Hobbsee : Asac knows this kind of problems? [07:39] he's the network manager guy [07:39] as for *why* you're setting up wpa manually now, i'm not sure. [07:39] Hobbsee : It's the only way to get a Static IP adress on a WPA network [07:39] oh, right, yes. [07:39] forgot about the static IP [07:40] Hobbsee : Not often used :) But often enough for a lot of people to report this problem [07:41] Hobbsee : I think that asac isn't here at this hour right? When can-I talk with him? [07:41] hm, few hours i guess. not sure where he is. [07:42] Hobbsee : Thanks, I'll wait.. even if it's 2:42 am here :P [07:43] good morning [07:44] asac : When you're here, can-you ping me? [07:44] saivann: he's on holidays [07:45] uh oh, it's morning for these people I only see at night [07:45] *looks at clock* [07:45] Haha [07:46] dholbach : Ho.. by holiday you mean that he's going to be absent for weeks? [07:46] jdong : Same here :) [07:46] saivann: I think he's away for 2 weeks or something, but I don't know for sure [07:46] now all I need to trigger bed is pitti signing on and me poking him about Azureus [07:47] dholbach : Well thanks a lot for this information, I can now go to bed :) [07:47] saivann: sleep tight :) [07:48] jdong : Thanks for all your great work on azureus! [07:48] saivann: thank you, and thanks for combing through all the bug reports for me :) [07:48] dholbach : hehe [07:48] jdong : I'm a new bug triager, so it's my pleasure :) [07:49] saivann: haha, the glamor of launchpad will soon wear off ;-) [07:49] haha [07:49] Well anyway, good night everyone [07:49] likewise, this sleep pattern isn't good for me ;-) [07:50] jdong : :D you can say good morning if it's more appropriate [07:50] haha don't depress me like that, I've got a long day ahead of me tomorrow ;-) [07:51] jdong : Own sorry :P Well that's depressing to sleep, I must force myself to sleep, so I must go right now before I stay here another hour [07:52] ha, if I don't have this silly school thing during the day I think I'd be here 24/7 :D [07:52] I can understand this :) @++ [07:52] * jdong checks e-mail and forums *one last time* [07:52] jdong: See you in an hour, then [07:54] StevenK: haha fortunately all are empty :) [07:54] jdong: Heh [07:57] you know, a thought just occurred to me [07:57] surely someone's thought of an aalib compiz plugin [07:58] Oh, *TWITCH* [07:58] mplayer supports libaa, and libcaca === jamesh_ is now known as jamesh [07:59] yes, but what about fullscreen aalib'ing? [07:59] Fullscreen the movie, done [08:00] i dont think you understand the point [08:00] or rather, you're looking to hard for one === mpt_ is now known as mpt [08:01] an aalib compiz plugin doesn't fill some need by being used, it mainly works by merely _existing_ === ion__ is now known as ion_ === \sh_away is now known as \sh [08:06] <\sh> moins === DrPepperKid is now known as MacSlow [08:32] v [08:38] u [08:40] ups [08:40] so much for Ctrl-v [08:58] hi seb128 [08:58] Morning seb128 [08:58] hey lool, hey mvo, hey seb128 [08:58] dholbach! [08:59] * Hobbsee waves to mvo [08:59] Hi everybody [08:59] hey seb [08:59] * mvo hugs Hobbsee [09:00] hey Hobbsee [09:01] * Hobbsee hugs mvo, and attempts not to drown in all the paper [09:02] * Mithrandir throws Hobbsee a paper shredder and a life buoy [09:03] thanks Mithrandir! [09:03] Mithrandir: i'll need that for some of it, anyway [09:15] jdong: FWIW, there are a few things ubiquity does after installing grub: installing extra packages (may be a no-op, depending), removing extra packages (definitely not a no-op - removes stuff like ubiquity itself), and copying logs [09:16] StevenK: did you add some preinst magic for gimp? that's required otherwise dpkg will not replace the documentation symlinks (see hal for one example) [09:21] hi all. is heron stable enough to run yet? [09:21] m3ga: no. [09:22] problems? [09:22] m3ga: lots [09:23] m3ga: if you are asking whether it's stable enough to use yet, it's not stable enough for you. It's in early development and stuff breaks and is fixed daily so people are expected to be able to recover their system from breakages. [09:23] ok, now is too early, but how do i prevent getting too late. I got gutsy about a month before release and there are about a dozen bug that affect me and were never fixed before release. [09:23] where is the sweet spot? [09:24] there's no magic point as to when you'll get your bugs fixed. We have more bugs than we manage to fix [09:25] m3ga: Id you've a spare (non-production) machine, and want to test early, I'd suggest starting around Alpha 5 (just after feature freeze). There's still lots of bugs then, but the base system is less likely to be changed on the fly. [09:26] the sweet spot for getting your bugs fixed is a lot sweeter if you help to contribute the fixes :-) [09:26] persia: is there an estimated date for that? [09:26] m3ga: mid to late February [09:27] cjwatson: I have submitted bugs, with patches, that have been ignored (possibly because they were after release) [09:27] m3ga: see https://wiki.ubuntu.com/HardyReleaseSchedule [09:27] m3ga: But, as cjwatson says, if you can track down bugs and causes in gutsy, and check against what changed in hardy, and submit solutions or pointers to patches (or even patches themselves), that significantly improves the chances that they will be fixed. [09:27] patches submitted after release are certainly extremely unlikely to be applied in that release. The next one is a different story, of course. [09:28] cjwatson: what is the best way for those to be raised to the attention of a committer ? [09:28] we do not change stable releases except for a very small number of cases [09:28] lifeless: for stable releases? usually, not [09:28] cjwatson: for hardy; [09:28] we are not yet at the stage when I would expect people to be trawling for patches [09:28] developers are still engaged in merging changes from upstream [09:28] Sure. I mean, is there a tag that can be put on the bug? Or a status change? [09:29] there's already a search for bugs with patches available, which I think should be sufficient [09:29] lifeless: the "patch" tag helps change patches into debdiffs, and debdiffs can be sent to the sponsor queues. [09:29] (of course not every patch is suitable for immediate application) [09:30] Right, I was asking to refresh my mind and get info for m3ga. Thanks. [09:31] I'm not sure I entirely believe the search for bugs with patches in all cases, mind - for instance bug 9006 has no patch attached, nor do its duplicates [09:31] Launchpad bug 9006 in grub-installer "system unbootable due to old BIOS" [Medium,Confirmed] https://launchpad.net/bugs/9006 [09:33] https://bugs.edge.launchpad.net/ubuntu/+source/grub-installer/+bug/9006/attachments/1133 [09:33] Launchpad bug 9006 in grub-installer "system unbootable due to old BIOS" [Medium,Confirmed] [09:33] ^ says it's a patch ;-) [09:34] stretching the point a little, I feel [09:35] thanks, at least that shows me how to remove the patch tag [09:35] how so? That's how the patch query works [09:35] it shows you all the open bugs with things attached that mark themselves as patches [09:35] for me, dapper was the last good release, followed by feisty. for me, gutsy is almost as bad as edgy. with ubuntu the fine details of ubuntu-core are good, but I need a lot of other stuff around it and when that stuff breaks, i'm up the creek without a paddle. at least with debian testing, that stuff does get fixed, but with debian, the core stuff lacks the attention to detail i like with ubuntu. [09:36] I mean that it was stretching the point for the submitter of that attachment to tag it as a patch [09:36] oh, right [09:36] yes [09:36] that's very true [09:41] m3ga: generally, if your patches aren't being handled it's not that there wouldn't be interested to fix bugs, but that there are not enough people to handle everything. the easier you can do it for the maintainers, the better - ie. you would supply patches, get other users to test the patches and supply comments, check that you can generate packages with those patches, work on getting patches to Debian, too (from where they'll be merged to [09:42] Hi all! [09:42] personally I like to fix things in Debian and then get them merged to Ubuntu, that way it benefits the most and is the least amount of work for Ubuntu developers [09:44] Mirv: i do find the debian process better (reportbug vs launchpad is the obvious one), but to fix bugs in debian I should be running debian. [09:44] m3ga: ah yes, my point was assuming one also has a Debian installation somewhere [09:45] m3ga: you know about apport right? [09:45] apport? [09:46] m3ga: ubuntu bug tool (what is used to submit crash bugs, etc) [09:46] ok found apport. not really for me, i'm much better off with gdb and valgrind === seb128_ is now known as seb128 [09:47] i'm a developer, i can dig into code and fix stuff [09:47] * seb128 kicks xchat-gnome [09:48] m3ga: just mentioning because of your comment about reportbug against launchpad [09:50] indeed, apport ships a tool called 'ubuntu-bug' ... [09:57] cjwatson: is ubuntu-bug anything like debian's reportbug? [09:58] m3ga: it does send useful informations to launchpad (version of packages installed, architecture, etc) [09:58] it does not share code. it fulfils much of the same function [10:00] jdong: hello! I want to do some backports with the latest compiz from hardy to gutsy-backports. should I talk to the backports team first or can I just go ahead and ask the archive-amdins for it myself? [10:07] c [10:07] ugh wrong tab [10:09] cjwatson: i just played with ubuntu-bug. it does a small part of what debian's reportbug does, but doesn't allow reporting of bugs like https://bugs.launchpad.net/ubuntu/+source/cupsys/+bug/163704 or https://bugs.launchpad.net/ubuntu/+source/usbutils/+bug/159189 [10:09] Launchpad bug 163704 in cupsys "'cups-config --libs' gives spurious output" [Undecided,New] [10:13] m3ga: -> pitti [10:13] (when he's around) [10:14] i've met pitti, but what am i supposed to see him about? [10:17] m3ga: he's the apport developer [10:17] and he kicks ass [10:18] not mine i hope ;-) [10:19] so, does apport run outside gnome/kde? if so how? there's no man page :-( [10:19] m3ga: there is an apport-cli [10:20] m3ga: and it has a manpage [10:20] man apport-cli [10:20] m3ga: https://bugs.launchpad.net/ubuntu/+source/gnome-utils/+bug/164249 [10:20] Launchpad bug 164249 in gnome-utils "Mozilla crashes" [Undecided,Invalid] [10:20] that's an example of bug sent using apport [10:21] m3ga: what does reportbug do in your opinion that apport doesn't? [10:21] hey coNP[uni] [10:21] hey seb128 [10:21] hey coNP[uni] [10:21] heya dholbach [10:22] ok looking at apport-cli manpage now. that seems to cover most of what reportbug does. [10:25] mvo: jdong certainly wants to backport compiz [10:26] Hobbsee: ok, that is all cool then [10:27] seb128: 164249 is not a particularly good bug report :-) [10:28] and i have a program that crashes, how do I make it generate a .crash file? [10:30] seb128_: I didn't remove the preinst magic, since I thought doko added it. [10:30] StevenK: ok, it was already there [10:30] StevenK: for quite some packages he did symlink the directories from debian/rules [10:31] and the symlink are not replaced with the real directories on upgrade [10:31] seb128_: Yeah, I think doko learnt that lesson. :-) [10:31] m3ga: Hey! [10:31] m3ga: I didn't know you ran Ubuntu. :-) [10:31] StevenK: there's nothing to learn; just keep the preinst stuff where you like it. [10:32] m3ga: no, but that was the first example in my box with apport informations [10:32] doko: Actually, maybe cdbs should do the preinst magic [10:32] StevenK: long time no see/hear/chat! [10:32] * Hobbsee wonders when we get a new kernel. [10:32] m3ga: Indeed! [10:32] been running ubuntu for about 2 years, debian before that [10:32] m3ga: you don't do anything, the crash should be written in /var/crash, just double click on it with nautilus to send the bug (or use the apport command line on the crash) [10:33] m3ga: I knew you were a Debian bloke, didn't know about Ubuntu [10:33] m3ga: I became an Ubuntu member about 2 years ago. :-) [10:34] StevenK! [10:34] mpt! [10:34] How's HardyAboutUbuntu? [10:34] mpt: So, what are we going to do about this spec? [10:34] * StevenK grins [10:34] using it at work, basically building an ubuntu derived distro for our embedded boxes [10:35] m3ga: I'm using Ubuntu at work. :-) I started working for Canonical about seven weeks ago. [10:35] congrats! [10:35] m3ga: Thanks. :-) [10:35] StevenK, I propose simplifying it to something you have time to implement for 7.10 [10:35] mine is www.bcode.com [10:35] mpt: That's rather time-efficient. [10:36] Canonical breaking new ground, as usual :P [10:36] mpt: I saw that edit to the wiki page, what do you have in mind? [10:36] Fujitsu: Shush you [10:36] StevenK, the original design, in GTK and Qt versions? :-) [10:37] Heh [10:37] Keybuk: Ping [10:38] * StevenK checks if he wins at gimp [10:38] seb128_: Did you see my report? [10:39] StevenK: yes ;-) [10:39] * seb128_ hugs StevenK [10:39] * StevenK grins and hugs seb128_ back [10:40] StevenK: you didn't do the preinst thing though, I just looked [10:41] Hrm. [10:41] StevenK: http://paste.ubuntu.com/2125/ [10:41] StevenK: you need something similar to that [10:42] Yeah, that requires knowing what packages were changed. [10:42] StevenK: gimp and gimp-python [10:42] those are the ones symlinked in gutsy [10:42] * StevenK sighs, wanting dpkg -I to take multiple .deb arguments [10:42] oh, maybe some other ones as well [10:43] those are the ones installed on my laptop in fact [10:43] likely most of the binaries [10:43] Drat, CDBS should do it for me! [10:43] there is very few packages b0rked by doko [10:43] * thom guffaws [10:43] (Since CDBS is the one that does the symlinking) [10:44] but you have a point [10:44] then fix cdbs at your peril :P [10:44] I had to add like 15 preinst to evolution-data-server [10:44] seb128_: Okay, so I should I fix CDBS, and then re-upload a rebuild of gimp? [10:44] StevenK: cdbs does the symlinking, the preinst is to workaround what doko broke in gutsy [10:44] thom: Shush you [10:45] StevenK: how do you want to fix that in cdbs? [10:45] Actually, it's hard to do it in CDBS, now that I think about it [10:45] StevenK: you want to add a similar preinst to every single package in the archive when it's required only for like 10 packages? [10:45] I think we should just fix those manually [10:45] * StevenK nods. [10:46] StevenK: gnip [10:46] you likely need a preinst by binary [10:46] seb128_: Likely, I'm just going to see what doko did [10:46] Keybuk: Can you read backscroll - the conversation between me and mpt about HardyAboutUbuntu? [10:47] he basically symlinked all the /usr/share/doc/binaries to /usr/share/doc/one-binary [10:47] and dpkg preserves the symlink on upgrade now [10:47] StevenK: I have no backscroll [10:47] <\sh> keescook, jdstrand : bug #164007 ready for review, thx [10:47] Launchpad bug 164007 in net-snmp "[net-snmp] remote Denial of Service vulnerability" [Undecided,In progress] https://launchpad.net/bugs/164007 [10:47] Keybuk: Oh, hang a sec. [10:47] seb128_: Oh, I see! I was too overzealous with my culling of debian/rules. :-( [10:48] StevenK: well, you can drop the debian/rules changes [10:49] seb128_: Using "gdm-new" will probably look weird when we have a new new; perhaps "gdm-snapshot"? I proposed exactly this during the UDS session, but people seemed to prefer uploading the gdm package itself [10:49] StevenK: but now cdbs symlink individual files and not the directories so you need to remove the directories symlink in the preinst so the new version can install its real directory and the symlinks there [10:49] lool: new new? [10:49] Riddell: shouldn't KDE 4.0 RC1 be "Kalamity" ? ;) [10:49] seb128_: ... I can? [10:49] StevenK: you need too [10:50] seb128_: using the name "new" looks bad after a while [10:50] StevenK: the debian/rules was doing directory symlink, now we have real directories and files symlinking [10:50] seb128_: Right. So the preinst hacking needs to stay, but I now I remove the symlink and mkdir [10:50] Keybuk: the 4.0 codenames have a deliberate policy of avoiding forced Ks, something to which I entirely approve [10:50] lool: the idea was to not keep it a while, but I'm fine with snapshot which is the standard naming for such thing anyway, thanks for the comment [10:51] StevenK: yes, cdbs does everything for you but removing the previous symlink [10:51] lool: I was not for having a different package because I was under the impression that the new gdm was testable which is not really true at the moment [10:52] lool: you can't even select the language and session, there is no gdmflexiserver, no gdmsetup, etc [10:52] seb128_: Right. Give me a few seconds, and I'll belt with a debdiff [10:53] seb128_: Right; I can only be happy that you meet my original proposal :) [10:53] Keybuk: Relevant bits at http://paste.ubuntu.com/2126/ [10:53] lool: happy that new gdm sucks? [10:54] lool: I would be happy if it was usuable rather [10:54] * StevenK ponders how to fix his webserver [10:54] seb128_: Happy with deciding to package it as a separate package as I really believe it's not ready to replace the regular gdm either [10:54] But then I was mostly alone in this position during the session [10:55] StevenK: err, riiight? [10:55] I don't see why you ping'd me though [10:56] Keybuk: Because we spent like an hour discussing this, and now HardyAboutUbuntu looks suspiciously like AboutUbuntuRevisited [10:57] StevenK: except that we changed it to have a summary of your hardware and be "about your computer" [10:57] and not "about ubuntu" [10:57] Keybuk: But mpt's point is that seems too complicated for an about dialog. [10:58] why? [10:58] it depends what the purpose of the dialog is, no? [10:59] mpt: Come back :-) [10:59] re [10:59] it comes back to use cases [10:59] my dsl linux sucks today [10:59] lool: I was saying [10:59] Keybuk: Mmm [10:59] lool: happy that new gdm sucks? [11:00] lool: I would be happy if it was usuable rather [11:00] the use case we discussed in the meeting was "answering the most common questions about the computer" === seb128__ is now known as seb128 [11:00] seb128_: Your DSL linux? === pedro is now known as pedro_ [11:00] seb128__: I was clarifying that I'm not happy that it sucks... Just happy that others come to realize it does suck ATM [11:00] StevenK: s/linux// [11:00] Keybuk: Right, which the current specs deals with by showing ... [11:00] lool: nobody played with it at UDS so we were not discussing the current state [11:01] lool: can't have an opinion on what you didn't try [11:01] seb128: http://people.ubuntu.com/~stevenk/gimp_2.4.1-1ubuntu2.debdiff [11:01] seb128: People seemed to have a strong opinion on basing on it [11:02] lool: desrt from an upstream point of view had one, yes [11:02] StevenK: why do you generate the preinsts from debian/rules rather than adding normal .preinsts? [11:02] StevenK: I have a phone call now [11:02] I'm not sure I understand the objection to the spec I'm afraid === cprov-away is now known as cprov [11:03] Keybuk: Let me reclarify with mpt [11:03] either we have an "About Ubuntu" dialog which just says "Ubuntu 7.10" [11:03] He wasn't alone in pushing for the new gdm; people are seduced by its new architectures which is nicer and more versatile but disregarded its stability [11:03] (or functionality) [11:03] StevenK: I'm not sure I like have a "rm -rf" in a preinst, especially that's not required to remove a symlink [11:03] or we have "About your Computer" which helps you answer common questions (and one of the incidental details there is Ubuntu 7.10) [11:03] we discussed the latter in the BOF [11:03] lool: not really, we said that if the new one was usuable we should try pushing for it [11:03] * StevenK nods [11:04] seb128: Right, let me rip that apart again [11:04] lool: we did pronounce opinions on whether it was or not since we didn't try it [11:05] lool: what came out of the discussion was "package new version quickly, look at it, and revert at feature freeze if that's not satisfactory" [11:05] (my personal opinion is that "About Ubuntu" is entirely useless; if all we do is display the lsb info, we can almost certainly drop it) [11:05] seb128: Yes; and the plan was to replace the package in the archive with a 2.20.x+really-2.21 version; I do recall that [11:06] lool: right, *if the current version was good enough for hardy testing*, which is the best way to get testing [11:07] lool: I still plan to do that when it'll allow to select a language and session [11:07] Too bad nobody took notes during the discussions [11:07] we likely have notes on gooby and wiki no? [11:07] anyway my understanding was that we want to get as much testing as possible [11:07] The wiki doesn't mention the packaging plans or the security review [11:08] and the way to get that is to update in new version in hardy as soon as we consider it ready to been pushed to unstable users [11:08] s/in/to the [11:17] seb128: debdiff updated. [11:22] * Hobbsee waves to mdz_ [11:24] StevenK: I like the http://paste.ubuntu.com/2125/ variant better but yours should work too, no need to mkdir though, the packages installed files to this directory so it'll be created automatically [11:24] s/installed/install [11:25] seb128: Oh right, so do it in just gimp.preinst? [11:25] Geez I'm slow tonight [11:25] StevenK, can I decide to start merging, and subscribe sponsors teams? [11:26] xhaker: Start merging what? [11:26] i'm looking for instructions on merges procedures [11:26] StevenK: no, by binary, the hal way or listing several package is wrong because you don't have guarantee on the order I think [11:27] seb128: So that snippet is in each preinst? [11:27] StevenK: I would say remove the mkdir and add a || true to your version and that's alright [11:27] we don't want to fail the install if the rm doesn't work [11:28] maybe rm -f symlink too [11:28] seb128: Er, read the if link [11:28] && [ -L /usr/share/doc/gimp-dbg ] === Adri2000_ is now known as Adri2000 [11:29] StevenK: right, I did read this one [11:29] just remove the mkdir then ;-) [11:29] seb128: Right [11:29] StevenK, universe packages... i don't know.. whatever i can [11:29] seb128: Let me just give you the debdiff again so we can be certain. [11:29] ok [11:30] xhaker: You want to ask in #ubuntu-motu [11:31] seb128: debdiff updated. [11:31] At least I don't need to upload a 25Mb tarball. [11:33] StevenK: that version looks good now ;-) [11:33] seb128: Great! I'll upload it now. [11:33] how do I get something not to show in the autoremove list ? [11:33] StevenK: thanks [11:33] Mez: sudo apt-get install it? [11:33] seb128, it's installed, but it shows up saying [11:34] The following packages were automatically installed and are no longer required: [11:34] Mez: --reinstall maybe then [11:34] I dont want to accidentally remove it [11:34] (It's iptables) [11:34] mvo: ^ [11:35] Mez: apt-get install packagename [11:35] Mez: that will mark it explicitely installed [11:35] Mez: what package is it? [11:35] mvo: even if it's already installed? [11:35] * Mez shrugs - I did a --reinstall and that worked... [11:35] seb128: yes [11:35] mvo - iptables [11:35] mvo: too to know [11:35] mvo: good to know [11:36] Mez: interessting, its a rdepends of ubuntu-standard, is that a machine that got upgraded all the way back from edgy? [11:37] er, I believe it was a fresh feisty install [11:38] might have been an edgy install - but I doubt it [11:38] Mez: hm, iptables should be marked as manual install in this case. odd [11:39] mvo, is there a way I can check if I upgraded it ? [11:40] this was a pretty fresh server actually.. only a month old.. [11:40] Mez: I can't think of a easy way to check other than checking /var/log/dist-upgrade [11:41] no such file [11:41] Mez: ok === dholbach_ is now known as dholbach [11:41] mvo, checking with my server guy, he'll tell me how it was provisioned [11:42] I have a feeling it wasnt a default install [11:42] Mez: thanks, that would explain it :) [11:42] I think he too a pbuild and played with that ;) [11:42] heh :) [11:42] so probably built using essential stuff only [11:43] seb128: Uploaded. Bloody gimp's 7Mb .diff.gz :-) [11:53] StevenK: holy crap, what's in there? [11:53] Amaranth: you shouldn't need to ask. [11:53] Amaranth: crack, and misspellings. [11:54] Amaranth: Uncompressed, it's 38Mb, which is larger than the tarball [11:55] Amaranth: ~ 70% is translations [11:55] i thought we stripped those [11:57] keescook, hi :) [11:58] Amaranth: From the binary, yes [12:02] doko: is pycentral maintained in bzr? I would like to change it so that it does not use /usr/bin/dpkg-querry but the dpkg-querry that is in the PATH. I'm happy to pass a special environment like DPKG_QUERRY too if you prefer that. the reason is that during dapper->hardy we will use a private dpkg that supports triggers and pycentral may be called before /usr/bin/dpkg-querry gets updated with the new dpkg - this would rsult in failures because [12:02] dpkg does not understand the new trigger keywords yet [12:03] that sounds so scary [12:04] mvo: either that, or just read /var/lib/dpkg/info/*.files directly (using an environment variable) [12:04] * Hobbsee attacks fabbione with a herring [12:04] * fabbione waves his 20T Thor Hammer in front of Hobbsee [12:05] doko: I don't mind either way, just let me know how you want the change from me. just uploaded as a new package or in a bzr repository? [12:05] fabbione: you need longer arms. [12:06] Hobbsee: buh [12:06] * fabbione heads offline [12:06] mvo: just upload, and maybe change it (make that conditional on an env var): #lines = [s[:-1] for s in file('/var/lib/dpkg/info/%s.list' % self.name).readlines()] [12:06] cmd = ['/usr/bin/dpkg-query', '-L', self.name] [12:06] should be faster anyway [12:07] doko: thanks! that sounds good. I will do that after lunch === stu1 is now known as stub === TiMiDo_ is now known as TiMiDo === Kranz is now known as DktrKranz === seb128_ is now known as seb128 === svolpe_gerrath is now known as svolpe [12:40] <\sh> keescook, jdstrand : bug #149616 ready for review [12:40] Launchpad bug 149616 in ruby1.8 "Net::HTTPS Vulnerability" [Undecided,In progress] https://launchpad.net/bugs/149616 [13:36] TheMuso: erm, did i ever upload gnome-speech for yoU? if i didn't, remind me tomorrow [13:57] oh, WOW, debmany(1) is cool [debian-goodies package] [13:57] why did I not think of doing that? [14:03] cjwatson: read man pages from unextracted debs? cute [14:04] unextracted, or installed, or go and download them from the archive [14:04] in general, "show me a list of the man pages in this package and let me read them conveniently" [14:06] oh wow [14:06] the description doesn't do it justice then [14:11] Hi there, can someone take a look at https://bugs.edge.launchpad.net/ubuntu/+source/openoffice.org/+bug/132583 [14:11] Launchpad bug 132583 in openoffice.org "python-uno can't be imported" [Low,Confirmed] [14:11] it's milestoned for hardy [14:11] but there's a one-line fix, and gutsy has been released one month ago [14:12] I find it unbearable that we ship buggy software, when we know the fix and that's trivial to upload [14:12] expecially because I am actively pushing ubuntu on an institutional server where I will produce a package which uses python-uno [14:13] Le-Chuck_ITA: that doesn't look like something high importance on a normal installation [14:13] Le-Chuck_ITA: you are welcome to work on a SRU [14:14] ok, thanks, that can be my next move but I will have to re-learn lots of burocracy when a developer can just upload the fixed package [14:14] I don't see why I should waste my energy doing that, even though I did it in the past and can do it again [14:14] Le-Chuck_ITA: there is no such thing as "can just upload" to stable [14:15] Le-Chuck_ITA: The developers follow the same processes as anyone else in order to get a patch applied to a release. it's not easy on purpose. [14:15] Le-Chuck_ITA: any maintainer need to follow the SRU process to get things to stable updates [14:15] ok [14:15] I didn't know [14:16] in any case, the package may have little importance, but the fix is vital for the package itself [14:16] it's an ld.so.conf error [14:17] seb128: I am not even a MOTU, do you think I will get the same chances as an ubuntu developer to see my SRU go on? It's not polemic, I just have to decide if it's worth [14:18] Le-Chuck_ITA: attach a debdiff to the bug and let somebody do the official SRU === cjwatson_ is now known as cjwatson [14:18] Le-Chuck_ITA: the sru's get blocked because people dont find testers. [14:18] SRUs for openoffice.org are a rather big deal [14:18] the rest of it's not so hard [14:19] we prefer to stack up fixes there, since the download is so big [14:19] Come on, it's an added path to ld.so.conf [14:19] ah ok [14:19] I misunderstood you [14:19] I dropped off the network for a moment, so perhaps I missed a bit [14:19] cjwatson: image change to non-branded form, anyone? [14:19] Hobbsee: ... which won't go into -updates until there's something to stack it with [14:20] ok so i will try to provide a debdiff, I repeat point is that I want to push ubuntu on a server, but I can't do that and then require a patch to the default install, because they would say it doesn't look serious [14:20] cjwatson: oh, it's not even in -updates? right. [14:20] cjwatson: i upgraded from gutsy a long time ago :) [14:20] stable releases are boring, and all that. [14:21] Le-Chuck_ITA: for large packages, it definitely requires signoff from the main responsible developer [14:21] in your case, it's not obvious to me why you can't just set LD_LIBRARY_PATH locally as a workaround [14:21] just a moment, cjwatson: the fix should be done in python-uno [14:21] that doesn't require a patch to the default install [14:22] $ apt-cache showsrc python-uno | grep Package: [14:22] Package: openoffice.org [14:22] ah ok I surrender on this :) However, I don't want to write "if DISTRO==ubuntu then LD_LIBRARY_PATH+=" ecc ecc in my code [14:22] the unit of changes to Ubuntu is source packages; you can't change just python-uno without uploading the whole of openoffice.org [14:23] ok [14:23] I can eventually provide an ubuntu package with the missing ld.so.conf.d [14:23] it seems to me that it would be preferable to set the path to the OOo libraries in the uno module rather than adding an ld.so.conf.d file [14:24] (the latter would affect *all* binaries, and probably slow down all library lookups) [14:24] (just by a little bit of course, but still) [14:24] however, calc should make that decision === cprov is now known as cprov-lunch [14:25] Le-Chuck_ITA: Please submit a change for openoffice.org. It may take a bit to get in, but in the worst case, it will be in for hardy, and be useful to all clients of python-uno [14:26] (the diff is probably fairly trivial, so I doubt submitting one will make a huge difference) [14:26] persia: now I will have to wait on the decision about using ld.so.conf.d or setting the path in python scripts [14:26] and I agreed with cjwatson in principle [14:26] if the diff is trivial the problem of a developer won't be having a debdiff [14:26] but finding time to take a look at the bug [14:27] In any case I bothered you all a little more than I wanted to :) I just hope to see that fix in gutsy [14:27] not that I want to discourage people from contributing of course; just to point out that the world doesn't revolve around diffs :-) [14:29] cjwatson: True. My experience has just been that diffs often demonstrate willingness to help fix or test things, but I'm not the person making the oo.o decision :) [14:31] keescook, ping [14:37] bye all [14:48] doko: fyi: http://paste.ubuntu.com/2132/ if you are happy with it, I will upload [14:49] mvo: QUERRY? double R ? [14:51] doko: thanks! fixing this now [14:51] * mvo blushes [14:54] seb128: you doing source new today? [14:54] or syncs? [14:54] some [14:54] I'll not likely do everything [14:54] anything you would like to get? [14:55] I have some pending syncs, let me dig out the names [14:55] mvo: if they are filled I'll do them [14:55] mvo: I do syncs almost daily ;-) [14:55] seb128: sync the world plz. [14:55] source NEW is an another issue [14:55] Hobbsee: already done this morning [14:55] seb128: they are filed [14:55] I do that almost daily too ;-) [14:55] seb128: there are probably more people there now, though :) [14:55] mvo: ok, will do them then, no need to bother [14:55] seb128: sync the universe too, plz. [14:56] thanks seb128 [14:56] seb128: They're not particularly urgent. If I just know that you probably won't get around to it today, I won't be twiddling my thumbs waiting for it. :) [14:57] seb128: It's virtinst and virt-viewer, if you feel like doing them anyway :) [14:59] soren: will have a look later [15:04] Hello [15:04] hiya pitti! [15:08] Hi pitti! [15:08] hi warp10 === gouki_ is now known as gouki [15:22] pitti: quick question about that gcompris .pot, I'm generating it, but it doesn't get installed to any .deb, is that necessary? [15:22] in appearance->visual effects its none...if i click on normal it says composite extension is not available...what it mean?in ubuntu 7.10 [15:24] LaserJock: it's not necessary to be in the .deb [15:24] LaserJock: as long as it's in the built source tree, that's fine [15:25] dholbach: ah, ok [15:40] LaserJock: what dholbach said (sorry, missed your ping) [15:44] mvo: if you haven't already, file bug reports in product gutsy-backports, and we shall take a look. I think Compiz is a great Backports candidate :) [15:54] pitti: ping about language pack update for Gutsy, did you have time to start the process? [15:55] carlos: I'd like to do that with Arne [15:55] no time yesterday, but I should be able to start that today [15:55] ok, thank you === cjwatson_ is now known as cjwatson [15:59] jdong: I haven't filed a bug yet, my plan was to just bribe some of the archive-admins to get a backport build going. but I'm fine with filing a bug too (I want to get it done somewhat quick so that we can start pushing compiz into hardy that will be no longer backportable) === cprov-lunch is now known as cprov [15:59] mvo: the archive admins typically like a backport request ticket they can close when pushing a backport [16:00] I hav ubuntu 7.10 whith desktop effects working on xgl...how to install compiz fusion? graphics card ati radeon x200 [16:00] mvo: but if you can find an admin to shortcut the process I'm all up for it :) [16:00] ganeshhegde: -> #ubuntu; this is not a support channel [16:00] jdong: thanks :) I see what I can do [16:01] mvo: I want a written statement from you that this won't break gutsy, in triplicate, signed with your blood, and a beer as bribe :) [16:01] mvo: but I also prefer a bug just to have some papertrail [16:01] :D [16:01] * pitti hugs mvo [16:02] heh :) [16:02] pitti: I think bug 57875 is ready to copy from backports to updates. So far I've received no signs that the backports is any regression from the countless positive votes on the prior testing package [16:02] Launchpad bug 57875 in azureus "Azureus hangs or crashes showing splash screen at start" [High,Fix released] https://launchpad.net/bugs/57875 [16:05] jdong: I agree [16:06] pitti: the one regression report towards the end of the bug seems to be more of a problem with that one user's system setup; it's not happened for anyone else [16:07] carlos: gutsy-updates packages are building [16:07] pitti: ok, cool [16:07] I'll wait until the PPA has binaries, then we copy them to -proposed, and send out a call for testing [16:08] pitti: ok, thank you [16:09] jdong: *bzzt*, it's in -updates, bug updated [16:10] pitti: sweet. Thanks so much for you help through this 2-year bug squash! :D [16:10] you're welcome, thanks to you for enduring it [16:10] anything for our users, right? :) [16:10] * \sh needs a faster computer for mono :( [16:12] take two that can get you stereo then :) [16:13] <\sh> ogra, good idea...working on two computers at the same time...no..bad idea, I need two more arms for that ,-) [16:29] * Keybuk prepares to reboot into hardy [16:31] cool, nobody screamed "WAIT!" :p [16:32] Keybuk: way should we do that, no reason to not share the breakage fun with you ;-) [16:34] to paraphrase SteveA ... "because I have changelogs.ubuntu.com, and I can fire people" [16:35] (s/way/why, I should learn to type correctly or read what I type rather ;-) [16:43] slangasek: are you planning a new samba upload in debian in the next few days ? [16:44] mathiaz: 3.0.27a, sure [16:45] slangasek: ok. So there is no point starting working on a merge [16:45] * dholbach hugs soren :) [16:46] * soren hugs dholbach back :) [16:48] <\sh> keescook, jdstrand : bug #162826 ready for review [16:48] Launchpad bug 162826 in mono "[Mono] Buffer overflow in Mono 1.2.5.1 and earlier" [Undecided,In progress] https://launchpad.net/bugs/162826 [16:49] bryce: I subscribed you to bug #163850; it renders the touchscreen on Q1 reference hardware unusable and has no trivial solution; if you know how to write an evdev config from the /proc bits, perhaps we can work on this, or we could check whether input axis scaling could be readded in xorg-server; some hints on the path to follow would be appreciated :) [16:49] Launchpad bug 163850 in xf86-input-evtouch "Uncalibrated and can't calibrate with 0.8.7-2 on Samsung Q1 Ultra" [Undecided,New] https://launchpad.net/bugs/163850 [16:51] lool: I subscribed ubuntu-x to evtouch bugmail, good catch :) [16:53] tjaalton: TY [16:54] lool: does evdev support those devices? [16:55] <\sh> so..time to go home... === \sh is now known as \sh_away [16:55] slangasek: BTW where is the samba 3.2 package in Debian ? in experimental ? [16:56] yes [16:56] tjaalton: I don't know, but the evdev syntax seemed to be generic enough to allow this [16:56] slangasek: because merges.ubuntu.com lists 3.2 as debian version. [16:56] tjaalton: Can't tell for sure [16:57] Keybuk: so Hobbsee was saying that MoM looks at Debian experimental selectively, is that right? [16:57] lool: ok, that would be cool [16:57] slangasek: err.... [16:58] slangasek: how about you tell me what you want, and I'll tell you whether it can or can't do that :) [16:58] Keybuk: it's in general not obvious to me why we would want mom to list experimental-only merges as outstanding, and e.g. for samba we wouldn't want to merge from there [16:59] it doesn't list experimental-merges as outstanding [16:59] oh [16:59] it just gets the wrong version for "Debian version"? [16:59] no [16:59] not that I'm aware of [16:59] 3.0.26a-1ubuntu2 3.2.0~pre1-1 3.0.26a-1 [17:00] ^^ samba from http://merges.ubuntu.com/main.html, the middle number is the version in experimental [17:00] I had another example which I can dig up if it's helpful, but I've lost track of it in the list at the mo [17:01] that's odd [17:01] debian: 3.2.0~pre1-1 [17:01] samba_3.2.0~pre1-1.dsc [17:01] samba_3.2.0~pre1.orig.tar.gz [17:01] samba_3.2.0~pre1-1.diff.gz [17:01] it generated the merge from experimental [17:01] oh, cock [17:03] lool: ok I'll take a look in a bit [17:04] MoM has gone senile. [17:04] The men in white coats are coming to pick her up and take her to therapy. [17:04] bryce: Thanks; I can hand you the /proc values as reported by the device if that's helpful [17:04] bryce: (Or do you have a Q1?) [17:04] hello bryce [17:04] no I don't have one [17:04] hi airjump [17:05] orga from the irc chan. edubuntu give me your nick name [17:05] you have a eee pc from asus? [17:05] lool, the /proc values may be useful; I haven't written evdev rules before so am not certain what'll be needed exactly [17:06] airjump: I *cough* still need to pack it up and send it to bryce ... [17:06] bryce: My problem is that I couldn't find much information on the format of the /proc entry, but the xorg.conf part seems more clear and documented [17:06] my bad, I've had too much to do [17:06] ok [17:06] right [17:06] lobotomy applied to MoM [17:07] poor MoM [17:07] slangasek: check again in a few days when it's rebuilt [17:07] Keybuk: ok :) [17:09] hmm. bzr upgrade --dirstate-tags sftp://blah not the fastest thing in the world [17:09] cjwatson: s/sftp/bzr+ssh/ [17:09] last time I tried that with bzr upgrade it didn't work [17:10] I'll try it again at some point ... === asac_ is now known as asac === emgent_ is now known as emgent_work [17:12] bzr log --line | grep 'releasing version .*ubuntu' | sed 's/^\([0-9]*\):.*version \(.*\)$/bzr tag -r\1 \2/' | sh [17:12] mm [17:12] dirty hacks 'r' us === emgent_work is now known as emgent_ === emgent_ is now known as emgentwork [17:15] Keybuk: bzr upgrade bzr+ssh:// just claims it's at the most recent format. bzr upgrade sftp:// actually does the upgrade [17:16] mathiaz: do you really think bug #163194 warrants another debconf question? [17:16] Launchpad bug 163194 in samba "Disable creation of weak lanman hashes by default in samba" [Undecided,Triaged] https://launchpad.net/bugs/163194 [17:16] mathiaz: I wanted to take the easy route, and just patch the default behavior in the binary :) [17:17] Keybuk: bug 125166 [17:17] Launchpad bug 125166 in bzr "Smart server doesn't suppport upgrading" [Low,Confirmed] https://launchpad.net/bugs/125166 [17:17] ahh [17:17] you're using LP [17:17] ok [17:18] slangasek: I'm not really sure. I don't know if there is a lot Win95/98/Me clients out there. [17:18] slangasek: I would put the debconf priority to low. [17:20] mathiaz: if it's only going to be low priority, I would argue against doing the work on it before there's clear demand for the feature, since it makes work for translators, bloats the package with translated templates, etc. [17:20] slangasek: yes. Makes sense. [17:21] slangasek: on the same subject, IIRC to keep the passwords in sync, we need to install pam-smbpass [17:21] yes [17:21] slangasek: would this be part of a default desktop install ? [17:21] but I'm still working on the pam framework for that :) [17:21] I think it should be [17:22] slangasek: I wouldn't put on the defautl server install tought. [17:23] slangasek: what about the case of a server that installs samba ? [17:23] as part of the "filesharing" task, or installing the package direct? [17:24] slangasek: for the fileshareing task, I think pam-smbpass should be installed. [17:24] agreed then [17:25] well, unless we want the filesharing task to depend on krb5+ldap instead :) [17:26] slangasek: but I don't think that the samba package should depend on pam-smbpass. [17:26] slangasek: we'll wait a little bit for the krb5+ldap integration... [17:26] ack === dendrobates is now known as dendro-away === emgentwork is now known as egentili [17:40] Can someone give-back https://edge.launchpad.net/ubuntu/hardy/+source/faad2/2.6-1 on the archs it FTBFS'ed on? [17:40] the xmms-dev culprit has been resolved === BenC_ is now known as BenC === cprov is now known as cprov-away [18:43] Could someone retry anjuta build? It failed in some archs a week ago because of missing dependencies, but it wasn't set to DEPWAIT: https://edge.launchpad.net/ubuntu/+source/anjuta/2:2.2.2-1ubuntu1 [18:59] pitti: am I likely to need this patch soon? http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=450603 [18:59] Debian bug 450603 in kdm "kdm: Please add support for ConsoleKit" [Normal,Fixed] [19:02] Riddell: right, I think you do [19:02] Riddell: ah, that means current Hardy breaks power management and device mounting for Kubuntu? [19:06] pitti: not sure, I havn't tried it in a full session [19:20] pitti: have you seen pochu's give-back request for anjuta? [19:20] geser: hm, seems I missed it [19:20] ah, right, there [19:21] pochu: done [19:21] pitti: ty === zen-afk is now known as zenrox === pedro__ is now known as pedro_ === Pici` is now known as Pici [19:49] mjg59: hi I am merging our belowed laptop-mode-tools. Should we ship apm scripts or should we take them out as we do for acpi scripts? [19:50] Can you skip merging it until we know what we're doing with it? [19:51] I can also just merge it, until we know better. Isn't much work. [19:51] http://merges.ubuntu.com/main.html --- erm... [19:51] did we just sync the entire hardy to sid? :) [19:51] haha [19:52] pitti: we rock apparently ;-) [19:52] we said we would push patches to Debian, and WE REALLY MEAN IT [19:52] * ajmitch should still have a couple of outstanding merges there [19:52] pitti: "but.. they kept asking" [19:53] I asked Keybuk about the experimental confusion, and he did something, and now everything's gone plaid [19:54] 09:06 < Keybuk> lobotomy applied to MoM [19:54] 09:07 < Keybuk> slangasek: check again in a few days when it's rebuilt [19:54] just another reason to use DaD http://dad.dunnewind.net/main.php [19:54] I didn't realize he meant we'd be without merge info for two days :/ [19:55] tormod: oh, DaD speaks main now? cool [19:55] tormod: support efforts duplications? not really required [19:55] tormod: no sort by age? :) [19:56] but DaD doesn't seem to support initial vs. updated merges [19:56] pitti: it lists main (since a long time) [19:56] seb128: it has a "comment" field [19:57] slangasek: it's open-source so you can fix it ;) [19:57] tormod: I usually don't need to comment on merges I'm doing [19:57] and if that's useful that should be added to merges.ubuntu.com rather than duplicating [19:58] tormod: sure, if I want to run my own copy of a service, pfff [19:58] yeah, duplicating core infrastructure like that seems entirely sub-optimal [20:00] is MoM open-source now? [20:00] seb128, elmo: I think Adri2000 and Lutin volunteered to do that in MoM, but they can't ;) [20:00] tormod: that's your justification? [20:00] pochu: did they talk to Scott about it? [20:01] pochu: are you going to use an another bug tracker than launchpad next because you can't fix bugs? and then something else than soyuz? [20:01] seb128: yeah, it's called 'debian' :) [20:02] elmo, that's the most important element [20:02] seb128: nope, but I'm not justificating them, only letting you know :-) [20:02] ajmitch: use Debian if you prefer, that's your choice [20:02] ajmitch: that's smarter than creating yet another service [20:02] seb128: pfft, you know that I'm stuck with ubuntu :) [20:02] ;-) [20:03] * ajmitch is using a sid desktop right now though, at work [20:03] I doubt many people will work on dak though ;-) [20:03] I'd prefer not to [20:04] I'm wondering if they will also work on they own google ;-) [20:04] though I like having comments on DaD, they're not much use if people are using 2 separate pages, that won't even give the same merge results [20:05] I'm wondering if they try to sort the issues with Scott [20:05] they have talked with him [20:05] or if they just went "that's closed source let's do our own version" [20:05] as someone said "We’re committed to reinventing everything we need until the free software stack is a genuinely complete computing universe" [20:05] I can't remember what the outcome was, I think scott suggested just using the MoM output as-is & reskinning it somehow [20:06] tormod: you are free to reinvent the wheel, that doesn't mean it's really required or useful, especially if the tool you are using is likely to be opensourced one day [20:06] tormod: obviously, it's your time, but there's ways to be clever about it. and reinventing services when there's really no need, doesn't seem like an efficent use of time, to me, but whatever [20:06] ajmitch: well, adding a comment feature on top on merges.ubutnu.com would likely be doable [20:06] yep === dendro-away is now known as dendrobates [20:07] * ajmitch already has some scraping & commenting stuff [20:07] not sure why comments are useful for merges [20:07] usually you claim a merge and just do it [20:08] and if you need sponsoring you use launchpad [20:08] mainly useful for universe, where merges can take a little bit of time, and people like to claim them on the page [20:08] right, I do that. [20:08] there'd been some duplication of work in the past, and it does still happen a bit [20:09] for the desktop team we use the wiki to claim those [20:09] I'm sure MOTU could come with an easy system to claim merges without reimplementing everything [20:10] I must admit I didn't invent DaD, but I think it would be nice if everybody could use the same tool, and an open-source one by preference. [20:10] the wiki used to be used, and the pages ended up rather large [20:11] tormod: please don't be so dramatic, everyone can use MoM [20:11] whether they wish to or not, is another matter [20:11] hmm, another MoM/DaD debate [20:12] btw, I'm still waiting for sabdfl to answer a mail about that issue I sent him back in June [20:13] it probably got lost in a pile of email then [20:14] ajmitch: probably, but I reminded it him many times, and once again yesterday or two days ago, via mail and irc [20:15] we prefer open source everywhere else, I fail to see why we shouldn't prefer it in our infrastructure as well [20:15] tormod: "a genuinely complete computing universe" -- you're familiar with Gödel? :) [20:15] hi -- I don't know what phase ubuntu's development is in currently, but could someone please make sure that my (source) package nbd is updated in ubuntu? [20:15] AAIU, the current version is held back from Debian since it's got some ubuntu-specific patches, but I incorporated those into my Debian package [20:15] slangasek: I was quoting sabdfl :) [20:15] also, is there a way for me to make this more obvious to whatever Ubuntu tool looks at differences between packages? [20:16] Burgundavia: when there is an opensource alternative sure, reinventing the wheel when you have working tools is an another question [20:16] seb128: by that argument, why do we have git when we could all be using perforce or bk? [20:16] tormod: ah, heh [20:16] Burgundavia: I don't care about git and don't use it ;-) [20:16] Burgundavia: because upstream uses git? :) [20:16] Yoe: it's in the merge queue, see dad.dunnewind.net/universe.php, you can add a comment saying it likely needs a sync [20:16] Burgundavia: you're being disingenuous [20:17] pitti: ping (about language packs build...) [20:17] elmo: no. Infrastructure is no different to anything else [20:17] carlos: yeah? [20:17] Yoe: oh actually it's in main, so main.php [20:17] pitti: how's that going? [20:17] Burgundavia: do you also refuse using google? [20:17] carlos: buildds are grinding for a while [20:17] Burgundavia: I'm sorry, let me introduce you to this product called Launchpad. maybe you've heard of it? [20:17] carlos: sources are all up for some hours already [20:17] Adri2000: yes, I was just going to say that :) [20:17] carlos: (all PPA) [20:18] pitti: still building packages? [20:18] elmo: I am well aware of LP. I don't like that it isn't open source either [20:18] Burgundavia: please don't pretend it doesn't exist then [20:18] how long does it take to be built? [20:18] seb128: there is a difference between a tool we use to build a free distribution and a mere tool [20:18] carlos: yeah; it'll take until tomorrow morning, I suppose; we have three buildds now, but it's quite a large number of updates [20:18] pitti: I see [20:18] pitti: ok, thanks [20:19] elmo: I am not. I am merely saying that in this specific instance, there is no reason for the project to keep using a closed source tool when an open source one exists [20:19] Burgundavia: sigh [20:20] Burgundavia: I hate it when you get like this [20:20] Burgundavia: free archive management tools exist (dak) [20:20] Yoe: you're welcome to use the procedure on https://wiki.ubuntu.com/SyncRequestProcess btw, or if you prefer I can go through it for you [20:20] Burgundavia: free bug tracking tools exist (bugzilla) [20:20] Burgundavia: ubuntu right now does not use them [20:20] Burgundavia: please get in sync with that reality, and restate your argument [20:21] elmo: I know we don't use all those things. I didn't say I agreed with the decision [20:21] slangasek: oh, I thought that procedure was "ask on this channel" ;-) [20:21] Burgundavia: do you think that reimplementing those for the sake for reimplementing will benefit anybody? [20:22] the difference is that DaD was designed and developed for Ubuntu, just like MoM (the difference being freeness), while dak and bugzilla are not [20:22] seb128: by that argument, LP is a giant reinvention of the wheel that, IMO, add no value and takes away one key one, freedom [20:22] Yoe: either way... :) Just pointing you it out if you'd prefer to have your hands on it directly [20:22] Burgundavia: don't you think we should better spend the efforts on the distribution rather than reimplementing the tools? [20:22] hmm, pointing you it out [20:22] * slangasek rewrites him this sentence [20:22] Burgundavia: point me to something that does what LP is doing? [20:23] seb128: sure it'd would take time and efforts to now switch to bugzilla or dak, but that's not true for switching from MoM to DaD [20:23] seb128: I think LP is a giant rathole, tbh [20:23] seb128: it's for the sake of having open-source tools, not merely reimplementing [20:23] tormod: we have better things to do that recoding the tools [20:24] slangasek: well, if you want to do it, not stopping you. [20:24] we have tools there working and available, let's use them and focus on users issues [20:24] or jcastro (who's /query'ing me) could. Whatever. [20:24] as long as it happens :) [20:24] jcastro: put your hands where we can see them! [20:24] seb128: recoding tools because they are non-free is EXACTLY what we have been doing for 2 decades [20:25] seb128: why not open-source MoM, any compelling reasons those claimed for LP? [20:25] Burgundavia: you certainly haven't been doing it for 2 decades [20:25] * like those [20:25] tormod: dunno, you need to ask Scott [20:25] seb128: Scott told me to ask sabdfl, that's why I sent him a mail back in June [20:25] and still got no answer [20:25] elmo: no, I have not. It does not change the facts of my argument [20:25] Adri2000: dude, send the mail [20:26] elmo: ? [20:26] Burgundavia: no, it's just fun to knock you off the pedestal you like to put you on [20:26] Adri2000: + again [20:26] Adri2000: strangely enough, Mark gets a lot of email ,it's not hard to believe it got missed, if you only sent it once [20:26] elmo: heh. If nobody will point out the elephant in the room, who will? [20:26] jcastro: so are you doing the sync request for nbd? [20:27] slangasek: I was going to ping sbalneav about it [20:27] ok [20:27] elmo: not sure that will help. I already sent him twice, and pinged/queried sabdfl many times. sometimes I got answer "I will look at it" sometimes nothing [20:27] Burgundavia: nobody really focused on rewriting online webservices until now [20:27] s/him/it/ [20:27] jcastro: alternatively, I could prepare the sync request and subscribe sbalneav for sign-off? [20:27] seb128: and that is a problem we have yet to solve [20:27] Burgundavia: is there any effort to do a free google? [20:27] slangasek: I didn't want to interrupt the conversation with things actually relevant to this channel [20:27] slangasek: that would be great [20:28] seb128: I fail to see what the google argument has to do with LP? [20:28] Burgundavia: that's the same issue, why do you refuse to use LP but are happy to use google? [20:28] corey doesn't refuse to use it, he just likes to whine about how it's not yet OS [20:28] (if he even remembers to add the yet) [20:29] seb128: I use both, because they work. I dislike using both [20:29] seb128: there is an initiative to make a free "google". You think it's stupid to have a transparent search engine outside corporate control? [20:29] I mean at least the DaD people did something, I can respect that, even if I think it's a waste of their time [20:29] tormod: no, but I think we have better thing to do right now that rewritting websites and tools [20:29] elmo: took us only two or three days, + all the bug fixing of course :) [20:29] elmo: are you saying I am not worth listening to because I cannot code? [20:30] Burgundavia: nope [20:30] Adri2000: then i sort of doubt it's feature complete compared to MoM [20:30] (or you guys are just geniuses and keybuk's a slack jawed hippy, but I guess either of those is possible) [20:30] tell me what's missing then please [20:31] seb128: for many people, the Operating System is just a tool [20:31] tormod: we don't argue with those guys, there is just no point [20:31] I know that comments, automatic maintainer update, etc. are missing in MoM, but... [20:31] automatic maintainer update? [20:32] seb128: yes, for the DebianMaintainerField spec, though it's probably not very useful today [20:32] as said before, comments ... who cares [20:32] usually you claim a merge and do it [20:32] that's a bit more complicated in universe usually [20:32] use launchpad like we do for everything else? [20:33] it's easy enough to open a "claim the merge" bug [20:33] open a bug to comment on a merge? [20:33] what sort of comment do you need? [20:33] sync requested, please don't touch, wait for new debian version, X working on it, etc... [20:34] just take a look at http://dad.dunnewind.net/universe.php [20:34] seb128: with comments it's easier to see if someone works on a merge, a sync got filed, any problems preventing the merge, etc. [20:34] you could get a comment system around merges.ubuntu.com without reimplementing everything [20:35] that's not a valid reason to reinvent the wheel [20:35] I'd be very happy to help doing that, but I'm still looking for the code [20:35] no need to get the code, the datas are public [20:35] slangasek: I'll send sbalneav a heads up email now [20:35] it's easy to get a list of merges [20:36] ah yes, but that would mean reinventing all the UI wheel [20:36] Adri2000: afaik the frontend only "parses" the directory structure on merges.u.c [20:36] Adri2000: better than reinventing the logic and UI wheels [20:37] geser: it should parse that http://merges.ubuntu.com/tomerge-universe [20:37] but now this file is empty... [20:37] it used to look like dad.d.n/tomerge-universe [20:38] Adri2000: it has been resetted today, it's not like it happens every day [20:40] ok === Adri2000_ is now known as Adri2000 [21:10] hi, I'm looking for someone who could give some information on how the gfxboot files onn the ubuntu cd were created [21:12] or any pointers to such info would help too... [21:14] mvo: hi [21:14] hi calc [21:14] mvo: i have an apt patch for you once i verify it actually works === mpt_ is now known as mpt [21:14] calc: sounds good [21:14] calc: lzma? [21:14] mvo: it adds some missing lzma support for apt-ftparchive, etc [21:14] mvo: yea [21:15] calc: just mail it to me once you are happy with it. I will go to bed soon [21:15] ok have a great night :) [21:49] hmm, hardy is a mixed bag for me [21:52] bryce: you remember I said my brightness sensor didn't work? [21:53] Keybuk: I remember you saying that at least. :-) [21:54] well, it works perfectly in hardy! \o/ [21:54] Xteven: it's in the gfxboot-theme-ubuntu source package [21:54] and the brightness keys on the keyboard even pop up a little ☀ OSD [21:55] the fact that the OSD progress bar always remains at zero is ... well, minor [21:55] Hahaha [21:55] cjwatson: those are the files, but how were they created ? or how do I use them ? [21:55] gfxboot has no man-pages [21:55] Xteven: like any other source package, you build it ... [21:55] in the standard way [21:55] it has a very shallow "reference manual" that doesn't give much information at all [21:56] the reference manual is INVALUABLE [21:56] I couldn't maintain gfxboot-theme-ubuntu without it! [21:56] that xml file ? [21:56] once built properly, yes [21:56] there's a Makefile in there that builds it into HTML [21:56] it is a language manual [21:57] it may not be the information you need, but it's entirely wrong to say it doesn't give much information [21:57] interestingly, I don't have hal-addon-dell-backlight running [21:57] anyway, 'debuild -b' in an unpacked copy of gfxboot-theme-ubuntu with the build-dependencies installed on your system should build the files, just like any other source package in Ubuntu [21:58] hmm [21:58] there's a very small amount of code in our deployed debian-cd that extracts /usr/share/gfxboot-theme-ubuntu/bootlogo.tar.gz into the /isolinux/ directory on the CD [21:58] is any of this information documented somewhere ? because I couldn't finnd it [21:59] I don't believe it's documented, no [21:59] although, in general, how to build a package from source should certainly be documented [21:59] (I don't know offhand where) [21:59] I didn't know it was a source package actually :/ [22:00] could you tell me how much work it is to create a new theme inn gfxboot ? [22:00] the reference doc hints towards a postscript-like language [22:01] and the .inc files in gfxboot-theme-ubuntu look complicated enough that I don't want to build anything from scratch :/ [22:04] thx for the info [22:06] Xteven: unfortunately, lots [22:06] I based the Ubuntu theme on the SuSE theme [22:07] it was several weeks of fairly solid work [22:07] it's not an easy language, I'm afraid [22:07] * mpt wonders if it would make sense for a shell to have a "Chime when anything non-interactive that's taken longer than 2 minutes is finished" option [22:07] I imagine smaller modifications would be less work, of course [22:08] cjwatson: I guess I could do with small modificaions :) [22:08] cjwatson: I'd like to strip out all the non-english languages, since we won't be using them [22:08] and then I'd like to change the menu entries and the image [22:08] thats pretty much it [22:08] I know how to put a new image and how to change the menu entries [22:08] mpt: something like that should probably be generalized, ala the geyser project [22:09] (dinnertime, back later) [22:09] but the translated files are not easy to understand [22:09] ok, enjoy your meal [22:09] stripping out the languages is easy, just remove the translations and take them out of langlist or whatever it is [22:09] mpt: basically, the terminal would say "something is done, notify the user" and the system would do what it was told to do, which on a server might be SMS or email the admin [22:10] Yeah, doing it just for the shell would be too easy to implement :-P [22:11] yep, you need to build a bigger bikeshed and *then* you paint it [22:11] cjwatson: ok thx, I'll try it [22:11] Coincidentally I commented on a few days ago [22:12] mpt: he should have called it MessageKit [22:12] We're going to need a KitKit to organize all these Kits [22:13] oh indeed [22:13] and HAL+udev is soon to be DeviceKit === `23meg is now known as mgunes [22:18] Hello. I'm following the osdev Bare Bones tutorial for writing a simple kernel. http://www.osdev.org/wiki/Bare_bones I'm a complete noob at this, so please don't assume I know all the terminology you guys do. I'd like to be able to boot the kernel shown in the tutorial in either Bochs or QEMU. How would I go about this? [22:19] theunixgeek: This isn't really the right place to ask [22:19] This channel is for development of Ubuntu, not development using Ubuntu [22:19] mjg59: sorry :( [22:19] No problem [22:19] #ubuntu is a better place to ask [22:20] mjg59: even for development questions? [22:20] Yes [22:20] ok [22:20] thanks [22:20] If they're not about developing Ubuntu [22:20] heh [22:21] mjg59: guess who just popped up on #edubuntu? [22:22] ? [22:22] our little friend, theunixgeek [22:22] If he's asking the same question, explain that it's inappropriate [22:23] And that it's inappropriate in every Ubuntu channel except #ubuntu [22:27] oh, and other mixed-bag on hardy ... performance is really dire [22:28] is hard to tell whether it's compiz being slow as shit ... or the computer [22:28] since my primary interaction is through compiz [22:28] Keybuk: tracker seems to fire up right on login, with predictable results [22:29] nope, isn't tracker [22:29] that's the FIRST thing I blamed :p [22:29] heh [22:29] deskbar-applet is using unreasonable amounts of CPU for something that's not supposed to be doing anything [22:30] we should probably add a deferred startup bit to tracker [22:30] Keybuk: what driver? [22:30] tjaalton: intel [22:31] Keybuk: ok, try using XAA [22:31] tjaalton: -v [22:31] Burgundavia: It's supposed to have a 45 sec startup delay already. [22:31] johanbr: 45 secs is not enough [22:31] on a slow harddrive like mine, 45 secs is still in the panel load time [22:31] Well, it's adjustable. [22:32] tjaalton: googling for "enable xaa" is surprisingly unhelpful [22:32] Burgundavia: And it takes you more than 45 seconds from gdm to gnome?! [22:32] Burgundavia: Are you turning the platter by hand, or what? [22:33] StevenK: tiny japanese hamsters, this is a toshiba [22:33] Keybuk: ok :) 'Option "AccelMethod" "XAA"' [22:33] Ugh, Toshiba [22:33] tjaalton: which section? [22:33] Device [22:33] johanbr: it takes me longer to get from gdm to gnome than it does from bios to gdm [22:33] it's EXA by default now [22:33] if only X supported setting driver params on the fly [22:33] but hey, that'd involve it being dragged kicking and screaming into the 20th century [22:34] Keybuk: It needs to be dragged kicking and screaming into the 19th first :-P [22:34] Burgundavia: That doesn't sound right. [22:35] johanbr: login has always been slow for me. Is there a way to instrument it to figure out where the time is being wasted? [22:36] Burgundavia: I think bootchart can do that. [22:36] johanbr: I didn't think bootchart worked after gdm [22:36] tjaalton: highly subjective not-a-metric, but this is a hell of a lot faster [22:37] Burgundavia: sure it can [22:37] tjaalton: oh, it's a hell of a lot faster AND tracker is running [22:37] Keybuk: heh, I couldn't really tell the difference with i945 [22:37] this is a 945 [22:38] ok.. interesting [22:38] without XAA (with EXA?) simple animations were visibly jerky [22:38] and larger ones like workspace switching only rendered two or three frames [22:38] and the whole desktop felt sick and slow [22:38] with XAA, it's back to normal -- windows slide around, etc. [22:39] Login is insanely slow on anything with high seek latency [22:39] actually, workspace swithes were jerky sometimes [22:39] From a warm cache, it takes a couple of seconds [22:39] tjaalton: did you try scrolling web pages, or PDF files? [22:39] Which tells you where the problem is [22:39] Keybuk: I know there are bugs in that area, but it wasn't too bad [22:39] mjg59: so basically I have a crap HDD and there isn't much I can do? [22:40] Reduce the number of seeks performed during login [22:40] Burgundavia: no, there's plenty you can do, but it all comes down to reducing the number of seeks and reads [22:40] right [22:40] it's not one specific process you can blame [22:40] You can strace gnome-session and all its children [22:40] mjg59: I straced X once, that was funny === jamesh__ is now known as jamesh [22:42] mjg59: If I recall, gnome-session seems to react badly when strace'd. If you kill strace(), gnome-session doesn't seem to wake up, even when you kill -CONT it [22:42] Michael Meeks' iotrace is probably more worthwhile [22:45] * StevenK kicks openssl for being inconsistent and stupid [22:45] how do i look at a prior revision of a wiki page on w.u.c ? [22:45] ah 'get info' heh [22:47] Quoting Dave Jones: "when it isn't busy scanning non-existent pci busses, X likes to reopen files it already read" === mc44_ is now known as mc44 [23:00] Keybuk: great to hear about the brightness sensor! [23:05] lionel: Has there been any movement on merging gtk-engines-murrine? [23:07] rexbron: I think your're looking for mr_pouit :) [23:08] oh, ok [23:08] sorry about that [23:08] rexbron: no pb :) [23:15] lionel: ^^' [23:16] hey mr_pouit [23:17] rexbron: iirc no, you can merge it of you want (the ubuntu package has a patch to fix a bug with a gimp theme afaik). [23:17] s/of/if/ [23:17] hey lionel :) [23:19] mr_pouit: Cool [23:39] Please confirm my nomination of the gparted top crasher for Gutsy (bug 141516). [23:39] Launchpad bug 141516 in gparted "[MASTER] Gparted crashes when refreshing devices" [Medium,Fix released] https://launchpad.net/bugs/141516