/srv/irclogs.ubuntu.com/2011/08/08/#ubuntu-desktop.txt

RAOFHm.  Why isn't my espresso machine on!01:42
micahgRAOF: can we actually use lzma debs to save space on the CD?  Does that help for the debs separately than the livefs?01:47
TheMusomicahg: Using lzma will save space on the alternate disk/the DVD, but not the desktop CD, as the livefs is only zlib compressed.01:55
micahgah, so it only helps for stuff that's only on the alternate and not the live CD since isn't the alternate usually smaller?01:55
RAOFmicahg: Right.  Well, it also helps keep the download smaller, too.02:04
TheMusoThere is that too.02:35
TheMusot/c02:35
micahgdoes anyone want to merge my vala-0.14 uploads into the desktop branch?02:38
micahgafter 2 more publisher runs, it should be built with itself on all archs02:39
micahgthere's a merge proposal waiting02:43
=== kyu is now known as kengyu
rickspencer3guess it's quiet as everyone is heading to Berlin for Desktop Summit05:38
rickspencer3or there already and asleep :)05:38
RAOFI think the latter.05:41
RAOFHow many people of this timezone are going?05:41
rickspencer3hiya RAOF05:43
RAOFHey there rickspencer305:43
rickspencer3RAOF, I dunnoe05:43
rickspencer3I'm not going this year :/05:43
rickspencer3I'll call into the adboard meeting, though05:44
RAOF:(05:44
rickspencer3RAOF, it's all good05:44
rickspencer3I've been running around so much this summer, I'm not sorry to not be taking another trip05:45
RAOFPlenty of conferences to go around? :)05:45
rickspencer3I've gone Dublin -> Kansas -> Delaware -> Isle of Man -> Amsterdadm -> Londong -> Toulouse05:45
rickspencer3I admit the weekend of Amsterdam was a bit self-inflicted ;)05:46
rickspencer3Londong is surprisingly a lot like London, just more typos05:46
rickspencer3TheMuso, hos is vinux coming along05:47
rickspencer3?05:47
* rickspencer3 saw it referenced on Reddit this morning05:47
RAOFHeh.05:48
RAOFToulouse is very nice.05:48
RAOFWhat was happening there?05:48
TheMusorickspencer3: haven't really been involved muc recently. I am going to start doing more with them pos feature freeze, as I want to help hem move to a better build process and using more packaging bits for a oneiric based release.05:50
RAOFGARH.  Whose fault it is that it takes *seconds* for alt-tab to pop up on this SandyBridge with plenty of ram and spare CPU?05:50
rickspencer3RAOF, uh, shouldn't you be the one person who can answer that?05:50
RAOFrickspencer3: It'll be someone in DX, mark my words :P05:51
RAOFAnd the tools which I'd normally use to debug this seem to have broken.  Huh.05:53
rickspencer3ah, Sandybridge05:54
rickspencer3how I love thee05:54
RAOFOneiric is *much* better than it was.05:54
rickspencer3RAOF, go on05:55
RAOFThis laptop used to have time-to-hardlock measured in minutes.05:55
RAOFNow it's been behaving well enough to get the SSD and 8GB of ram from my thinkpad.05:55
RAOFI think some of that has actually been the newer mesa; everyone can now benefit from that.05:56
RAOFIt's one of HWE's laptops, so I was thinking that it was a hardware issue rather than05:56
RAOFOneiric being screwy.  Now that everything's working, though, I think it may have just been SandyBridge being horribly broken on oneiric for a while.05:57
RAOFWooo!  Mesa's transitive dependencies are once again multiarched.06:06
AfCEvery time I try to upgrade, I get told:06:36
AfCCalculating upgrade... Done06:36
AfCThe following packages will be REMOVED:06:36
AfC  gir1.2-mutter-3.0 gnome-shell gnome-tweak-tool06:36
AfCCan anyone suggest what I'm doing wrong?06:36
RAOFWe're probably half way through a transition.06:37
AfCi.e. Obviously there's a dependency problem somewhere, but how can I figure out what it would be? `apt-get dist-upgrade -s` isn't really helping06:37
AfCRAOF: It's been like this for > 10 days06:37
RAOFOk.  Well, I have gnome-tweak-tool installed and nothing's being particularly crazy.  Perhaps it's a local issue?06:37
RAOFYou can always explicitly go "apt-get install gnome-shell" and it should give some indication as to why it can't be installed/upgraded.06:38
chrisccoulsongood morning everyone06:38
chrisccoulsonit's going to be quiet in here this week ;)06:39
AfCRAOF: ah, right, thanks06:39
* AfC runs `apt-get install gnome-shell -s`06:39
RAOFchrisccoulson: Hey, ho!06:39
AfCHere we are:06:39
AfCgnome-shell : Depends: gir1.2-json-1.0 but it is not going to be installed06:39
chrisccoulsonhi RAOF, how are you?06:39
AfCHm06:39
RAOFchrisccoulson: How's the import-my-evolution-filters branch of thunderbird going? :)06:39
RAOFAfC: And gir1.2-json-1.0 isn't going to be installed because...?06:40
chrisccoulsonheh ;)06:40
AfCRAOF: not sure, but I now see that that is what is trying to pull in all of Unity & Compiz.06:40
AfCRAOF: [thanks, I'm further along now]06:41
AfCNo to figure out why the hell is that trying to pull in the entire Unity+Compiz+UbuntuOne stack...06:41
mvoAfC: I missed part of this, but there is apt-get install/dist-upgrade etc with -o Debug::pkgDepCache::AutoInstall=true, that will tell what pkg pulls in what other06:42
AfCHm. gir1.2-json-glib-1.0 is to be removed. Replaced by gir1.2-json-1.0? Could that be the problem, perhaps06:42
AfCmvo: thanks06:43
AfCmvo: [for about 10 days now I've been unable to upgrade because the dependency resolution wants to remove gnome-shell and wholesale install the entire unity stack]06:44
AfCmvo: trying...06:44
AfCmvo: NICE!06:44
AfCmvo: [I've been looking for something like that for ages!]06:44
mvoAfC: your welcome06:46
AfCNow to figure this out. gir1.2-json-1.0 does the conflicts/replaces/provides dance for gir1.2-json-glib-1.006:48
AfCfine06:48
AfCSo the bulk of the problem would seem to be not so much installing the new gir1.2-json-1.0 as removing the existing gir1.2-json-glib-1.006:56
AfCI can get the horrible install graph up if I attempt to remove that specifically.06:57
AfCShould a package both Replaces: and Conflicts: its predecessor? I thought it was one or the other?07:03
AfCSo, the problem is that I can's safely uninstall gir1.2-json-glib-1.0. New package gir1.2-json-1.0 purports to provide it, but removing the original results in a gooey mess.07:04
AfCNot sure where to go from here. Should I be trying to roll my own gir1.2-json-1.0?07:04
=== tkamppeter_ is now known as tkamppeter
seb128hey07:16
AfCA package that Provides (and Replaces) something shouldn't Conflict with it, should it?07:24
chrisccoulsonhi seb128, how are you?07:26
chrisccoulsonhow is the desktop summit?07:26
seb128hey chrisccoulson07:26
seb128I'm great thanks07:26
seb128desktop summit is good so far07:26
seb128though it's only monday morning but it feels like friday07:26
seb128chrisccoulson, how are you? had a good w.e?07:26
chrisccoulsonseb128, yeah, i'm good thanks. my weekend was quiet, apart from spending time trying to debug a firefox crash ;)07:27
seb128chrisccoulson, ready for f.f?07:27
chrisccoulsonseb128, yeah, i think so. i still need to land the contact integration work from m_conley in thunderbird07:28
seb128that would be nice07:28
chrisccoulsoni should be able to do that before f.f07:28
seb128chrisccoulson, do you know if there is any unity light theme integration btw?07:28
seb128or is there only the default theme variant?07:28
chrisccoulsonseb128, did you see the theme andreasn did? or are you asking something different?07:29
seb128chrisccoulson, yeah, I saw this one, I was asking if there is a light variant07:34
chrisccoulsonseb128, oh, i'm not sure. perhaps it works on the light variant too, although i've not tried that ;)07:34
chrisccoulsonseb128, oh, it works fine :)07:35
AfCCan someone do an `apt-cache show gir1.2-json-1.0` and tell me if their output says to both Provide, Replace *and* Conflict on `gir1.2-json-glib-1.0`? I'm hoping this is just a problem here locally, but maybe not07:35
seb128oh, so maybe I need to get that installed ;-)07:35
seb128when will it be default in the distro?07:36
chrisccoulsonthe toolbar and tabbar blends in with the titlebar07:36
chrisccoulsonseb128, when the main issues are fixed. the compose window is currently a bit messed up atm07:36
chrisccoulsonseb128, ah, the main issue is that the icons aren't very visible ;)07:37
chrisccoulsonseb128, is andreasn at the desktop summit?07:37
seb128chrisccoulson, I think he is yes07:37
chrisccoulsonyou could probably ask him about that ;)07:38
chrisccoulsoni'm not sure if he's aware of the radiance theme07:38
chrisccoulsoni'd forgotten about that, as i've never used it ;)07:38
seb128ok07:38
chrisccoulsonoh, wow, my session is really messed up now07:41
chrisccoulsongtk-window-decorator has crashed twice already this morning, and now i've restarted it again, my windows have the wrong decorations :/07:41
chrisccoulsonthis is messed up - http://people.canonical.com/~chrisccoulson/gtk-window-decorator-suckiness.png07:42
seb128chrisccoulson, when do you start fixing compiz !07:43
RAOFAfC: Yes, it does here.  That's a normal idiom for virtual packages, though.  Conflicts/Replaces says "I replace that other thing, so install me and remove it" and Provides says "Actually, I am that thing"07:43
chrisccoulsonseb128, ;)07:44
chrisccoulsonwhere do i start?07:44
RAOFchrisccoulson: Heh, cool.  Very old-school.  That's, what, ion?07:44
seb128hey RAOF07:44
RAOFseb128: Hey, ho!07:44
chrisccoulsonhmm, compiz isn't recovering now. i think i'm going to have to restart my session already07:45
chrisccoulsonnot a good start to the week ;)07:46
lifelessRAOF: conflicts-replaces is deprecated07:46
lifelessRAOF: because it makes the package graph between releases unmigratable without --force-depends07:46
lifelessRAOF: the new(like 3 years now) shiny is conflicts/breaks07:46
RAOFlifeless: Not according to http://www.debian.org/doc/debian-policy/ch-relationships.html07:47
RAOFConflicts should be used... "in conjunction with Provides when only one package providing a given virtual facility may be unpacked at a time (see Virtual packages - Provides, Section 7.5), "07:47
RAOFBut yeah.  When you're just moving files from one package to the other, Breaks is the new hawtness.07:48
RAOFlifeless: I seem to recall doing this dance before :)07:48
lifelesshmmm07:49
lifelessso, anything depending on the virtual package has to be removed, then the deps need to be removed, and then reinstalled, unless either:07:49
lifeless --force-depends07:49
lifelessor07:49
lifelessdpkg has a special case around it07:49
RAOFI suspect the latter.07:50
lifelessRAOF: thats provides/conflicts anyhow07:50
lifelessRAOF: not replaces/conflicts07:50
lifelessRAOF: so, your appeal to authority is denied :)07:50
RAOFAh, yeah.  That's true.07:51
RAOFCurse you, gadget!07:51
lifeless:)07:52
lifelessRAOF: 7.4 is fairly clear that conflicts-replaces is no longer a use case - replaces is listed under 'use breaks'07:57
pittiGood morning everyone!08:00
RAOFlifeless: 7.6.2 still suggests Provides/Conflicts/Replaces.08:01
RAOFlifeless: I suspect we've hit a bug in policy :)08:01
RAOFpitti: Gooooood morning!08:01
pittinow the holiday slacking is over, and the desktop summit slacking starts :)08:01
RAOFHeh.08:01
RAOFOh, excellent.  You've popped up just as I was filing a pkgbinarymangler bug!08:02
lifelessRAOF: WIN08:03
RAOFpitti: I suspect pkgbinarymangler's changelog mangling is breaking multiarch; dpkg considers the i386 and amd64 libglapi-mesa changelog.Debian.gz as different and hence bails on the install.08:05
pittiRAOF: is that blocking you?08:05
pittiRAOF: if so, let me know the bug # once you filed it, then I can look at it right away08:05
pittiscrew that massive holiday email backlog08:05
RAOFNot blocking, because we don't really do much with multiarch.08:06
RAOFJust annoying, because I'd like up-to-date mesa with wine :)08:06
RAOFAnd time for pilates!08:06
lifelessRAOF: it should say break or conflicts if the unpacking limit is what is truely desired, AFAICT08:06
rodrigo_morning08:07
pittimicahg: ah, I just uploaded another vala-0.14, apparently that conflicted with your's (apparently you didn't push to bzr)08:09
pittimicahg: but as it built now anyway, I guess I now just restore it to plain "valac"?08:09
seb128hey rodrigo_08:12
rodrigo_hi seb128, how are you?08:12
pittiah, you did08:13
seb128rodrigo_, I'm fine thanks, how are you?08:14
seb128rodrigo_, had good holidays?08:14
rodrigo_seb128, yes, very good :)08:14
rodrigo_so, is everyone in berlin?08:14
seb128yes08:15
seb128you need to hold desktop while we are away :p08:15
seb128btw seems RAOF managed to get colord in debian08:15
AfClifeless, RAOF: so does this mean I need to uninstall my entire GNOME stack, then reinstall it, to get around this?08:16
seb128RAOF, did you file a mir and sync request for it?08:16
pittihey rodrigo_, how are you?08:16
AfC[or start doing lots of manual installs with --force-depends, or...?]08:16
rodrigo_seb128, ok, I will guard the ship :)08:16
pittiRAOF: http://packages.qa.debian.org/c/colord.html -> great job!08:16
rodrigo_hi pitti, I'm fine, a bit exhausted still after the trip back home on Saturday though08:17
pittimicahg: I updated vala-0.14 bzr now, thanks for fixing it08:17
rodrigo_seb128, ok cool, I'll enable the color plugin in g-s-d and g-c-c as soon as the MIR is accepted08:17
lifelessAfC: well, what did aptitude say the root of the problem was?08:17
seb128_re08:18
seb128_crappy conference internet08:19
seb128_RAOF, pitti, rodrigo_: did you get what I was saying about colord?08:19
rodrigo_:)08:19
rodrigo_seb128_, I did08:20
pittiseb128_: that it's in Debian, and that we need to wait for the MIR, yes08:21
=== seb128_ is now known as seb128
chrisccoulsonhi pitti, how are you? did you have a nice vacation?08:22
pittichrisccoulson: it was great, yes (see http://piware.de/fotos/Innradtour-2011/)08:23
pittiwas a nice cycling trip again, and this time with Alps mountain profile08:23
pittithat was quite an exercise :)08:23
rodrigo_pitti, wow08:24
chrisccoulsonpitti, the pictures look nice :)08:25
rodrigo_pitti, where was that?08:25
rodrigo_ah, St. moritz08:25
pittirodrigo_: we started in St. Moritz (Switzerland), through Austria, until Passau in Germany08:25
pittialong the whole Inn river08:25
rodrigo_cool08:25
seb128_grrrr summit internet08:26
pittiseb128_: looks like my 2G connection in the train at 200 km/h is more stable than your wifi? :-)08:26
seb128_pitti, wifi there is crapland08:28
seb128_pitti, it takes over a minute to load a webpage, irc lag keeps raising08:31
seb128_still not sure if you guys received what I wrote before08:31
seb128_I'm waiting for the irclog website to catch up to see that ;-)08:32
rodrigo_seb128_, we got your message about getting the week off, yes :)08:33
=== seb128_ is now known as seb128
pittiseb128_: last thing you said was "seb128_ | RAOF, pitti, rodrigo_: did you get what I was saying about colord?"08:33
rodrigo_seb128, about the colord thing, yes we did get it08:34
rodrigo_talking about broken things, I broke my laptop during holidays, on an update08:34
seb128ok08:35
seb128rodrigo_, broke like physically drop off and broke it?08:35
seb128or like distro breakage?08:35
rodrigo_seb128, software brokenness08:35
seb128did you fix it?08:35
rodrigo_need to recover it and copy all stuff to my PC08:36
rodrigo_seb128, in the process of08:36
rodrigo_seems to be some kernel 3.0-related08:36
seb128or is it still broken?08:36
seb128ok08:36
rodrigo_so, let's see if I can boot into another kernel and copy all stuff08:36
seb128so back to my colord question, did somebody filled a sync request and a mir bug?08:39
seb128rodrigo_, ^08:39
seb128once that done you can build g-c-c with it08:39
rodrigo_seb128, not me, so will ask RAOF08:39
AfClifeless: all the aptitude "solutions" involved me uninstalling gnome-shell, gnome-tweak-tool, empathy, and epiphany.08:41
seb128ok08:41
pittiRAOF, seb128: i can sync it right now08:42
AfCRAOF: so, based on lifeless's input, does that mean we need to redo the gir1.2-json-1.0 package?08:42
seb128pitti, seems like a first step, then RAOF or rodrigo_ will need to write a mir for it08:46
AfCRAOF: If so, I can attempt repackaging it locally, but before I do that I'd like to know it's the right thing to try :)08:46
lifelessAfC: right but it should explain why08:47
lifelessRAOF: as far as installaility, conflicts and breaks are identical08:48
lifelessAfC: this sort of thing is usually due to an incompatible library bump and only some users of it being migrated08:48
pittiRAOF, seb128, rodrigo_: can't sync colord; launchpad doesn't support .tar.xz yet08:50
seb128pitti, it should08:50
seb128cjwatson, said it was ok before natty08:50
pittiplease do an ubuntu1 upload as tar.bz2 or so, or just do that in debian08:50
pitti2011-08-08 08:47:54 DEBUG   Rejected:08:50
pitti2011-08-08 08:47:54 DEBUG   File colord_0.1.11.orig.tar.xz mentioned in the changes file was not found.08:50
pitti2011-08-08 08:47:54 DEBUG   Unable to find colord_0.1.11.orig.tar.xz in upload or distribution.08:50
pitti2011-08-08 08:47:54 DEBUG   Files specified in DSC are broken or missing, skipping package unpack verification.08:50
pittithe file does exist in ~/syncs/08:51
pittiso whatever it tries to tell me, it doesn't like it08:51
seb128he fixed the remaining issues08:51
seb128not sure that get real world testing since though08:51
seb128I wonder if that's a sync tool issue08:52
pittisync-source.py seems fine08:52
pittiit's flush-syncs which broke08:52
seb128i.e if a manual upload would work08:53
seb128if only opening a webpage was taking less than 3 minutes here...08:53
seb128https://bugs.launchpad.net/launchpad/+bug/61915208:54
ubot2Ubuntu bug 619152 in launchpad "Add data.tar.xz support" [Low,Triaged]08:54
pittiseb128: ah, so "pending"08:55
pittiRAOF: eh, new mesa pulls in libwayland0?08:56
pittiRAOF: is wayland meant for main now?08:56
seb128hum, timeouted this time09:03
seb128https://bugs.launchpad.net/launchpad/+bug/61915209:04
seb128well I guess RAOF will need to do an ubuntu upload with a bz209:04
seb128that's what I was saying before timing out09:04
ubot2Ubuntu bug 619152 in launchpad "Add data.tar.xz support" [Low,Triaged]09:04
pittiseb128: ah, so "pending"09:04
seb128seems so09:05
Laneyare you talking about colord?09:08
Laney(can't do much, on 3g)09:09
pittiLaney: yes, in particular tar.xz support09:09
LaneyRAOF and I tested that on launchpad, and it seemed to work09:09
Laney.data.tar.xz is figgerent to .orig.tar.xz09:10
Laneysee my PPA for an uploadwhich worked09:10
Laneyer, different*09:10
Laneyhttps://launchpad.net/~laney/+archive/ppa/+sourcepub/1844994/+listing-archive-extra09:10
mvohey pitti ! when you go on the pygi hackfest, it would be great if you could have a look at https://bugzilla.gnome.org/show_bug.cgi?id=656014 - i attached a patch that works for me(tm) but as noted in the bug it may be incomplete09:11
ubot2Gnome bug 656014 in general "Missing GIR annotation/wrapper for pango_layout_get_line_readonly()" [Normal,Unconfirmed]09:11
seb128Laney, the sync tools don't handle it fine it seems09:16
seb128but it could be an issue in the sync tools, not launchpad itself09:16
seb128hey mvo09:16
mvohey seb128, how is berlin?09:18
Laneyindeed09:18
AfClifeless: right; the "reason" is because gir1.2-json-1.0: Conflicts: gir1.2-json-glib-1.0 and gir1.2-json-glib-1.0 is installed since lots of things depend on it.09:18
lifelessAfC: so, is gir1.2-json-1.0 newer ?09:20
lifelessAfC: and do the things that depend on gir1.2-json-glib-1.0 depend on that name, or on the virtual package name ?09:22
rodrigo_ok, ready to recover my laptop and the data in it, hopefully, so bbiab09:28
seb128pitti, ready IRC logs not sure that was replied to but not logged yet09:30
seb128bug #81021709:30
ubot2Launchpad bug 810217 in wayland "[MIR] wayland protocol package" [Medium,Fix released] https://launchpad.net/bugs/81021709:30
pittiseb128: ah, so it is for main indeed09:32
pittihey mvo!09:32
pittimvo: opened that in a tab, I'll get to that09:32
mvogreat, thanks pitti, its not urgent, but would allow me merging some code that uses this in the s-c gtk3 port09:32
pittimvo: note that you didn't actually attach a patch, did you mean to?09:33
mvopitti: indeed, let me fix that, its pretty trivial, just a "(transfer none)" for the two functions09:34
cjwatsonpitti: I'd like to investigate that .orig.tar.xz sync problem09:51
cjwatsonso please don't manually upload that yet09:51
cjwatsonfixed now09:58
AfClifeless: [sorry, was out]10:19
AfClifeless: 1) gir1.2-json-1.0 newer10:20
AfClifeless: 2) How do you differentiate between something depending on a virtual and a normal package?10:22
AfC[I'd have to guess not virtual since it wasn't previously a virtual package]10:23
lifelessby knowing what the package is10:23
lifelessanyhow10:23
lifelessthis isn't a virtual package10:23
lifelessthe conflicts/replaces pair is for virtual packages10:23
AfC(didn't think so)10:23
lifelessso10:24
lifelessthe case here is that the old things are depending on the old library10:24
lifelessand the new package does not fulfil stuff10:24
AfClots of old things, yeah10:24
lifelessthe old packages need to be rebuilt10:24
lifelesseither to change their dep line, if they would work, or to link against the new lib, if they wouldn't.10:25
lifelesseither way, rebuild.10:25
lifelessfor all the rdepends of the package being removed.10:25
AfCso I thought the whole idea of them having done Provides: in the "new" package is that everything else under the sun doesn't have to be rebuilt?10:25
* pitti grabs some lunch, bbl10:25
lifelessAfC: whats the provides statement say ?10:25
AfC# apt-cache show gir1.2-json-1.010:26
AfC...10:26
AfCProvides: gir1.2-json-glib-1.010:26
AfCReplaces: gir1.0-json-glib-1.0, gir1.2-json-glib-1.010:26
AfCConflicts: gir1.0-json-glib-1.0, gir1.2-json-glib-1.010:26
AfCI tried manually force removing the "old" package, but the dependency resolution isn't any better.10:28
AfClifeless: ^10:29
lifelessAfC: provides cannot provide a versioned dependency.10:30
lifelessAfC: what does apt-cache show for one of the dependencies that wants to be removed show.10:30
AfClifeless: so, eg gnome-shell (to be updated), Depends: gir1.2-json-1.010:31
lifelessAfC: no, one of the ones that needs to be removed10:32
AfClifeless: so, eg gnome-shell (not updated), Depends: gir1.2-json-glib-1.0 (>= 0.12.0)10:32
AfClifeless: this is the problem: it wants to remove gnome-shell10:32
lifelessok10:33
lifelessgir1.2-json-glib-1.0 (>= 0.12.0)10:33
lifelessthe bits in brackets mean that this relation can only be satisfied by non-virtual packages.10:33
lifelessvirtual packages satisfy no versions10:34
AfCinteresting, didn't know that10:34
AfCwhich is why Provides isn't working, and the Conflicts is screwing us10:34
lifelessthis is covered in the policy doc linked earlier, if you want the gory details10:34
AfCso the ... if not solution, at least workaround ... will be to remove the Conflicts line and rebuild the "new" gir1.2-json-1.0 package10:35
AfC?10:35
AfC[darn, out of time, have to go offline]10:38
lifelessuhm, depends on the package contents10:39
AfClifeless: thanks for your help, been illuminating.10:39
AfCRAOF, I can certainly try & test a rebuild locally here if you don't have time to rebuild the package in PPA per Robert's suggestions.10:39
AfCI think it's just a package rename, but not sure. Bit of a black box.10:39
AfC→ off10:39
=== MacSlow is now known as MacSlow|lunch
pittimdeslaur: FYI, new polkit is now in oneiric which supports the X11 pkexec stuff11:45
pitti(as well as your multiline patch)11:45
mdeslaurpitti: yeah, I saw that...thanks! I'll try and find some time this week to try and get rid of gksudo11:45
chrisccoulsonhi mdeslaur, how are you?11:46
mdeslaurhi chrisccoulson!11:46
mdeslaurchrisccoulson: not bad, you?11:46
chrisccoulsonmdeslaur, tired ;)11:47
chrisccoulsonand i feel like my head is going to explode. i've spent several days looking at the same problem and not getting very far ;)11:47
mdeslaurchrisccoulson: oh, I hate when that happens :(11:48
chrisccoulsonyeah, me too :(11:48
pittichrisccoulson: maybe stop looking at it for a day or two and do something else?11:56
micahgpitti: re vala-0.14> thanks for merging11:57
geseris the "vala-0.14" source package the successor of "vala"?12:05
pittiyes, but we keep several major versions in parallel for an easier transition12:10
geserdoes it need a proper MIR to get to main? currently it's in universe and some packages build-depend on valac (now build from vala-0.14)12:11
pittino, just new version; I promote it12:12
pitti(done)12:13
geserthx12:14
* ogra_ is stunned ... so we dont ship a single tool anymore whith which you can resize a picture ?12:27
* ogra_ just desperately tried to shrink a photo with an ubuntu default install 12:27
pittihm, apparently not :/ f-spot used to be able to, but shotwell isn't12:30
ogra_right12:32
ogra_i wasted 1h to try to do that on my android tablet ... all apps there only have crop either ... so i decided to quickly do irt on my lappie and copied it over ... to just find the same situation12:33
ogra_its weird that nobody seems to have thought about that function, i cant belive thats not a common task12:33
ogra_(in either OS)12:34
GunnarHjpitti: Hi Martin, hope your vacation was fine.12:34
pittihey GunnarHj12:34
pittiGunnarHj: indeed it was, thanks! catching up on stuff/mail now12:34
pittihow are you?12:34
GunnarHjpitti: I'm fine, thanks.12:34
GunnarHjpitti: I have three pending MPs that are related to each other. Probably I'd better ask Rodrigo to review and sponsor the g-c-c MP. That way most of it may be done upstream instantly.12:35
GunnarHjBut I'd appreciate if you could help process the l-s and accountsservice MPs.12:35
GunnarHjhttps://code.launchpad.net/~gunnarhj/language-selector/shared-tools/+merge/6919012:35
GunnarHjhttps://code.launchpad.net/~gunnarhj/ubuntu/oneiric/accountsservice/lang-to-profile/+merge/6908512:35
GunnarHjFor a quick briefing you may want to read the latest comment on https://code.launchpad.net/~gunnarhj/gnome-control-center/langfix/+merge/6899712:35
=== MacSlow|lunch is now known as MacSlow
=== m_conley_away is now known as m_conley
pittiGunnarHj: thanks, saw the MPs; will get to it as part of catching up with mail12:52
GunnarHjpitti: Ok, thanks.12:53
cyphermoxgood morning!13:07
cyphermoxhey pitti, happy coming back from vacation ;)13:08
pittihey cyphermox, how are you?13:08
cjwatsoncyphermox: do you think we could migrate network-manager et al to libnl3 without too much pain?  I would love to merge http://packages.qa.debian.org/w/wpasupplicant/news/20110730T134712Z.html13:09
cjwatson(well, that's a sync actually)13:09
cjwatsonit would be least paperwork if we could just swap out libnl for libnl313:09
cyphermoxcjwatson: sure. I can see if it builds fine13:10
cyphermoxthere was discussion about libnl3 jsut recently on the mailing list, too13:10
cyphermoxpitti: doing good :)13:10
cjwatsonthe build-deps in main are  bluez ipvsadm libvirt network-manager ntrack quota wireless-crda wpasupplicant13:11
cyphermoxcjwatson: nothing like trying now, nm builds quickly13:12
cjwatsonI can have a quick look at the others13:12
cyphermoxbluez I can look at too13:13
cyphermoxntrack isn't that asac's project?13:13
cjwatsonipvsadm fails out of the box13:15
asaccyphermox: libnl3 should be ready for ntrack in latest release + bzr13:15
cyphermoxasac: cool, thanks13:16
asaccyphermox: needs packaging bump and a build-depends update afaict13:16
cyphermoxalrighty13:17
ricotzcyphermox, hi13:17
cyphermoxricotz: hey13:17
ricotzcyphermox, do you know if the "use as hotspot" button in g-c-c is suppose to work?13:17
ricotzor are there known problems13:17
cyphermoxricotz: no idea, I haven't tried13:18
cyphermoxricotz: please open a bug if it doesn't work and I'll take a look at it later :)13:18
ricotzok, it just crashes, but it doesnt seem to be a nm problem13:18
ricotzcyphermox, are you able to test this? for me it crashes in glib (but i am using a git version)13:19
cyphermoxsure, I'll test it, but later13:19
ricotzthanks13:20
cyphermoxricotz: trying to figure out what needs to be done for libnl3 :)13:20
ricotzno problem, good luck with that :)13:20
cyphermoxcjwatson: seems there is a git branch in progress for libnl3 in NM; but it's not merged, and there only seems to be one commit; I'll try it and see how far I can get13:20
cjwatsonit might just be linkage tweaks, with luck?13:21
cyphermoxah, i see, it's not just a few commits, it's probably mostly done, and removing old code13:21
tjaaltondesrt: hmm, I read that dconf "doesn't work over NFS", is it true?13:24
tjaaltondesrt: ok found the upstream bug13:29
cjwatsoncyphermox: ipvsadm seems to amount to http://paste.ubuntu.com/661134/13:31
cyphermoxcjwatson: thanks13:32
cyphermoxI think dcbw was doing some refactoring in the netlink stuff as well; but I'll see, trying to build now with a patch that contains all the commits from the libnl3 branch13:33
cyphermoxcjwatson: it built and seems to work "fine"; minus a few error messages but nothing appears to be omgwrong (e.g. I still got a connection, routes look okay, etc.)13:51
cyphermoxI'll just reboot and kick it around some more to be sure13:51
ricotzcyphermox, fyi, fixes the crash http://git.gnome.org/browse/gnome-control-center/commit/?id=48dec984dbfe5e69c6717c16f69a6f52e08cd1f214:11
ricotzsorry, no it doesnt :\14:12
cyphermoxyeah, that seemed a little weird ;)14:14
cyphermoxricotz: if you fix the fix, wanna prepare a package (and close the corresponding lp bug)? I'll be happy to sponsor if I can14:14
ricotzcyphermox, i just wanted to play with my wireless settings creating network, and it seems it cant handle this if there isnt something set up already14:16
cyphermoxricotz: ok. I'll give it a quick shot in a few minutes... this really all should work but I haven't had any hand in g-c-c, I only know there is a page that somewhat works14:17
ricotzcyphermox, ok14:18
ricotzit crashes to while getting the length of a null string14:18
cyphermoxricotz: ok, final build test for NM with libnl3-dev before uploading; now it's really your turn ;)14:18
ricotznice, is it a new snapshot too?14:19
cyphermoxnah14:19
ricotzok14:19
cyphermoxyou need a snapshot for something?14:19
ricotzno14:19
ricotzif my problem isnt a nm problem though ;)14:20
cyphermoxright14:20
cyphermoxthere's no big things in git, except for that "possible dbus activation errors" commit, but I was really more planning on taking another snapshot when Dan merges in the libnl3 stuff14:21
cyphermox(or releases 0.9)14:21
cyphermoxricotz: did you report the bug in lp after all?:14:25
cyphermoxI just managed to reproduce the crash after removing my ad-hoc networks already configured14:25
ricotzno14:26
cyphermoxplease ;)14:28
ricotzok14:28
cyphermoxreally interesting how it crashes, I wonder what that code is trying to do14:33
ricotzhttps://bugs.edge.launchpad.net/ubuntu/+source/gnome-control-center/+bug/82272214:37
ubot2Ubuntu bug 822722 in gnome-control-center "Crashes on using "Use as Hotspot"" [Undecided,New]14:37
ricotzcyphermox, ^14:37
cyphermoxthanks!14:37
cyphermoxricotz: #2  generate_ssid_for_hotspot (panel=0x7ffff82d4660) at cc-network-panel.c:2685   ///        ssid = 0x014:39
cyphermoxricotz: this probably doesn't help: (gnome-control-center:6327): network-cc-panel-WARNING **: Getting pretty hostname failed: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.14:40
cyphermoxhostname1 was not provided by any .service files14:40
ricotzcould be the/a cause for this14:48
cyphermoxpretty sure it is, testing a patch now14:48
cyphermoxcjwatson: just uploaded NM with the libnl3 changes14:51
geseris it normal that unity-panel-service segfaults every 60s? (oneiric, AMD64, unity-2d)15:02
cyphermoxgeser: nope15:05
cyphermoxmight be a known issue though15:06
cjwatsoncyphermox: OK, I'd better get uploading other things then so it can be swapped over and actually build :-)15:11
cyphermoxah, woops, didn't think it needed much coordination ;)15:14
cyphermoxcjwatson: d'oh. I see now.15:15
cyphermoxbluez to be ready soon too.15:16
rodrigo_hmm, can't find the gnome-online-accounts mir bug15:19
cyphermoxrodrigo_: closed?15:19
rodrigo_cyphermox, yeah, found it -> https://bugs.launchpad.net/ubuntu/+source/gnome-online-accounts/+bug/81292715:22
ubot2Ubuntu bug 812927 in gnome-online-accounts "[MIR] gnome-online-accounts" [Undecided,Fix released]15:22
cyphermoxrodrigo_: that's good, now just one mir to go and I can enable goa in evo15:23
rodrigo_cyphermox, what else do you need?15:23
chrisccoulsong'ah, i hate gtk-window-decorator15:24
chrisccoulsonplease stop crashing15:24
cyphermoxrodrigo_: ah, liboauth. there's already bug 808765 about it, waiting for cdbs to upload a new revision15:24
ubot2Launchpad bug 808765 in liboauth "[MIR] liboauth" [Undecided,Incomplete] https://launchpad.net/bugs/80876515:24
rodrigo_cyphermox, ok15:24
cyphermoxrodrigo_: if I have a patch for g-c-c, can I just ping you with it?15:25
rodrigo_ah, seb already uploaded a g-c-c with goa enabled, good seb!15:25
rodrigo_cyphermox, yes15:25
cyphermoxcool15:25
dobeyanyone know why this: http://pastebin.ubuntu.com/661194/ would happen building in a PPA? python-gobject is installed via build-depends.15:29
dobeymvo: ^^ you maybe? :)15:30
mvodobey: hm, that looks like aptdaemon is for some reason trying to talk to X :/ I guess you use this in your tests?15:34
dobeymvo: yes, but why would setup.py build do any of that?15:48
jbichabug 822464 would be nice to get fixed16:00
ubot2Launchpad bug 822464 in gjs "Upgrade gjs to 1.29.16" [Wishlist,New] https://launchpad.net/bugs/82246416:00
jbichaupstream never released a new python-gobject, I don't know if it's required or not16:01
cyphermoxrodrigo_: we may want to fallback to g_get_host_name()'s return value if the dbus call to ...hostname1 fails: http://paste.ubuntu.com/661213/16:05
cyphermoxoops16:05
cyphermoxjbicha: want to try to update the package to find out?16:07
=== cypher is now known as czajkowski
ricotzcyphermox, this should probably go upstream16:09
cyphermoxricotz: yes, that's why I was pinging rodrigo :)16:10
ricotzsince in the current case this g_warning leads to a fatal condition16:10
ricotzoh right ;)16:10
jbichacyphermox: I can try building and see what happens but I think testing it would be a bit over my head16:10
cyphermoxbut he got scared away so I'll open a bug16:10
=== ayan_ is now known as ayan
cyphermoxjbicha: ok16:11
rodrigo_cyphermox, ok thanks16:11
mvodobey: oh, so this does not run any tests, its really just from the build? I need to look into this16:11
cyphermoxah, so you saw my ping anyway16:11
mvodobey: do you have a link to the full build log for me?16:11
rodrigo_cyphermox, yeah, xchat crashed when I clicked on your link :)16:12
cyphermoxfun16:13
cyphermoxis that reproduceable? ;)16:13
rodrigo_cyphermox, not anymore for now16:15
ricotzcyphermox, rodrigo_, while this patched g-c-c gets uploaded it could need a symbols update for 3.1.4 too -- there are two new16:18
rodrigo_cyphermox, pushed to git16:19
rodrigo_hmm, can't we just wait for the next release?16:19
rodrigo_ricotz, which symbols are missing?16:19
ricotzrodrigo_, it could wait -- http://paste.debian.net/plain/12550916:20
ricotzbut still it is a crash ;)16:20
rodrigo_ah, if it's affecting users, yes, sure, let's upload it16:22
rodrigo_cyphermox, are you uploading it?16:22
cyphermoxrodrigo_: yeah, i just wanted to make sure this went upstream too16:24
cyphermoxsorry, I was reading stuff16:24
rodrigo_cyphermox, ok, so please do the symbols update from ricotz also, please :)16:24
cyphermoxricotz: yup, saw the symbols changes16:24
rodrigo_out for a bit, bbl16:31
=== MacSlow is now known as MacSlow|break
jbichacyphermox: here's a build of gobject-introspection if you want to test https://launchpad.net/~jbicha/+archive/dev/16:42
jbichabut I don't understand the invoke-rewrite stuff16:47
=== cypher is now known as czajkowski
charlie-tcaAny word on fixing lightdm to work for everything?17:56
charlie-tcaWe could start with bug 81960917:57
ubot2Launchpad bug 819609 in casper "Oneiric live CD boots to login screen" [High,Fix released] https://launchpad.net/bugs/81960917:57
charlie-tcaThat really isn't fixed for any flavor except Ubuntu17:57
charlie-tcaThe rest of us can not even force lightdm to auto-login17:57
cyphermoxcharlie-tca: would be https://bugs.launchpad.net/ubuntu/+source/lightdm/+bug/806408 no?18:00
ubot2Ubuntu bug 806408 in lightdm "After xubuntu upgrade or installation, default session on greeter must be xubuntu" [Medium,Confirmed]18:00
cyphermoxI'd maybe bring up the importance so that robertancell can look into it when he logs in18:01
charlie-tcaThat doesn't include the live sessions, does it?18:01
charlie-tcaSince we do not have Ubuntu session, we can use auto-login for live cd18:01
charlie-tcaSince we do not have Ubuntu session, we can *not* use auto-login for live cd18:01
cyphermoxcharlie-tca: well, it's refered to by stgraber from bug 819624 which is the bug refered to by 81960918:02
ubot2Launchpad bug 819624 in casper "casper doesn't configure autologin for lightdm properly" [High,Confirmed] https://launchpad.net/bugs/81962418:02
cyphermox(messy)18:02
charlie-tcayeah, messy and getting worse18:02
charlie-tcaI need a live session to auto-login when not calling Ubuntu, and I need to be able to default the session to Xubuntu18:03
charlie-tcaIt really comes back to bug 79975418:03
ubot2Launchpad bug 799754 in lightdm "Please let vendors easily provide their own config" [High,Triaged] https://launchpad.net/bugs/79975418:03
charlie-tcawhich needs to be resolvedd18:03
scott-work_charlie-tca: i just want to point out that you rock!  carry on ;)18:04
cyphermoxaye.18:04
charlie-tcaI need these fixed, and it seems it is getting nowhere fast18:05
cyphermoxI'd really like to avoid muddying the water and let robert handle this :/18:05
cyphermoxI can certainly prod him when I see him online though18:05
charlie-tcaPlease do so18:07
charlie-tcaunfortunately, it keeps getting put aside, since Ubuntu has it working for them18:07
charlie-tcaLubuntu, Kubuntu, Mythbuntu, Xubuntu, can not use lightdm18:07
charlie-tcaIt is worse for us now then gdm, and that's pretty bad18:08
jbichawell gdm mostly works18:08
charlie-tcaYes, with a lot of effort18:09
cyphermoxcharlie-tca: alright. I subscribed to the bugs so I'll know if they move... or not18:09
charlie-tcaIt doesn't really "just work"18:09
cyphermoxricotz: confirm you get g-c-c 1:3.1.4-0ubuntu5 and it fixed the hotspot thing for you?18:09
charlie-tcaThank you, cyphermox18:09
jbichaok, it works for Ubuntu though, don't know about how difficult it is for derivatives18:09
cyphermoxbrb18:10
charlie-tcawhich is the entire issue here18:10
jbicharodrigo_: especially with the extra icons we're adding to g-c-c, it would be nice if it expanded horizontally18:11
jbichaon smaller screens it does vertical scrolling when just becoming a wider window would work18:12
cyphermoxuh-oh, no power cycling at reboot?18:21
ricotzcyphermox, will do if it gets rolled out, but i confirmed it works with a local build ealier18:24
cyphermoxok18:25
cyphermoxricotz: well, it should get published any time now18:25
dupondjehttps://bugzilla.redhat.com/show_bug.cgi?id=590202 => seems like we have same bug here also ?18:30
ubot2bugzilla.redhat.com bug 590202 in NetworkManager "NM spontaneously reconnects, seemingly for no reason" [Medium,Assigned]18:31
cyphermoxdupondje: doubtful that this is really a bug... care to share your particuliar experiences?18:35
cyphermoxdupondje: if you file a bug in Launchpad (using ubuntu-bug), it's going to be easier for me to tell what might be acting up18:36
dupondjeAug  7 23:18:18 laptop-jl kernel: [ 4783.051128] wlan0: deauthenticating from c0:c1:c0:0b:13:34 by local choice (reason=3)18:37
cyphermoxactually, what I wrote there is better written as "doubtful that this is a bug in *NM* ;)18:37
dupondjewireless disconnects quite often :)18:37
cyphermoxdupondje: can you paste the whole syslog?18:37
cyphermoxthat message can happen for a variety of reasons :)18:37
dupondjewell cyphermox, its because ipv6 settings somewhere :)18:38
cyphermoxpaste in a pastebin of course18:38
dupondjehttp://paste.ubuntu.com/661321/18:40
cyphermoxanything else above that?18:42
dupondjenope, but i'll run NetworkManager in debug now18:42
cyphermoxah18:42
dupondjelets wait for a disconnect ^^18:42
cyphermoxsure.18:43
cyphermoxthis looks like it might be failing dhcp renewals, but it's hard to tell, all I see is the first point where the device gets failed18:43
cyphermoxI got some tweaks and fixes to make to the patch that handles ipv6/ipv4 in parallel, so maybe this will actually get fixed there, too18:44
=== Amaranthus is now known as Amaranth
dupondjeback :)18:52
dupondjestill running debug18:52
cyphermoxok18:53
dupondjebut it doesn't set my nameservers now somehow :s18:53
cyphermoxuh-oh18:53
cyphermoxdupondje: are you on oneiric?18:53
dupondjeye18:53
GunnarHjrodrigo_: Hello, did you see the updates of https://code.launchpad.net/~gunnarhj/gnome-control-center/langfix/+merge/68997 ? Not ready to be uploaded yet, since I'm waiting for pitti to sponsor the related l-s MP, but in the meantime we might try to agree on what's upstreamable. My current suggestion is simply the proposed 04_language_settings.patch.18:56
ricotzcyphermox, g-c-c works18:57
cyphermoxricotz: thanks19:03
dupondjeSeems like in DEBUG mode it doesn't have nameservers :p INFO works fine :s19:18
cyphermoxthat doesn't make sense19:19
cyphermoxit's the same code, minus the debug messages19:19
dupondjerunned 3x with DEBUG, no nameservers19:19
dupondjerun it again with INFO, have nameservers19:19
dupondjeweird19:19
cyphermoxso it's some race with getting your namservers19:20
cyphermoxdo you get them from the RA or from DHCPv6?19:20
dupondjeRA19:20
dupondjeNetworkManager[32387]: <debug> [1312830928.24751] [nm-ip6-manager.c:708] process_nduseropt_rdnss(): (wlan0): refreshing RA-provided nameserver 2001:1af8:4400:a020:1:: (expires in 70 seconds)19:21
cyphermox(and dah, we tested this thoroughly with multiple different cases of v6, v4 and dhcp)19:21
cyphermoxah19:21
cyphermoxdupondje: then I would suggest you open a bug in LP if you haven't already done so (ubuntu-bug network-manager) and attach a packet capture if you can19:21
cyphermoxa packet capture for the failure case, because we need to know if you receive the RA on time or not, and if NM is being unhappy with it for some reason)19:22
=== MacSlow|break is now known as MacSlow
=== _LibertyZero is now known as LibertyZero
=== pedro__ is now known as pedro_
dupondjecyphermox: got a disconnect: http://paste.ubuntu.com/661423/20:58
=== m_conley is now known as m_conley_away
bassognome shell is quite unstable :(23:45

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