[02:05] === IMAGE 136 building (started: 20150316-02:05) === [03:30] === IMAGE 136 DONE (finished: 20150316-03:30) === [03:30] === changelog: http://people.canonical.com/~ogra/touch-image-stats/136.changes === === marcusto_ is now known as marcustomlinson_ === marcustomlinson_ is now known as marcustomlinson === sil2100 changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: - [09:32] Mirv: wait for the authorization error [09:32] Mirv: afterwards sound is back [09:33] jibel: meeting! [09:35] cihelp: jenkins seems unwell http://91.189.93.70:8080/job/generic-mediumtests-utopic/2349/console via https://code.launchpad.net/~carlos-mazieri/ubuntu-filemanager-app/samba-browsing-11/+merge/252982 === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [11:17] popey, look like the jenkins slave was temporarily disconnected. The job is restarted and looking better now. [11:17] thanks fginther === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === MacSlow is now known as MacSlow|lunch [12:38] Mirv: would you reconf the silo13 please? I have just added there a browser fix MR [12:45] bzoltan_: sure [12:54] Mirv: danke === chihchun is now known as chihchun_afk === alan_g is now known as alan_g|lunch === MacSlow|lunch is now known as MacSlow [13:49] dobey: ping, did you see ToyKeeper's +1 on the payui? [13:50] sil2100: yes, i'm moving it the rest of the way through now [13:50] dobey: excellent, give me a sign once it's published [13:50] sil2100: i'll update the spreadsheet status when it's done [13:52] Thanks [13:56] bfiller: Silo 2 is now approved. [13:56] rvr: thanks === chihchun_afk is now known as chihchun === alan_g|lunch is now known as alan_g [14:15] ralsina, Hi! [14:15] hi om26er [14:15] ralsina, How can I verify the memleak fix for silo 16 ? [14:15] om26er: there's no way to verify it directly :-P [14:16] Running the test plan ensures that it's not triggering a crash [14:16] it was a very slow memory leak, so it's not really noticeable [14:16] ralsina, hmm, test plan is quite simple, not sure of the level of its coverage [14:17] om26er: it's a very simple package [14:17] ralsina, how about some kind of automated test for the leak ? [14:18] om26er: I can give you a shell script that will trigger the formerly-leaking code [14:18] ralsina, yeah, that would be nice [14:18] om26er: ok, give me a few minutes and I'll send it to you [14:47] hmm, the spreadsheet seems to be giving me grief, telling me what I do causes a fatal error [14:47] bregma: ugh [14:48] bregma: what are you trying to do? [14:48] I added a line [14:48] huh [14:48] any updates I do give me the error [14:48] Ok, I see the spreadsheet is in agony right now [14:48] still seems to work, it just complains a lit [14:48] *lot [14:48] I get error notifications on every 5 minute update interval :| [14:49] kill it with fire [14:49] Let me stop the auto-refresh scrips for a moment and let it settle [14:49] ralsina, any update on the script ? [14:50] om26er: almost there [14:54] om26er: I have a click app that has a button. When you click that button, it will call the leaking functions 10K times. That shows that the fix doesn't introduce any new crashes. Sadly, the leak was so small that even when calling it like that it's not noticeable :-P [14:54] ralsina, wow, nice. :) [14:55] om26er: mailing you the click now :-) [15:00] ralsina, nothing, crashes, I'll give it a +1 [15:00] om26er: awesome, thanks! [15:00] om26er: and sorry about fixing invisible stuff ;-) [15:00] heh. [15:23] rsalveti, Is there any test plan for hybris [15:24] om26er: not yet, because it's a core piece, if it gets broken, the UI will not even start [15:24] so we covered it with exploratory tests before [15:24] and it depends entirely on the hardware [15:25] rsalveti, by exploratory testing you mean general checking the sanity of things ? [15:25] om26er: yeah [16:28] om26er_: do you know what is the situation with the UITK in the silo13? [16:30] bzoltan_, we'd like to wait the verdict on Qt 5.4.1 [16:32] sil2100, I won't attend the landing meeting, from the list this morning, only 5 is remaining, everything else landed. [16:32] jibel: ok, excellent [16:32] No worries [16:32] sil2100, landed or testing is in progress [16:33] bzoltan_, not really, I see jibel made a comment there. [16:33] jibel, if I understand correctly, you don't want to land UITK till qt5.4.1 lands ? [16:34] jibel: I thought that Mirv wants the UITK to land first on Vivid ... the silo was built against the [16:34] Mirv, ^ ? [16:35] jibel: Mirv: i am confused .. I though that the Qt landing is blocked by the UITK and not the other direction [16:37] jibel: the silo13 UITK was built against the 5.4.0 Qt [16:37] jibel: Mirv: I do not fully understand why it would be blocked by the newer Qt [16:38] Mirv: or do you want to hold all landings what depend on Qt and the rebuild all of them with 5.4.1? [16:38] bzoltan_: I think what jibel has in mind is that they want to first assess Qt 5.4.1 as it is now - if they sign-off UITK they might need to re-build the 5.4.1 silo and re-test [16:47] huh [16:47] heh [16:47] Ok, I see the spreadsheet again ate some data (probably) [16:48] hm, let me compare with the backups [16:48] sil2100: dashboard looks sane, I think queuebot just barfed. [16:49] robru: I just got an error in the spreadsheet and probably saw some data getting lost, so I want to double-confirm [16:50] fair [16:50] phew, no, it's just my local spreadsheet instance was busted [16:52] Ok, but I still can't assign silos [16:53] robru: can you try assigning a silo through the spreadsheet? [16:53] I don't want to infinitely do this manually [16:53] sil2100: which one? [16:54] robru: any of the ones ready for a silo [16:54] sil2100: ah, no I get fatal error [16:54] sil2100: I like how the "details" window gives the same message verbatim without a single extra piece of info on that failure. [16:55] Yeah, and the spreadsheet seems b0rken really badly [16:55] sil2100: what other symptoms are you seeing? [16:56] cihelp: can you please upload https://jenkins.qa.ubuntu.com/job/generic-click-builder-14.09-armhf/16/artifact/output/com.canonical.payui_15.01.113_armhf.click to the store? thanks [16:57] dobey: let me find out how to do it [16:57] cprov: usually it's plars or fginther who do it [16:58] cprov: https://wiki.canonical.com/UbuntuEngineering/CI/Playbook#Uploading_a_click_app_to_the_click_store [16:58] Ursinha: thanks [16:59] robru: it seems the spreadsheet is encountering fatal errors in various seemingly random places of the refreshSilosStatus function [17:00] hm [17:00] robru: I tried debugging it now and I got the error in 2 different places on 2 different runs [17:00] sil2100: seems like some kind of transient/load issue with google servers. maybe just try turning the sheet off (eg, disable all scripts) for a couple hours? [17:01] rsalveti, Hi! [17:01] robru: I already turned the triggers to 30 minutes like a few hours ago [17:01] Didn't help [17:01] buh [17:01] I turned them back to 5 minutes [17:01] sil2100: well, that's it then. I'll start training the carrier pigeons. [17:01] Ok, it seems google has issues in overall [17:02] Since I again have HO problems [17:02] The same authorization thing, I can't see or hear anyone [17:02] rsalveti, while testing libhybris silo I found that sound in recorded videos is broken/robotic [17:02] dobey: done, please check if everything looks correct. [17:03] om26er_: is it ok without the silo installed? [17:03] sil2100, yes, rvr verified [17:03] dobey: i.e. -> The new version is being reviewed. Please check again at a later time. [17:04] cprov: yeah i can't see until it gets approved and is available as an update :) [17:04] dobey: uhm, who is responsible for the reviews ? [17:06] cprov: i'm not sure who all has privs for that. popey does, beuno and probably others on his team. [17:07] hmm? [17:07] popey: you can review packages in the store no? [17:07] doing payui now [17:07] popey: thanks [17:07] great, thanks [17:08] done [17:20] om26er_: which device, which image? === sil2100 changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: the CI Train spreadsheet is experiencing problems [17:21] krillin, vivid-proposed 147 [17:21] rsalveti, ^ [17:25] om26er_: and that only happens with the new libhybris? first testing with RTM now [17:26] rsalveti, yes, I found the issue with the silo and rvr verified on vivid-proposed without silo [17:30] hmm, actually it might not show up as an update on my phone [17:31] om26er_: yup, found a bug, nice, could also reproduce on RTM [17:31] om26er_: will investigate, thanks for the feedback, will make sure to write down a wiki page describing what to be tested [17:31] and adding this one to the list [17:31] rsalveti, thanks [17:35] oh, my mako shows it, cool [17:41] sil2100: so pay-ui is "landed" now. i tried to change the status in the spreadsheet to "Landed" but i am not sure if it actually saved it and did the magic [17:42] bzoltan_: jibel: no strong opinion whether to land uitk first not not. qt silo can rebuild it if it lands. [17:42] uitk is a no-change rebuild in the qt silo [17:43] dobey: what magic? nothing gets triggered by setting the spreadsheet to 'landed' [17:43] so it's only what matters to testing (both) silos that counts [17:44] robru: doesn't some script move things that are "Landed" to the archive page? [17:44] dobey: when we run it manually, yeah. it doesn't get triggered by the status changing to "Landed" [17:45] robru: sure, there are way too many green boxes for it to be automatic. i just presumed there was a cron job somewhere [17:46] dobey: nah we do that by hand. I just thought you were expecting something really magical to happen. if it says 'Landed' it's fine. [17:46] jibel: bzoltan_: so, last week I was thinking that maybe uitk should just go in before QA starts testing Qt silo. now that they've been testing it since Friday evening, I guess they indeed may want to finish that properly before entering into new silos. [17:51] robru: ok, seems like Steve removed the later meeting, I guess that means I can go to sleep after I'm done with work here ;) [17:51] sil2100: sweet, time for breakfast then ;-) [17:51] robru: nah, i was just saying google was being fickle about me typing "Landed" in the status cell :) [17:51] robru: hah, have fun ;) [17:52] dobey: oh yeah the spreadsheet is a bit borked at the moment, google servers are just crapping all over it or whatever. if it's not saving your changes just try again later. [17:53] robru: ah ok. well sil2100 told me to ping him when it was done, so i was doing that too :) [17:53] anyway, go have some proper USian bacon or something [17:53] hehe === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun [18:03] Kaleo, Hi! [18:03] Kaleo, Looking here: https://code.launchpad.net/~fboucault/qtvideo-node/fix_rotate_z_order/+merge/252265 [18:04] seems the newly added tests are disabled ? [18:09] om26er_, yes [18:09] om26er_, can't get them to work on jenkins [18:09] om26er_, and it would not be much value: they only fail on device [18:09] om26er_, but at least the test exists [18:09] om26er_, and can be run manually [18:10] om26er_, and a corresponding manual test was added to the camera test plan [18:10] Kaleo, hmm, it would be nice to have the automated test work [18:10] Kaleo, did you get help from the CI guys ? [18:10] om26er_, I know, spent enough hours on it though [18:10] om26er_, some [18:11] Kaleo, hmm, ok. Can you please report a bug to track that, so that we could track it [18:11] om26er_, track that the test is disabled? [18:12] Kaleo, yes, so that we could enable it at some point [18:13] om26er_, https://bugs.launchpad.net/ubuntu/+source/qtvideo-node/+bug/1432766 [18:13] Ubuntu bug 1432766 in qtvideo-node (Ubuntu) "unit test tst_videooutput is disabled because it cannot run on jenkins" [Undecided,New] [18:13] done [18:13] Mirv: jibel: so what is the verdict ... the silo13 is there for a week waiting for QA validation. I had time to update it, so it has all fixes what I had on my backlogs for the last week. [18:13] super === om26er_ is now known as om26er|food === alan_g is now known as alan_g|EOD [18:21] dobey: https://code.launchpad.net/~unity-api-team/unity-scopes-shell/scope-harness/+merge/250612 has this mp changed since your 'needs fixing' review? can't publish it until it's approved... [18:22] the mp hasn't changed [18:22] hm [18:23] the other mp which adds the symbols changed, but i'm not happy with the "fix" that was done there either really, as it doesn't really fix the problem [18:27] dobey: no worries, I won't publish. too bad pstolowski isn't on to get poked about that [18:27] yeah, nor is pete [18:27] hopefully michi will be on soon, and i can at least discuss with him about it === om26er|food is now known as om26er [18:34] om26er: found the issue, will push another update in a few [18:58] rsalveti, great, can you share the diff [19:12] om26er: just reverting the change that caused this: http://paste.ubuntu.com/10611435/ [19:12] om26er: just pushed a new package to the ppa [19:13] om26er: also started https://wiki.ubuntu.com/Process/Merges/TestPlan/libhybris [19:14] mterry: kenvandine: anybody around for a packaging ack? https://ci-train.ubuntu.com/job/ubuntu-landing-018-2-publish/lastSuccessfulBuild/artifact/qtvideo-node_packaging_changes.diff/*view*/ got some new deps [19:15] rsalveti, that's good to see :) [19:16] robru, sure seems fine [19:19] mterry: thanks [19:44] robru: Where do you keep the citrain bot code? I'm trying to figure out the right place to put the qa silo bot code... [19:45] om26er: mind checking latest packages from https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-004 ? [19:45] rsalveti, yes, I can start on it. [19:45] ToyKeeper: lp:queuebot and it's controlled by stgraber. I'd advise against putting a lot of effort into that because I've started a port to a different bot but that stalled a bit [19:46] om26er: thanks [19:48] robru: This is an independent bot, but it's related... Mostly, I'm looking for the right logical place to put it on Launchpad. :) [19:48] john-mcaleely: device tarball is good marked it so and then the spreadsheet went awol :) [19:48] sil2100: ^ [19:48] ToyKeeper: oh just start a new project I guess. [19:49] That would explain why I haven't been able to find a central place for this stuff. [19:52] ToyKeeper: well lp:queuebot is actually a new thing, stgraber used to just store that in a +junk branch and I couldn't even MP against it for a while. [19:52] robru: That's exactly what I'm fixing. :) [19:58] bfiller: https://code.launchpad.net/~michael-sheldon/ubuntu-keyboard/fix-1421411-uncrisscross/+merge/252789 just need this merge approved [19:59] ToyKeeper: what irc bot framework do you plan to use for your irc bot? [19:59] robru: done [19:59] bfiller: thanks [19:59] robru: It's not an IRC bot. I was hoping we could add features to queuebot if necessary. [20:00] ToyKeeper: oh you said it was an independent bot [20:00] robru: For now, I'm just adding stuff to the silo/trello bot QA uses. [20:01] ToyKeeper: yeah I'm not a fan of queuebot, it's ostensibly modular but you have to write a lot of your own threading boilerplate to make it work. I found a new bot framework called willie that seems to allow a significantly easier plugin architecture, but I only ported some of the plugins, and then the work got deprioritized. [20:09] Woot, we already have a project for misc qa tools. Seems like a good place for it. [20:12] alright, I'm off for lunch [20:12] ToyKeeper: feel free to ping me if you have any questions about queuebot, I wrote the plugins specific to the train, stgraber wrote the larger framework. === chihchun is now known as chihchun_afk [21:46] robru: please don't yet land the new libhybris, want to coordinate the landing with the next image [21:46] rsalveti: sure, i saw your name on it and assumed you'd land it [21:47] cihelp bug 1431365 - see latest comment from carla - is there a firewall issue ? [21:47] bug 1431365 in Ubuntu RSS Feed Reader App "Tests fail at warning dialog "A topic with that name already exists"" [High,New] https://launchpad.net/bugs/1431365 === chihchunl is now known as chihchun [22:34] popey, I recorded your issue, someone will probably have a look tomorrow === chihchun is now known as chihchun_afk