/srv/irclogs.ubuntu.com/2012/05/25/#ubuntu-devel.txt

infinityxnox: If this boost merge is yours, why does it ScottK's name in the changelog? :P00:05
infinityxnox: Err, nevermind, I was looking at the changelog after patching with --dry-run.  I'm not awake.00:07
infinityLa la la00:07
* micahg pokes infinity with the -v stick00:23
cjohnstoncjwatson: fwiw, on the 31st we will delete the quantal db and the graphs will be started over, so you probably don't need to worry about the trend line for now00:38
=== funkyHat_ is now known as funkyHat
=== tumbleweed_ is now known as tumbleweed
=== Myrtti_ is now known as Myrtti
=== Myrtti is now known as Guest71494
=== panda is now known as Guest12542
=== zumbi is now known as Guest10379
cjwatsoncjohnston: ack, it was just a temporary tweak really01:18
cjohnstonthats fine.. just letting you know01:24
vibhavGood Morning01:38
scienteswho has the link to the unity hello world app03:18
scientesi remember it being in python03:18
micahgscientes: https://launchpad.net/hello-unity03:19
pittiGood morning04:08
=== robert_ancell_ is now known as robert_ancell
=== yofel_ is now known as yofel
=== larsdues1ng is now known as larsduesing
=== adrient is now known as finch
roshanHello, I was compiling a kernel in Ubuntu 12.04 as specified in one of the Ubuntu help pages.05:51
roshanA particular section says to enale "A particular section on the tutorial askes me to enable "Kernel module loader" , but I couldn't find any05:52
roshanWhere could I find that ?05:52
=== tkamppeter_ is now known as tkamppeter
larsduesinggood morning, is here anybody who is sort of firm with apport-hooks?06:37
pittilarsduesing: I should be :)06:38
vibhavheh, pitti is being humble06:39
larsduesingAh :)06:39
larsduesinghi pitti :)06:39
larsduesingarghs, we had pastebin here?06:39
larsduesingpitti: would you please have a look at that: http://pastebin.com/Y8D9u5yY06:41
vibhavlarsduesing: Which package is this hook for?06:42
larsduesingthe last two lines - how can I find out, if the specified reports are actually made, so that I only start masking for available reports?06:42
larsduesingaiccu06:42
pittilarsduesing: (looking in a bit, still reviewing an SRU)06:42
larsduesingoh, sorry, take your time :)06:42
pittilarsduesing: what do you mean with "if the reports are actually made"?06:46
larsduesingthe first report is only made, if a dpkg install fails06:48
vibhavIs it fine to SRU "Wishlist" bugs?06:48
dholbachgood morning06:49
larsduesinggood morning dholbach06:50
pittilarsduesing: I'm afraid I still don't understand the problem06:50
pittilarsduesing: you can /msg me in German if that helps06:50
dholbachhi larsduesing06:50
pittihey dholbach, wie gehts?06:51
vibhavhi06:52
vibhavdholbach: Is it fine to SRU "Wishlist" bugs?06:52
dholbachpitti, hey hey - alles gut und bei Dir?06:52
dholbachvibhav, no06:52
dholbachvibhav, either they are important or they're not SRUs :)06:52
vibhavoh ok06:52
dholbachhttps://wiki.ubuntu.com/StableReleaseUpdates has some decision making help06:52
=== smb` is now known as smb
pittidholbach: prima, danke!06:57
larsduesinghttps://code.launchpad.net/~lars.duesing/ubuntu/quantal/aiccu/aiccu-apport-fix <- so I should use "Propose for merging" now, shouldn't I?07:46
alkisgAny good tools for writing man pages? Is docbook + po4a a good choice, in order to have them translatable in launchpad etc?07:57
=== myrtti is now known as Myrtti
vibhavCould anybody nominate https://bugs.launchpad.net/ubuntu/+source/glom/+bug/736913 for maverick?08:18
ubottuLaunchpad bug 736913 in glom (Ubuntu) "libglom dev package missing dependency on libxml++ dev package" [Undecided,Fix released]08:18
vibhavoh wait, maverick's EOL08:19
vibhavCould somebody nominate it for natty?08:20
Davieyvibhav: Can you explain the impact?  It feels like a 'won't fix' for Natty IMO.08:22
pittiyes, I agree; this is absolutely unimportant for natty now08:23
vibhavDaviey: Ill agree with you then08:32
dupondjeSome question about virt-manager. Are all the current deps needed? Cause I don't know why I should need bridge-utils and ebtables for example if I just want to admin a remote server.08:35
seb128bdmurray, hey, those "Missing SRU information" bot comments, what do you check for?08:42
henrixpitti: hi, any chances of having a look at kernel linux-ec2 2.6.32-345.48?08:42
henrixpitti: we need to upload a new version into proposed, but this one is on our way08:43
pittihenrix: I'll try; please note that I formally left the SRU team08:53
henrixpitti: oh, sorry. didn't knew that. any idea about who shall i bother next time? :)08:54
pittihenrix: RAOF and SpamapS do regular SRUs08:56
henrixpitti: ack, thanks.08:56
onosendiI'm using mate with marco as my WM. How would I go about getting the geometry of a window when it is closed? It's x, y, w, h, etc?08:57
pittihenrix: promoted09:00
henrixpitti: great, thanks09:01
=== popey_ is now known as popey
larsduesingbtw, dumb question: does anybody use any other urgency as "low" in changelogs (maybe except security ones?)09:10
larsduesingand if yes: when it is advisable to do so?09:11
pittilarsduesing: the urgency is entirely ignored in ubuntu; it should be kept as "low" in general09:12
Davieycjwatson: I assume you saw mjg59's EFI related blog post?09:13
larsduesingok, thanks pitti09:16
larsduesingbtw, pitti: branch is done: https://code.launchpad.net/~lars.duesing/ubuntu/quantal/aiccu/aiccu-apport-fix/+merge/107351 (maybe it should be even a "security" fix, because of possible information disclosure...)09:17
seb128diwic, hey09:26
diwicseb128, pong09:39
seb128diwic, hey, how are you?09:39
diwicseb128, I'm okay, how are you?09:40
seb128diwic, asac reported bug #1004384 and I can confirm it with a bt device, could you advice on whetever pulseaudio debug infos would be useful for it?09:40
ubottuLaunchpad bug 1004384 in gnome-control-center (Ubuntu Precise) "[soundnua]: segfaults when selecting bluetooth input device" [High,New] https://launchpad.net/bugs/100438409:40
seb128diwic, I'm good thanks09:40
diwicseb128, hey, I fixed that an hour ago, as I found it myself :-)09:40
seb128asac, ^09:41
diwicI think so at least, judging from the title only09:41
seb128diwic, great!09:41
asacdiwic: have the patch?09:41
asaci can tryu ... hav ethe built tree here09:41
asacdebdiff preferrred09:41
diwicseb128, I spent much of yesterday rewriting the profile get/set stuff, tested it some more this morning, that's when I found the bug09:42
cjwatsonDaviey: have now ... if only I had time to do something useful about it09:42
cjwatsonpitti: the urgency isn't *entirely* ignored, it has a tiny effect on build scoring - but yeah, not much :)09:43
Davieycjwatson: yeah :(09:43
seb128diwic, do you have a patch,diff somewhere we could try?09:44
diwicseb128, asac, now pushed to http://bazaar.launchpad.net/~diwic/+junk/soundnua-lp984637/revision/3309:45
seb128diwic, thanks09:45
diwicseb128, note that r32 of that branch is my profile rewrite, which might need more testing before releasing, so just take the r33 patch09:45
cjwatsonNeed more installer team people who care about Macs09:46
cjwatsonAnd have any time at all09:46
diwicseb128, at least that makes input look like output codewise, and it fixes the crash, so assuming it's the right thing to do, might wanna double-check with cjcurran when he comes back09:46
xnoxcjwatson: i have an old intel 32bit macbook, but it can't run Lion and up.09:52
seb128diwic, could you add the diff to the bug report? that will be needed to SRU it, I will look to that09:57
seb128diwic, thanks!09:57
seb128asac, let me know if you try the patch09:57
pitticjwatson: oh, I didn't know that; thanks!10:00
diwicseb128, what type of diff would you like10:02
seb128diwic, standard unified diff over the current precise version (i.e over ronoc's vcs)10:03
seb128diwic, i.e something we can apply with patch to our current package10:03
diwicseb128, asac debdiff for precise-proposed added to bug10:13
Davieylool: https://launchpad.net/ubuntu/+source/btrfs-tools/0.19+20100601-3ubuntu2 still seems to be a delta with Debian... is this really warranted ?10:24
cjwatsonDaviey: xnox is working on that10:25
cjwatson(the merge I mean)10:25
xnoxDaviey: I did a merge it's sitting in the sponsor's queue. The delta is still warranted10:26
xnoxsee the merge bug discussions.10:26
xnoxbetween me and other contributors.10:26
Davieyah, neat10:26
xnoxDaviey: there is also mdadm and e2fsprogs bugfix/merge in the queue if you are up for it =))))10:27
Davieyxnox: i see you asked slangasek for a review... i wouldn't want to step on his toes or anything :P10:28
xnoxDaviey: I'm sure slangasek wouldn't mind if somebody else does review/upload10:29
asacseb128: diwic: ack. crash is gone10:29
seb128asac, things work as well?10:29
seb128asac, i.e can you configure your bluetooth?10:29
asacseb128: yes, i can select it and it seems it uses the bluetooth input afterwards ... let me try a few more things10:29
xnoxDaviey: btrfs-tools is on this list http://qa.ubuntuwire.org/oldmerges/10:29
asacseb128: all good ... all input and output channels work ... and i can use bluetooth and switch between high fidelity and telephony profile10:32
asacseb128: i tried the patch though by manually moving the return;10:33
asaci have no input of the complete branch10:33
seb128asac, diwic: thanks, I will test build it there and get it SRUed10:33
asacrock10:33
asac:)10:33
asaci have a good one with debug symbols now :)10:33
pittitseliot: hey Alberto, how are you?11:14
tseliotpitti: hey Martin, fine thanks, you?11:14
pittitseliot: you said you had a hybrid nvidia system, right? i. e. right now "ubuntu-drivers list" should show the nvidia drivers, but we are supposed to hide them, right?11:14
pittitseliot: I'm great, thanks! Looking forward to the long weekend (Monday is national holiday)11:14
tseliotpitti: yes, I have a hybrid system here. Well, we shouldn't always hide them11:15
pittithat's at least what the current jockey handler does11:15
pittiif "intel" is loaded by X.org, it doesn't offer the nivida driver11:16
tseliotpitti: lucky you (no nationaly holday here)11:16
pittiI'm going to add that special-case to u-d-common11:16
tseliotpitti: right, let keep doing that for now. There's a special case we need to cover but it can wait11:16
=== MacSlow is now known as MacSlow|lunch
Davieyxnox: comment added11:32
xnoxDaviey: thanks11:33
Davieyxnox: if you are happy with my comment, i'll just upload it here.. no need to bother fixing in bzr11:34
pittitseliot: are you currently using the intel or nvidia driver in X.org?11:34
pittitseliot: I pushed this to git now with a test case, but if you could pull, build, and double-check "ubuntu-drivers debug", that'd be nice11:34
xnoxDaviey: yes please upload. I don't like leading slashes either... and it was not me who introduced the link =))) I will comment on the bug report to the other contributor.11:35
tseliotpitti: the nvidia driver but only because I can select the graphics card from the BIOS. Not all laptops have this feature11:35
pittitseliot: runnign "PYTHONPATH ./ubuntu-drivers debug" will do fine11:35
tseliotpitti: sure11:35
pittitseliot: ah, cool11:35
xnoxDaviey: yes, please upload! it will unblock a couple of workitems for the btrfs spec ;-)11:35
pittitseliot: so in your current setup it should still show the nvidia driver11:35
tseliotpitti: would this work in precise?11:36
pittitseliot: it should11:37
pittitseliot: the tests currently fail, I'll fix that11:37
tseliotpitti: ok, as the laptop has precise on it11:38
Davieyxnox: uploaded11:38
* xnox 0/11:38
xnoxDaviey: last time I have checked CEPHs mailing list w.r.t. btrfs I saw loads of people with bug reports & errors/warnings from btrfs as well as degraded performance after some time11:40
pittitseliot: erk, git skills failure; really pushed now, with test case fix11:41
tseliotpitti: unmet build deps: python-aptdaemon.test (>= 0.43+bzr810-0ubuntu2~) gir1.2-packagekitglib-1.011:41
tseliotpitti: I should probably upgrade to quantal11:41
pittitseliot: that's fine11:41
xnoxDaviey: do you have interest in btrfs for particular workloads/services?11:41
pittitseliot: just run it out of the source tree11:41
pittitseliot: PYTHONPATH ./ubuntu-drivers debug11:41
tseliotpitti: ah, ok11:41
Davieyxnox: tentative :)11:43
loolDaviey: the delta that you point at was required to have btrfs work ok with upstart + mountall11:44
loolDaviey: Back then I was playing with ARM btrfs images, and was getting at least one of the two bugs fixed in this upload; the other one had enough comments supporting the proposed fix that warranted the patch to be included IIRC11:45
loolDaviey: maybe upstream/the Debian package have been changed similarly in the mean time though11:46
xnoxDaviey: there were not many folks from server/cloud at the https://blueprints.launchpad.net/ubuntu/+spec/foundations-q-btrfs-requirements11:46
pittitseliot: so it should list nvidia if you are on nvidia, and hide it if you are on intel11:46
xnoxDaviey: any comments about btrfs in Ubuntu are welcome11:46
pittitseliot: the test cases reproduce this fairly well, so I'm rather sure it's going to work, but oh well, theory and practice ... :-)11:47
tseliotpitti: it seems to work. Let me get you the output11:47
tseliotpitti: here you go: http://paste.ubuntu.com/1006259/11:49
tseliotvery nice, BTW11:50
Davieyxnox: wilco11:51
* xnox ?11:51
* xnox Wilco is an American alternative rock band based in Chicago, Illinois. 11:51
Davieywill comply. :)11:52
xnoxah =) ok.11:52
pittitseliot: ok, that's with nivida?11:52
tseliotpitti: yep11:52
pittitseliot: good, thanks11:52
pittiI'll upload that and then wave good bye11:52
tseliotpitti: we have two more cases: 1) intel on and nvidia off; 2) both intel and nvidia on11:53
pittitseliot: I thought 2) wasn't possible with current X11:53
tseliotpitti: well, both cards can be on but only one card can drive the display11:54
tseliotpitti: i.e. only one driver can be used11:54
* tseliot -> lunch11:55
pittisplendid11:55
tseliot;)11:55
pittitseliot: so, see you next Tuesday, and enjoy the weekend!11:55
tseliotpitti: thanks, you too!11:55
pittitseliot: I'll add apport hook (with u-drivers debug) etc. next week11:55
tseliotgreat :)11:55
asacdiwic: noticed something odd ... i have two microphone input devices (webcam + built in) ... i need to set both explicitely to mute to really mute the input level in the sound preference area11:55
diwicasac, uhm, can you tell me what you did in more detail?12:01
diwicasac, how were you recording e g?12:01
asacdiwic: i am not recording ... i am just looking at the control center. what i see is this:12:02
asaci make noise and the input indicator goes up (shows signal)12:02
asaci check the "mute" box12:02
asacstill signal on noise12:02
asaci go to the other input device and check the mute box and signal is gone12:03
asacdiwic: makes sense?12:03
diwicasac, right, sounds like the input level is taking its input from the wrong card essentially12:04
asacdiwic: more background: i have a thinkpad in a dock; a built-in mic, a usb webcam mic and a bluetooth mic12:04
asacdiwic: this problem seems to only affect the built in and usb mic12:04
asacdiwic: well, but it feels it takes it from both, because it doesnt matter which card i mute12:04
diwicasac, can you verify which mic it seems to take its input from12:04
diwicasac, by poking on the actual mic12:04
asacdiwic: no... as i said: doesnt matter in which order12:04
asacok12:04
asaci can try12:04
asacone sec12:05
asacdiwic: yeah. always from the usb mic12:05
asacdiwic: but if i have both mute checkboxes selected it gets disabled :)12:06
asacthats the fun part12:06
asacmaybe something caused by dock rerouting magic?12:07
diwicasac, so if you have the usb mic selected it takes the signal from the usb mic, but muting *only* the usb mic would not mute the signal?12:08
asacdiwic: restarting fixed it i think :/12:09
diwicasac, ok, let cjcurran know if it's reproducable12:09
asacyeah. how weird :)12:10
mr_pouitcjwatson: quantal/i386 daily builds for Xubuntu fail because cdrom/non-pae/initrd.gz doesn't exist (the non-pae kernel was dropped I think). Could you do some magic again please? (do you prefer that I file a bug report?) Thanks. ;-)12:12
cjwatsonmr_pouit: Oh yes, I can just fix that now12:13
cjwatsonThought I'd caught everything, evidently not12:13
cjwatsonmr_pouit: done12:16
mr_pouitcjwatson: awesome, thanks!12:16
=== _salem is now known as salem_
=== MacSlow|lunch is now known as MacSlow
=== elmo_ is now known as elmo
bdmurrayseb128: the bug matches my expectations now although I expanded the test case section a bit13:21
seb128bdmurray, thanks13:22
seb128SpamapS, hey, is there any chance you could do some SRU reviews today? ;-)13:24
psusicjwatson: there are a few bug reports that I think are caused by a size increase in the grub2 core image that makes it too large to fit in 62 sectors when including raid and lvm modules.  Is this just a too bad, you need more room, in which case it bears mentioning in the release notes?13:25
zulmterry: ping dwarves builds fine now13:28
mterryzul, was just about to look at it13:28
cjwatsonpsusi: Sometimes it's fixable, it depends ... I generally think it's a bug although not necessarily a straightforward one13:29
cjwatsonAlthough I'd advise people to put their first partition at 1MiB rather than at sector 63 these days; much more efficient on modern disks, and no less efficient on disks since about oh 1985 or something13:30
cjwatsonOf course I realise it's tedious to move partitions13:30
psusicjwatson: indeed... but it seems some people are upgrading old setups from 10.04 and still have partitions starting at sector 6313:30
psusicjwatson: it seems that the image with both raid and lvm built in is now something like 200 bytes too large to fit in that space13:31
cjwatsonThis is probably fixable but I'm afraid I don't have time for the next month or two13:34
apwjodh, is there a simple way to tell from upstart --verbose output which thing we are waiting on13:48
apwjodh, for a han13:48
apwjodh, for a hanging boot13:48
jodhapw: try http://upstart.ubuntu.com/cookbook/#establish-blocking-job13:48
jodhapw: can you send me a log when booting with --debug ?13:50
apwjodh, the end user is talking to us on #ubuntu-kernel now13:50
apwjodh, http://sprunge.us/iGaU13:54
apwjodh, for me i think its wiating on mountall ... perhaps mismatched UUID13:56
tjaaltonis it ok to add build-depends on a SRU release, to fix obvious omissions in the packaging? :)14:00
seb128tjaalton, yes14:00
seb128tjaalton, well adding build-depends is ok, turning on feature which were not is subject at review and rational14:01
tjaaltonseb128: yeah, of course. minor features but something that should've been enabled.. and have been on other distros14:01
seb128tjaalton, well, the fact that they should doesn't change that they didn't get tested so it's not a trivial wave in change14:02
tjaaltonseb128: it'll get tested before the push14:03
seb128tjaalton, right, well anyway I was just saying, adding build-depends is not something against the rules ... are those build-depends in the same pocket btw?14:04
tjaaltonseb128: the package is in universe, so no problem there14:04
tjaaltonpackage which is being sru'd14:05
seb128ok14:05
stgrabermvo: hey there, I was talking with jibel about getting automated upgrade testing up and running for Quantal. What's still required to get do-release-upgrade happy?14:17
mvostgraber: the release-upgrader is currently defnct due to the py3 porting in trunk once that is resolved there is nothing blocking it14:18
stgrabermvo: ok. Who's doing the py3 port?14:20
cjwatsonThat would be me14:21
SpamapSseb128: I did some w/ bdmurray who is training to be an SRU team member yesterday. Between he and I, and slangasek, we didn't get your SRU's reviewed?14:21
cjwatsonI'll try to get that sorted out - mvo, is there an easy way I can reproduce the problems?14:21
mvostgraber: I meant to look at it this week, but couldn't quite find the time14:21
=== dannf` is now known as dannf
mvocjwatson: yeah, cd DistUpgrade; sudo ./dist-uprade.py will fail14:21
cjwatsonOK, thanks14:23
bdmurraySpamapS: we did review one of his and he's updated it now and it looks good to me14:24
SpamapSbdmurray: bug #?14:26
apwcjwatson, i just installed a kernel and extras together and got an unexpected result: http://paste.ubuntu.com/1006472/ see the last line there14:26
cjwatsonapw: Looks like a bug in your postinst14:27
bdmurraySpamapS: bug 980317 for gnome-control-center14:28
cjwatsonapw: Though actually, maybe not, try 'sudo dpkg --configure -a'14:28
ubottuLaunchpad bug 980317 in gnome-control-center (Ubuntu Precise) "details -> overview disk size includes network mounts" [Low,Fix committed] https://launchpad.net/bugs/98031714:28
cjwatsonMight just be a pending trigger14:28
apwnothing happened14:28
cjwatsoncase "$DPKG_MAINTSCRIPT_PACKAGE" in14:29
cjwatsonlinux-image-*)14:29
cjwatsonhm, so maybe update-initramfs needs to exclude linux-image-extra-* from that since that behaves differently14:29
cjwatsonI don't suppose I can convince you to not call it something matching linux-image-*? :-P14:30
apwcjwatson, yeah i was seeing the same lines ...14:30
apwcjwatson, well i guess we can call it anything you like :)14:30
cjwatsonIf it didn't match that wildcard you wouldn't have a probem14:30
cjwatson*problem14:30
SpamapSbdmurray: so, +1 to accept that upload?14:30
apwcjwatson, so either we change the name or fix the trigger ... ok thanks14:31
apwcjwatson, its possible we should be calling the hooks later anyhow as we carry half of the kernel14:31
=== mbarnett` is now known as mbarnett
bdmurraySpamapS: yes14:32
SpamapSbdmurray: I concur. Pulling lever... you want to run sru-accept?14:32
bdmurraySpamapS: yes will do14:33
cjwatsonapw: Maybe, though I'd have thought the ordinary trigger would be sufficient14:33
apwcjwatson, well let me think, iirc the generic trigger is unperameterised and triggers a rebuild of the running kerenl initramfs, and this is specificall not for the running one in the majority of cases14:34
SpamapSbdmurray: did you notice bug 1004384 was also still missing SRU headers?14:35
ubottuLaunchpad bug 1004384 in gnome-control-center (Ubuntu Precise) "[soundnua]: segfaults when selecting bluetooth input device" [High,In progress] https://launchpad.net/bugs/100438414:35
cjwatsonOh, true, you want a specific version here14:35
cjwatsonSo maybe you might as well just use your initramfs hook and then it would all be well14:35
cjwatsonSince indeed the trigger won't do the right thing14:35
SpamapSbdmurray: I accepted anyway, but that one probably needs impact/regpotential/etc. as well14:36
SpamapSseb128: ^^14:36
apwcjwatson, ok will work on that, i presume i just need to run the kernel postint.d hooks as normal14:36
SpamapSI think its probably time we sent an email out to ubuntu-devel about SRU headers14:36
SpamapStoo many people are uploading without following that bit of the process.14:36
* SpamapS goes to make pbreakfast14:37
SpamapSpbreakfast - a clean chrooted breakfast14:37
cjwatsonapw: I think so14:39
apwcjwatson, i suppose this will mean we will rebuild the initramfs twice for every kernel which isn't ideal14:41
apwcjwatson, i suspect to avoid that properly we will want to have the initramfs trigger perameterisable14:41
cjwatsonYeah, I'm having a slightly difficult time thinking about how to avoid that14:41
cjwatsonWell, it's technically possible; we could queue up the initramfs versions to rebuild in a file in /var and have the trigger process that14:42
cjwatsonThat kind of thing is the usual way to pass data through the trigger interface14:42
apwcjwatson, would it be reasonable to generate a list somewhere and ... yeah that14:42
apwcjwatson, ok so for now i will bodge it to make it do both and take an action to make it perameterisable; where is appropraite in /var, is there a defined spot ?14:43
cjwatsonI wonder if there are things that currently assume that some kind of initramfs exists immediately after the kernel is configured14:43
cjwatsonHopefully not ...14:43
cjwatson/var/lib/initramfs-tools/ I guess14:43
apwok ... thanks ... sigh why is nothing every easy14:43
cjwatsonIt should be per-package in any event14:43
apwcjwatson, i suspect if we delay initramfs build to the end, then grub configure will have to be as well, so there is in initramfs for it to link to14:46
apwthere is bound to be untold fallout ...14:46
apwbut anyhow i can do the framekworks and we can iterate on the issue14:46
cjwatsonapw: Mm, that's a bit scary14:48
cjwatsonapw: The trigger could put an empty initramfs in place, maybe14:48
cjwatsonJust so there's *something*14:48
apwcjwatson, yep ... thats what i am thinking ... yeah that might work, ick14:48
=== tsdgeos_ is now known as tsdgeos
seb128SpamapS, bdmurray: hey, some GNOME ones are sitting in the queue for some days, no hurry but they are mostly trivial ones and fix reported bugs ... if they are stalled because the "paperwork" is insuffisant to your taste please state it on the bugs rather than just keeping them sitting in the queue ;-)15:02
seb128cjwatson, slangasek: is there a way to turn off multiarch for dh compat 9 level packages?15:33
slangasekseb128: pass -- --libdir15:33
slangasekto dh_auto_configure15:33
seb128tkamppeter, ^ or that15:33
slangasekwhy are you turning it off, though? :)15:34
seb128slangasek, I suggested to downgrade compat to 8 but I figured there would be a better way :p15:34
seb128slangasek, because it's a cups driver and cups is not multiarched15:34
slangasekah15:34
slangasekgood reason :)15:34
seb128slangasek, i.e it doesn't look in the multiarch dir15:34
slangasekyeah, this is probably a case where it doesn't make sense to ever multiarch15:35
slangasekalso, it's probably --libexecdir wanted here, not --libdir, right?15:35
slangasek(helper executables - not DSOs)15:35
seb128right15:37
seb128slangasek, thanks15:37
slangasekI've actually wondered if pointing --libexecdir at the multiarch dir by default was a mistake... too late now to change it for compat 9 though :/15:37
seb128bdmurray, why is your bot randomly going through bugs and deleting some of the apt log files?15:43
bdmurrayseb128: its not random and its because of an update-manager cve15:44
seb128bdmurray, ok, thanks, I was rather curious of the reason ;-)15:46
bdmurrayseb128: sure, no problem15:47
tkamppeterslangasek, seb128, I have already tried "dh_auto_configure -- --libexecdir=/usr/lib", "dh_auto_configure -- --libdir=/usr/lib", "dh_auto_configure --libdir", "dh_auto_configure --libexecdir", all to no avail.15:47
slangasektkamppeter: can you link me the package?15:48
=== nhandler_ is now known as nhandler
cjwatsonThe third and fourth of those are unambiguously wrong15:48
cjwatson(I don't have an answer for you, but you might as well not waste your time on options that definitely won't work)15:49
cr3Is there a recommended PPA that might contain more python3 packages for Precise, like python3-openssl if such a thing exists?15:59
SpamapSseb128: we actually have been stating in the bugs when the paperwork is insufficient. ;) There have been well over 100 packages uploaded to precise-proposed in the last 2 days.. so please just be patient.15:59
seb128SpamapS, 85 of those being kde-l10n ones? ;-)16:00
SpamapSseb128: probably :)16:00
seb128SpamapS, but yeah, sure no hurry, though I blame a bit KDE for swamping the queue like that :p16:01
SpamapSseb128: agreed, KDE always seems to make a lot of work. Anyway, make sure you have all your impact/testcase/regression potential filled out, and we'll get through your bugs faster. :)16:05
seb128SpamapS, noted, thanks for the work!16:05
tkamppeterslangasek, it is ptouch-driver. Simply "apt-get source" it. Then try to add an override_dh_auto_configure to it.16:17
cjwatsoninfinity,cyphermox: Debian seems to have merged the wpasupplicant and hostapd source packages into a single wpa source package.  You two have outstanding Ubuntu modifications to the prior source packages; do you think one or both of you could merge those into the new one?16:25
cjwatsonIt's showing up for auto-sync, which I'm holding off on due to the Ubuntu changes.16:26
cyphermoxcjwatson: sure.16:26
roshanHello16:27
cjwatsoncyphermox: thanks16:27
cyphermoxinfinity: I'll take your hostapd changes as well16:27
cjwatsonslangasek: You're TIL for autoconf; do you think you could merge it, which will allow me to process the autoconf-nonfree removal?16:28
roshanI would like to build a custom Ubuntu from the minimal Ubuntu vrsion.16:28
roshanI plan to use the 11.10 version16:28
roshanMy question is, how to I dd the basic necessary packages like networking, desktop, apt-get , and all the basic stuff ?16:28
roshan*add16:29
sladenroshan: look up "Ubuntu Customisation Kit"16:32
roshanthank you. Let me check that16:33
infinitycyphermox: Sure, it was a small change.  Thanks.16:34
* infinity wasn't even aware those were from the same upstream, though it makes some sense.16:34
roshanDoes UCK support command line ?16:34
gyotoit22http://www.reddit.com/r/nsfwhot/comments/u0upg/young_teen_brutal_fucked_by_two_big_cocks/16:35
cjwatsonapologies for that intrusion16:37
cyphermoxinfinity: wouldn't the migration /lib/init/rw -> /run migration check work as well with just adding || true?16:37
cyphermoxI mean, so that postinst doesn't bomb if it's trying to move a file to the same destination as it already was16:38
infinitycyphermox: Possibly, I don't remember the original code (or my modifications, for that matter).  I don't usually like to mask over errors in symlink/directory issues with || true, though, as there are corner cases that will give you nested directories and such that you really weren't expecting.16:39
infinitycyphermox: This may not have had one of those cases.  Like I said, I don't remember. :P16:40
cyphermoxwell, anyway I see this would probably avoid error messages anyway16:40
cyphermoxgah, I'm repeating words today repeating16:40
infinityHeh.16:40
infinity(Heh.)16:41
cr3barry: if I'm wondering whether I can rely on some python3 package being available in quantal, like python-gst for example, where should I look for an answer instead of pinging you each time? :)16:42
tumbleweedxnox: so, what's happening with boost-source1.49?16:43
xnoxtumbleweed: I am working on it. Didn't finish it yesterday evening.16:43
xnoxtumbleweed: and I wanted to unbreak the rest of boost.16:43
tumbleweedcool. vtk is currently broken and waiting on boost to be unbroken first :)16:44
xnoxtumbleweed: I am "in charge" of boost1.49 transition =) and I hope to get it finished before alpha1. Note some of the build-failures are due to gcc4.7 and affect debian as well.16:44
xnoxtumbleweed: awesome =) I bet everyone loves broken vtk =)16:44
tumbleweednaah, that's it's steady state16:44
* xnox "we will not break quantal" meh =)16:44
barrycr3: if you know the python 2 version of the package, just search for python3-<whatever>16:45
roshanDoes Ubuntu Customiation kit allows us to change the look of the Ubuntu ?16:48
cr3barry: of course but, if it's not there now, is there a roadmap for quantal?16:48
sladenroshan: you can change anything you want in Ubuntu.  The amount of customisation you'll get  depends on the amount of effort you're able/prepared to put into it16:50
barrycr3: only if it's required for the desktop cd.  if not, work isn't scheduled, but i am always happy to review ports16:50
roshanokay16:50
infinitycr3 / barry: In the case of python-gst it's on the desktop CD (apt-cache show python-gst0.10 | grep ^Task), so that answers your question.16:52
barryinfinity, cr3 it's in the spreadsheet: https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0AiT4gOXSkmapdFA1anRkWERsaXgtWnllUG9QWXhDVWc&pli=1#gid=016:53
infinity(That too)16:53
cr3barry, infinity: awesome, thanks folks!16:53
barryinfinity, cr3 the suggestion though is to port anything on the desktop to pygi and get rid of the gst dependency :)16:53
xnoxbarry: and no, the tracker update is not trigger from the spreadsheet (yet)16:53
ScottKxnox: Does that mean you're working on boost-mpi-source1.49?16:54
* xnox as you probably figured, due to commits from you16:54
barryxnox: that's okay, i figured out how to update the tracker :)16:54
xnoxScottK: unless you really want to =)16:54
ScottKxnox: No.  I'm glad you're doing it.  I took a first stab at it, it didn't work, and I've not had time to get back to it.16:54
xnoxbarry: now that I have access to people.c.c I will set up a cron job there.16:54
ScottKxnox: I have done the split before though, so if you have questions, I'm glad to assist if needed.16:55
xnoxScottK: ...thanks =)16:55
barryxnox: aweome, thanks16:55
cr3barry: I'll try to rewrite scripts depending on gst to use pygi instead of just migrating them to python3, thanks for the heads up16:55
xnoxScottK: ok, thank you.16:55
* xnox is this the time when timezones meet16:55
barrycr3: fantastic.  i'm always happy to review16:56
cr3barry: I'm equally unfamiliar with both, so should be the same amount of work :)16:56
infinitybarry: How does pygi relate to gst?  Or is it just that you can use pygi to talk to anything with gi bindings, thus elminating the need to have python bindings for every library?16:59
dobeyinfinity: the latter sort of17:01
barryi am not a pygi expert ;)17:02
=== nxvl_ is now known as nxvl
dobeythough i don't think gst is really usable from gi yet. though maybe gst 1.0 fixes that17:02
dobeyhi barry17:02
slangasekcjwatson: yep, will do... sorry, I'd so far only looked at the list that launchpad's +localpackagediffs was showing my name for, and that list is... small17:09
barrydobey: hi17:10
dobeybarry: so i am at the brink of having python-oauth 1.0.1 working on python317:11
dobeybarry: at least, it does work, except for 1 tiny problem which may be some insanity in the base http server stuff changing in 317:12
slangasektkamppeter: this works for me:17:12
slangasekoverride_dh_auto_configure:17:12
slangasek        dh_auto_configure -- --libdir=/usr/lib17:12
ScottKdobey: By definition, if it's different in Python 3, the insanity was in Python, not Python 3.  You've got it backwards.17:13
ScottK;-)17:13
dobeyScottK: my standards dictate that if it's python, it is insanity, regardless of version. ;)17:14
dobeyalso, python-oauth is a pretty good example of things to not do in python17:14
ScottKFair enough, but you'll make barry sad.17:14
dobeyon the contrary, i think making python-oauth work in python3 will make barry quite happy :)17:15
ScottKTrue.17:15
=== deryck is now known as deryck[lunch]
=== allison is now known as wendar
cjwatsonIf you want to print Unicode text to sys.stdout, the insanity required is merely different in Python 3 :-/17:17
cjwatson(And not as bad, I'll grant)17:17
=== wendar is now known as allison
cjwatsonsys.stdout = io.TextIOWrapper(sys.stdout.detach(), encoding="UTF-8", line_buffering=True)  # is the least awful approach I've found that's safe against LC_ALL=C, although not exactly pleasant17:18
=== allison is now known as wendar
dobeybarry: are you in the debian python module maintainers team?17:21
slangasekcjwatson: heh, can that be documented somewhere?17:21
slangasekbecause that's going to come in handy in the future I'm sure :)17:21
ScottKdobey: He is.17:27
cjwatsonbarry: ^- perhaps you could document that on Python/3, or some alternative that's more right17:27
dobeyScottK: thanks17:27
cjwatsonthis is basically what you need if you have something that might ever run in LC_ALL=C (i.e. the default locale) and processes non-ASCII text17:28
cjwatsonwell, at least in a pipeline-based way17:28
cjwatsonif it opens files and writes to those it's fine; Python makes an illogical distinction here17:28
barrycjwatson: good idea17:28
barrydobey: re python-oauth - that's fantastic, can i take a look at a branch?17:29
dobeybarry: i'm working on getting a branch pushed17:30
cjwatsonOh, wait, that's not true about files17:30
cjwatsonBut at least with files you can open with an explicit encoding straight off17:30
cjwatsonStill a bit of a gotcha though; I wonder if we need to do more explicit encoding= :-/17:31
dobeybarry: do i need to do all the dh_overrride and python%-foo stuff to make python2 and 3 installs work, or does --with python2,python3 handle that automagically?17:31
cjwatson(Which is evil in its own way, though)17:31
cjwatsonIn ubiquity I perpetrated a hack that forces the locale to C.UTF-8 on startup if it's C; but that won't be appropriate for all applications17:31
barrydobey: for now, you'll need the overrides for py3, since dh doesn't handle that automatically17:32
dobeyok17:32
barrydobey: http://wiki.debian.org/Python/LibraryStyleGuide17:32
tumbleweed--with doesn't ever affect dh_auto_* AFAIK17:33
barrycjwatson: maybe you should subscribe to https://wiki.ubuntu.com/Python/3 ?17:34
dobeybarry: yep, i'm looking at that. :)17:35
cjwatsonSure, done17:35
dobeydpkg-deb: building package `python-oauth' in `../python-oauth_1.0.1-4_all.deb'.18:00
dobeydpkg-deb: building package `python3-oauth' in `../python3-oauth_1.0.1-4_all.deb'18:00
dobeybarry: ^^ :) pushing the branch now18:00
barrydobey: awesome.  so the main development branch will be the ubuntu branch?18:01
barry(dobey: that's fine, as long as vcs-bzr is up-to-date i think)18:01
dobeybarry: oh no, i just made a patch18:02
barrydobey: k18:02
dobeybarry: was hoping it would just be uploaded into debian that way :)18:02
barrydobey: how are you getting it into debian?  bug report?18:03
dobeybarry: of course, this is completely untested in the Real World (TM) so far.18:03
dobeybarry: i haven't filed a bug yet, but was hoping you could upload it there18:03
dobeybarry: anyway, it's at lp:~dobey/ubuntu/quantal/python-oauth/python3-packages18:03
barrydobey: actually, i'm not a dd.  i can upload to ubuntu though18:03
tumbleweedit's usually best to give ports to upstreams first18:04
dobeyoh, ScottK lied then :)18:04
barrytumbleweed: upstream is long abandoned18:04
tumbleweedah18:04
dobeyand i am not trying to become the upstream18:04
barryno no, me neither :)18:04
=== dendro-afk is now known as dendrobates
barryi'll test it and file a debian bug18:05
ScottKdobey: No.  I didn't.  Being a member of DPMT doesn't imply being a DD.  You can even join yoursef.18:05
dobeyoh18:05
ScottKbarry not being a DD is a bug in the system that he should, however, fix.18:05
barryScottK: i am apparently "in the queue" for an am, but i think the queue is deep18:05
ScottKbarry: If it's in DPMT, you can just commit to svn.18:06
dobeyhaving a team that maintains a package, but for which people in taht team can't upload to, seems silly18:06
barryScottK: good point, i'll check that18:06
ScottKbarry: OK.  That's good to know.18:06
dobeyah, well if can commit to the svn that's a start i guess18:06
ScottKdobey: Having people work as a team and having them all be able to commit to the VCS aids sponsorship and getting stuff done.18:07
dobeynow i wonder if i didn't totally break python-oauth with that patch18:07
barrydobey, ScottK http://anonscm.debian.org/viewvc/python-modules/packages/python-oauth/18:09
ScottKYou might want to email TANIGUCHI Takaki as well.18:10
dobeyoh, fail18:10
dobeyah, of course. i'm an idiot18:11
dobeybarry: forgot the install files. but just pushed them to the branch18:16
slangasektkamppeter: did that override_dh_auto_configure rule help?18:17
tkamppeterslangasek, yes, that works, thank you very much.18:22
=== deryck[lunch] is now known as deryck
=== dendrobates is now known as dendro-afk
dobeybarry: looks like at least 1 test fails in ubuntu-sso-client as a result of my patch to python-oauth18:34
apwcan someone remind me where an installed package's postrm is on disk on the machine?18:37
dobeyapw: in /var/lib/dpkg somehwere18:38
=== dendro-afk is now known as dendrobates
dobeybarry: hrmm, but this may actually be a bug in ubuntu-sso-client. the test has a unicode char in the url's path18:43
barrydobey: nice18:45
tkamppeterslangasek, fixed ptouch-driver uploaded to Q and also uploaded for P-proposed.18:47
slangasektkamppeter: excellent :)18:51
dobeybarry: hrmm. though it looks like after fixing that test, it still fails in the oauth signing, even though the URL looks correct outside of oauth.py. so will have to look at that some more19:00
barrydobey: cool, no worries19:00
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
=== dendrobates is now known as dendro-afk
mterrymvo, hello!  What does "update-manager --no-update" do?  I see where it affects the code (calling "self.list.update(self.cache)" or not), but I don't understand what affect that ends up having19:41
vsingh165anyone here familiar with bug #822993? https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/82299319:57
ubottuLaunchpad bug 822993 in nautilus (Ubuntu) "Applying Permissions to enclosed files in Nautilus' Folder Properties window doesn't work" [Low,Triaged]19:57
vsingh165I'm a new developer and still figuring this out.19:57
vsingh165basically, there's a button in nautilus file properties that allows you to apply a folder's permissions to its contents.  but the button does absolutely nothing19:58
vsingh165it was originally found in 11.10, but I've verified that it still happens in 12.0419:58
=== dendro-afk is now known as dendrobates
mterrympt, on the software updates spec (https://wiki.ubuntu.com/SoftwareUpdates), most of the mockups have the ubuntu logo.  But the last two mockups use a package icon.  Is that intentional?20:14
mptmterry, I'm planning to change the others to match those last two20:14
mterrympt, OK, so all should have package icons20:14
mpt(the application icon, not a package icon per se)20:14
mterrympt, sure20:15
mptthanks for checking :-)20:15
ahasenackhi, can someone please review and if all good upload landscape-client to quantal? This is the bug, and I attached a branch to it based on the current quantal branch (ubuntu:landscape-client):21:43
ahasenackhttps://bugs.launchpad.net/landscape-client/+bug/100467821:43
ubottuLaunchpad bug 1004678 in Landscape Client "Release 12.05" [Undecided,Fix committed]21:43
ahasenackthanks21:44
* ahasenack bbl, walk the dog21:44
=== dendrobates is now known as dendro-afk
=== salem_ is now known as _salem
andreas__back22:17
=== andreas__ is now known as ahasenack
ahasenackback :)22:18
bkerensacyphermox: Any more advice to resolve Bug #97206323:25
ubottuLaunchpad bug 972063 in bluez (Ubuntu) "Bluetooth Headset pairs but does not show up in Sound Settings profile" [Medium,Confirmed] https://launchpad.net/bugs/97206323:25
infinityNCommander: Uhm, dude.  debian/rules clean? :P23:52
infinityNCommander: Your flash-kernel diff includes debian/flash-kernel (and then on the next upload, you patched the contents of said directory!)23:53
cjwatsonDon't you have to work pretty hard to do that?  dpkg-buildpackage -S runs clean.23:54
cjwatson(Also, *always* read the debdiff before uploading.)23:54

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