=== vrruiz_ is now known as rvr === _salem is now known as salem_ === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [13:12] from -bugs: [2014/10/01 06:14:54] Dear all! Today I have installed Ubuntu Utopic from beta2 CD and discovered that many bugs are still not fixed. These bugs are: 1245137, 1242572, 1263228, 1244090, 1270574, 1270579, 1280759. [13:12] (current time is 09:12 UTC-0400, here, for timestamp reference) [13:16] it would be really useful if they sent a test report to the ISO tracker [13:17] you can forward them here if they need help in doing that === chihchun is now known as chihchun_afk === _salem is now known as salem_ [14:44] knome: if they were still on i would, i'll keep an eye in -bugs [14:44] thanks [14:45] knome: i think the two channels should maybe merge, or at least have relays, but meh. *shrugs* [14:45] yep. [14:45] since they kinda end up in the same radar coverage closer to release times, in a manner of speaking [14:46] agreed [14:47] i wonder how big the overlap is [14:48] knome: i think there was discussion about an irc channel merge a while ago, I forget where that went... [14:49] knome: i know that i see a lot of bug issues for utopic in my radar, but most aren't ISO issues [14:49] i think it was the usual "didn't get to it" [14:49] (two were, one was a VBox install issue, one was that last one i copied) [14:49] i mean overlap on the people front [14:49] mmm [14:49] i lurk here, for the occasional triage that overlaps, but other than that I dunno the overlap [14:50] but also the issue coverage starts to overlap near release time with ISO tests and such and bugs in the images [14:52] absolutely [14:53] i ignore most of the discussions on this channel anyway (since they aren't related to xubuntu), and even i wouldn't mind some more discussions here if even one of them occasionally helped the xubuntu team [14:53] at least all the quality-related information would go through one channel then [15:10] mhm === chihchunl is now known as chihchun === wxl_ is now known as wxl [16:28] evening all [16:30] teward: that one was also in +1 saying exactly the same thing [16:30] I for one don't see that merging this channel and -bugs would be useful [16:31] -bugs? that would be evil [16:32] well I'm glad it's not just me then :) [16:33] there needs to be something saying "Issues with installing on the Ubuntu+1 Image or ISO? #ubuntu-quality" or something :P [16:33] in -bugs, so that those related things end up HERE [16:33] ubuntu-qa: can I get a review here, please? https://code.launchpad.net/~canonical-platform-qa/ubuntu-ui-toolkit/fix1365674-optional_containers/+merge/236745 [16:33] teward: maybe so :) [16:34] elopio: I will, I have been triaging so it will be nice to look at some code. [16:34] bugs is too general, and about already released software wouldn't really quality here so much [16:34] but for pre-release, defently in here [16:35] robotfuel: It's not interesting code, sorry :) [16:35] :) [16:36] but then again - if someone comes here talking about a bug in +1 - is there anyone actually about here who's going to a)be able to help b) be available to do so [16:36] I know that in general if it doesn't affect Xubuntu I've no interest [16:38] teward, elfy, knome I've been following along to some extent, but if there is something worthy of change, you know I'm happy to help make it so [16:38] well perhaps there's a need to point people here to discuss +1 bugs [16:39] changing channel headers is certainly something easy we can do [16:39] I will say I generally don't follow -bugs.. to many channels as it is. So I didn't see the note about bugs still found [16:40] but as I said - if we do point people to here to discuss +1 bugs - will they be then talking to the wind [16:40] I don't think they find much of an audience in -bugs; unless it's related to bugwork itself and not the bug [16:41] mmm [16:41] so maybe the right place to discuss issues is the +1 channel then [16:42] elopio: it would be nice to have tests for containers=None and containers= [16:43] robotfuel: there are tests for containers=NotNone, that's how the other swipes work. [16:43] robotfuel: so should I add two more with containers=None explicitly? [16:43] elfy: i specifically meant the QA related ones, failed installs, ubiquity, etc. [16:43] you're right, all the +1 bugs would be a headache [16:44] elopio: if there are other tests that's okay, I just didn't see them at first glance [16:44] also the message from this particular user was along the lines of "there's loads of bugs still" with an underlying "someone else fix them" [16:45] teward: well up to a point those get discussed if there's a big issue - eg the vbox fail things [16:45] right [16:45] robotfuel: the swipe to top and swipe to bottom call swipe to show more with a value for containers. [16:45] one of those bugs is " [16:45] icons in indicators are small (difficult to view on FullHD display and smaller than was on 12.04)" [16:45] that's really going to get a discussion going :p [16:46] robotfuel: they don't have an explicit test, but they are already covered. Do you think that's enough? [16:46] teward: also - I've not seen this type of issue come up previously - does it happen a lot in -bugs? [16:46] elopio: I think an explicit test will be better, because someone might change them and it will be hard to tell that it's providing coverage. [16:46] hi PaulW2U [16:47] robotfuel: ack. On it... [16:47] elfy: recently [16:48] not normally i don't think [16:48] teward: right [16:48] oop, FWIW the beta image Fails To Install in VMware, but don't beat me for using something that i can easily configure VM LAN and firewall segregation for :P [16:49] (might be because E:LowSpecVM) [16:49] teward: :) I think it should be fixed tomorrow - was expecting it today [16:51] robotfuel: pushed. [16:53] teward, I actually tested the beta image on vmware and found it working.. I had to install vmware just to do it [16:53] anyways, not to stray the topic :-) [16:54] balloons: might be the specs on the VM, they're very low since my Lubuntu packaging environment VM is also running [16:54] (any more power to the 14.10 test VM, and E:CRASH) [16:54] s/CRASH/HOST_CRASH/ [17:18] balloons, the real solution to that problem? enough people in all timezones who follow -bugs closely enough and who can/know where to point people, including this channel [17:18] having two channels and a bot to proxy the discussions from one to another is the most disastrous option [17:20] one option that would require some work and would only potentially improve the situation would be a bot that would gather the bugs mentioned in channels X,Y,Z and then output them to either a website or channels A,B on regular intervals (like twice a day, or so) [17:20] that way we'd know what the "heat" for a bug in irc would be [17:34] can't help feeling this is a solution looking for a problem [17:53] elfy, heh, maybe [17:53] elfy, but it might also give some nice information we don't currently have :) [17:53] mmm [17:56] I'd rather have information of bugs easily available on the tracker - than random pings of bugs that *someone* thinks is really important - but isn't actually even reported on the tracker [17:57] not sure that a list of bugs is nice information I guess [17:57] though the point would be that if a bug is mentioned 3 times, it gets more heat than one that's mentioned one time [17:57] and furthermore, the bot could collect the information in whatever channels [17:57] I still think it's meh [17:57] we could even leverage one that follows just xubuntu channels [17:58] yeah, maybe [17:58] it's not a replacement to anything, that's for sure === salem_ is now known as _salem === roadmr is now known as roadmr_afk [19:17] ubuntu-qa: can I get a review here? https://code.launchpad.net/~canonical-platform-qa/ubuntu-ui-toolkit/fix1340227-expandable_autopilot_helper/+merge/236770 === _salem is now known as salem_ [19:36] balloons: bug 1376414 :/ [19:36] bug 1376414 in ubuntu-release-upgrader (Ubuntu) "upgrade from xubuntu 14.04 to 14.10 fails, no user feedback" [Undecided,New] https://launchpad.net/bugs/1376414 [19:48] dkessel, looking [19:50] knome, I think I may side with elfy on this one as a solution looking for a problem. It's great people share bugs, but we should make sure they are reported in the tracker. On the tooling side, improving the tracker bug displays probably makes more sense [19:50] we can certainly educate people to report bugs in this way [19:51] dkessel, your image is OLD! 20140316? [19:51] ohh, lol.. [19:51] nvm [19:51] balloons: ... :p [19:52] I was just trying to see if you used the daily or the beta [19:52] tried upgrading my real hardware which has 14.04 since march [19:53] balloons, sure, but as i said, the real solution (to get people report things on the tracker) is more people following channels, and that's hard to do [19:54] dkessel, so normally I would pass -d in the call since 14.10 is in development [19:54] ow did you call it? [19:55] update-manager -d -c as it says in: http://iso.qa.ubuntu.com/qatracker/milestones/315/builds/80270/testcases/1310/results [19:55] knome, ahh you mean getting people into the tracker to begin with. [19:55] sure [19:56] dkessel, I'm confused why the -c is in there. That only checks [19:57] bah, that run siwth update manager, not do-release-upgrade [19:58] is that a deprecated way of doing it? [20:00] balloons, yeah, but if that's not possible/doesn't work out, i don't think it *hurt* us if we got some bug mention stats. [20:01] the downside of that is that it takes some work to get done... but the upside in that work is that it's hardly dependent on anything, and anybody can simply jump in and do it. [20:03] dkessel, no do-release-upgrade is intended for servers [20:03] either way, it should only be -d [20:03] they have the same core [20:03] and mostly the same args.. but it's command line only [20:03] balloons: using "update-manager -d" gives the same result [20:03] maybe intended, but many people use do-release-upgrade with desktop installations, and it works just as well [20:04] ^ right, I use it [20:04] for everything === roadmr_afk is now known as roadmr [20:09] dkessel, ack thanks. I think we should file a bug to change the testcase. would you mind? [20:12] knome balloons - I just fail to see what we gain by spamming the channel with random bug numbers [20:13] shouting out is just one option, we can also make that on-demand [20:14] well whatever - I guess I'll just /ignore whatever bot it is [20:14] I'm not convinced either elfy [20:14] i'm just throwing ideas around [20:14] yea - I know that :) [20:15] i'm not forcing anybody to use anything [20:15] right.. :-) knome isn't selling anything, but is offering discussion which is good [20:15] I just fail to see why we even need it [20:16] this is just the result of one user wondering why a handful of bugs aren't fixed - the answers the same as it is for the majority of bugs [20:16] balloons: yea - I've not got a problem with it at all :) [20:17] i guess the problem is also that people do not even file bugs [20:17] balloons: i'm not sure... maybe the "-c" really _is_ needed... after changing the setting from step 2 [20:17] and unfiled bugs are harder to catch from any discussion... [20:17] having bugs pop up in a channel because someone somewhere else keeps pinging it doesn't achieve much imo [20:17] it's just random noise in a channel that 99% of people are just going to ignore [20:18] probably. but so is a lot on a lot of channels.. [20:18] it's just another way to measure another thing. [20:18] hey, you can call me stats junkie ;) [20:18] yea - but we're discussing one issue in one channel [20:18] dkessel, check the --help.. but yes, I could be wrong [20:18] heck, I probably wrote that testcase! [20:19] it doesn't measure anything other than someone pinging a bug number into a channel [20:19] if you write me a bot that counts the times i mention different board games and lists them by times mentioned, heck, i'm in! [20:19] balloons: the problem is I tend to be the one who fixes them and it always looks like I write the things :( [20:19] knome: lol [20:20] is it useful? not if you don't want to know which game is trending inside my head... [20:20] stats are great - when they are of use - otherwise they're just stuff [20:20] of course [20:21] and I don't see this being anymore than stuff :p [20:21] and it's completely possible that this list would more or less correlate with the bug heats on LP... but if not, THEN it's interesting [20:21] and yes, i said interesting, and didn't mean "useful" [20:21] heh [20:21] but it MIGHT be useful as well... ;) [20:22] i don't mind if we don't have that bot either... [20:22] I'd find trending bugs on tracker fairly useful [20:22] yep. [20:22] my point is: [20:22] but [20:22] while we should get everybody to use the tracker, not everybody does it, whatever we tell them [20:22] so there are always people who only mention bugs on irc [20:23] so [20:23] educate them [20:23] tell them not to spam the channel but to report them [20:23] whether it's a group that's big enough to be substantial enough to take into account is a different thing [20:23] that was my "real solution" #1 ;) [20:23] but that requires people who are hanging out at -bugs [20:23] and telling users that [20:24] apparently, there aren't enough people doing that since we get reports of people who said something on -bugs but didn't file a tracker report [20:25] once or twice [20:25] one or two reports more for xubuntu so far on this cycle would have been relatively quite big percentage (sadly) [20:25] that's not any different than foo jumping into xub-dev and asking why some bug they reported 3 years ago is still not fixed [20:25] would you want stats for that? [20:26] probably, because that would communicate which old bugs are still important for people [20:26] would it get fixed? probably not [20:26] so why should everyone else see the bug [20:27] at that point it's just spam - well intentioned but spam nonetheless [20:27] now you're talking about shouting out the bugs on a channel [20:27] it could just as well be a website [20:27] or a factoid-like thing you could get on demand [20:27] and what would this bot do - exactly the same - shout out bug numbers [20:27] that's just technical issues [20:27] with times mentioned since the beginning of the cycle [20:28] personally I think there are more pressing issues with QA reporting/websites than this [20:28] sure. [20:28] but that's what my other point was... [20:28] the QA website is not progressing because we have no people who are familiar enough with the technologies involved [20:29] yea [20:29] but maybe we could have somebody else free for work who knew something else, and could work on the bot or sth [20:29] not sure I agree with that [20:29] or rather the likely outcome [20:29] well, [20:29] i would be able to write the bot script. [20:30] but i'm not familiar enough with the tracker to actually change *too much* there. [20:30] if i had 2 hours of time i could contribute the project, but didn't want to run tests, i could write the bot. [20:31] (in theory! in theory! i have nothing against running tests!) [20:31] but is the bot going to be able to get information from where it needs to come from [20:31] like? [20:31] which IS the tracker and LP - not random pings in random channels [20:31] well, no, that's not the point [20:31] it *wouldn't* be a replacement for the tracker or the tracker improvements. [20:31] from launchpad... sure. [20:32] but what's the point of digging up launchpad stuff and making it available in IRC while everybody can just go to launchpad and get the information from there directly? [20:32] so - someone shouts about bug in libreoffice-impress in some channel and everyone gets to know about it ? [20:32] just forget that the bots needs to shout that out. [20:33] I don't know - not sure I care much tbh - I see no need for it [20:33] sure [20:34] and again, i agree that fixing the tracker would be a much better idea [20:35] but since i've literally waited over a year for some changes in the tracker, i'm not keeping high hopes up that it suddenly got improved by a lot. [20:36] ^ including the testcase not needing to be expandable (and hidden by default) [20:37] stgraber, ^ re: that, any time in sight that you could use on working on the UI changes for the tracker we talked about like a year and a half ago? [20:38] knome: hardly, I'm pretty much not home till December and I'm assigned fulltime on some fancy confidential projects... [20:39] stgraber, yeah, i wasn't expecting "tomorrow", just some timeframe we can somewhat rely on... [20:39] or if you have anybody else in mind who would be familiar enough with the tracker [20:40] or if it looks bad enough that it's fair to consider build that from the scratch up [20:40] or anything [20:40] because some tiny things that you mentioned that are easy to fix are being really annoying [20:41] the tracker has an api [20:41] balloons, that doesn't fix the tracker [20:41] * balloons is not following te conversation [20:42] lol [20:42] balloons, unless the api supports all operations that you can do on the tracker itself, and somebody is willing to write a separate frontend for it [20:44] sorry, what do you want, too many threads. The api can pull and push results [20:44] I never found something I couldn't do with the API [20:44] what do i want... i want something that works as the current tracker but has bits here and there fixed. [20:45] is there any documentation for the API? [20:47] http://iso.qa.ubuntu.com/qatracker/api [20:47] the docs are more or less the code and the example on that page [20:47] so no, no docs [20:48] The API isn't considered stable at the moment [20:48] yea, it all works fine, but since we can hack on it, I think that disclaimer is there to say we reserve the right to break things [20:48] ok, so [20:48] who would hack on it? [20:49] is it the same people who could/would hack on the tracker itself? [20:49] since it's python, it's not so insane [20:49] because at that point it doesn't make any difference if we fork their time to the tracker or the API [20:49] i can't python [20:49] would balloons be willing/able to hack on it? [20:49] if it's something critical I can try and give some cycles [20:50] I would certainly try [20:50] well as critical as any bug on the tracker is.. [20:50] heh.. If there was some missing data you needed exposed I assume that would be easily done [20:50] i mean the problem with simply using the API is that we would have to build another UI first [20:50] well again, I missed why you wanted the api, heh [20:51] i didn't "want" the API... [20:51] we were simply discussing the realistic timeframe when we would get things fixed on the tracker [20:51] and stgraber's answer was "definitely not before december" [20:51] so i brought up the question if it was sane to go another route [20:52] especially if the situation in the future regarding the time the developers could contribute would look as bad as it does now [20:52] to try to shift to something that's more maintainable for (much) more people [20:52] ahh, on fixing the tracker itself the key piece I undertook was getting a test setup. I've done that and need to document it and have someone else repeat [20:53] like simply getting the python API be stable, and separate the API from the UI [20:53] with that in place, it should be possible for others to contribute [20:54] even then i'd imagine it has a relatively high threshold for new developers to start [20:57] I don't think it's as hard as I thought if you know django. [20:57] but anyways, I will try and finish the writeup soon [20:57] balloons: ping me once you have that documentation. i will verify the steps. [20:57] in that case we should fiercely start looking for a django developer [20:58] dkessel, you have some django skills? [20:58] I followed this; but it was a bit off in places, so I'd like to clean the page up: https://wiki.ubuntu.com/Testing/ISO/DevEnv [20:59] please do.. [21:00] balloons: not really. but i want to improve my python and django is an mvc based web framework, i read. so, that is not exactly exotic. [21:00] dkessel, that would be awesome.. See, some further motivation :-) [21:01] one problem i see with the whole thing is that you need drupal 5. [21:01] dkessel, I will let you know when I'm happy with the cleanup.. [21:01] knome, that's one of the things, it uses drupal7 not 5 :-) [21:01] aha. [21:01] that's a relief. [21:01] ok balloons. i will see what i can do then :) [21:01] i mean, only slight! :P [21:01] i still don't like drupal... [21:01] LOL, poor knome [21:01] still, the js tweaks are useful [21:02] well, that's good === salem_ is now known as _salem