/srv/irclogs.ubuntu.com/2010/06/25/#ubuntu-devel.txt

=== jono_ is now known as jono
=== ubott2 is now known as ubottu
=== jjohansen is now known as jjohansen-afk
=== jtechidna is now known as JontheEchidna
psusiso.... with the debian import freeze going into effect today is there any chance of getting lvm2 updated this cycle?02:27
lifelessyes02:30
lifelessjust requires manual requests02:30
lifelessnot automatic02:30
RAOFIt's still open season on merges and sync requests.02:30
lifelessits not the end of syncs, its the end of the firehose02:30
ebroderlvm2 was never going to get automatically synced anyway, because Ubuntu modifies heavily from Debian02:32
psusiok... anyone planning on doing that anytime soon? :)02:34
psusiyea, I looked at the bzr repo for it and the last merge is a nightmare02:34
psusibut the new upstream has support for merging a snapshot back into the origin so you could make a snapshot, test an upgrade, then if it goes tits up, revert to the snapshot02:34
lifelesspsusi: that would be nice; care to do the merge ? ;)02:36
ebroderpsusi: The difficulty isn't really in dealing with LVM metadata restoration. The problem is that the way that Ubuntu packaged things is substantially different (because the startup processes for Debian and Ubuntu are fundamentally different)02:37
psusiI tried already... none of the ubuntu specific patches have proper documentation within the quilt patch file, and when I tried to track down their history in bzr, I got stuck on the nightmare merge and could not penetrate that history02:38
ebroderpsusi: Exactly. That's why nobody's done the merge :)02:38
psusiebroder, how so?02:38
psusiI ended up just disabling all of the ubuntu specific patches iirc... except for one or two that seemed to still apply to the new upstream without any fudging02:39
psusiwhat are the critical points of divergence ubuntu needs to maintain?02:40
* psusi pets make -j 2... if it takes this long to compile the kernel with it... boy...02:41
=== nenolod_ is now known as nenolod
=== Dr_Who is now known as tgall_foo
=== oubiwann is now known as oubiann
=== yofel_ is now known as yofel
siretartSarvatt: just rebuild it so it picks up libavutil5006:20
pssw0rtmaybe in this room06:37
pssw0rtone day, ubuntu would be xfce4??? i don see future with gnome 306:37
pssw0rtand the use of another lang called Vala06:38
pssw0rtand mono and C#06:38
Chipzzpssw0rt: no, not in this room ;P06:49
pssw0rtxD06:55
pssw0rtdevelopers  without opinion xD06:56
pssw0rtbb06:56
dholbachgood morning07:23
=== almaisan-away is now known as al-maisan
dholbachcan an archive admin please take care of the sync bugs? particularly 598274?07:54
=== al-maisan is now known as almaisan-away
=== oubiann is now known as oubiwann
=== hrw|gone is now known as HRW
=== HRW is now known as hrw
dholbachpitti: do you know what GSI means on https://wiki.ubuntu.com/UbuntuPackagingChanges?08:14
=== enrico_ is now known as enrico
dholbachccheney: is GSI still a change that should be listed on https://wiki.ubuntu.com/UbuntuPackagingChanges?08:20
Laneythat page seems somewhat out of date08:23
Laneybut I guess that's what you're cleaning up right now :P08:23
dholbachLaney: no, everybody's invited to update it08:24
dholbachLaney: don't rely on me - it was just a question that just came up :)08:24
Laneyalright08:25
dholbach:)08:25
LaneyI'll nuke the outdated ones (breaks, dh_iconcache)08:25
dholbachthanks08:25
pittiGood morning08:31
pittidholbach: GSI is the name of the OO.o translation format; don't ask me what it's spelt out08:32
nigelbpitti: the question rather is if its still significant08:32
pittiwell, it has never really been08:33
pittioo.o has had its own langpacks forever08:33
pittiwe have our own source package oo.o-l10n08:34
pittibut that's about it, the binary structure is as in Debian08:34
=== oubiwann is now known as oubiann
=== oubiann is now known as oubiwann-away
=== oubiwann-away is now known as oubiwann
pittiapw: good morning09:57
pittiapw: debian bug 509089 is such a gold nugget, thanks for digging this out!09:58
ubottuDebian bug 509089 in dhcp3-client "dhcp lease negotiation takes longer than necessary" [Normal,Open] http://bugs.debian.org/50908909:58
BlackZpitti: hmm, I can't find that file in cdbs09:58
pittifor my ethernet connection, it drops DHCP time from 2.5 s to 0.1509:58
BlackZI downloaded the currently version in maverick09:58
pittiBlackZ: ah, perhaps the Kubuntu team dropped it and it was just a copy&paste error in the merge changelog09:59
pittiseems so, yes09:59
BlackZpitti: so could it be dropped?09:59
pittifrom the merge changelog, right09:59
BlackZBTW it fails locally too, pitti10:00
BlackZI have no idea10:00
=== Amto_res`OFF is now known as Amto_res
RiddellBlackZ: what's this?10:02
BlackZRiddell: the dropped file? or the FTBFS?10:03
Riddellwhatever the bit the kubuntu team should care about is :)10:03
BlackZRiddell: - Add 1/class/kde4.mk.in: Generic KDE4 build rules. Install it in Makefile.am;10:05
BlackZI can't find it in the package10:05
Riddellyes it's gone10:06
Riddelluse /usr/share/pkg-kde-tools/makefiles/1/cdbs/kde.mk from pkg-kde-tools10:06
BlackZOK, dropping it then10:06
pittiBlackZ: as I said, run one of the tests locally and see what it's failing on10:08
BlackZpitti: .: 21: testsuite_functions: not found10:10
pittiyou have to cd tests10:11
BlackZI'm in it10:11
pitticd test; ./langpack-1.sh -> works fine for me10:11
pittido you actually have the testsuite_functions file?10:12
BlackZhm, yes10:12
BlackZand the package from debian works fine10:12
BlackZmaybe I've broken something but I don't think so10:12
pittiwell, it fails all tests, so _something_ was broken10:12
BlackZpitti: http://pastebin.ubuntu.com/454884/10:15
pittiBlackZ: so whatever "debhelper.mk.in:239: *** Riferimento incompleto a variabile.  Arresto.10:16
pitti" means, that's likely the error10:17
pitti"variable does not exist" or so?10:17
BlackZ"reference to variable incomplete"10:17
BlackZOK, looking there10:18
ograpitti, ... could you set the trend line again for http://people.canonical.com/~pitti/workitems/maverick/canonical-mobile.html (57 items) and http://people.canonical.com/~pitti/workitems/maverick/canonical-mobile-maverick-alpha-3.html (16 items) ?10:21
ograpitti, sorry, we got a new spec yesterday10:21
pittiogra: alpha-3 will auto-reset once alpha2 is done10:21
pittiso I won't touch that10:21
ograah. great then only the overall view10:21
pittithe idea is that you plan a milestone before it begins and the chart grows up10:22
pittiand you see how much stuff you have10:22
ograyeah10:22
pittiand once the milestone begins, the chart begins "clean" again10:22
pittiogra: note that the entire idea is to _not_ change the trend line when you add a spec10:22
pittithe point is that you see what/when the additional workload was added and thus what is to blame for dropping other work instead :)10:23
pittibut I can change it if you really want10:23
DktrKranzmvo: gdebi uploaded in Debian (with some little trivial changes), and sync request filed. you can mark branch as merged :)10:23
ograpitti, yes, but i was asked to since the initial value is all wrong for the total at least10:23
mvoDktrKranz: sweet, thanks!10:23
pittiogra: sure, was just a suggestion; committed10:23
ograpitti, thanks, i have another issue :)10:24
ograpitti, we need jockey to show an eula for a driver, is something for that implemented or do i need to do it on a dpkg/debconf level10:25
pittiogra: it's an open wishlist thing, bug 27128810:26
ubottuLaunchpad bug 271288 in jockey (Ubuntu) "Require the user to confirm the license before downloading a driver if it is non-free or if it has patent issues" [Wishlist,Triaged] https://launchpad.net/bugs/27128810:26
ograok10:26
pittiogra: jockey doesn't show debconf, sorry10:26
ograno, buut dpkg does if gtk-perl is installed, no ?10:26
ograi.e. if i put it in the package and make sure gtk-perl is there it should use the gtk frontend10:27
pittijockey doesn't use synaptic10:28
pittiit uses python-apt in the dbus backend10:28
pittiit doesn't have anything to display stuff to, and isn't meant to have10:28
ograhmm, and that omits debconf questions ?10:28
pittiif a package requires debconf, you can't use jockey, sorry10:28
* ogra thought that would work parallel to jockey just using the debconf gtk frontend10:29
mvopitti: have you considered switching to aptdaemon as the backend? then you get debconf support10:30
pittimvo: it's an option indeed, but I haven't found much time to hack on jockey since aptdaemon exists10:31
pittibut it's also kind of a design decision10:31
pittidrivers shouldn't need debconf or anything scary to work10:31
ograsadly some driver manufacturers disagree :)10:32
pittiI do acknowledge the need for displaying an EULA at some point, but that should be in jockey's UI, not debconf10:32
pittiapw: I uploaded dhcp3 with those patches to maverick now10:44
pittiwell, slightly adapted (without conffile changes)10:45
cjwatsonkirkland: merged your uec-live seed; as far as I'm concerned you can hack on it directly in the main seed branch at this point11:14
geserdoko: would it be possible to get our python-defaults package merged/synced with Debian ones? The first packages are already in DEPWAIT on python >= 2.6.5-2~11:26
cjwatsonmvo: is the rest of foundations-maverick-buy-something likely to land for alpha-2?11:27
cjwatsondoko: arm-m-tool-chain-selection has a work item for alpha-2 on it; is that still feasible?11:29
cjwatsoncking: you have "Investigate situation with Intel graphics drivers on EFI" as a work item for alpha-2, which is showing up on the foundations list; have you had a chance to look into that at all?11:29
ckingcjwatson, not yet, been a bit swamped. Will find some cycles today to start thus11:36
ckingcjwatson, want to you need to know?11:37
cjwatsoncking: at the moment, I'm just checking in so that I have information for the release meeting11:38
dokogeser: ScottK is working on this, waiting for the ia64 build11:38
geserok11:39
dokocjwatson: yes, I think the test rebuild is still doable, the decision probably later11:45
mvocjwatson: the buy-something work-items are not 100% reflecting reality anymore, there were some changes in the server side. I will update them today with noodles (who works on the server side of things)12:05
cjwatsonthanks12:07
=== ghostcube_ is now known as ghostcube
=== MacSlow is now known as MacSlow|lunch
hrwsomeone remember where I can find uds-m presentation about udev/hal?13:03
hrwkeybuk's one13:06
cjwatsondholbach: processing syncs now13:20
* dholbach hugs cjwatson13:20
nigelbhrw: ubuntu miro community should have them13:25
hrwnigelb: I already have video version13:26
hrwbbl - called for lunch13:26
ogracjwatson, for enabling preinstalled images i'm wondering if i should either mangle bin/cron.ports_daily-live to export CDIMAGE_PREINSTALLED instead of CDIMAGE_LIVE or should i rather create a new command i.e. bin/cron.ports_daily-prenistalled and leave ron.ports_daily-live alone ?13:38
ogra(without the typo indeed :P )13:38
dholbachcould somebody imagine giving a session at Ubuntu Developer Week about the release schedule, freezes and stuff we do? https://wiki.ubuntu.com/UbuntuDeveloperWeek/Prep still has a few open slots13:39
cjwatsonogra: I don't mind either way really; what's clearest in the code?13:39
ogracjwatson, well, i think cron.ports_daily-preinstalled would be less intrusive13:40
cjwatsonthat's fine by me, then13:40
ograeffectively its only that one variable that needs to change13:40
ccheneydholbach, yea support is still there its not working correctly atm13:47
dholbachccheney: does the page need an update?13:47
ccheneydholbach, i think its accurate enough about the gsi item13:48
dholbachok13:49
kirklandcjwatson: ah, great, thanks14:10
kirklandcjwatson: and as for building a server image directly for the hardware, I have gotten vmbuilder working for that purpose now :-)14:10
kirklandcjwatson: when will it build on cdimage?  anything else I need to do on that front at this point, to get an ISO together that we can start working from?14:13
cjwatsonkirkland: do I have to have more things on cdimage?  it's really tight on space14:14
cjwatsonif it's at all possible to do it externally, that would be preferable14:14
kirklandcjwatson: okay -- can you point me to some docs as to building an ISO from the seed itself now?14:17
cjwatsonkirkland: that said, how big does it end up being?14:17
kirklandcjwatson: ~800MB roughly14:18
cjwatsonhm, not small then14:18
kirklandcjwatson: yeah, well, i need to work on shrinking it down some14:18
kirklandcjwatson: there's more stuff that can go14:18
cjwatson/dev/mapper/cdimage-srv14:18
cjwatson                     1693107968 1560767584 132340384  93% /srv14:18
cjwatsonwe might be able to cope with another14:18
cjwatsondoes it need to be i386 and amd64?14:18
kirklandcjwatson: and that estimate was from my modified desktop livecd -- i've not yet built an ISO based directly on this seed14:19
kirklandcjwatson: no, only amd64 IMHO14:19
cjwatsonok, that helps14:19
kirklandcjwatson: and 800MB is an over estimate14:19
cjwatsonleave me with the action to get it done, then14:19
cjwatsonalpha-3 work item or whatever?14:19
kirklandcjwatson: sure14:19
=== sconklin-gone is now known as sconklin
cjwatsondoes anyone know where the code that updates changelogs.ubuntu.com comes from?14:46
dholbachcjwatson: mvo might14:47
cjwatsonhm, lp:extract-changelogs maybe14:48
mvocjwatson: that is a mix of code living on changelogs.ubuntu.com plus fixup code that adds symlinks where binver != srcver. is there a problem with it?14:49
seif_any1 know where anmar ius14:49
cjwatsonmvo: when processing syncs I noticed that some packages didn't seem to have changelogs, e.g. http://changelogs.ubuntu.com/changelogs/pool/universe/m/mc/mc_4.7.0-1ubuntu2/14:50
cjwatsonmvo: and I was wondering if this was something to do with them being 3.0 (quilt), since all the examples I noticed were14:50
cjwatsonmvo: but then current man-db has a changelog and that's 3.0 (quilt), so maybe dpkg-source was just out of date?14:51
mvocjwatson: that is likely, it does use a simple dpkg-source -x iirc but we requested a backport14:51
cjwatsonthat would make sense then, thanks14:51
cjwatsonjust wanted to check that it wasn't a current problem14:51
mvocjwatson: it maybe that its stuff extracted between the time when v3 arrived and when we got the backport installed14:51
cjwatsonyeah, could well be14:51
LucidFoxYay, finally, for most applications, the global menu seems to Just Work(tm)!15:05
LucidFoxno reordering or vanishing issues so far15:05
LucidFoxOkay, Qt applications are still reordered15:09
LucidFoxbut GTK ones are fine15:10
jcastroLucidFox: can you check gnome-terminal and tell me if you see a partial menu?15:11
LucidFoxjcastro> Okay, the Help menu is different in the global menu applet, and the Terminal menu has some different items entirely15:13
jcastroyeah, that seems to have regressed, people are looking into it15:14
LucidFoxthe Help menu has only Contents and About15:14
jcastroyeah in some cases it's still showing partial menus15:14
jcastrothat's lp #59422815:15
ubottuLaunchpad bug 594228 in Application Menu Indicator "[Master] Application is showing a partial menu" [High,Confirmed] https://launchpad.net/bugs/59422815:15
LucidFoxBut there are separators, correct order in everything apart from VLC, and it no longer loses the menu when the application is restarted15:15
LucidFoxIt's almost usable now :)15:15
jcastroLucidFox: feedback on any of those open bugs on the menu would be appreciated15:15
LucidFoxOut of curiosity, why did you implement this from scratch instead of using the gnome2-globalmenu code?15:15
jcastrothe DX guys talked to the gnome2-globalmenu guys before they started15:16
jcastroi don't know the details other that they decided that ted/bratches way was the way forward15:17
LucidFoxted/bratches?15:19
LucidFoxOh, and the panel periodically crashes now -_-15:19
jcastroted and bratsche I mean (cody)15:19
jcastroLucidFox: join us in #ayatana if you want to help us with the menu!15:20
MacSlowbryce_, ping15:34
LucidFoxOkay, the status of xchat-gnome upstream is concerning15:36
LucidFoxThey're apparently a hairline away from release, yet nobody's answering bug reports or mail, the IRC channel is almost empty, and my post to the mailing list has been held in moderation for days.15:37
dholbachsorry for spamming, but we have one slot for UDW left: would anybody like to talk about distributed development 16th July at 17 UTC?15:43
=== virtuald_ is now known as virtuald
=== bjf[afk] is now known as bjf
=== sami is now known as samiz
unknownwhere can i find out how the jockey-gtk package is built?17:09
=== hrw is now known as hrw|gone
ScottKunknown: apt-get source jockey17:11
unknownScottK: what if i'm not running ubuntu? (i have a ubuntu vm, but just curious)17:14
ScottKGetting the source package and looking is still the best way.  Do that in the vm is the easiest.17:15
ScottKIt's also available at https://launchpad.net/ubuntu/+source/jockey17:16
unknownahh, all i found was this: http://packages.ubuntu.com/source/lucid/jockey17:18
ebroderunknown: Yeah, Launchpad is a little less discoverable than packages.u.c, but it has about a gajillion times as much information17:18
=== yofel_ is now known as yofel
ckingcjwatson, "Investigate situation with Intel graphics drivers on EFI" - I'm on the ball, waiting for some more input from Intel on this17:26
cjwatsoncool, ta17:27
ckingapologies for the delay, hope to get some info back next week17:29
=== beuno is now known as beuno-lunch
=== MacSlow is now known as MacSlow|capoeira
=== deryck is now known as deryck[lunch]
arandHow are the dailies of the alternate CD faring atm, there seems to have been none the last three days, and we're a bunch that's quite eager to try out btrfs...17:57
cjwatsonarand: I fixed a problem earlier today that affected them17:57
arandcjwatson: Cool, so hopefully one for tomorrow then?17:58
cjwatsonmaybe, I wouldn't like to promise17:58
* cjwatson scores up d-i17:59
cjwatsongah, one hour to start even at 2000018:00
cjwatsontoolchain ate our builders18:00
ScottKcjwatson: Looking at the other packages, I think it's safe to say doko ate our builders.18:02
cjwatsonopenjdk is sort of toolchain :)18:02
ScottKpython2.6 has been at 999999 for 7 hours on ia64 and didn't get a chance yet ...18:03
=== sconklin is now known as sconklin-gone
=== beuno-lunch is now known as beuno
=== beuno_ is now known as bueno
=== bueno is now known as beu\no
=== beu\no is now known as beuno
=== oubiwann is now known as oubiwann-away
=== deryck[lunch] is now known as deryck
=== bjf is now known as bjf[afk]
=== Amto_res is now known as Amto`Off
[reed]any Makefile experts around, Firefox could use your assistance -- http://christian.legnitto.com/blog/2010/06/25/are-you-a-makefile-guru-would-you-like-to-make-firefox-more-secure/19:43
=== jtechidna is now known as JontheEchidna
=== dendro-afk is now known as dendrobates
hungerIn maverick I get "root filesystem check failed" on the console... mounting fails, btrfsck says all is well, the lucid kernel boots fine. Any ideas what I can try to debug this?20:36
=== bjf[afk] is now known as bjf
=== oubiwann-away is now known as oubiwann
=== MacSlow|capoeira is now known as MacSlow
rohanwasn't ubuntu 8.04 onwards supposed to receive latest versions of firefox from upstream (3.6.4) last week?21:07
rohanwas the plan dropped?21:07
rohan_wasn't ubuntu 8.04 onwards supposed to receive latest versions of firefox from upstream (3.6.4) last week?21:09
rohan_was the plan dropped?21:09
rohan_i'm talking about https://blueprints.edge.launchpad.net/ubuntu/+spec/desktop-lucid-new-firefox-support-model21:09
=== rohan_ is now known as rohan
arandrohan: They were testing it in the security PPA, not sure if they rolled it out yet..21:13
rohanarand: they haven't..21:14
ScottKrohan: Yes. No. Yes. No.21:20
jdstrandrohan: it is being worked on21:24
jdstrandrohan: see https://wiki.ubuntu.com/Testing/Firefox3.6.4Upgrade/SecurityPublication21:24
jdstrandrohan: hardy and lucid should go out early next week. jaunty and karmic sometime after21:25
rohanScottK, jdstrand : thank you21:42
rohanhmm is this going to cause one more point release of 8.04?21:44
cjwatsonas far as anyone's told me, no more point releases of 8.04 are planned21:44
ebrodercjwatson: Are they not going to be done every 6 months anymore?21:44
rohancjwatson: then what is the point of updating the "release notes" for 8.04?21:45
cjwatsonebroder: https://wiki.ubuntu.com/HardyReleaseSchedule - "Point releases will cease once Ubuntu 10.04 is released."21:46
rohanubuntu 8.04 support for desktop has expired right?21:47
ebrodercjwatson: Ah, ok. I hadn't really been paying attention; my apologies21:47
cjwatsonwe don't have the resources to do point releases for multiple LTS series at once21:47
cjwatsonrohan: no idea - but of course saying there are no more point releases is not the same as saying there are no more updates21:47
cjwatsonrohan: no, 8.04 is supported on desktop until April 201121:47
cjwatsonwe're just not planning more CD image refreshes, that's all21:48
rohani understand, but the page which jdstrand linked to mentions updating the release notes for 8.04, which doesn't make sense21:48
rohanand i thought LTS support was 2 years for desktop and 5 years for server..?21:48
cjwatson3 years for desktop21:48
jdstrandrohan: you misunderstand21:48
jdstrandrohan: those release notes are special notes that need to be added to the USN21:49
jdstrandI could have perhaps been clearer, but the idea is that there are additional notes for the 3.6.4 release in hardy, et al21:50
rohanjdstrand: oh, i am sorry.21:52
rohanthanks for clarifying, cjwatson21:52
=== almaisan-away is now known as al-maisan
=== al-maisan is now known as almaisan-away
achianganyone know what happened to gdk-pixbuf in lucid?22:39
gordachiang, nothing? its still there22:40
achianggord: hm, i'm actually looking for the package that would supply the gdk.typelib file in /usr/share/girepository-1.022:41
gordachiang, ah, i have no idea about that22:42
achiangand apt-cache search pixbuf doesn't reveal anything interesting22:42
achiangnor does apt-cache search gdk22:42
gordachiang, there is a gir-repository package (or something like that) that installs a lot of the glib/gnome typelibs22:42
achianggord: yeah, i've got that installed and it doesn't provide anything related to gdk22:43
geserGdk-2.0.typelib is in gir1.0-gtk-2.022:44
achianggeser: ah, there it is! ta!22:45

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!