[00:37] balloons: still around? === blankley_ is now known as blankley [07:01] Good morning [07:03] morning jibel [07:04] morning elfy [08:20] morning all [08:25] Hello DanChapman [09:21] morning all [09:21] morning slickymaster [09:21] hi DanChapman === vrruiz_ is now known as rvr [09:31] Good morning [11:52] afternoon people [11:53] hello elfy [11:53] hi knome [12:14] hi elfy [12:14] knome: :) [12:16] elfy: can you tell me please when is going to be the meeting you mentioned on the xubuntu-devel mailing list, last 17th [12:17] slickymaster: my qa update mail? [12:17] that meeting was the same day as my mail - I left it as late as possible to be as up to date as I could be [12:18] elfy: my bad, i've misread your mail [12:18] slickymaster: logs not posted yet - but you can read them if you want at http://irclogs.ubuntu.com/2013/09/17/%23xubuntu-devel.html#t15:02 [12:19] there was talk of another yesterday - but that didn't happen [12:20] elfy: thanks, you're always helpful [12:20] I'll get up to date with those logs [12:20] not everyone would say that :) [12:21] * slickymaster thinks that everyone would be completely wrong if they see it otherwise [12:26] balloons: am I right in assuming that there won't be a mail from -release about Final Beta participation, assumption being everyone will? [12:26] slickymaster: next time I get a PM that doesn't say that - I'll show you a copy :D [12:27] elfy, technically, it is everyone "must":P (But not sure of Touch) [12:28] :) [12:28] elfy: well, here in Portugal we have an old saying that roughly translated goes like donkey voices do not reach the sky [12:29] I can understand the meaning behind that :) [12:29] I'm just glad the forum has a language filter lol [12:30] * slickymaster agrees [12:30] time for lunch I think [12:31] yeah, same over her [12:32] good lunch everybody [12:35] Good morning [12:35] morning pitti [12:36] hey DanChapman [12:37] elfy, slickymaster I just finished dinner:P [12:37] morning pitti [12:38] hi smartboyhw [12:38] elfy, you are hi-ing me again? [12:38] -.- [12:38] * smartboyhw thought we just talked about the Beta 2 stuff [12:38] wut? [12:39] elfy, technically, it is everyone "must":P (But not sure of Touch) [12:39] :) [12:39] (That's after I finished dinner:P) [12:39] oh ... [12:39] didn't read the name - thought that was slickymaster lol [12:39] elfy, ha ha [12:40] and it sort of made sense in that context too [12:40] well I've not seen anything that says everyone must [12:41] and why you're answering someone else's question I've no idea ;) [12:42] * smartboyhw finds "must" double-quoted:) [12:47] morning [13:04] elfy, yes I woul dsay you are correct. Final beta is everyone :-) [13:04] thanks balloons [13:04] thomi: calling the phone, I think that might currently be easiest with umockdev, so that we can test through the real ofonod stack [13:05] thomi: /usr/share/doc/umockdev/README.rst.gz explains how to do that [13:05] thomi: eventually I think we should create an ofonod mock with dbusmock, which is more flexible; but that doesn't exist yet [13:05] thomi: if that's a blocker, I can work on that [13:16] pitti: would I need a sim card in the phone in order to use that? [13:16] pitti: also, perhaps we should integrate this with autopilot [13:16] somehow... [13:42] thomi: you need a SIM card to do the recording; for replay you don't even need Android or a phone [13:42] thomi: I'm still in the US, so I can't do a proper record here; can do one next week, or anyone else produces one; umockdev is in the archive and the steps are documented [13:43] DanChapman, going to do a review of autopilot-gtk before the day sweeps me away [13:43] pitti: so, when you replay it, I guess you're forced to replay *exactly* the same thing, right? [13:44] thomi: yes, i. e. dial the same number [13:44] pitti: so, for example, you'd want 1 recording of a call where the remote end hung up, one recording where you answered etc. etc. [13:44] thomi: of course you can have/load different records for different scenarios [13:44] right, so there's not a lot of flexibility about controlling that programmatically [13:44] thomi: but I'm not even sure that we actually have that much control over that; rild abstracts a lot of that away [13:45] thomi: as I said, for more flexibility, and if we are only concerned about UI testing, it would be better to mock ofonod [13:45] I mean, how would that work in the scenario where I want to answer an incoming call? I push the "answer call" button, and the recording has to start replaying different events? [13:45] ok [13:45] that might be something to talk about [13:45] that's harder to do, but you can then easily test corner cases [13:46] thomi: you'd still umockdev-record ofonod, so in that case on replay ofonod would get the same answer back from rild and then emit its own dbus signals (I guess, I didn't try that case) === vrruiz_ is now known as rvr [14:07] good morning friends. [14:08] elopio, hola amigo [14:08] cgoldberg: hola compadre. [14:10] thomi, found an AP bug that's causing the CI failure on my branch... question: if I have a CI failure, and now I push a fix, will Jenkins automagically re-run it, or do I have to trigger something after I push? [14:19] cgoldberg: it will rerun it. But is faster if you go to the retrigger URL left on the bottom of the comment. [14:20] elopio, i saw that, but the url it gives doesn't resolve.. do I need to be on VPN or something? (it points to s-jenkins:8080) [14:23] cgoldberg: yes, QA vpn. [14:28] cgoldberg: my hard drive is dying, so I'm getting an SSD [14:29] what would you recommend? Samsung, intel, kingston? [14:31] cgoldberg: s-jenkins is a dumb alias for 10.97.2.10 [14:33] thomi: re-thinking your "obsolete package" problem: it sounds a bit weird why apt would know about newer versions of individual packages in the first place [14:34] thomi: it seems we still run the daily apt-get update cronjobs on phones although we don't even support apt-get there? [14:37] elopio, i usually get whats on sale :) but I've had good luck with samsung, crucial, and samsung [14:38] samsung, crucial, and Intel rather [14:39] elopio, what size SSD are you looking for? [14:39] cgoldberg: a cheap one. 120GB sounds fine. [14:40] elopio, and most importantly.. do you have SATA3 in your machine? if not, you can prob save a few bucks and get a SATA2 drive [14:42] elopio, I'm using this one now in my main machine: http://www.amazon.com/Samsung-MZ-7PD128BW-SATAIII-Solid-State/dp/B009LI7CKI/ [14:42] samsung 840 Pro (128 gb) [14:42] it's super fast and have had no probs [14:43] cgoldberg: I have sata3 [14:44] elopio, there is a newer version of mine. this one looks really good and the price is right: http://www.amazon.com/Samsung-Electronics-EVO-Series-2-5-Inch-MZ-7TE120BW/dp/B00E3W15P0 [14:45] cgoldberg: that's the one I was looking at. $100 sounds perfect. [15:01] balloons: awesome thanks :-) [15:25] DanChapman, it's weird though I get errors running the tests [15:25] balloons: what errors? [15:25] DanChapman, 7 errors to be exact I'll have to go back and look quickly, but I assume it's error free for you [15:25] you use the sandbox? [15:25] no i havn't tried it yet [15:26] Yeah it runs sweet for me. [15:29] someone remind me: what's the magic invocation to get the current build number? [15:29] getprop something.something.something? [15:30] thomi: on phones? [15:30] yeah [15:30] thomi: cat /etc/media-info for classic, system-image-cli -i for readonly system images [15:31] pitti: thanks! [15:35] DanChapman, hmm.. Well I can give you the specific logs of what's failing, but I wonder why the difference [15:35] I'm on saucy [15:35] what's up balloons, we are close to Beta 2! [15:35] balloons: me too. let me just run them again [15:36] DanChapman, I'll paste the log [15:36] hey smartboyhw , yes yes we are close [15:36] the release is very close now [15:37] balloons, time to move your world-known press effort to promote Beta 2 QA. [15:39] DanChapman, http://paste.ubuntu.com/6128772/ [15:39] smartboyhw, yes, there is much to talk about [15:44] balloons: your running the wrong tests. Thats the libap-gtk tests the emulator tests are in further in /tests/autopilot/emulators/tests [15:50] balloons: did you find them? :-) [15:53] DanChapman, yes.. thanks for noticing.. still the ap-gtk tests shouldn't be failing :-) [15:55] Ok, so now for the more fun part, running commentary [15:56] DanChapman, have you found the labels to be consistent across applications? You have selects against english text names spread about [15:58] I take it it works out since something like say the file dialog is fairly well reused -- any issues with non-english locales? [15:59] also, is logger.info the right level for some of these statements? perhaps they should be logger.debug? [16:09] balloons: The selects against labels are only in the GtkFIleCHooserDialog I believe. I hadn't thought about non-english locale. :-S [16:10] DanChapman, I left my comments in the review, since, well others will be using it [16:10] and for logging Its the first time i have used it so wasn't sure on what levels should be what [16:10] balloons: ok cool will go and look [16:11] might consider a couple tweaks, but overall, it's a +1 from me [16:14] balloons: great thanks. Well its a good start for it anyway. SO with logging, would the level mainly be .debug ? The py docs aren't to clear on when to use what [16:15] DanChapman, I would consider it to be all on the debug level.. however, it gets interesting when logging a module [16:18] DanChapman: this autopilot-gtk stuff going to be useful for us on xubuntu? [16:32] balloons: Ok well i will change them around. [16:33] elfy it should be useful for everyone. It makes writing tests alot easier :-) [16:35] elfy, yes, it makes the process much sane [16:35] *saner [16:37] thanks both [17:15] jibel: I've been poking arm retraces for a while now; I keep getting "wrong size gregset struct in core file" with gdb-multiarch on amd64 :( [17:16] that used to work (for most crashes anyway), but it seems our toolchain has changed somehow [17:19] I'll try that on a phone tomorrow (after lunch I'll have Plumber talks) [17:19] there are still some missing ddebs, but it shouldn't be that broken [17:37] ping thomi [17:39] balloons: yo [17:41] thomi, so I'll be to the point, though we can go deeper if needed. Calendar app tests have been failing for a long time, for unknown reasons. One of the tests seemed to hang for a long time. Turns out there was a big select_many that grabbed many things at once (it had a filter, but probably was client-side). Can we someone generate a message when people do things like this? A warning that you selected a huge amount of objects? [17:42] balloons: what was the query? [17:42] Increasing the performance would also be good, but a warning in the log would alert people invesigating things [17:42] balloons: 1.4 (due to land in T) will help this a lot [17:42] thomi, labels = self.select_many("Label", text=title).. You know how many labels there are :-) [17:43] yeah [17:43] depending on what the title is, that may be much faster in 1.4 [17:43] it was a bit of a silent killer.. the fix was to limit it to a specific view.. in other words grab a child node first and look under that [17:44] * elopio <- lunch [17:44] yeah. [17:45] balloons: I'm not sure how we could tell that the query is large before you execute it [17:45] thomi, yes I'm perfectly ok with a warning saying hey, this query returned 500 objects [17:45] it could be after the fact, it would just let those trying to figure out what the issues are do so more easily.. [17:46] ahh, I see - AFTER the query returned :) [17:46] you can't save the test author from themselves in this case, but you can provide a helpful hint I think for those looking at what's wrong [17:46] balloons: yeah, I think we can do that [17:47] thomi, shall I file a bug? [18:26] thomi, https://bugs.launchpad.net/autopilot/+bug/1227830 [18:26] Ubuntu bug 1227830 in Autopilot "Provide warning after returning large queries" [Undecided,New] === Ursinha is now known as Ursinha-afk [19:11] thomi... hey... is there anything specific for 13.10 release that needs to be done for any tools? I'm looking for some work besides the automation of found bugs. Anything for AutoPilot (or anything else)? feel free to throw stuff my way [19:12] cgoldberg: I remembered something yesterday... what was it. ummm... [19:12] uhhh... [19:13] I'll go make a coffee, hopefully that'll jog my memory [19:13] I told jfunk, maybe he remembers? [19:13] balloons, hey :) [19:14] knome, hello! [19:14] balloons, what's up? [19:14] well, trying to finish off some things before a long weekend [19:14] there's some cool stuff going on in QA, gotta share it all :-) [19:14] heh, trying to avoid the freeze fuzz? [19:15] balloons, hi [19:15] yea, freeze smeeze [19:15] :-p [19:15] letozaf, buonasera [19:15] balloons, buonasera :D [19:16] cgoldberg: there are certainly some bugs that would be good to fix. I can't remember what the thing from yesterday was [19:16] how about yourself? Trying to think about and share the manual/automated story moving forward knome. [19:16] thomi, i'll check the bug list [19:16] cgoldberg: want me to suggest a few? [19:17] knome, basically trying to move manual testing forward to the exciting exploratory break it testing and out of the repetitious stuff :-) [19:17] balloons, have been quite hectic with getting last uploads in for the beta (and not all have landed yet), but it seems we're again making it! [19:17] balloons, I was trying to run phablet-test-run -n ubuntu-clock-app on my device but I get a error: closed message :( [19:17] thomi, sure... if you can ping me a list of bug numbers, I'll triage and get on them for fixing [19:17] letozaf, can you share what you got exactly? can you adb shell in ok? [19:17] cgoldberg: there's this, which may end up being already fixed, but there's no test for it: https://bugs.launchpad.net/autopilot/+bug/1206805 [19:17] Ubuntu bug 1206805 in Autopilot "Autopilot Eventually matcher gives an obscure message inside the Eventually matcher" [Critical,Confirmed] [19:17] balloons, do you think a blog article or something would help? if you had some questions thought out, could easily answer to them [19:18] cgoldberg: but you should be able to mock out the dbus call that's in that stack trace and see what happens to wait_for [19:18] balloons, yes adb shell is ok [19:18] knome, I've got a least one article in draft talking about it.. Trying to get it all out :-) [19:18] thomi, ok.. i'll write a test for it and fix up [19:19] balloons, after installing yesterday's software, I tried to run the ubuntu-clock-app with [19:19] cgoldberg: awesome! maybe I'll remember what the other thing was :-/ [19:19] balloons, we probably should sit down together some time and think about that, probably after b2 testsing [19:19] *testing too [19:19] hah.. let me know if you do [19:19] balloons, phablet-test-run -n ubuntu-clock-app [19:20] balloons, I just get two error: closed messages and the shell again [19:20] balloons, maybe I forgot to install something ? [19:20] knome, yes, I would like us all to have a think about it. I'm just mentioning it to you so you can think about it. We have autopilot tests for ubiquity for many of the iso tests. We can probably re-think how we handle respins, particularly for non-impacting packages [19:20] letozaf, ohh, lol [19:20] don't adb sheel in first.. run it from your desktop :-) [19:21] balloons, would what you are saying be simply not updating the "build" on the iso tracker, but only in packages tracker [19:21] balloons, ahhhh! :P [19:22] balloons, thanks :P [19:27] balloons, :( ImportError: No module named ubuntu-clock-app [19:27] letozaf, that is the type of error you should see. It means you didn't install ubuntu-clock-app-autopilot on the device. Ohh, also watch the dashes '-'.. it should look like this: phablet-test-run -b ubuntu_clock_app [19:29] balloons, oh! I wrote ubuntu-clock-app :( anyway ubuntu-clock-app-autopilot was non installed on the device, I will do it now [19:29] yes, the module name uses underscores [19:29] keeps you on your toes [19:29] balloons, yeah! [19:30] pitti, thanks for looking into this. Tell me I can be of any help. [19:31] balloons, cool it's running ... the test :P [19:36] elopio: I think I've fixed this bug, but I wonder if you could confirm for me please? https://bugs.launchpad.net/autopilot/+bug/1225026 [19:36] Ubuntu bug 1225026 in Autopilot "It takes a lot of time from the application launch until autopilot can excercise it" [Undecided,Confirmed] [19:40] letozaf, awesome.. So you got the hang of the device now? [19:41] balloons, "the hang of the device" :) [19:41] balloons, I'm starting to figure out how things work [19:45] balloons, is there a way to be sure that I have the latest updates on the device, I mean I see "version 56" under settings/updates but how do I know I have the last one ? [19:49] letozaf, if there is no update available, you are running the latest :-) [19:49] letozaf, https://plus.google.com/104307250302998042813/posts/f4viizsY2v5 [19:49] letozaf, sorry, https://wiki.ubuntu.com/QATeam/TouchTesting#Which_build_am_I_running.3F [19:51] balloons, mmm Ubuntu Saucy Salamander (development branch) - armhf (20130916.3) [19:52] balloons, doesn't look like the last, shouldn't it be 20130918.1 ? [19:52] letozaf, indeed. [19:52] balloons, I will try to update again [19:53] if needed, you can phablet-flash ubuntu-system, and that will do that same thing, albeit it you'll have to download the full image [19:55] balloons, if updating from the device won't work I will phablet-flash ubuntu-system [19:55] balloons, wow, no, it updated it's rebooting now! thought i hit a bug :P [19:57] balloons, this is weired, after reboot I have the same previous version [19:58] cgoldberg: I wonder if you have the time to process a couple of AP code reviews? [19:58] balloons, could it be a bug or should I just phablet-flash ubuntu-system [19:59] cgoldberg: also, I notice you changes tox.ini to run the AP functional tests, but it doesn't really work very well, since the applications it launches still appear in the desktop, so I think we should revert that change [20:01] Hi all :) Hi letozaf, do you have 5 minutes for a query? [20:01] WebbyIT, hello! [20:02] WebbyIT, yes [20:03] thanks [20:03] letozaf, what does adb shell cat /etc/media-info now? [20:03] letozaf, ohh, why lookey there, hehe [20:04] balloons, Ubuntu Saucy Salamander (development branch) - armhf (20130916.3) [20:06] letozaf, yea, it should have updated. did the phablet-flash download a new build? [20:06] what did it say it was installing? [20:07] can you do a system update? [20:07] I just got an orange button saying install and reboot [20:07] balloons, when I clicked it the device rebooted in recovery mode [20:08] balloons, so I chose just to reboot, if I choose install it asks me what to install, not sure if I had to select the .zip [20:08] balloons, file ... the name was [20:09] letozaf, ohh no, let phablet-flash do it's thing [20:09] you shouldn't need to touch the phone at all.. it'll do everything and when it's done it will boot into ubuntu [20:10] letozaf, did system-settings, updates not work for you however? [20:10] it should [20:11] balloons, when I choose to update from system-settings updates, it askes me to install and reboot, just that it reboots in recovery, so I select just to reboot, ubuntu starts but the version is the old one [20:12] letozaf, that all sounds correct. it will reboot into recovery and update itself, then reboot into ubuntu [20:13] balloons, well maybe my problem is that, when I first installed ubuntu touch, with the phablet-flash ubuntu-system --channel devel --no-backup [20:14] balloons, command the device rebooted but recovery did not launch, I just got the green android figure with a yellow triangle [20:14] letozaf, well I'd say flash with phablet-flash ubuntu-system. then tomorrow when an update hits, do the update [20:15] balloons, yes I was also thinking to do a phablet-flash [20:15] the update goes to a weird android looking screen with a progress bar along the bottom.. takes a bit, but then it reboots and your updated [20:15] you're [20:15] balloons, not for me [20:15] knome, :-) [20:15] balloons, it reboots into recovery :( [20:16] letozaf, k, well just reflash and see if the update works tomorrow.. if it doesn't, that's a bug, hah [20:16] mine have actually worked well [20:17] balloons, ok I will do so [20:22] thomi: yes, I'll give it a try. [20:24] elopio: thanks [20:36] balloons, reflashing worked fine :) [20:37] balloons, let's see tomorrow that happens when I update [20:37] :-) [21:22] hey slickymaster [21:23] hi knome. Hope everything is fine with you [21:23] yep [21:24] knome, I haven't forgot the Xubuntu Documentation translation, but I've been pretty busy with editing the upcoming 13.10 manual [21:25] slickymaster, no problem, we're not going to be able to use the docs translations this cycle anyway [21:26] knome, yeah, I know. Documentation string freeze today [21:26] slickymaster, that's not the culprit :) [21:27] slickymaster, there is another freeze for translated strings anyway [21:27] knome, didn't knew that [21:30] knome, anyway as it seems that there's nothing besides that for me to help in the xubuntu team [21:31] knome, I will try to finish it before the end of next month [21:31] slickymaster, no problem, take your time :) [21:32] slickymaster, " nothing besides that for me to help in the xubuntu team" ? test things, fix bugs, learn a coding language. [21:32] ;) [21:33] Noskcaj, I spent my work days coding in Java and in SQL [21:35] Then if you ever run out of things to do, we've got a gint bug tracker [21:35] *giant [21:36] Noskcaj, well I was thinking to take a plunge at wiki.documentation as they seem to need a few extras pair of hands [21:36] slickymaster, That would also be great [21:37] i need food, i'll be back later [21:37] Noskcaj, like the French say: Bon appetit [21:49] thomi: I'm fully updated with the autopilot/experimental ppa and the issue is still happening. === salem_ is now known as _salem [23:27] cyu all tomorrow. good night