UbuntuIRC / 2014 /08 /05 /#ubuntu-desktop.txt
niansa
Initial commit
4aa5fce
[02:30] <willcooke> morning all
[02:30] <willcooke> O_o
[02:30] <willcooke> olli, you want to know what tomorrow has in store? We're in the future
[02:30] <didrocks> waow, seems really similar to the shape of your eyes right now :)
[02:30] <willcooke> :D
[02:31] <willcooke> or rather...
[02:31] <olli> willcooke, what are tomorrows lottery numbers?
[02:31] * TheMuso waves.
[02:31] <willcooke> X/
[02:31] <didrocks> yeah, +1 ;)
[02:31] <willcooke> olli, 1 2 3 4 5 6 7 and ...... 8
[02:31] <sarnold> do we have flying cars in the future?
[02:31] * willcooke looks out the window
[02:31] <TheMuso> Not yet.
[02:31] * TheMuso notes he is further in the future. :p
[02:31] <willcooke> ... hrm. Not yet, but it's still the morning, so maybe later on today
[02:31] <didrocks> who knows…
[02:32] <olli> heh, willcooke... Beijing right... as if you'd see something outside your window with all that smog
[02:32] <sarnold> oh okay :) I wouldn't want the future to be bland and boring..
[02:32] <TheMuso> But the authoriatitve figure is actually robert_ancell.
[02:32] <willcooke> olli, :)
[02:32] <TheMuso> Robert is further in the future than I am.
[02:33] <willcooke> TheMuso, bad news... I have been pointed at: http://people.canonical.com/~platform/desktop/ubuntu-desktop.html
[02:33] <willcooke> TheMuso, ;) Top of the list is alsa
[02:33] <willcooke> TheMuso, anything we need to be concerned about? Seems like its only a minor point release
[02:34] <TheMuso> willcooke: Oh right, I forgot to update that. I'm patch piloting atm, will take care of after.
[02:34] <willcooke> TheMuso, sweet, thanks!
[02:42] <seb128> good morning desktopers
[03:43] <seb128> rsalveti, hey, do you know if there is any known ofono issue with recent updates?
[03:44] <rsalveti> seb128: not that I know, why?
[03:44] <seb128> the settings tests started being unhappy since yesterday
[03:44] <seb128> e.g CI run on https://code.launchpad.net/~mterry/ubuntu-system-settings/get-user-by-id/+merge/229510
[03:45] <rsalveti> I know alfonso landed a new ofono, maybe that caused the regression
[03:45] <rsalveti> https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-utopic/1978/console
[03:46] <rsalveti> this was because of a jenkins issue though
[03:46] <seb128> there is a timeout
[03:46] <seb128> but some of the tests are FAIL before that
[03:46] <seb128> Laney had another run showing similar failures
[03:47] <rsalveti> right, quite a few failures
[03:47] <seb128> rsalveti, https://jenkins.qa.ubuntu.com/job/autopilot-testrunner-otto-utopic/1977/?
[03:47] <rsalveti> probably because of latest landing, yeah
[03:47] <rsalveti> seb128: abeato should be on soon
[03:47] <seb128> k
[05:10] <pitti> didrocks: you can use the "needs-recommends" restriction if you want the test Depends: recommends installed
[05:10] <pitti> didrocks: you say you are running this on trusty, does that mean trusty's autopkgtest?
[05:11] <pitti> didrocks: that exception when trying to print an exception looks familiar, but I think I fixed that months ago
[05:35] <didrocks> pitti: hey, sorry, we were reconnected automatically :)
[05:35] <didrocks> pitti: thanks for the needs-recommends trick, I'll do that right now
[05:35] <pitti> didrocks: pas de probléme :)
[05:36] <didrocks> when I say on trusty, I have a trusty host (so trusty adt), but the chroot is utopic
[05:36] <pitti> didrocks: I'll answer to your mail in more detail, but that was the quick thing (and whether you run autopkgtest from trusty)
[05:36] <pitti> didrocks: ah; please try utopic's autopkgtest (you can just install the .deb)
[05:36] <didrocks> yeah, no worry! :)
[05:36] <didrocks> ok, doing that now
[05:36] <didrocks> so, without the -o option?
[05:36] <pitti> didrocks: trusty's autopkgtest was just barely working with qemu, it was very new at that time
[05:37] <didrocks> ah, that would explain :)
[05:37] <pitti> didrocks: -o> as you like
[05:37] * didrocks download the .deb and tries
[05:37] <pitti> didrocks: -o just puts the log and per-test stdout/err etc. into that directory, if you need it
[05:37] <pitti> it's --output-dir
[05:37] <didrocks> hum
[05:37] <pitti> otherwise it just goes to stdout/err
[05:37] <didrocks> but then, the output should be realtime, still?
[05:37] <didrocks> (the test output)
[05:38] <pitti> didrocks: yes, it's always realtime
[05:38] <pitti> didrocks: well, it wasn't in trusty yet for qemu
[05:38] <didrocks> ahah
[05:38] <didrocks> that explains :)
[05:38] <pitti> didrocks: (I think; pretty much half of the code changed since then..)
[05:38] <didrocks> waow
[05:38] <didrocks> ok, new deb installed
[05:38] <didrocks> no need for a newer qemu?
[05:39] <didrocks> adt-run [13:39:13]: version 3.3.1
[05:39] <didrocks> let's see ;)
[05:39] <pitti> didrocks: no, shoudl be ok
[05:39] <pitti> didrocks: we are using trusty's qemu in production CI
[05:39] <pitti> didrocks: but autopkgtest straight from git head
[05:39] <didrocks> living on the edge!
[05:40] <didrocks> (I try to keep my box trust*ed* as the developer story main focus is there)
[05:40] <didrocks> but that's fine, if I only have newer pep8 and adt :)
[05:40] <pitti> $ git shortlog 2.14.1 |wc -l
[05:40] <pitti> 706
[05:40] <pitti> $ git diff 2.14.1.. |diffstat
[05:40] <didrocks> waow, some work is going on here!
[05:40] <pitti> 82 files changed, 8191 insertions(+), 5170 deletions(-)
[05:40] <didrocks> :)
[05:40] <pitti> didrocks: a few commits here and there :)
[05:41] <didrocks> ok, you were not kidding telling half the code changed :p
[05:41] <pitti> didrocks: no, there's a few times when I'm serious
[05:41] <didrocks> heh
[05:41] <didrocks> like in the early morning apparently :)
[05:44] * didrocks sees one byte after another coming on his laptop
[05:45] <pitti> didrocks: I hope you have a well filled apt-cacher-ng or something similar
[05:45] <didrocks> pitti: yeah, otherwise I would have lose patience :p
[05:46] <pitti> didrocks: it's pretty much a must for adt, otherwise you'll go crazy -- indeed
[05:46] <pitti> didrocks: adt-buildvm-ubuntu-cloud and adt-build-lxc auto-detect and use it, too
[05:46] <pitti> (guess which cacher I'm using :) )
[05:47] <didrocks> enough bet for this week, even if the answser is obvious ;)
[05:47] <didrocks> answer*
[05:47] <pitti> didrocks: oh, how many beers did you already lose?
[05:48] <didrocks> pitti: nothing, seb is only interesting in betting moneys. He doesn't go down to less than a thousand euros :p
[05:48] <didrocks> and then, he wonders why noone is accepting his bets :p
[05:49] <pitti> seb128: o_O
[05:49] <seb128> lol
[05:49] <seb128> guys!
[05:49] <seb128> how you make me a reputation there
[05:53] <didrocks> pitti: the output is slightly misordered (like I get the coverage and test results before some other output), otherwise, it seems to behave :)
[05:54] <didrocks> it exits with 4 though, is that expected? (the output told no test failing)
[05:55] <pitti> didrocks: no, that means some test failed; log | pastebin, SVP?
[05:55] <pitti> didrocks: I still had the PEP-8 error yesterday
[05:56] <didrocks> pitti: yeah, only with utopic pep8, I fixed that in ~utopic4
[05:56] <didrocks> rerunning and pasting
[05:56] <pitti> didrocks: that's where -o comes in handy :)
[05:56] <pitti> (or -l at least)
[05:56] <pitti> didrocks: it didn't just complain about lots of stderr?
[05:58] <didrocks> pitti: no, I didn't get anything like that
[05:58] <didrocks> let's see, I just try to 2>&1 | tee to get the exact same output order
[05:59] <pitti> didrocks: the log file by and large has that already
[05:59] <didrocks> ok, will do for next runs :)
[05:59] <didrocks> oh, we have a meeting now, will move there and paste it to you later (still running)
[06:05] <didrocks1> pitti: so, output is 1 and traces are: http://people.canonical.com/~didrocks/tmp/foo (I wonder if it doesn't try to run them twice or if it's the output not being in order)
[06:05] <didrocks1> I couldn't use pastebin, it's always timing out
[06:05] <didrocks1> in the middle, you can see:
[06:05] <didrocks1> Ran 255 tests in 42.031s
[06:05] <didrocks1> but: all FAIL stderr: 2014-08-05 06:00:40,901 [nose.plugins.manager] DEBUG: DefaultPluginManager load plugin cov = nose_cov:Cov
[06:06] <didrocks1> and I guess it's the wrapping bash script that is set -e producing that?
[06:06] <pitti> all FAIL stderr: 2014-08-05 06:00:40,901 [nose.plugins.manager] DEBUG: DefaultPluginManager load plugin cov = nose_cov:Cov
[06:06] <pitti> adt-run [14:01:24]: test all: - - - - - - - - - - stderr - - - - - - - - - -
[06:06] <pitti> 2014-08-05 06:00:40,901 [nose.plugins.manager] DEBUG: DefaultPluginManager load plugin cov = nose_cov:Cov
[06:06] <didrocks1> and then adt repaste what was in stderr?
[06:06] <pitti> didrocks1: right
[06:06] <didrocks1> ah ok ;)
[06:06] * didrocks1 removes -e
[06:07] <pitti> didrocks1: so, either you fix your tests to output to stdout, or you add "allow-stderr" restriction
[06:07] <pitti> didrocks1: unfortunately python's unittest defaults to stderr
[06:07] <didrocks1> yeah
[06:07] <didrocks1> ok, will add that
[06:07] <pitti> unittest.main(testRunner=unittest.TextTestRunner(stream=sys.stdout, verbosity=2))
[06:07] <pitti> most of my tests use that to write to stdout
[06:08] <pitti> dear unittest, test progress is not an error!
[06:08] <didrocks1> indeed :p
[06:08] <didrocks1> I'll just add one of the 2
[06:09] <didrocks1> connection being bad here, I won't go on with the docker tests (requiring downloading a docker image) and the integration tests
[06:09] <didrocks1> I guess that will be for next week
[06:09] <didrocks1> that's great already that way
[06:10] <pitti> didrocks1: ok, I think we settled everything from your mails then?
[06:10] <didrocks1> pitti: yeah, thanks a lot for all your help!
[06:10] <pitti> didrocks1: de rien, tu es un bienvenue !
[06:11] <pitti> (or something similar?)
[06:11] <didrocks1> parfait "de rien" is enough ;)
[06:12] <didrocks1> pitti: let's sync next week on getting network up then :)
[06:12] <didrocks1> thanks again
[06:12] <pitti> yes
=== seb128_ is now known as seb128
=== didrocks1 is now known as didrocks
=== MacSlow is now known as MacSlow|lunch
=== alan_g is now known as alan_g|lunch
=== MacSlow|lunch is now known as MacSlow
=== alan_g|lunch is now known as alan_g
=== sil2100_ is now known as sil2100
=== m_conley_away is now known as m_conley
[15:11] <Dickens> I DID IT!!!
[15:17] <tsdgeos> tedg: do you know who could i ask to review https://code.launchpad.net/~aacid/indicator-power/fix_autostart/+merge/229325 and https://code.launchpad.net/~aacid/indicator-bluetooth/fix_autostart/+merge/229324 ?
[15:18] <tedg> tsdgeos, I saw those, but I haven't looked, what is the issue?
[15:18] <tsdgeos> tedg: they get autostarted if you log into Plasma (or any non unity desktop) which doesn't make sense afaics
[15:19] <tedg> tsdgeos, So we can filter them out of plasma then. The problem is folks like XFCE do use them.
[15:19] <tsdgeos> tedg: do they? i had a look at my autostart and looks like this
[15:20] <tsdgeos> tedg: http://paste.ubuntu.com/7961955/
[15:20] <tsdgeos> that's why i just did the same we seem to do everywhere
[15:24] <tedg> tsdgeos, Is plasma using Upstart sessions?
[15:24] <tsdgeos> i have no clue :D
[15:24] <tsdgeos> i guess the kubuntu guys made it use
[15:25] <tedg> tsdgeos, initctl list ← does that list system jobs or session jobs ?
[15:25] <tsdgeos> i'm not in Plasma at the moment
[15:25] <tsdgeos> i'll come back to you once i stop working :D
[15:25] * tedg gives tsdgeos the rest of the day off ;-)
[15:26] <tsdgeos> only 34 min more ;)
[15:27] <ogra_> geez ... did anyone tell you before that you sound really german ?
[15:27] <ogra_> :)
[15:27] <ogra_> 34min 12seconds and i'm done
[15:28] <tsdgeos> ogra_: nope
[15:28] <tsdgeos> ogra_: once a guy told me i didn't look like "an Albert"
[15:28] <tsdgeos> whatever that's suposed to mean
[15:28] <GunnarHj> Laney: ping?
[15:28] <ogra_> lol
[15:31] * tedg makes a note to always call tsdgeos "Jim"
=== gatox is now known as gatox_lunch
[16:04] <mitya57> tsdgeos: I think "OnlyShowIn: unity" is wrong. Indicators should *not* be started in Unity via xdg autostart, as they are started via upstart.
[16:05] <tsdgeos> mitya57: it's not to make it open in unity, it's to make it not open in !unity
[16:05] <mitya57> Will "NotShowIn: Unity;" ---> "NotShowIn: Unity;KDE;" suit you?
[16:06] <tsdgeos> mitya57: it'll fix the bug for Plasma, not for others
[16:06] <tsdgeos> but it is a hard situation i understand
[16:06] <tsdgeos> you can either blacklist all and fail because you blacklisted too much or blacklist some and fail because you blackilisted too few
[16:07] <tsdgeos> mitya57: i can go that way if people prefer
[16:09] <tsdgeos> i'll be back in an hour or so
[16:09] <tsdgeos> or just comment on the merges ;)
[16:12] <mitya57> According to https://launchpad.net/ubuntu/+source/xfce4-indicator-plugin/2.3.2-0ubuntu2, Xubuntu is now using upstart jobs as well
[16:12] * mitya57 comments on the merges
[16:13] <mitya57> Actually, maybe we don't need that desktop file at all
[16:25] <Dickens> yikes... i'm running low on HD space :-(
[16:25] <Dickens> i gotta find a way to compress some of my older movies down to a more reasonable size
[16:28] <chrisccoulson> I keep running out of space too, and I've almost convinced myself that the solution to this is http://www.amazon.co.uk/Samsung-inch-Solid-State-Drive/dp/B00LMXBMUQ/
[16:29] <Dickens> yea... umm why ssd though?
[16:30] <Dickens> i dont think your performance increase outweighs the cost if it's just for storing movies
[16:30] <Dickens> i have 4TB of storage and i'm running low
[16:30] <Dickens> it's these 50GB blueray rips that's doing it
[16:30] <Dickens> i gotta replace them with smaller files
[16:30] <Dickens> i wish there was a way to script handbrake to compress movies to a certain bitrate after a certain amount of time
=== robru_ is now known as robru
=== gatox_lunch is now known as gatox
=== alan_g is now known as alan_g|EOD
[17:57] <tsdgeos> tedg: which command you told me to run?
[18:40] <tedg> tsdgeos, initctl list
[18:41] <tsdgeos> tedg: http://paste.kde.org/pygwdmhn6
[18:41] <tsdgeos> oh i have forgotten to stop mediascanner today
[18:58] <mitya57> tsdgeos: I think it'll be a good idea to get someone else from desktop team to confirm that they don't need that file
[19:00] <tsdgeos> +1
=== m_conley is now known as m_conley_away
=== m_conley_away is now known as m_conley
[23:42] <RAOF> Laney: You on? I'm looking at the webkitgtk SRU, and have some questions.