[00:00] * balloons floats in [00:00] oi [00:00] see what i did to https://wiki.ubuntu.com/QATeam/DevelopmentInstall [00:00] I'm seeing the updates flow :-) [00:00] veebers: I assume we can drop it. But to do so we need to change the packaging too. That's certainly not something we have time to do in this release. [00:01] s/release/iteration. [00:01] elopio: ack, agreed. It's worth putting on the backlog thought [00:01] though* [00:01] knome, that's nice [00:01] balloons, "installing" should be called something else though [00:01] veebers: can you please make a card for it? for both projects, please. [00:01] and the last section isn't probably fit for that page [00:01] or very useful [00:02] I always thought 'development install' was a bit weird page, but it's random tidbits loosely tied [00:02] we can definitely rename it now if you want [00:02] I have the autopilot cards ready. I think the three of them :D [00:02] will mark it for review. [00:02] elopio: which cards are they? [00:02] balloons, taking the context into account, that page should describe how to set up an environment to test in [00:02] woot elopio [00:03] knome, right.. and how to maintain the enviroment I assume is what the bottom of the page is about [00:03] elopio: also, that MP looks good, the only style thing I would comment on is the closing paren on a new line (or lack of in this case :-)). diffline lo6 for example. [00:03] veebers: launching, assertions, and make. I just copied what was in rick's branch :D [00:03] elopio: I can make the cards, were abouts do I make them? [00:03] elopio: nice! [00:04] veebers: lets discuss about the closing parentheses when I return from the gym. I'll ping you. [00:04] balloons, isn't it about maintaining any installed ubuntu system really? [00:04] thanks for the review. [00:05] ack, nw [00:06] balloons, and tbh, with ISO testing, you don't really want to "maintain" anything, you just want to destroy after destroying [00:06] balloons, so maybe a more valid point there would be to tell people how to manage their virtual machines [00:06] knome, well the development install is more about actually running ubuntu devel on your box [00:06] or machine if you choose [00:07] judging by the context where we are linking to that page, not really [00:07] it's referred to as a guide to get set up for testing [00:08] veebers: in the UEQA: Projects board. [00:08] elopio: awesome, thanks sir [00:08] on the line that's not sanity. [00:08] heh, is there any sanity in qa? ;-) [00:08] balloons, but i'll happily move the setup guide to somewhere else if you think we need another page for maintaining a development install [00:09] balloons, but really, how does maintaining that differ from maintaining a regular release? [00:10] balloons, now that this bit is done, we can simply link to it from any page that needs instructions on how to set up a development ISO on hardware, or a virtual machine [00:11] balloons, i just changed the "installing" headers to something more suitable [00:11] knome, the only difference really is partial upgrades [00:12] balloons, technically those can happen with regular desktops as well [00:12] and you are right; I guess we don't necessarily need to maintain that type of documentation [00:12] balloons, and have happened for me [00:12] knome, yes I know :-) [00:12] it's much more likely when you are using a non-main server too [00:12] you are convincing me.. if we can simplify, let's do it [00:12] we can keep it there for now [00:13] until we find a better place or are sure we don't want it [00:13] that being fixed... [00:13] i just understood what is weird with https://wiki.ubuntu.com/QATeam/Roles/Tester [00:13] it talks about setting up a development version before it talks about the activities you can do [00:14] maybe we should list the activities first, then list all the guides that are/might be needed for them [00:15] the setup? yea, it's being pushed, as is the 'getting stuck' [00:15] i'll do a PoC page for what i think would work well for the role pages [00:16] but tbh, i don't know if we need the subpages for them after i've done this [00:16] :X [00:16] feel free to directly edit; you certainly aren't going to hurt my feelings :-) [00:16] no, i want to make sure it works [00:16] it might take some time [00:16] sure, just letting you know [00:16] maybe 15 mins :P [00:16] don't understand me wrong, but i don't mind if your feelings are hurt... if we can improve something ;) [00:18] knome, LOL.. very very true [00:28] balloons, https://wiki.ubuntu.com/QATeam/Roles/TesterN [00:28] the top chart.. fancy! [00:28] * balloons loves visuals [00:28] yeah.. one thing that i'm not sure of is the maintaining burden with that table [00:28] it isn't automatical [00:29] otoh, we can bump those banners to the main Roles page [00:29] and you can get a better overview of the activities [00:29] though, what i'm thinking is that the "other activities" should all go under one common page [00:30] or maybe not like that [00:30] but at least "reporting bugs" and "triaging bugs" could have different targets [00:44] balloons, wondering how much https://wiki.ubuntu.com/Testing/QATracker is different from the tracker walkthroughs linked at https://wiki.ubuntu.com/QATeam/Roles/TesterN ? [00:44] ...and whether there could be one page to serve them all [00:45] there's a qatracker page explaining generically what it is, then walkthrough for each type [00:45] except the link i pasted is quite a walkthrough itself [00:45] yea, indeed [00:45] so imo either cut that down, or, [00:46] it covers fairly unique info though [00:46] make it a bit broader, and then do some specific comments about the specific trackers [00:46] i'm pretty sure that once a person has learned to use one of the trackers, they have little or no problems jumping into the next one [00:46] especially if they learn the ISO tracker first, which is using the features to their max [00:48] i personally believe that one page could serve as a walkthrough for them all, especially with links to your video walkthroughs, which show off the specific trackers [00:49] knome, well let's consolidate then [00:50] then we also have these pages: [00:50] https://wiki.ubuntu.com/QATeam/Activities/ManualTesting [00:50] https://wiki.ubuntu.com/QATeam/Activities/ImageTesting [00:50] https://wiki.ubuntu.com/QATeam/Activities/ApplicationTesting [00:50] which i think are... [00:50] hmm [00:50] not very useful [00:50] weird.. I edited them at least [00:51] * balloons fires up the powershredder [00:51] they aren't linked to from anywhere else except the FAW [00:51] *FAQ [00:51] i'll remove the links from the FAQ then [00:52] so we should migrate logging in over to qatracker, not much else I'd guess.. maybe executing the testcase. you are right I think almost everything can be in one place [00:53] yep [00:53] so once that's done, the Tester page is much better and concise already [00:54] and i just figured what "Other activities" should be called [00:54] "Supporting tasks" [00:54] that makes sense and helps us categorize what belongs there [00:54] and what is essential for the nature of that role [01:00] thomi, ping [01:00] charles: hello [01:00] I've got a question about proper use of matchers [01:00] sure [01:00] Leo had a nice suggestion about using something like [01:00] self.assertThat(indicator.get_icon_name, Eventually(Equals(expected['icon_name'])) [01:01] in this line, is the first argument "indicator.get_icon_name()" periodically re-called and compared to expected['icon_name']? [01:01] balloons, https://wiki.ubuntu.com/QATeam/Roles/TesterN is better now [01:01] I mean, I understand the intent, that's clear & readable enough [01:02] I'm just wanting to confirm which parts of this line get re-evaluated during Eventually [01:02] balloons, and as i go with this, i realize we shouldn't try to teach the people reading these role descriptions what's involved with each activity or task, that they should be in a separate place, like the tracker guide [01:03] charles: so.. [01:04] charles: yes, if you pass a callable, it will be re-called inside the Eventually() matcher. HOWEVER, we realised that was a bad idea, so it's best not to use it [01:04] as you point out, it's not very clear [01:04] it also breaks the matcher API :D [01:04] get off this team [01:05] ;) kidding! with love, always :) [01:05] charles: however, you can use testtools.matchers.AfterProcessing [01:05] alesage: O.0 [01:05] charles: does that help? [01:07] thomi, yep, totally does [01:13] balloons, concerning https://wiki.ubuntu.com/Testing/VirtualBox ... i would think it's mostly useless now (could be replaced with a video tutorial on how to set up a virtual machine, which even probably exists already) [01:14] balloons, the streamlining section isn't useful as is, but as i said, we might want to cover managing the virtual machines *somewhere* (or maybe not, people will learn workflows that fit them easily) [01:15] balloons, but the one thing (tm) we could snatch from that page is the guest addition thingy - maybe we should cover how to install them briefly [01:16] balloons, all that being said, i'd just redirect that page to https://wiki.ubuntu.com/QATeam/DevelopmentSetup [01:17] can somebody confirm if you still manually need to add users to 'vboxusers' on first install? [01:18] you don't [01:18] ok, so redirect it is [01:18] oi, that page is quite old, but yea, info should be maintained there not within our docs [01:18] done [01:19] awesome [01:19] i mean ultimately, if people can't figure something out by reading the wiki, they will *ask* us anyway [01:19] or if they aren't bothered to do that, they aren't bothered to read a long wikipage either [01:19] there's also your favorite search engine, askubuntu, the forums, irc ;-) [01:19] irc ftw ;) [01:20] soo... [01:20] https://wiki.ubuntu.com/Testing/Virtual_Machines [01:20] redirect to same page? [01:21] meh, the sibling/children listings should not list redirected pages :) [01:22] hmm.. [01:23] care to mention and linking to kvm too? then yes [01:23] alesage, elopio, updates pushed for leo's round 2 comments [01:23] can do that, and vmware [01:24] charles, thx--veebers or thomi, more to say there? https://code.launchpad.net/~indicator-applet-developers/unity8/indicator-power-autopilot-test/+merge/247079 [01:24] knome, ahh right right.. [01:25] https://wiki.ubuntu.com/QATeam/DevelopmentSetup [01:25] updated that [01:25] ## page was copied from Testing/Testing_The_Devlopment_Release [01:25] ## page was renamed from Testing/Virtual_Machines [01:25] ## page was renamed from Testing/QATracker [01:25] haha :) [01:25] my inbox is officially flooded [01:26] that happens when you let a wild knome loose in the wiki [01:26] https://wiki.ubuntu.com/Testing/InstallMethods [01:26] ancient, mentions alternate CD [01:26] that... doesn't seem to bring much to the table either [01:27] agreed [01:27] https://wiki.ubuntu.com/Testing/Includes - woot! [01:28] should i point that to the QA Tracker page? [01:28] alesage: ack, will hit it in a moment [01:29] alesage: lookinh [01:29] balloons, ^ "that" being "InstallMethods" :P [01:30] balloons, https://wiki.ubuntu.com/Testing/FixesToVerify hardy-age stuff, delete? [01:31] balloons, you're lagging :( [01:48] * balloons is sleepy [01:48] elopio, I'm having some trouble with paul's branch on mako devel-proposed, is this where you've been running it? not getting a unity8 launch [01:48] balloons, sleepy? it's almost 4am here. :) [01:48] knome, yes delete [01:48] oi, you're utc + 2? wow [01:48] yeppers [01:48] I thought +1 maybe [01:48] nah [01:49] physically live in +2, technically living in a different TZ... [01:49] righto [01:49] so what to do with QATeam/phillw? [01:49] what's https://wiki.ubuntu.com/Testing/Includes? [01:49] looks like it's a lot of screenshots [01:49] from 2008 [01:50] probably something that was related to the ancient testcase website [01:50] knome, hmm.. it's his I wouldn't want to remove it [01:50] then let's not [01:50] wow, https://wiki.ubuntu.com/Testing/Includes is wild [01:51] yes... [01:51] we don't need it [01:51] going then [01:51] need a redirect? :P [01:51] nah, I've never seen it [01:51] and are you sure, we can't get attachments back :P [01:51] bedtime for me knome, lol.. maybe not for you but :-) [01:52] well gone already ;) [01:52] old school photos! sad [01:52] hah [01:52] you can re-enact them! [01:52] bahahah.. it's still there: https://wiki.ubuntu.com/Testing/Ubuntu/Screenshots?action=AttachFile&do=get&target=Aboutme.png [01:53] my eyes! https://wiki.ubuntu.com/Testing/Ubuntu/Screenshots [01:54] that's a fun page to load [01:54] hahah [01:54] my bandwidth! [01:54] anyway, nighty ;) [01:54] let's get back to this tomorrow or sth [01:55] huhu [01:55] https://wiki.ubuntu.com/Testing/TestCaseAdmins [01:55] i'll redirect that to QATeam/Contact [01:55] we'll need some reorganization there anyway [02:05] nighty! [02:05] --> [02:32] alesage_: still around? [02:32] thomi yes [02:32] strange, I got booted [02:32] can I ask you some questions about that merge proposal? [02:32] thomi go ahead [02:33] don't know if charles is around also [02:33] charles: too, I guess :D [02:33] so, the MockBattery class [02:33] may need to catch up on that [02:33] yo [02:33] I'm a little concerned that it should just be a function instead. What's the rationale for making it a class? Why not just have a function that takes three arguments? [02:33] heh [02:34] thomi wouldn't that be a new pattern for us, if it's a 'helper'? [02:34] in general, a class with an __init__ and one method is like a massive warning bell for me [02:34] alesage_: I don't think it's controversial [02:35] remember that you can always use functools.partial to bind the first N arguments, ifyou want to call it many times with different values for the third argument [02:35] thomi, it was kind of a compromise approach: [02:36] thomi, elopio's suggestion was that the battery be owned inside the MockUPower class, so that some of the complexity such as its object_path can be encapsulated there [02:36] which makes sense [02:37] but, that limits some of the tests that we might want to write with MockUPower, e.g. on laptops we want to be able to mock two batteries [02:37] hmm, ok [02:37] so MockBattery is there to enapsulate some of the dbus impl, but also to leave a little wiggle room for future tests [02:37] thanks [02:44] alesage_: charles: review posted. still has a few issues, but much closer now :D [02:44] thomi thx for that [02:44] thomi could I get an "in principle" kind of review for this pdf-generation in a few? [02:44] alesage_: charles: one thing I didn't mention is that you fail pep8 in quite a few places [02:44] alesage_: sure [02:44] alesage_: not sure what that means, but I'll do my best LD [02:44] :D [02:45] that was kind-of a tears of laughter emoticon [02:46] alesage_: I've been running the tests in krillin unity8 devel-proposed. What is your error? [02:46] elopio, I just don't get a unity8 session on mako, updated and am trying again in a sec [02:47] This will be an interesting question. Not good with words. How is one suppose to file a bug found in an iso if it was never installed? [02:47] I was testing the ubuntu server i386 and was unable to install the system as one of the test requirements. [02:49] veebers: ping. About the closing parenthesis... [02:50] elopio prognosis better after reflash :) [02:50] veebers: we have already discussed about it, but I'm not sure if you were part of that team. So feel free to reopen the discussion. [02:50] what we decided at that time was that, as pep8 mentions both styles as valid, we can use anyone. But trying to be consistent with what the rest of the code does. [02:51] thomi, I was using to flake8 to look for pep8 failures, and only see one warning. To what are you referring? [02:52] elopio don't mean to keep you up but could we get some responses to charles' questions? https://code.launchpad.net/~indicator-applet-developers/unity8/indicator-power-autopilot-test/+merge/247079 [02:52] alesage_: of course. [02:52] charles: http://pastebin.ubuntu.com/9813456/ [02:52] elopio: ack, fair enough, I'm pretty certain that 99% of autopilot code puts the parens on it's own line (i.e. that's what I'm used to at least) [02:53] I agree that either are fine and that being consistent is the considering factor [02:53] thomi, should we do pep8 fixes to code that isn't otherwise touched in the MP? [02:54] veebers: yeah, and being consistent with what is an issue. In unity8, a good bunch of code have the closing parenthesis on the same line as the last argument. [02:54] thomi, this MP doesn't touch tests/autopilot/unity8/shell/ [02:54] veebers: I actually don't care. If you prefer, I can follow the autopilot style everywhere, starting from this branch. [02:54] IMO that would be more cleanly handled by a different MP [02:55] thomi https://code.launchpad.net/~canonical-platform-qa/autopilot/pdf-build/+merge/247249 also veebers elopio [02:55] elopio: oh wait yeah you're right this is for unity8 rats sorry. That's all good then :-) [02:56] elopio: remind me which MP this is for and I can re-review and approve if you like [02:57] alesage_: whitespace changes make me crazy :( [02:57] thomi I feared as much [02:57] alesage_: so.. please remove. as to your problem... [02:58] balloons: if you're still around and have a moment could you please re-visit this MP (after our discussion before) https://code.launchpad.net/~veebers/autopilot/adding-run-instructions-readme/+merge/247088 [02:58] Make isn't bash (although it is fairly close). You can byass the problem with "sh -c find ...." [02:58] thomi of course, have tried, get hung up on the "$1" item within, doesn't parse [02:59] thomi those are em-dashes or some such being changed to hyphens [02:59] alesage_: wha? [02:59] why? [03:00] veebers: https://code.launchpad.net/~canonical/unity8/pair_with_paul/+merge/246330 [03:00] thomi LaTeX error log concerned not supporting that unicode char [03:02] alesage_: remind me why we care about PDF output again? [03:02] elopio: coolio thanks. Also, did my email make sense or did I miss the point [03:02] thomi hah [03:02] thomi ask our stakeholders? balloons ^^ ? [03:03] alesage_: otherwise, this looks promising: https://sites.google.com/site/nickfolse/home/sphinx-latexpdf-output-with-svg-images [03:04] alesage_: charles: thomi's comments make sense to me, and I think I have nothing else to add. [03:04] just that this will need to be merged with the display branch. Well, the one that is landed last needs to be merged. [03:05] thomi I think that's effectively the same, making pdfs out of .svgs, i.e. depending on inkscape instead of imagemagick [03:05] veebers: yes, your mail makes sense. [03:05] which reminds me that I need to add that build-dep [03:05] thanks. [03:05] alesage_: right, but that page shows how to do it in a Makefile [03:05] elopio, thomi, thanks [03:05] thomi ah but yes the Makefile schizzni would be helpful [03:05] working through thomit's comments now [03:05] alesage_: which I don't understnad, but you should be able to copy that as a starting point [03:05] charles: no problem [03:06] funny that was tedg's suggested solution, to invest more heavily in make :) [03:06] thomi ok otherwise please do give more comments, may not get to tonight [03:10] alesage_: charles: actually, I'm missing some things: [03:10] Rename Indicator(main_window, name) to PowerIndicator(main_window) [03:11] Move it so it to the indicators module so that it can be called as indicators.PowerIndicator. [03:11] Get back the py2-compatible super calls. [03:11] you already had some of this things, it seems that on one review we ask for them and in the next we ask to revert them. Sorry about that. [03:12] elopio, particularly as it's late, it'd help to get the definitive review first ;) [03:14] alesage_: definitive, we need py2 compatibility because we are installing the package in the py2 path. [03:14] about removing the helpers module, veebers agreed so if you agree, lets do it. [03:15] about renaming Indicator to PowerIndicator, that improves readability. But we haven't discussed this one, so feel free to tell me if you think it's not good. [03:32] another round of feedback fixes pushed for https://code.launchpad.net/~indicator-applet-developers/unity8/indicator-power-autopilot-test/+merge/247079 [03:32] come at me! :D [03:32] lol [03:33] now it's a challenge :D [03:34] charles: I think the get_icon_name method on diff line 32 would be nicer if it actually returned a list, rather than a comma delimited string - what do you think? [03:35] what's more, icon_matches wouldn't need to change at all [03:35] Contains() does a substring search on strings, and looks for matching list items in a list, yes? [03:35] I like that suggestion [03:35] Contains() doesn't, but the 'in' keyword does [03:36] hmm, actually, maybe Contains() does [03:36] charles you got there: the uncertainty threshold [03:36] charles: also, multiple imports (diff line 57) should be split over multiple lines inside () [03:36] thomi, do a thought experiment for me [03:37] alesage_: I agree with your other comments in the pair_with_paul branch. Hopefull he'll fix them before I get in tomorrow :) [03:37] if Contains() didn't behave like that, how would you express it s.t. Eventually works [03:37] charles: I thought there might be ListContains() or similar, but I was too lazy to look it up [03:37] alesage_: I left a comment on why I don't agree about rotation_lock_is_enabled. [03:37] current line is self.assertTrue(indicator.get_indicator_name, Eventually(Contains(expected['icon_name'])) [03:38] elopio, saw that, thanks for the reminder [03:38] charles: yeah, should continue to work [03:38] is there a way in python to wedge an "in" in that assertion s.t. Eventually would work? [03:39] charles: Contains('foo').match(['foo', 'bar']) [03:39] you're good to go [03:40] hmm [03:40] actually after all that I'm going to walk it back, we shouldn't split that string into a list [03:41] charles: why not? [03:41] IMO it *is* a list (you even call it a list in the docstring) [03:41] that string is GIcon's semi-private format [03:41] the fact that it's human-readable as a string is a side-effect [03:41] well, in that case your entire test is based on that side-effect [03:42] or... do you think it liekly that the ',' will change to '|' or something? [03:42] thomi when's your EOD? [03:42] alesage_: very soon now [03:42] thomi, I don't think the delimiter will change; I'm thinking more of the leading "image://theme/" [03:42] thomi noooo [03:43] charles: ahhh, ok, good point [03:43] thomi anything else on the pdf MP? [03:43] alesage_: sorry, did you hve an updated version? [03:43] thomi nope [03:43] nothing else after I sent you that link [03:44] thomi can you predict whether or not there will be more? [03:47] thomi, multiline import tweaked to do pep 328 [03:47] alesage_: sorry, I don't follow [03:47] alesage_: the MP seems fine, except for... you know... it not working :D [03:48] thomi ok got it thanks [03:49] charles: LGTM [03:49] thomi, you should mark it as Approved before you EOD then [03:49] there's not enough green on that page [03:49] woo [03:50] charles: I have! [03:50] \o/ [03:50] charles hi-five o/ [03:51] * alesage_ is 'left hanging', in US slang [03:56] alesage_: :-( [03:58] \o [03:59] :( [03:59] hah, that's even better ^_^ === chihchun_afk is now known as chihchun [09:26] balloons: re what to use - tested disks in trusty/utopic and vivid - all bootable for me [09:39] elfy, did you see some updates on the wiki ? :P [09:39] nope [09:40] start with these [09:40] https://wiki.ubuntu.com/QATeam/ [09:40] https://wiki.ubuntu.com/QATeam/DevelopmentSetup [09:40] https://wiki.ubuntu.com/QATeam/Roles/TesterN (WIP) [09:41] heh [09:41] just about to wander to work ... [09:41] i was up until 4am or sth, had to stop because balloons went to bed [09:41] :) [09:42] mmm [09:42] anyway, have a nice day @ work [09:42] and tell me if DevelopmentSetup answers some of the questions for xubuntu :) [09:42] I did say last night that the ubuntu tool didn't work doing a trusty burn of a vivid image [09:42] i noticed that [09:43] as you can see, the USB part is still quite barebone.. [09:43] good enough if it works :) [09:43] but i would hope to have the one place (tm) for those instructions, so if there are exceptions like that, we should probably mention them there [09:44] was there a workaround for that? [09:44] or recommend something that works ... just saying :) [09:44] or was the workaround to just use another tool? [09:44] i've had both good and bad experiences with SDC, but when it works, it's a nice tool to use [09:44] yea - same as any workaround for the com32 error - then use live or unetbootindefault depending on what used [09:45] mhm [09:45] yep - agreed [09:48] bbl [09:48] yup [09:48] hf [11:44] jibel: are we still doing automatic EFI install/boot tests? [11:45] jibel: I just installed current vivid with "-bios OVMF.fd" and the omvf package; install works fine, but boot fails as it can't find ubutu [11:45] jibel: do we use the omvf package, or a download from tianocore? [12:09] pitti, I didn't check if it is still running since the move. I was using tianocore [12:19] jibel: hm, I tried with manual partitioning and automatic, and fails to boot both times :/ [12:19] * pitti tries with a different BIOS [12:37] pitti, in a VM with vivid? [12:38] jibel: right, with today's daily [12:39] * jibel tries [12:39] jibel: boot from cd works fine, and the install works too, but booting the installed system doesn't [12:39] jibel: don't waste any time on it for now please; I mostly just wanted to know which OVMF.fd we use in the automatic tests, and whether they still run [12:41] pitti, well, I've known good BIOS, if the image is broken it's good to know and report it [13:19] pitti, you tried today's desktop image? installation fails for me because extra.ubunut.com doesn't exists [13:19] extra.ubuntu.com* [13:20] * jibel tries without network [13:20] I also had a crash in ubi-console-setup [13:26] pitti, it's definitely uninstallable. I'll try with a server image. [13:42] jibel: hm, console-setup was fixed with the latest casper [13:42] jibel: are you sure it's from today? that fix landed earlier this week [13:42] jibel: I installed with network [13:43] pitti, ah current has not been updated for the past few days [13:43] I'll try with pending/ [13:44] jibel: oh, I see -- I changed my rsync script to /pending a few months ago [13:44] and then forgot about it [13:46] pitti, promotion to current failed because systemd is uninstallable [13:46] Jan 22 07:00:57 in-target: The following packages have unmet dependencies: [13:46] Jan 22 07:00:57 in-target: systemd : Depends: libsystemd0 (= 218-3ubuntu3) but 218-3ubuntu4 is to be installed [13:46] http://d-jenkins.ubuntu-ci:8080/view/Vivid/view/Smoke%20Testing/job/vivid-server-amd64-smoke-default/84/console [13:46] err, huh? mirror out of date? [13:47] how odd -- how would it have different versions of binaries for the same source [13:47] this is with build 20150122 [13:47] on amd64 [13:48] pitti, actually this is for server, for desktop it's different [13:48] pitti, looks like a CI issue [13:49] * jibel sees utah in the log and runs away [13:52] pitti, so nothing to do with the image there is a connection refused to a server in the logs. [14:19] pitti, this bios works with latest server and desktop images http://people.canonical.com/~j-lallement/junk/bios.bin.20141014 [14:20] pitti, the version in the archive has to be refreshed I guess [14:21] jibel: I got http://cznic.dl.sourceforge.net/project/edk2/OVMF/OVMF-X64-r15214.zip and that failed too; that seems to be the latest version [14:22] jibel: thanks, I'll try your's then [15:42] elfy, knome thanks for the updates. And yes elfy knome went through the wiki with a fine toothed cinv [15:42] *comb === chihchun is now known as chihchun_afk [16:05] rhuddie: I did the comments again on your process control branch. [16:05] balloons: saw that :) [16:06] elopio, oh thank you. I'll take care of that. weird they seemed to disappear! [16:07] I don't like inline comments, I'm not sure why I'm still doing them. [16:09] elopio, yea I always feel like people won't see them [16:12] balloons: can you please merge this one with trunk? [16:12] https://code.launchpad.net/~nskaggs/autopilot/apdocs-restruct/+merge/246599 [16:13] elopio, sure.. I rebased all my branches last night, do I need to do so again/ [16:13] balloons: not sure. The failure says 11 hours ago. [16:13] I'll re-merge them all [16:13] balloons: that's the last one of yours in review. [16:14] elopio, ahh, that'd be the ticket then. [16:14] I have this too: https://code.launchpad.net/~nskaggs/autopilot/apdocs-install/+merge/247232 [16:15] yea, since the page object one went first it messed up the others [16:15] I should have had it depend on the first [16:16] elopio, ready for another review https://code.launchpad.net/~indicator-applet-developers/unity8/indicator-power-autopilot-test/+merge/247079 [16:16] balloons: that one doesn't have a card. I'll review it. [16:16] elopio, yea, more bug fodder I found while diving in. I needed to close it so I did [16:18] balloons, my work is not done yet, but looking good so far ;) [16:25] elopio, I rebased both mp's, should be good I hope [16:25] brb [16:25] elopio, bah, sphinx warnings after merging :-( [16:27] elopio do we need another review of paulliu's branch from our team, I wonder? https://code.launchpad.net/~canonical/unity8/pair_with_paul/+merge/246330 [16:37] alesage: oh, they approved it without making your two changes. [16:38] elopio, they were teensy style things :) [16:38] elopio, more importantly we need to get charles' branch approved, can you review? [16:39] alesage: I'm on it. [16:40] elopio, ok again I believe everything should be ready on my remaining two mp's. [16:41] balloons, needing reviews? [16:41] alesage, yes; https://code.launchpad.net/~nskaggs/autopilot/apdocs-restruct/+merge/246599, and then https://code.launchpad.net/~nskaggs/autopilot/apdocs-install/+merge/247232 which depends on the first [16:42] ohh, you approved the one already [16:42] balloons, thx for the convenient linkings [16:48] elopio: I fix one of the changes. [16:49] elopio: But for the _is stuff. there are actually 3~4 there already. start with "_is" or "is". [16:49] paulliu: oh, thanks. [16:49] paulliu: actually, the recommendation is to start with is. So they are ok. [16:57] elopio, what is happening with meetings today? do we still have a sprint review meeting starting in an hour? [17:05] rhuddie: just figuring that out. [17:05] rhuddie: I'll move our delivery meeting to tomorrow. Lets just do the retrospective today. [17:05] that way I'll have time to merge the pending things. [17:06] balloons: charles: can you meet in one hour for the retrospective? [17:06] elopio, I planned to be around for the first half of the meetin' [17:08] balloons, when do you plan doing the merging work for the tracker wiki pages? [17:09] elopio, ok. if we're starting at the earlier time in about 50 mins I'll need to go now and will be back then [17:09] balloons: the retrospective shouldn't be long. [17:10] elopio, last fixes https://code.launchpad.net/~indicator-applet-developers/unity8/indicator-power-autopilot-test/+merge/247079 [17:10] alesage: give me a minute. [17:11] knome, I 'm happy to be around and help when it's convenient for you. Re-doing the video tutorial is ?, but the rest should be quite doable [17:11] balloons, i thought you would do it [17:12] balloons, you seemed to have a clear image of what needed to be done [17:12] knome, sure, happy to. I guess it doesn't matter if you are around when I do, you can see it after, heh [17:12] (for once i know that the screenshots need to be smaller on the wikipage to be followable, but i can take care of that... [17:12] * alesage pokes elopio repeatedly [17:12] so I'll do the rough edits [17:12] i'm on and off all evening and into the night, so feel free to ping me if you need any feedback [17:12] perfect thanks knome [17:13] poke poke poke [17:14] alesage: looks good. [17:14] yayz [17:14] alesage: now we need to figure out which should be the prerequisite of which. Heads or tails? or as paulliu's was approved 1 hour ago, it should go first? [17:15] elopio, I'm already dancing in the endzone [17:16] elopio, actually I'm not a football fan, for the record [17:16] elopio, no matter [17:16] I don't understand why you use your hands while playing football, but I like it. [17:20] balloons, i think https://wiki.ubuntu.com/QATeam/Roles/TesterN is as good as it'll come for a while [17:21] balloons, happy with it? (eg. should i copy it over the Tester page) [17:21] balloons, and/or should Ubuntu Touch testing be a main activity or a supporting task/activity? [17:21] knome, yes. I'll likely do some more tweaking work at some point [17:22] unity8 testing should move to main likely [17:22] ok [17:22] the unity8 pages needs some wiki page cleanup of it's own [17:22] yep.. [17:23] ok, I'll start on the qatracker merge now; I have a few mins [17:23] heh [17:38] balloons, do you have time to give this a test drive? pdf output https://code.launchpad.net/~canonical-platform-qa/autopilot/pdf-build/+merge/247249 [17:38] charles: ping. Will you be available to meetin in 20 minutes? [17:39] s/meetin/meet [17:39] elopio, sure === pgraner is now known as pgraner-otp [17:39] note to self, take holidays on thursdays :) [17:40] elopio, hey have time to try this pdf review? https://code.launchpad.net/~canonical-platform-qa/autopilot/pdf-build/+merge/247249 [17:41] * alesage still needs coffee [17:41] alesage: can you add to the docs how to build the pdf? [17:41] and there are some changes on docs/man.rst on your diff that I don't understand. [17:42] elopio, I'll mark up the MP [17:42] alesage: why are you leaving that #find commentedd out? [17:43] elopio, corrected, pushed [17:44] ok, let me give it a try, because I don't understand it. [17:45] elopio, comments on MP [17:46] alesage: my imagemagick crashed. [17:47] elopio, ok that's bad news :) , does it convert manually? [17:47] alesage: do you have a command for me to try? [17:47] elopio one sec, just want to test again [17:49] convert foo.svg foo.pdf , or find ./docs/images -name "*.svg" -exec sh -c 'convert "$1" "${1%.svg}.pdf"' _ {} \; elopio [17:54] veebers my good man, quick review my MP! https://code.launchpad.net/~canonical-platform-qa/autopilot/pdf-build/+merge/247249 [17:54] veebers: and, please merge this one with trunk [17:54] https://code.launchpad.net/~veebers/autopilot/adding-run-instructions-readme/+merge/247088 [17:54] :) [17:54] what a warm welcome. [17:54] nothing says happy friday like being slammed with reviews before you've had coffeee [17:54] thomi too! https://code.launchpad.net/~canonical-platform-qa/autopilot/pdf-build/+merge/247249 [17:55] actually out of coffee here [17:55] where by coffee I mean diet cola [17:56] cutting it a bit fine aren't we alesage? ;) [17:56] alesage: I think you need a packaging review [17:56] thomi from whom should I procure such? [17:56] alesage: a DD or UD [17:57] plenty of choices at this time of the day :D [17:58] hey alesage o/ [17:58] veebers, woo! [18:01] elopio, not seeing the meeting in my calendar -- got a link? [18:01] wait, is the meeting in 2 hours or now elopio ? [18:01] it's now? [18:01] brillant, others have come [18:01] charles: balloons: alesage: veebers: thomi: https://plus.google.com/hangouts/_/canonical.com/qa-scrum-sprint?authuser=1 === pgraner-otp is now known as pgraner [20:18] balloons: so - what do we do, if sdc is not working to produce bootable vivid images currently - let it go and use the other one which does appear to be a constant? [20:19] the other one? [20:19] unetbootin [20:19] Letozaf_, howdy [20:19] did you see the troubles with rssreader? [20:20] balloons, https://www.youtube.com/watch?v=rqAkPZgIGjU [20:20] no knome - not unetbootin [20:20] oh? :) [20:20] what other one then... [20:20] gnome-disks [20:20] uhh [20:20] i've never even heard that name, promise! [20:20] must have been reading with your eyes shut then [20:21] i haven't always been logged in this channel, nor ever read most of the discussions :) [20:21] the most recent was at 20:02 UTC in x-offtopic :) [20:22] oh [20:22] balloons, hey [20:22] welllll [20:22] balloons, yes I just ran the tests now [20:22] balloons, the work fine on my desktop [20:22] elfy, lol, yeah, referred as "the disks thingy" :D [20:22] yep - that one :D [20:22] balloons, I had seen something with Roman yesterday evening [20:23] * balloons listens [20:23] balloons, I will review the mp now, looks good to me [20:23] knome: obviously I'm only one person testing it - but it did work properly and constantly for me - the one time I did it from each :) [20:23] ;) [20:24] which is more that can be said for anything else I tried yesterday/today other than dd [20:24] yeah [20:24] and I'm not recommending that to people [20:24] lol elfy . If it doesn't work, I guess it doesn't work. I'm not sure what to say about that. I'd prefer we pointed people at the official tool.. and it *should* work [20:25] well yea [20:25] so would I [20:25] but I can't as it doesn't work :) [20:26] balloons, how do you get tests to be launched by Jenkins ? specifically Shorts app ones I was curios to see if the worked there [20:27] Letozaf_, any new commit will trigger it, or I can do it [20:28] elfy, heh, but just for trusty.. [20:28] balloons: the main issue for me is that for *us* we have to tell people to actually install something, as neither of those are defaults for us - and I would rather recommend to *my* testers something I know works [20:28] balloons, do you think it's the case to see if this works on Jenkins ? : https://code.launchpad.net/~mrqtros/ubuntu-rssreader-app/ubuntu-rssreader-app-new-devel-period/+merge/244656 [20:28] balloons: yes - but we're possibly going to see people using the LTS [20:28] elfy, ahh right, indeed they have to install sdc [20:28] or gnome-disks [20:28] elfy, in theory although it is affected by a bug at the moment, it should work [20:29] elfy, of course. that's the #1 use case, the lts [20:29] yea - I realise that - but in less than a fortnight - pleia2 is running a ugj specifically for xubuntu testing [20:29] if we go that route, in theory, i am a millionaire.. [20:29] elfy, so I don't think an alternative is out of the question, just in case someone encounters a bug [20:29] that's my priority currently [20:30] which at the moment they most certainly would [20:30] elfy, ack [20:30] elfy, yes I know too. [20:30] balloons: yep [20:30] Letozaf_, do they work running in a sandbox? [20:30] balloons, let me try that now [20:30] Letozaf_, you can see the one failed test on that mp.. it should be good though right? [20:31] elfy, so I don't like unetbootin, but otherwise I'm impartial I guess as to the alternative [20:32] balloons, yes I saw that yesterday evening with Roman and he fixed it, it's ok now [20:32] * balloons runs [20:33] balloons, autopilot is python, right? [20:33] balloons: I'm not going to recommend that either - currently 2 (3) things work for me regardless of where I burn it from - dd (and mksub) and gnome-disks [20:34] balloons, so yes tests run without failures also in sandbox [20:36] Letozaf_, hmm, locally I had 2 failures actually [20:36] * balloons re-runs with logs [20:36] knome, yes autopilot is python [20:37] balloons, where ? I did not have failures I ran the tests on desktop and desktop in sandbox [20:39] balloons, look like Jenkins passed the tests [20:39] looks [20:40] Letozaf_, so it might be occassional failures. I'll push to trunk, but we should get a bug filed and have a look at some point [20:40] Letozaf_, http://paste.ubuntu.com/9824606/ is my sandbox run [20:41] * Letozaf_ is looking [20:42] balloons, so looks like getting text field focus fails when adding a feed [20:43] balloons, but did you run the tests on a VM ? [20:43] veebers: instead of opening the file during the init and closing it on del, isn't it better to open and close it on every method? [20:43] balloons, I was wondering how to reproduce this error [20:46] elopio: perhaps, that is another good idea. It could still be a fixture though as that is a good pattern [20:46] balloons, oops now I saw you ran it in sandbox [20:46] balloons, should I file a bug or are you going to ? [20:47] Letozaf_, right I ran it in the sandbox, but the jenkins bug I suppose is relevant also [20:47] Letozaf_, go for it [20:47] veebers: that's the thing I wanted to ask you. Making it a fixture would be really nice to get it back to the original orientation. [20:47] however, according to ricmm, we can't query the orientation from the sensors library. [20:47] elopio: heh, as per my comment. I might be wrong there that it belongs at a higher level. It would be good to happen here [20:47] that's something that unity calculates. [20:48] elopio: hmm, perhaps then there can be a helper in unity (or somewhere) that wraps this and does the cleanup [20:48] veebers: right, sorry, you wrote your comment faster than my question. [20:48] you can have a fixture that has a fixture [20:48] heh ^_^ [20:49] balloons, you mean this failure: http://91.189.93.70:8080/job/generic-mediumtests-vivid/790/testReport/junit/shorts_app.tests.test_rssreader/TestMainWindow/test_edit_topic/ [20:49] veebers: that's what I was thinking. but then it doesn't make sense to return to the original orientation. It only makes sense to stop using the fake sensor and get back to the orientation the phone is in. I think. [20:50] elopio: hmm good point [20:50] veebers: that's why I wanted to put it in the test, to get a better picture of what to request. [20:50] I need some minutes. [20:50] ack nw [20:50] veebers: is there a way to get the process id from the proxy object? [20:51] elopio: um, yes? perhaps not. let me check the code [20:54] Letozaf_, yes [20:55] Letozaf_, you can probably generically say the trunk tests sometimes fail, and give some examples. It will need more investigation certainly [20:55] elopio: yes and no, this is a bad part in AP. If you use NormalApplicationLauncher the proxy object will have set_process called on it, but not for clicks etc. [20:55] balloons, oh, so I will report just one bug for both issues, right ? [20:56] veebers: um, this is started by calling initctl unity8 directly, which probably is wrong. [20:56] elopio: where do you see that? [20:56] anyway, I can get the pid. [20:57] veebers: unity8/process_helpers.py [20:57] elopio: ah right ok [20:59] Letozaf_, yea, I would. The bug is more or less "trunk tests are flaky" [21:00] balloons, ok thanks [21:02] did the wiki just die? [21:02] knome, not for me [21:03] i've been in the community wiki blacklist several times for "editing too fast" [21:03] :( [21:03] ahh.. hmm [21:03] balloons, bug 1413746 [21:03] bug 1413746 in Ubuntu RSS Feed Reader App "Autopilot trunk tests are flaky" [Undecided,New] https://launchpad.net/bugs/1413746 [21:03] probably not that this time, just lagging [21:03] I updated the community.u.c/quality page [21:03] we need an image for it.. any thoughts? [21:04] * knome ponders what a rural english "gentleman" could reply to that... "image my ..." [21:04] but seriously speaking, anything that isn't hosted on facebook ;) [21:05] knome, I'll have it site hosted :-) [21:05] the englishmens bottom? i'm up for it! [21:05] elfyyyyy [21:06] I was born in Malta ... [21:06] does that make your bum as smooth as baby's? :P [21:06] might have done years ago :) [21:07] huhu [21:07] elfy, really?! [21:07] really which? [21:07] elfy, hah, the malta thing [21:07] yep [21:07] you know what's weird? when your ringing tone starts playing randomly from the music player [21:07] balloons, lol! [21:08] elfy, and you too.. ;) [21:08] balloons: parents in the RAF - born in a RN hospital [21:08] :) [21:16] balloons, https://wiki.ubuntu.com/QATeam/Roles [21:16] WHHOOOOOP [21:16] sorry for breaking the monotonous marketing speech page :( [21:18] too much grey [21:19] absolutely [21:19] that's a disgrace for the whole team :( [21:19] i'm so sorry for that [21:19] should be more blue [21:19] :p [21:19] knome, I'm loving all the changes [21:19] or teal ... [21:19] it's like you have a magic wand . . . [21:19] balloons, you know what you have to do next? [21:20] get the content in shape in all pages we are linking from the role subpages [21:20] it's worthy of it now! [21:20] definitely [21:21] balloons: well I'll hammer at the manual testcase pages at least [21:22] knome: ^^ [21:22] elfy, that would be lovely.. As you may have noticed, we've been working on fixing the automated testing docs and the developer.u.c website [21:22] so it's documentation month it seems :-) [21:23] I'm afraid I'd not have notice that :( [21:23] dragons there for me ... [21:23] elfy, ahh right, why did I saw you would've.. silly me [21:23] elfy, you're welvome :) [21:23] ugh [21:23] welcome too! [21:24] heh [21:25] elfy, ok circling back quickly though.. the image testing stuff. The wiki is in better shape, but the tool concerns. I guess we all agree we should recommend a single alternative tool. is it gnome-disks then? [21:26] if yes we should recommend an alternative, and gnome-disks is it, let's get to editing those wiki pages then. I need to update the ugj/testing page also [21:26] balloons: I'd guess that once we're sure it works - then sdc, but currently it's causing issues so - gnome-disks [21:27] elfy, do we feel we need an alternative in the wiki? [21:27] it's still possible for sdc to be fixed in time [21:27] right [21:27] but beyond this incident, do we think it's needed? [21:28] I don't think that an alternative is a bad idea [21:28] and given the problems with unetbootin and dd perhaps being a bad idea for someone who *might* be new(ish) [21:29] gnome-disks would be a sensible alternative [21:33] elfy, ok, so we should mention it on https://wiki.ubuntu.com/QATeam/DevelopmentSetup [21:33] yep [21:41] balloons: other than the little bit on https://wiki.ubuntu.com/Unity8DesktopIso I can't find any docs for gnome-disks [21:43] elfy, lovely; heh [21:52] I'd add something - but I hesitate to make knome's nice page look all used again :p [21:53] done for today anyway - back tomorrow :) [21:53] I'll start looking at those manual testcase pages tomorrow too [21:54] i'll fix it if you spill the milk on it ;) [21:54] nighty! [21:54] nighty night! [21:54] * balloons is editing already [21:54] nice [21:55] elfy, I should have a fixed up ugj page for you to look at in a second [21:55] I'll hang about for a short while then ... [22:03] can't wait - I'll look in the morning :) [22:04] hehe [22:04] good night elfy [22:04] elfy, :-) [22:04] elfy, have a look then: https://wiki.ubuntu.com/Jams/Testing [22:05] there's some cleanup to do, but that's more or less it [22:11] balloons, i would try to cut down on the headings [22:11] there are too many of them now, and it's hard to follow a page which is formatted like [22:11] HEADING [22:11] 1 line paragraph [22:11] HEADING [22:11] 1 line paragraph [22:11] etc... [22:12] knome, I can agree. I've been cutting content and trying to make it flow better. Feel free to edit it, but I think I'm going to let it sit for a bit and come back with fresh eyes [22:12] there's still stuff out of place and stuff to merge [22:12] I however lack your magic wand to make it look pretty :-) [22:13] the old page was horribly long [22:13] thanks for looking! [22:13] hah! [22:13] let's see some time ;) [22:13] i guess my main interest lies in the wide end of the branch... [22:13] ..to get the first few steps right [22:14] in a way, the stuff behind that are more or less area for longer descriptions [22:14] and that's fine... but i don't feel like i know enough of them to write them [22:14] yes, very true.. just need to make sure the net is proper [22:15] nor do i think every one of those pages needs to look awesome [22:15] tehe [22:15] as long as the content is in good shape and it flows well (visually as well) === txspud is now known as txspud|away [23:47] veebers: on your display proposal, you mention an __orientationAngle [23:47] how do I get it?