[00:48] I am back, on a phone, but back [00:50] Is someone able to confirm the oes install bug? [00:50] *oem === Ursinha-afk is now known as Ursinha [01:40] is anyone able to run an oem install? [02:09] whew.. back.. umm, I will check tomorrow Noskcaj === Ursinha-afk is now known as Ursinha [02:49] balloons, ok, see you then === mmrazik is now known as mmrazik|afk === yofel_ is now known as yofel === mmrazik|afk is now known as mmrazik === Ursinha-afk is now known as Ursinha [15:51] balloons, you awake yet? [15:51] phillw, you too. [15:51] can't speak for phillw [15:52] but I'm about [15:52] balloons, you need to read something, give me 3 seconds [15:52] balloons, https://lists.ubuntu.com/archives/ubuntu-quality/2013-January/003000.html and related responses [16:15] (and i betcha the IRCC would agree) [16:15] however, I appreciate om26er's efforts in raising awareness.. We do utilize this channel for other things besides bugs, and generally we're discussing and confirming bugs we find in dev releases if we do dicuss bugs [16:15] which is why i lurk here. [16:16] :) [16:16] you know... [16:16] i have a proposal: [16:16] and i would LOVE if you could comment, bdmurray: [16:17] balloons, bdmurray I have come to senses as well, given I was a bit unclear about this channel's purpose now its cleared. So we'll try to revive our BugSquad efforts the other way [16:17] the classroom is runing User Days in February. I think that members of QA and bugs should run a session to raise awareness (a "Bugs and QA 101") for both teams and how they coordinate with each other, and how they differ. [16:17] also, bdmurray, there should be a "Bugs for the End User" session, giving an entry level session about bugs. [16:18] (nobody's ever rallied for that thoug) [16:18] or, we should have a higher QA/BugSquad presence during a more technical run of -classroom [16:18] (heck, even have a UDS presence) [16:18] sounds good om26er I wish you the best [16:19] ... oh COME ON [16:19] how hard is it to keep sbuild working... [16:19] * TheLordOfTime walks off [16:19] TheLordOfTime, we are having sessions this month.. gema will actually be talking about bugs [16:19] TheLordOfTime, end users better not be introduced to bugs ;) [16:19] om26er, i meant bug reporting [16:19] om26er, you'd be surprised how many bugs i see filed against unrelated packages :p [16:19] not from a bug squad perspective though [16:19] balloons, thanks :) [16:19] balloons, when're you holding the session(s)? [16:20] (its not on any calendar) [16:20] * TheLordOfTime may want to lurk [16:20] we have to finalize dates [16:20] it's on the list this week to punch them down [16:20] that'd be why its not on any calendar :P [16:20] I owe liz an email [16:20] heh [16:21] bdmurray, thanks for stopping in, apparently this was resolved prior to you stopping by, in case there was additional input needed from the -bugs side of thing, i wanted you to be nearby :) [16:21] thanks much. [16:21] * TheLordOfTime goes to fix his local builders now. [16:47] actually, I propose we move the discussion to -bugs [19:13] Noskcaj, howdy [19:13] I updated the bug with my findings [19:25] balloons, hello, i saw that. [19:26] my reason for thinking it's a bug is that is where manufacturers name the machine e.g. dell-52845 [19:27] yes, I agree [19:27] xnox would have to tell us the intended behavoir [19:27] also, I noticed that if you change the default in oem, it's blank during user setup [19:28] interesting === TheLordOfTime is now known as LordOfTime === LordOfTime is now known as TheLordOfTime [19:51] i just filed bug 1095777 [19:51] bug 1095777 in telepathy-logger (Ubuntu) "package libtelepathy-logger3 (not installed) failed to install/upgrade: Versuch, »/usr/lib/x86_64-linux-gnu/libtelepathy-logger.so.3.2.0« zu überschreiben, welches auch in Paket libtelepathy-logger2:amd64 0.6.0-1 ist" [Undecided,New] https://launchpad.net/bugs/1095777 [19:52] to reproduce, you just need to try and update a raring vm to the current package versions. hm. thinking about it.... is this not a bug but a normal temporary issue in a development version? [19:52] dkessel, yes libtelepathy is in flux atm [19:52] I believe there is a bug on it [19:52] empathy fails to open atm, among other things [19:53] balloons: oops... and a duplicate of it now :) [19:53] no worries [19:53] let's find it and dupe it [19:53] I was just looking at it a couple days ago [19:53] however, in general, this stuff can happen during the development cycle [19:55] btw apport is reporting bugs automatically again in raring [19:55] no mor 404 [19:55] yay [19:56] I didn't get an answer, but maybe they saw my request [19:56] hehe [19:57] hm. is it better to run my vm in english? it would make the log output more useful to non-german people :) [19:57] heh.. yea, I suppose that [19:57] that's not a bad idea.. testing non-english stuff is always good [19:57] but reporting and logs in english is good [19:58] hm. too bad i can't change the language because my package database because of libtelepathy :/ [19:59] found the original on, marking my one as dupe (original: 1095745) [20:00] awesome, good stuff [20:04] balloons: which behavoir are you after? =) [20:06] hey xnox -- umm let me nab the bug [20:06] it's oem setup tuff [20:07] https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bug/1095465 [20:07] Launchpad bug 1095465 in ubiquity (Ubuntu) "Ubuntu raring oem install series name is ignored" [Undecided,Confirmed] [20:08] so in short, what should the series name do for the oem install? it doesn't change the machine name default [20:08] balloons: right series name has nothing to do with machine name. But it's interesting to know why that user is expecting it to be. [20:09] and additionally, if you change the machine name default it doesn't carry over to the user install [20:10] xnox, kk.. I wasn't sure. what should the series name affect? Anything? [20:13] balloons: it is suppose to store series id somewhere under /var/log/installer/*, and include them in apport bug reports as a means of identifying the batch of identical machines. [20:14] balloons: it should not change anything else at all. [20:17] got it.. so since we're chatting about oem installs.. can you answer the question about machine name defaults? will it only default to a machine name on the user install if you leave it alone (accept the default) on the oem install? [20:17] I expected the user install to mirror the oem install machine name.. [20:20] well. if there is network it tries to do dns lookup and use that name, if that fails we use something by default "$username-laptop" or something like that. [20:20] right.. so, your not going to use the oem machine name at all during user [20:20] so it's weird then why it came up blank for me [20:21] balloons: it is really a bad idea to ship thousands of machines all defaulting to hostname "model1000" they will collide. [20:22] balloons: black? even after you started typing your name in? [20:22] s/black/blank/ [20:22] xnox, my expectation was to default to my username - OEMname of something [20:24] regardless, yes, it was blank [20:24] I would be happy to repeat it, and file a bug.. assuming this is not expected [20:24] sounds like :-) [20:25] balloons: OEM series should be something like "thinkpad-t43-555546877-batch32-factory65-line2-shift4", e.g. nothing human sensible, but something that is uniquickly identifiable and tracable back to production line logs/batches/orders. [20:25] that would be a very ugly hostname =/ [20:26] balloons: it should not be black during the actual oem-config (i.e. when you reboot the second time as "end-user" after finalised & prepared install) [20:31] xnox, k.. I'll close with your explaination, and file a new one for the blank issue [20:31] ty ty! [20:32] ack. [20:47] hey letozaf_ [20:48] hello balloons [20:48] howzit ? [20:48] I'm looking at your merges [20:48] the evince and nautilus tests do not work unfortunatly [20:49] *unfortunately* [20:49] formatting is a bit weird tho [20:50] python has to have it's indenting be exact :-) [20:50] oh! let me look at them I thought they were ok [20:50] but can you run them or have you errors ? [20:51] I could run them [20:51] might just be my client [20:51] hmmm [20:51] nvm [20:51] yep, it's my client identing [20:51] hmm [20:52] are you using tabs or spaces? ideally 4 spaces [20:52] well usually I use tabs, sometimes spaces [20:52] let me see... [20:52] anyways, i'll try running them in a sec [20:53] yes 4 [20:53] they do not work :( [20:53] looks like evince's window does not have focus so [20:54] when you press F11 or F9 nothing happens [20:54] same with screenshot and nautilus [20:54] :'( [20:54] are you guys talking about manual tests converted to autopilot tests? [20:54] gotcha.. [20:54] dkessel, yes indeed [20:55] letozaf_ has been writing some autopilot conversions [20:57] letozaf_, are you going to convert the gedit testcase to leafpad, kwrite, ect. [20:57] bah.. geany is making me sad [20:57] i read about it the other day. it looks like a nice idea. i was surprised to see that autopilot is a part of unity though. i would've thought it would be better to evolve autopilot independently of unity, and have autopilot as a dependency when running unity tests [20:57] dkessel, the latter is what's happening now [20:58] Noskcaj, if balloons says so yes, otherwise no [20:59] balloons, oh - nice [20:59] hence, why we can utilize it now :-) [21:01] hmm... the python version used is still 2.x, right? but what the hell - i might help creating/converting tests too. gives my some practice in python :) [21:01] ever written python or another language? [21:01] hehe [21:01] yea, give it a whirl [21:01] learnt c/c++/c#/java at university. only just learned python3 in the christmas holidays ;) [21:03] gotcha [21:03] yea, this is relaxing stuff [21:04] :-D [21:05] is there a "worklist" like at the hackfest? with who is doing what, what is needed, etc...? [21:07] dkessel, hmm.. we could easily make such a thing [21:08] bascially the list atm is to convert our existing [21:09] since we have an lp project now, I'd like to use it and not a wiki to do this [21:09] ok :) [21:10] meaning, I'd like to put that work and assignments, etc into the project on lp if possible [21:10] not used lp in that way.. [21:10] so it's new to me, but I'm hoping/assuming it's possible [21:13] after I finish this merge I'll have a look at doing that [21:13] right now though, let me give you the list [21:14] and you can see what's in bzr and what's not, so it's easy to see what's been done / working on and what hasn't [21:14] dkessel, this is the list of package tests; http://packages.qa.ubuntu.com/qatracker/series/30/testsuites [21:15] for our purposes, we'll ignore the unity and webapp tests [21:16] by the way, balloons, but how are we supposed to test network manager with autopilot ? [21:16] I mean you need to plug in and unplug the network cable [21:17] to test it [21:17] suppose autopilot cannot do that ;) [21:17] you underestimate the power of autopilot ;) [21:17] ROFL [21:18] you could simulate it, but I agree [21:18] hence it's why the drones can't yet replace us [21:18] yeah! maybe we just disable and enable networking... [21:18] they've done some neat things [21:19] you could run the test on a dedicated machine with a simple robot attached ;) that moved the network plug in and out [21:19] faking sounds cards, etc [21:19] faking monitors [21:19] making hardware bits to fake things [21:19] well I would really like to see how they made the faking things... must be interesting [21:20] it is.. I should ping someone to talk a bit about it [21:20] I don't have intimate details sadly [21:20] and if I did, I don't remember them [21:20] hah! [21:20] :) [21:24] balloons, i guess i should be able to convert the fileroller test, as a start. i will try it on the weekend. just so you know and nobody does duplicate work... [21:26] got it dkessel thanks! [21:31] hm to create tests, will i still need unity source like it says here: https://wiki.ubuntu.com/Unity/QA/Autopilot? or can i get autopilot "standalone" from somewhere else? [21:32] no, you don't need unity [21:32] * balloons goes to look at link [21:33] dkessel, if possible have a look at my blog posts [21:33] goes through writing a testcase (converting one actually) [21:34] http://www.theorangenotebook.com/ [21:34] ah i see, nice [21:37] *sigh* me and my "old" precise... [21:38] umm.. [21:39] well, you might skate by running precise [21:39] use the autopilot ppa [21:39] *checks ppa* [21:39] you may have to use the quantal version.. but I'm guessing you could kludge it [21:40] dkessel, what about using a virutal machine... should work [21:40] create a VM in Virtualbox for autopilot tests [21:40] with quantal [21:41] alright :) you guys are right [21:41] nothing wrong with precise btw.. I use it on my netbook [21:41] and quantal on my wife's box [21:41] but I'm always running ubuntu+1 [21:43] i'm doing it the other way around... on my normal private laptop (this one) i run precise, for stability and less daily frustration ;) on the netbooks i run raring [21:43] yeah on my notebook I've got Quantal and Raring, but there's always a free partition for tests so if I need I can install Precise [21:44] the advantage of a VM is that you do not have to reboot [21:44] lol.. yea, daily frustration i suppose is part of the fun.. haha [21:44] but yea precise is really nice [21:44] balloons, could you run the evince test ? [21:46] on a netbook, would you normally run x64 or i386? [21:47] I run 64tput.pdf [21:49] yeah you're running evince test :) (tput.pdf) [21:49] lol [21:49] yes I am [21:50] :D [22:05] hmm just re-installing raring daily on one of my netbooks and the trackpad is not working [22:06] let's see if that stops me from installing [22:08] keyboard commands ftw [22:09] i guess there should still be options displayed in the boot menu when booting from the iso? [22:11] umm if you press something [22:11] else it just boots [22:12] ok [22:16] balloons, I'm going to bed now it's quite late. let me know about the tests :) [22:16] yes, I will [22:16] I'll check them in [22:16] no matter what [22:17] you did have tabs in there, so I'm killing them and using 4 spaces [22:17] I'll merge nautilus tests too [22:17] ok [22:17] ohh one thing [22:17] the screenshot, nautilus and evince tests have problems [22:17] yes? [22:18] this output.pdf thing [22:18] where is it created? [22:18] under Documents directory in the users home dir [22:18] yes, but how is it created? [22:19] ohh nvm [22:19] I see [22:19] kk [22:19] Well I open help with F1 [22:19] yep,. i get it [22:19] g-night [22:19] ok buona notte :) [22:20] hope we find a way to fix them :) [22:20] bye [22:20] kk [23:29] hey welcome ;-) [23:32] balloons: hello :-) [23:32] hey.. you missed our earlier discussions.. but it will be nice to have you lurking [23:38] alright, so bzr branch lp:ubuntu-autopilot-tests has letzoaf's code in tit [23:39] balloons: ok, got it. Which tests should I be looking at? [23:40] well, I was working on evince, which was the pending merge [23:40] ahhh, I see the issue [23:40] specifically, there's a few open ?'s comment lines in there [23:41] I and other folks have just left comment lines when we couldn't actually perform the task ;-) [23:41] basically, 'self.start_app_window' isn't what you want to use. I'll write a short test case and push up a separate branch so you can see it working [23:41] I'll do something simple - gedit maybe' [23:43] ok, there is a gedit case [23:44] umm in there, there's several gaps.. checking for windows or panels opening.. or for text, etc [23:44] regardless.. yes.. show me the light ;-0 [23:55] hmm, gedit is doing some really strange things. [23:55] like, when I open it, the default tab shows a 'busy' spinner in the tab header, and doesn't accept any keyboard input [23:55] * thomi investigates [23:56] ahh.. yea, when I tried messing with introspection on gedit, it was quite odd [23:56] * thomi swears [23:57] alesage: are you aware that the autopilot-gtk package (in raring anyway) is essentially empty? [23:57] alesage: this is the contents of the package: http://paste.ubuntu.com/1493751/ [23:57] thomi brrtwha? [23:57] no gtk module installed [23:57] thomi sorry let me investigate [23:58] no worries - I'll try building a local package [23:59] thomi.. well, it's late for me now