/srv/irclogs.ubuntu.com/2015/01/28/#ubuntu-devel.txt

=== RAOF is now known as tmpRAOF
=== _salem is now known as salem_
=== salem_ is now known as _salem
=== timrc_ is now known as timrc
=== rcjenn is now known as rcj
rbasakpitti: I think everything is built now. Please could you retest everything listed as a regression in http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html#php5? I just tried myself, but I don't seem to be able to reach anything via batuan.01:28
=== StevenK_ is now known as StevenK
ari-tczewcyphermox: Well, I think I'm ready with libcapi20-3 transition. I did a rebuild on all reverse depends. I'll look later whether are they all build fine.02:34
=== ssweeny` is now known as ssweeny
infinityrbasak: Still around?03:37
infinityrbasak: It's awfully late, so I'm going to assume no.03:37
infinityrbasak: Anyhow, I'm going to assume you tested this spamassassin change and it works, but if it had been me, I would have also commented out the line in 50_scores.cf03:38
infinityrbasak: All accepted.  Please verify ASAP, happy to fasttrack this once you sure it both solves the bug and doesn't otherwise break anything.03:41
rbasakinfinity: thanks. I'll verify after sleep. I did consider commenting out 50_scores.cf, but if I did that I'd also want to do the translations, and tested that it works across all releases without any warnings or anything by just doing the definition.03:58
rbasakI suppose in part it's because testing across all releases is tedious, so I started with the minimal, and that worked, so I didn't feel the need to go further.03:59
=== wolsen_ is now known as wolsen
pittiGood morning06:44
pittirbasak: yes, my retry button finger was busy last night and this morning :)06:45
pittirbasak: there are still a few regressions; e. g. jmespath.php looks like a victim of the same patch regression that hit systemd and glibc: debian bug 77625707:07
ubottuDebian bug 776257 in patch "Fails to apply patch with dangling symlink" [Serious,Open] http://bugs.debian.org/77625707:08
=== geser_ is now known as geser
dholbachgood morning08:09
=== dupondje_ is now known as dupondje
LocutusOfBorg1hi dholbach !08:45
LocutusOfBorg1hi all08:48
dholbachhi LocutusOfBorg108:49
mlankhorstcan I get libepoxy promoted to main for trusty?09:00
mlankhorsthttps://bugs.launchpad.net/ubuntu/+source/libepoxy/+bug/1342605 was the bug for utopic, but I need it in main for xorg-server-lts-utopic09:01
ubottuLaunchpad bug 1342605 in libepoxy (Ubuntu) "[MIR] libepoxy" [High,Fix released]09:01
mitya57pitti, jibel: vivid-adt-software-properties fails since Jan 20th, but for some reason it blocks pyqt5 which got built only yesterday09:03
mitya57Please do something :)09:03
pittiwell, best would be to fix software-properties :)09:05
pittiI'll look at it later, it also blocks gtk+3.009:05
=== mthaddon` is now known as mthaddon
=== jhenke_ is now known as jhenke
jibelpitti, it seems it's failing because the path to dbus-launch is wrong09:53
pittijibel: oh, indeed10:01
mitya57pitti, jibel: it probably needs to add dbus-x11 to Depends10:03
pittimitya57: yeah; running test locally now, will upload if that works10:09
=== dosaboy_ is now known as dosaboy
cjwatsonrbasak: You need to switch to the new company VPN; the old batuan one is AFAICT essentially dead/broken as of the end of last year.10:23
rbasakcjwatson: I thought it might be that. Thanks!10:24
rbasakpitti: Thanks - I'll follow up.10:24
mlankhorstdoko__: hm mesa 10.4.2 fails to build against llvm 3.610:25
pittimitya57, jibel: software-properties uploaded10:25
mitya57thanks :)10:26
hallynarges: hi, on your next sru stint could you please take a look at 1404388?  its fix has been baking for awhile...10:43
=== Chipaca` is now known as Chipaca
mlankhorstdoko__: I've grabbed some patches from git, but I would prefer to skip to 10.5 first10:52
mlankhorststill, it seems to build, no idea if it really runs..10:54
mlankhorstcrash crash11:26
=== _salem is now known as salem_
mlankhorsttons of stuff failing that failed before, I'll try with master..11:42
=== gnuoy` is now known as gnuoy
=== dgadomski_ is now known as dgadomski
=== MacSlow is now known as MacSlow|lunch
mlankhorstslightly better there, maybe I missed some backports :-/12:26
mlankhorstdefinitely did.. grabbed 1 more and llvm-3.6 works a lot better now12:50
=== doko__ is now known as doko
dokomlankhorst, cool, looks promising12:54
mlankhorstseems to have no regressions in llvmpipe compared to 3.512:54
mlankhorstI'll have to test i386 to be sure13:13
=== MacSlow|lunch is now known as MacSlow
Riddellmvo: I wonder if you could help me and harald with a apt resolver problem13:51
Bluefoxicyhttps://lkml.org/lkml/2015/1/23/68814:02
Bluefoxicythis is also going on on linux-mm14:02
Bluefoxicyhttp://marc.info/?l=linux-mm&m=142242638202925&w=214:03
Bluefoxicy.... oh.  Rik van Riel's response isn't on the archives yet14:04
Bluefoxicy"The Android low memory killer does exactly what you want, and for very much the same reasons.14:04
BluefoxicySee drivers/staging/android/lowmemorykiller.c"14:04
BluefoxicyTrying to get the reclaim code un-stupided so the computer doesn't just thrash like hell when you've eaten 99% of your RAM and have no space for page cache14:05
xnoxdidrocks: your emails came out as attachments, instead of inline =/14:10
xnoxunless my mail client is acting silly14:10
pittino, same here; but that's ok? avoids line wrapping and is easier to save?14:11
xnoxno idea if the standard is attachments or inline. I guess i like inline only, and most other people send things like that as well.14:13
xnoxsurely one syncs mailbox as mbox / maildir anyway thus they are already saved. (well integrators usually do)14:13
didrocksxnox: yeah, did that to avoid line wrapping14:14
didrocksxnox: did that last time as well, didn't seem to have been a biggy14:15
xnoxmeh fair enough.14:16
Bluefoxicystatic inline?14:18
argeshallyn: sure thing14:34
argeshallyn: hey we just need bug 1393548 verified to release that fix, its whats been blocking it14:41
ubottubug 1393548 in libvirt (Ubuntu Utopic) "libvirt's apparmor profile denies access to sgabios.bin" [High,New] https://launchpad.net/bugs/139354814:41
mlankhorstdoko: ok seems that on 32-bits llvm 3.5 -> llvm 3.6 has no regressions, just some fixes.. presumably from a commit that fixes unaligned access14:44
flexiondotorgcyphermox, Hi14:47
flexiondotorgcyphermox, What are the next steps for me to get Ubuntu MATE being built officially?14:47
flexiondotorgcyphermox, I see some of the merge proposals are approved but not merged yet.14:48
flexiondotorgcyphermox, How do I go about get me meta package integrated?14:48
hallynarges: oh, feh.  thanks14:59
cyphermoxflexiondotorg: yes, I'm fixing this in a bit, as soon as I'm done with what I'm currently working on15:04
cyphermoxflexiondotorg: is your metapackage prepared already? I could take a look15:05
flexiondotorgcyphermox, Sure - https://launchpad.net/~ubuntu-mate-dev/+archive/ubuntu/ppa/+packages15:06
cyphermoxgreat15:06
flexiondotorgcyphermox, It's called ubuntu-mate-meta. You want version 1.118~15.04~1215:07
cyphermoxack15:07
flexiondotorgcyphermox, I realise I need to clean up the version number for import.15:07
flexiondotorgcyphermox, Source is here - https://code.launchpad.net/~ubuntu-mate-dev/ubuntu-mate/ubuntu-mate-meta15:08
cyphermoxthanks15:08
xnoxcyphermox: you do know how to correctly do meta package updates?15:29
xnoxas in merge into seed branch, pull package, run update, upload.15:29
xnoxthey are semi-automatic15:29
=== roadmr is now known as roadmr_afk
rbasakpitti: all the remaining issues for php5 now are the dpkg symlink problem. Not sure what to do about that.15:38
pittirbasak: yeah, me neither; this is currently being discussed with patch upstream15:41
pittirbasak: we could revert the security fix for the time being of course;15:41
=== timrc is now known as timrc-afk
cyphermoxxnox: I know how to do it, but this would be a new package15:45
pittihm, just got a bug report: "I am running Ubuntu 12.05, and most things work fine."15:49
pittiwhere can I download it? :-)15:49
ogra_linaro ?15:49
ogra_:)15:49
xnoxcyphermox: =/ i've never done a new seed from scratch. as there are a few bits that need to happen elsewhere in the infra to get it working15:51
cyphermoxone step at the time, we'll make it work. I'm not worried15:51
rbasakbarry: around? I'm helping matsubara with bug 1412545. A reporter is asking about an embedded copy shipped in virtualenv.15:52
ubottubug 1412545 in python-urllib3 (Ubuntu Trusty) "proxy isn't used after a dropped connection" [Undecided,Triaged] https://launchpad.net/bugs/141254515:52
rbasakbarry: do you know about how the unvendoring and wheel stuff works, please? Does this need a rebuild of pip and then a rebuild of virtualenv to update virtualenv with a fix?15:53
cyphermoxflexiondotorg: one of the next steps I'd propose is to try to get more uploads sponsored, so that you can eventually get upload rights to at least some of the MATE packages, this would help you get stuff done.15:55
flexiondotorgcyphermox, Agreed.15:55
flexiondotorgcyphermox, Right now I am hoping to get all the "infrastructure" stuff completed. Then start to address the package uploads.15:56
cyphermoxvery well. I'm happy to sponsor things if you need15:56
flexiondotorgcyphermox, Great. dholbach has also sponsored stuff in the past.15:57
flexiondotorgI'll start preparing the few packages that require upload. Basically, settings and artwork.15:57
flexiondotorgI'll need a day or so to get round to that.15:57
barryrbasak: okay, you are either screen capturing me or reading my responses to LP: #141502815:57
ubottuLaunchpad bug 1415028 in python-virtualenv (Ubuntu) "virtualenv's included pip does not use system libraries" [Undecided,New] https://launchpad.net/bugs/141502815:57
barryrbasak: and unfortunately, i do know how all that works ;)15:58
barryor in trusty's case, doesn't work15:58
cyphermoxflexiondotorg: you may want to start working on ubiquity-slideshow-ubuntu-mate if it's not already prepared :)15:58
* rbasak reads15:58
flexiondotorgcyphermox, Already done.15:59
cyphermoxgreat16:00
flexiondotorgI've submitted a merge proposal which has been merged.16:00
barryrbasak: basically, this was all fixed in debian, and inherited in ubuntu, but not until *after* trusty.  pyvenv is deliberately broken in trusty because it violates debian policy. virtualenv violates it too, but that wasn't discovered until after trusty.  so trusty's virtualenv works by accident.  this is all done correctly in jessie, and thus inherited by ubuntu post-trusty.  i talk to *a lot* of people about this16:00
flexiondotorgHowever, it has not been built.16:00
rbasakbarry: understood, thanks. I didn't know about the other bug there.16:00
flexiondotorgcyphermox, If you can trigger a build that would be helpful because I have translator who can't work on it until it is built.16:00
flexiondotorg*translators16:00
cyphermoxok16:01
barryrbasak: i've changed my opinion on what should be done.  i now mostly agree this should be fixed properly in trusty (essentially what jessie/vivid has) but that's a lot of work and a lot of sru'ing16:01
rbasakbarry: so I'm mentoring some people on my team on Ubuntu development, and matsubara is tackling 1412545. I think I'll just go through and SRU python-urllib3 and not bother matsubara with the virtualenv of things, if that's OK.16:01
barryrbasak: so first, i want to see if i can get some resources to do that work, then discuss on u-d@ to see if the stack of sru's we'd need would actually get approved.  this includes adding things like python-wheel which didn't show up in debian until after trusty16:02
rbasakI can leave a bug task open in that bug for virtualenv.16:02
rbasakOK16:02
barryrbasak: i think that's reasonable.  probably as a workaround, you can create the virtualenv, and then `pip install --upgrade urllib3`16:03
barrywhich would pull the newer version from pypi16:03
rbasakThanks - we'll put that in the bug.16:06
=== roadmr_afk is now known as roadmr
=== timrc-afk is now known as timrc
=== wendar_ is now known as wendar
bdmurraycyphermox: Can you have a look at bug 1410573?17:48
ubottubug 1410573 in debian-installer (Ubuntu Vivid) "Expert mode installation fails with error" [Undecided,Triaged] https://launchpad.net/bugs/141057317:48
rbasakinfinity: bug 1412830 verified, if you want to fast track it.18:00
ubottubug 1412830 in spamassassin (Debian) "[AHBL] spamassassin is returning false positives by default" [Unknown,New] https://launchpad.net/bugs/141283018:00
infinityrbasak: Thanks.18:00
infinityrbasak: Will release now.  If it breaks my mail server on upgrade, I will return and yell at you. ;)18:02
rbasakinfinity: my mail server will get the update too :)18:03
infinityrbasak: Right, I'll be sure not to yell at you via email if it's broken, then.18:03
rbasak:-)18:03
xnoxlol19:21
=== ljp is now known as lpotter
=== roadmr is now known as roadmr_afk
=== salem_ is now known as _salem
bdmurraycyphermox: this SRU upload seems to have an accidental change - https://launchpadlibrarian.net/195843293/gnome-bluetooth_3.8.2.1-0ubuntu4.1_3.8.2.1-0ubuntu4.2.diff.gz21:24
cyphermoxoh. opps, that's true21:24
bdmurraycyphermox: okay, rejecting then. let me know when there is another one to review.21:25
cyphermoxsure, it will be in a second21:25
cyphermoxbdmurray: thanks, should be good now. I just uploaded it with the correction21:27
bdmurraycyphermox: it looks the same to me21:30
cyphermoxgah21:31
cyphermoxah, no?21:32
cyphermoxanyway, not the changelog issue21:32
cyphermoxare we both looking at http://launchpadlibrarian.net/196059612/gnome-bluetooth_3.8.2.1-0ubuntu4.1_3.8.2.1-0ubuntu4.2.diff.gz?21:32
bdmurray-+<device oui="7C:1E:52:" type="mouse" name="Microsoft Touch Mouse" pin="0000"/>21:32
bdmurray++<device oui="7C:1E:52:" pin="0000"/>21:32
cyphermoxthat's on purpose21:33
cyphermox7C:1E:52: devices aren't necessarily mice or calling themselves Microsoft Touch Mouse21:33
bdmurraythat's not called out in the changelog21:33
cyphermoxugh, I see21:33
bdmurrayso I thought it was an error because the last SRU did the opposite21:34
cyphermoxtake three21:34
cyphermoxyes21:34
cyphermoxit is most likely an error anyway to revert the previous SRU21:34
=== roadmr_afk is now known as roadmr
cyphermoxbdmurray: please reject it21:43
cyphermoxI'm sorry for wasting your time with this21:43
bdmurraycyphermox: no problem, so what's the plan for gnome-bluetooth then?21:45
cyphermoxto really fix that upload21:46
cyphermoxI've re-done it21:46
cyphermoxtriple-checked the debdiff21:46
cyphermoxI think it should be good this time :)21:46
cyphermoxflexiondotorg: have you built any kind of almost-daily image somewhere instead of your 14.10 images?22:55
=== robbiew is now known as robbiew-afk

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