[01:11] alex-abreu: https://code.launchpad.net/~abreu-alexandre/unity-webapps-qml/improve-embedded-ui-params-passing/+merge/250329 need this top-approved [05:24] Mirv: https://code.launchpad.net/~rsalveti/kubuntu-packaging/qgstreamercapturesession_avoid_race_eos/+merge/256881 [05:24] rsalveti: thanks! [05:32] robru: ohh, what a begginer mistake :) thanks... now the MRs are approved [05:32] Mirv: ^ [05:41] bzoltan_: nice! [05:42] Mirv: I am sorry ... I simple forget about approving the MRs .. [05:43] bah, now we'd need packaging ack [05:44] robru: the diffs fixing is still underway, right? [05:44] Mirv: should be fixed... [05:44] robru: it's saying the source is a new package https://ci-train.ubuntu.com/job/ubuntu-landing-004-2-publish/lastSuccessfulBuild/artifact/ubuntu-ui-toolkit_packaging_changes.diff [05:45] bzoltan is probably pretty sure he has landed ui toolkit before [05:45] few times [05:45] Buh [05:46] Mirv: go back to the old build job for now. Will investigate that tomorrow [05:47] robru: thanks! [05:47] Mirv: https://ci-train.ubuntu.com/job/ubuntu-landing-004-1-build/129/ this one I think [05:47] You're welcome [05:48] robru: yes, found it [05:48] Great [05:52] asking for packaging ack on #ubuntu-devel === ara is now known as Guest22932 [06:37] got ack [07:40] sil2100: cwayne bug 1446499 after updating today from store - it's gone. [07:40] bug 1446499 in unity8 (Ubuntu) "Updating today scope from store kills scope, doesn't restart" [Undecided,New] https://launchpad.net/bugs/1446499 [07:45] popey, does it come back after reboot ? [07:46] not rebooted yet [07:46] * ogra_ stays away from updating [07:54] I can confirm [07:55] and does not come back after reboot [08:02] gah! [08:03] victorp: ^ [08:03] huh [08:07] bzoltan_: we'll have to poke the release team about this one [08:08] bzoltan_: i.e. about qtcreator-plugin-ubuntu - it's critical, so should be good to get approved [08:09] sil2100: thanks. it is silly to catch a critical bug at that late phase and even sillier to shoot twice on the same bug. [08:12] popey, can we pull it from the store worst case ? [08:13] not sure, last time I did that, I kinda got chided by beuno [08:13] cihelp, the provision script's been aborting for me with this error http://pastebin.ubuntu.com/10860323/ for a few days now, that expected? [08:13] ah no, I can "unpublish" okayt [08:13] popey, well, or block it ... we should make sure not more people upgrade to it [08:14] (it was revert to previous version he didn't like) [08:14] ah, yeah [08:14] ~380 people have updated so far [08:14] damn [08:14] shall I unpublish do you think? [08:14] yeah [08:14] done [08:15] if it is reproducable we should prevent more damage [08:15] and i see you pinged victorp already ... i guess we need a quick fix to unscrew the users [08:15] * popey adds a comment to the bug to say I unpublished [08:15] well, they can reboot and then re-fave the scope [08:16] but that's not very pleasant, and no discoverable [08:16] i thought it is gone ? [08:16] popey: cwayne should be around in a few hours to help out if anything [08:16] * Saviq thinks it was renamed to "dashboard" wasn't it? [08:16] In the meantime let's unpublish [08:16] already happened :) [08:16] likely why it disappeared from the favorite list [08:16] ouch [08:17] * popey wonders how this was tested :( [08:17] yeah, we have no support for that [08:17] I wonder if it was signed off by QA at all [08:17] yup [08:17] weird [08:17] ok, well we have a bug and they can look when they wake [08:17] This might be another hole in our process [08:18] Normally testing of the today scope happens on new custom tarballs [08:18] thanks popey [08:18] np [08:19] popey, just to confirm, can you see in "gsettings get com.canonical.Unity.Dash favorite-scopes" [08:19] Saviq: i have already rebooted and re-added.. [08:19] popey, sure, that's fine, still interesting to know what the scope id is [08:19] ['scope://com.canonical.scopes.dashboard_dashboard', 'scope://unity-scope-nearby', 'scope://clickscope', 'scope://com.canonical.scopes.news_unity-scope-news', 'scope://musicaggregator', 'scope://videoaggregator', 'scope://com.canonical.scopes.photos_photos'] [08:19] popey, if you don't care for your favourites list, could you `gsettings reset` it? [08:20] I'd rather not. [08:20] * Saviq wonders if they updated the default [08:20] oh, it can be readded. but how do I get it to be the first on again instead of last one? [08:20] oh go on then [08:20] * sil2100 will upgrade the package [08:20] Mirv: you can re-order [08:20] long press [08:20] sil2100: found [08:20] Mirv, long-press, drag by handle [08:20] Mirv: just long-press it on the manage screen [08:20] Yeah ;) [08:20] * popey mails the phone list to let people know how to undo this. [08:20] almost discoverable [08:21] Mirv, good news: manage dash is likely going away [08:21] Saviq: resetting has done this:- [08:21] phablet@ubuntu-phablet:~$ gsettings get com.canonical.Unity.Dash favorite-scopes [08:21] ['scope://clickscope', 'scope://musicaggregator', 'scope://videoaggregator'] [08:21] :( [08:21] popey, weird, that's mako/rtm? [08:22] no [08:22] retail bq [08:22] huh [08:22] yeah, I got my retail bq yesterday [08:22] popey, fwiw, you can `gsettings set com.canonical.Unity.Dash favorite-scopes "['scope://com.canonical.scopes.dashboard_dashboard', 'scope://unity-scope-nearby', 'scope://clickscope', 'scope://com.canonical.scopes.news_unity-scope-news', 'scope://musicaggregator', 'scope://videoaggregator', 'scope://com.canonical.scopes.photos_photos']"` to get back your list [08:23] thanks [08:23] but the reset is scary, /me flashes krillin to see in a moment [08:24] this would mean that the custom tarball doesn't override the default [08:24] but that'd be *really* weird [08:24] Saviq, why would it [08:24] Strange, I don't see the today scope update on my krillin, huh [08:25] the custom tarball installs what is in it as upgrade time, thats it ... if there is a newer version of something it ships in teh store this will always override [08:25] sil2100, because it was unpublished already :) [08:25] ogra_, I meant the gsettings key [08:26] ogra_, the default list of favourite scopes is just apps, music, video unless it's overridden by custom tarball [08:26] the default is likely schipped as system wide schema ... [08:26] Ah, it was already, ok ;) [08:26] so if the user has the key set, it will override the system one [08:26] ogra_, which is why going `gsettings reset ...` should get it back [08:26] yeah [08:27] it doesnt ? [08:27] ogra_, and popey did the reset, and was presented with just the 3 scopes [08:27] http://people.canonical.com/~alan/clicks/com.canonical.scopes.dashboard_1.7.15_armhf.click is the offending click if you really want it sil2100 :) [08:27] this is what makes me wonder [08:27] popey: thanks ;) [08:27] Mirv, confirmed silo 15 makes CPU settle [08:28] Saviq: thanks! I tested both that use case and the one I found. [08:33] jibel: ping [08:42] ogra_, oh, they're not using gsettings overrides but a dconf one... [08:45] lovely [08:51] ogra_, popey, sil2100, IMO a bug in unity-scopes-shell, letting pete-woods know [08:51] pete-woods, https://bugs.launchpad.net/today-scope/+bug/1446499/comments/12 [08:51] pstolowski, ↑ [08:51] Ubuntu bug 1446499 in unity-scopes-shell (Ubuntu) "Updating today scope from store kills scope, doesn't restart" [Undecided,Confirmed] [08:52] ogra ping [08:52] victorp, hey [08:52] just saw the pings from you and popey, whats up? [08:53] victorp, today scope in the store is broken [08:53] victorp, we unpublished it, but it went out to ~350 users [08:53] ok.. [08:53] what are you installing it on? arale or krillin? [08:53] krillin stable [08:53] victorp: on krillin [08:53] retail devices [08:53] ogra_, victorp, I wouldn't say "the scope is broken" any more [08:54] well, the user exüperience is broken once you install the update :) [08:54] it's the update process that's broken [08:54] ("the scope is broken is shorter :P ) [08:54] yeah ok, details on the bug [08:54] Saviq: ok, might indeed be the case here [08:54] Saviq, well, it also didnt get any QA it seems [08:54] Although I think that updating the scope without QA is a bad thing in overall [08:55] ogra_, sure, FWIW I'd say the problem is race-y, too [08:55] yeah [08:55] so not everyone will hit it [08:55] Some people said it was fine for them [08:56] sil2100: where? at least on the bug report the reply was "fine if you readd it" if I understood it correctly. [08:56] Mirv: on the mailing list [08:56] sil2100: oh, ok. good if it doesn't hit everyone. [08:57] ogra_, sil2100 first of all , dont make things up [08:57] :) [08:57] Mirv: one person said that he upgraded and today scope remained on the favorite list [08:57] sil2100: so it seems [08:57] Saviq, commented, it's as you described [08:57] if you have a scope favourited and you updated it to a new version it *always* removes it from the fav list [08:57] Saviq, ^^ [08:58] but should still appear in your bottom edge list [08:58] victorp, ugh [08:58] victorp: so that's by design? [08:58] i thought that had been reported, it happens to me with the child scopes we favourited [08:58] victorp, yeah, that's correct, but shouldn't be the case should it [08:58] well seems like bad desing :) [08:58] ;) [08:58] that sounds pretty broken ... like if an update would remove the icons i added from the launcher [08:58] right, I thought it was a bug that had been reported , my bad for not checking [08:59] ogra_, agreed [08:59] so the more concerning part would be why is not appearing in the bottom edge with respect today scope [08:59] victorp, good that we catched it then ... better if a QA test had catched it before it went to the store though :) [09:00] do we have phased store updates in plans similar to image updates? [09:00] ogra_, it wouldnt, because this only happens if you update from the scope, if you update custom is doesnt [09:00] victorp, so it's not appearing on the bottom edge list under 'Also installed' ? [09:00] victorp, well, a new music app doesnt go into the store without QA run ... this scope apparently did according to QA [09:01] also i thought it is in the manage scopes tool, but under new name ? [09:01] ogra_, what are you talking about? [09:01] This is one of the holes in the process, every new click that's by default on the image and gets upgraded needs QA sign-off [09:02] Even the today scope [09:02] victorp, with what exactly ? [09:02] victorp, that the scope didnt get QA before oing to the store ? [09:02] I am telling you the scope is not broken, is the "update a scope from the store" process that is doing this , you just have notice it now [09:02] *goin [09:02] ah ! [09:02] right, i didnt get that ... [09:02] pstolowski, so reading the bug from popey it appears after reboot [09:03] victorp: only if you re-add it. [09:03] popey, in the manage... list [09:03] yes [09:03] popey, what do you mean by re-add it? [09:03] victorp: from user point of view it's gone for good and doesn't come back. I didn't find it in the management either untils someone told it that it's in there. [09:03] sorry, re-favourite it [09:03] popey, ack [09:04] ok, first of all - this has nothing to do with the today scope [09:04] popey, well, this was about: is it in the manage scopes area only after a reboot or is it there right after upgrade [09:04] that is the behaviour of all scopes you update from the sore [09:04] store [09:04] ok. [09:04] so it's two problems actually; manage list didn't refresh until reboot, and the scope needed to be favorited again [09:04] if it wasnt in the bottom edge after update and then shows after a reboot, that still has nothing to do with the scope [09:04] But the fact that it's the default first scope, it uncovers a bug [09:04] sure [09:05] pstolowski, +! [09:05] +1 [09:05] popey, exactly [09:06] as far as I can tell the team is just publishing to the store scopes that where added to custom-proposed , which is tested ogra_ [09:06] I didnt get an update notice [09:06] because we unpublished it [09:06] i pulled it from the store around 40 mins ago [09:06] but if they had to change the version for whatever reason, it would have trigger it [09:06] and you dont get noticxes for apps/scopes [09:06] i.e. change on mantainer name in manifest [09:07] so I dont think qaing this more would have helped, what we need is the beta/alpha channel in the store so we can test upgrades [09:07] I believe beuno's team is already working on that [09:07] well if someone had installed the click on a retail bq device they would have noticed the scope disappear [09:07] so, can we open 2 bugs to track the issues that pstolowski has highlighted [09:07] and hopefully file the right bug? [09:08] right, we would at least have catched one of the issues here [09:08] but I suspect nobody QA'd it on a _retail_ bq image. [09:08] because that's the only one shipping it by default [09:08] popey, I understand that this issues are new to you, but we have know about them for a while, I just thought they were being addressed [09:08] or they did, and ignored the fact that the scope disappeared [09:08] popey, more probably that [09:08] victorp: new to me and thousands of bq retail handset owners [09:09] yeah [09:09] popey, ok and? I am not sure what you are asking me? [09:09] ok. where is the bug tracking the fact that scopes disappear, and need a reboot? [09:09] I think we need two bugs for that [09:09] the assumption that things are being fixed because "we read:I already know about them" is a broken process. [09:10] can you please open them popey? [09:10] i opened one, I don't know what the second one is. [09:10] popey, sure - if you expect me to find all the bugs that is also broken :) [09:11] I didn't say you. I implied QA [09:11] If someone is QAing something and they spot something is broken, I'd expect a bug. [09:11] my point being nobody filed a bug for these issues because "we know about it" [09:11] victorp, if you found that bug before (knew about it), someone must have seen it but not filed it ... thats the broken part [09:11] but yeah, I'll file the bugs for you. [09:11] exactly [09:12] anyway, this is going in circles, there are two bugs in unity8 Saviq.. favouriting does not survive an update from the store , and the bottom edge/manage dash does not seem to always register updates. [09:12] victorp, when interacting with the users it helps if we can point to a bug [09:12] let me know if you need anything else [09:12] ogra_, 100% agreed [09:12] i.e. popey could have pointed it out in the warning mail he sent [09:13] victorp, both are unity-scopes-shell bugs [09:15] Saviq, ack [09:17] Saviq, are you happy just tracking the issues on 1446499 [09:18] i added a second [09:18] bug 1446536 [09:18] bug 1446536 in unity-scopes-shell (Ubuntu) "Scope disappears from bottom edge on update" [Undecided,New] https://launchpad.net/bugs/1446536 [09:19] victorp, popey, thanks [09:19] ta! popey [09:20] pstolowski, one more related, not sure if that's -scopes-shell exactly or the scope backing Manage Dash ↑ [09:20] so do we re-publish today in the store? [09:20] because this is a "known" bug? [09:21] I think it's really bad from user pov [09:21] well [09:21] yeah, but then otherwise we'd block any core scope updates until we release the next OTA... [09:21] we dont have a system update planned [09:22] well right, true [09:22] so the fix wont land before next OTA [09:22] people will learn to use scopes at least.. [09:22] i wonder if there is a way to hack around it from the scopeside [09:22] ogra_, not likely [09:22] Saviq, ok, so this is the second problem from the 1st bug [09:22] yeah, i was fearing that [09:23] ogra_, it's basically a case of "oh, the scope went away, let's drop it from favourites" [09:23] so either we live with the bug (very bad imho) and re-publish ... or we hold back all scope changes til it is fixed [09:24] pstolowski, [09:24] https://bugs.launchpad.net/today-scope/+bug/1446499/comments/16 [09:24] Ubuntu bug 1446499 in unity-scopes-shell (Ubuntu) "Updating scope from store removes it from favorites" [Undecided,Confirmed] [09:25] victorp, any opinion ? [09:25] Saviq, yeah, that would be simple to do, but would mean we never get rid of uninstalled scopes. unless we purge at startup [09:26] pstolowski, hook on uninstall won't do, as it's based off of the .ini file already? [09:27] Saviq, sorry, i can't parse what you're saying ;). i'm just talking about the plugin which purges favorites whenever registry is updated (and scope disappears) [09:28] pstolowski, sure, I'm saying that a hook on uninstall could trigger the removal from favourites, unless this is basically what happens already? [09:29] * Saviq wonders how relevant this discussion is in light of "scopes as apps"... [09:29] Saviq, ah, no. it's handled completly by the shell [09:29] * shell plugin [09:29] Saviq, yeah, becomes irrelevant then ;) [09:29] well, kinda [09:30] it just moves from Manage Dash to $wherever_we_store_favourites (like in click scope or something, to display at the top) [09:47] jibel: o/ i'm trying to get an idea whether silo 6 is considered for testing, or if there is something blocking landings in the overlay ppa [09:48] i may have another silo next, with branches targeted for desktop (w updates and potential srus); let me know [09:56] dbarth, it is in the triaged queue, it should be verified soon-ish [09:57] ok, nw [09:57] trainguards, point of procedure: do I need to get an ACK from the Ubuntu release team first to land silo ubuntu/landing-020 (#1446256) or does it just land and go into the UNAPPROVED queue and get ACKed from there? [09:57] jibel: thanks [09:58] bregma: we could publish it without an ACK, but it's best to coordinate beforehand especially if the bug that's being fixed is not set to critical [09:59] bregma: without the ACK it would en up in UNAPPROVED, but might get rejected due to lack of infromation etc. [09:59] sil2100, thanks [10:01] sil2100: bregma: since it's seeded I think it's even required to go via release team https://wiki.ubuntu.com/FinalFreeze [10:01] for example qtcreator-plugin-ubuntu was unseeded universe package but unity is a different beast [10:02] anyway, it's still >1.5 days so it's ok to go to vivid normal instead of proposed [10:04] davmor2: Bq now passes my official seal of approval requirements when I'm now writing this from Bq after using a menu shortcut to login + screen -dr and I've added Ctrl-D, Ctrl-P and Ctrl-N to the ctrl menu [10:04] sil2100, Mirv bregma if it is ready publish it, we need it on the image [10:04] jibel: ok [10:05] Mirv: let me publish the unity silo then [10:05] sil2100: ok [10:05] The release team will deal with it ;) [10:06] that's right, pass the buck [10:07] Mirv: it did anyway, I told you you were just using it wrong ;) [10:07] davmor2: that too :) [10:08] Mirv: did you like the osm touch app by the way? [10:08] Mirv: and don't play uu or we'll lose you forever :D [10:10] davmor2: the osmtouch is a familiar one to me, I've used it mako quite a bit. even though I didn't use mako "really really" (so this Bq is what I really give my "normal use case" for), I did use it for browsing and location [10:11] davmor2: it's ok but I need something closer to Osmand~ - offline vector maps, voice navigation [10:11] and if I would have click chroot creation not broken on my vivid I'd give a Navit or Marble a try to get something built [10:11] but alas I've not found the reason why click is totally busted on my machine [10:11] Mirv: here does turn by turn it you switch down to walking rather than driving :) [10:12] davmor2: here is closed source, I want OSM maps too :) [10:12] I prefer OSM in general even without closed/open over google maps and here maps [10:12] Mirv: whine whine whine ;) [10:13] just so much more "love" into details in there, like stairs, tree locations or cow locations when people go crazy in details :) [10:13] well maybe not cow unless there's real time tracking objects feature in OSM [10:13] davmor2: no I just want to make it win win win for me :) [10:13] Mirv: it's not working because it hates you, hate it back it works for me ;) It might be all the dev stuff you have in place for other stuff is interfering maybe? [10:14] davmor2: :) I won't install any dev stuff on the Bq (except for stuff like customizating terminal settings etc). but no apt etc hacking. [10:15] Mirv: what you really need is a new dell laptop and a fresh install of vivid that will fix it and your broken screen too :D [10:15] oh right the chroot, yes [10:15] davmor2: I'd like to understand the actual bug in there, I've already gone through two difficult ones but it still now fails at the end of it (used to be: in the beginning) [10:16] but it's annoying since at the end of click chroot creation it wants to unmount my encrypted home after which my desktop is busted [10:16] Mirv: blame pitti or ogra that normally gets things fixed ;) [10:16] davmor2: I don't blame anyone, I file bugs! [10:17] Mirv: you do it all wrong don't you :D [10:17] but true, I'll get a fresh machine soon (~month?). but I hate it when there's some bug that might really affect other people too. [10:18] I'm sure other people are running into that previously filed bug #1436835 too [10:18] bug 1436835 in click (Ubuntu) "click chroot build fails on vivid" [Undecided,New] https://launchpad.net/bugs/1436835 [10:20] Mirv: you could try throwing a kvm install or an lxc install and try running it in there and then figure out the difference between the base install and the version on your desktop to figure out why it hates you [10:20] Mirv: I wonder if it is the QT versions you are building playing havoc with the system maybe? [10:20] davmor2: true, although I can also just run 14.04 LTS from my server and do stuff there [10:20] davmor2: no, I [10:21] davmor2: 'm not building Qt on this machine, but in a chroot on my server machine [10:22] davmor2: similar to that filed bug ^ I've messed around with schroot before and configuration files a lot, and ideally click would detect the most common misconfigurations instead of blowing up [10:22] Mirv: are you testing the QT packages on that machine though? Ie is it expecting to install qt 5.3.1 and getting qt5.3.4 that you've installed or something crazy like that? [10:22] davmor2: one problem is that I've already checked everything I can regarding click & schroot against a clean install and I fail to find anything relevant. that way I found that missing fstab in that previous report. [10:23] davmor2: no, and since Click creates a fresh chroot it does not care what's on the host system [10:23] oh well yes, I do occasionally have Qt PPA:s here and test them, but I also purge them afterwards === MacSlow is now known as MacSlow|lunch [10:53] popey, hey, any idea how can i test a potential fix for the problem of today scope update, without you actually publishing it? [10:56] (i need any scope update to test it) [10:57] pstolowski: good question [10:58] pstolowski: maybe someone on the team has a scope of some kind that we can push to the store under a new test name? [10:58] popey: pstolowski: surely you can just install it locally? [10:58] that could work too, but does the problem trigger with pkcon? [10:59] popey: I'm assuming pkcon is doing the installing from the update [11:00] worth testing in both ways [11:01] davmor2, popey that sounds like a good suggestion; isn't it 'click install --local...' that i need? [11:02] I tend to "pkcon install-local --allow-untrusted foo.click" [11:02] ok [11:03] pstolowski: ^ what he said [11:03] yeah, thanks [11:03] popey, the fix goes only into vivid i presume? [11:04] well as I understand it there will be no more rtm-1409 / utopic based OTAs so vivid makes sense, but check with sil2100 [11:04] Yeah [11:05] sil2100, popey ack [11:05] victorp, who can provide me with an updated version of today scope .click? [11:05] cwayne or kylen [11:06] ty [11:12] victorp, so whats our plan forward, do we hold back scope landings until after next OTA ? [11:12] ogra_, landings to? [11:12] since the fix can only come from a OTA [11:12] scope landings to the store [11:13] yes, at least for the agregators [11:13] that are currently favourites [11:13] ok [11:13] pstolowski, whats up? [11:13] we can still update them via OTA [11:13] that seems to work fine [11:13] yeah [11:14] in general that shows a problem though ... [11:14] being stuck with store landings because we need an OTA for the underlying fix [11:14] i wonder if there is a way around [11:16] ogra_: include the fix in a store package but then that would get messy [11:17] davmor2, well, in this case we couldnt [11:17] cwayne, sent you an email. bbiab === pstolowski is now known as pstolowski|lunch [11:17] davmor2, a click cant ship system bits ... and the rootfs would even have to be writable to replace the broken parts [11:18] in snappy, where the system bits will be framewworks, this will be fixable via the store [11:18] ogra_: unless it is in core [11:19] well, core is small enough that it shouldnt affect us [11:19] but yeah === MacSlow|lunch is now known as MacSlow === chrisccoulson_ is now known as chrisccoulson === greyback__ is now known as greyback === _salem is now known as salem_ [12:06] sil2100: moving back over here. i see pstolowski|lunch is going to talk to cwayne when he wakes. [12:06] sil2100: lets wait for that conversation to happen. [12:06] sil2100, i've been here :) got him a click [12:06] oops popey ^ [12:06] there we go :) [12:07] thanks cwayne (sorry for all the pings) [12:07] 7:30AM meetings ftw [12:07] :) [12:07] or ftl i guess :P [12:07] wake up and the whole worlds on fire [12:08] cwayne: ;p [12:08] Ok, let me wait for pstolowski|lunch to give some input and then we decide [12:08] I want to know how hard this will be to fix [12:09] Well, hm, actually, since this happens in stable there's no way to fix it proper I suppose [12:10] So I'm sure we'll just re-publish it in a few moments === pstolowski|lunch is now known as pstolowski [12:33] sil2100, i'm working on a fix for first part of the issue (https://bugs.launchpad.net/today-scope/+bug/1446499), should have a fix for vivid today [12:33] Ubuntu bug 1446499 in unity-scopes-shell (Ubuntu) "Updating scope from store removes it from favorites" [Undecided,In progress] [12:33] cwayne: I know giving you that petrol can for christmas was a bad idea [12:33] knew even [12:34] sil2100, and there is a slight chance https://bugs.launchpad.net/ubuntu/+source/unity-scopes-shell/+bug/1446536 is not affecting vivid, but i haven't checked yet [12:34] Ubuntu bug 1446536 in unity-scopes-shell (Ubuntu) "Scope disappears from bottom edge on update" [Undecided,Confirmed] [12:34] pstolowski: I suppose all of those issues require changes on the scopes infra, right? [12:35] In this case I guess we can re-publish the scope to the store, nothing we can do short term to fix this [12:35] sil2100, only unity shell plugin [13:01] pstolowski: then I think there's no use holding up the today scope [13:01] pmcgowan: ^ I'll ask popey to re-publish the today scope to the store [13:03] sil2100, I thought there is an issue replacing favorite scopes, or did I miss something [13:03] pmcgowan: there is an issue, but to fix it we would need to promote an image with a quick-fix which is still in progress, not sure if we want that [13:03] sil2100, we shouldn't put it back until that issue is fixed [13:04] pmcgowan: indeed [13:04] pmcgowan: ok, so we agree that we want to re-publish today scope then [13:04] (it's not even a real update) [13:04] sil2100, I was saying not to republish it [13:04] Oh? [13:05] Ah, ok, in this sense you mean [13:05] hm [13:05] my understanding is folks get an update and it "disappeares" [13:05] correct [13:05] From what I know it can be re-added through the favorites, right? [13:05] also correct, after a reboot [13:05] Ah, after a reboot [13:05] we can talk during rtm call, but I think we keep it back [13:05] without a reboot it's just gone, from a user perspective. [13:07] popey: so it's gone from the list without rebooting? Then indeed it's a bit more serious, I got the wrong impression from what victor was saying then [13:07] We could re-spin a new image once the fix is ready [13:08] sil2100, well, se my conversation with victorp above ... there was agreement to hold back all favorite scopes [13:08] until after next OTA [13:08] ACK [13:08] they need to come with custom tarball updates for the moment [13:08] Yeah, missed this conclusion it seems [13:09] sil2100, but the fix needs to target vivid only, correct? [13:09] pstolowski: I suppose right now that's the plan, next OTA is planned to be vivid-based so no need to do a backport for RTM [13:10] rsalveti: Silo 2 approved [13:10] sil2100, great [13:10] sil2100, once it is fixed please let me know [13:10] I don't think pushing the scopes to the store is enough high-priority for OTA-3.5 [13:10] victorp: ACK [13:11] sil2100, no, we always have the custom tarball as fallback ... [13:11] and there it works [13:11] Indeed [13:11] Just as before [13:11] rsalveti: silo 002 is for touch specifically, right? [13:12] Ok, I suppose so, let me publish then [13:18] sil2100, mind if i kick a vivid image to check the space the hud frees up ? [13:18] ogra_: oh, you pushed the hud removal? :) [13:18] ogra_: sure, go on ahead [13:18] yeah [13:18] running [13:21] * sil2100 is still a bit detatched from the world [13:25] === IMAGE 180 building (started: 20150421-13:25) === [13:27] oh, even a round image number :) [13:29] thostr_: what's the status of line 13 scopes bg test silo, is it cancelled or do you want a test silo for it? [13:30] bfiller: what's the status of the syncevolution calendar fix landings, need silos? [13:30] Mirv: line 13 can be removed [13:30] thostr_: ok, thanks! [13:31] Mirv: already in silos, testing now [13:31] bfiller: oh.. so lines 27 and 33 are old ones for the same thing and can be removed? [13:32] * Mirv the cleaner [13:52] Saviq, the phone provisioning script relies on phablet-click-test-setup to install the test sources as directed by the click packages, but this doesn't work for some apps now. I've created an MP to address this: https://code.launchpad.net/~fginther/phablet-tools/skip-missing-click-tests/+merge/256931 [13:54] fginther, cool, I wasn't sure whether this was a critical error, it looked dangerous (it removed all the tests it just downloaded?) [13:54] Saviq, yeah, the error would cause the cleanup to be triggered before copying anything to the phone [13:55] fginther, right, so it only worked for me because I ran unity8 tests, not anything else [13:55] (I mean running the actual tests) [13:56] Saviq, that makes sense, calculator app is what it currently making it bomb on devel-proposed images. I'll work on getting the patch upstream, but it might not before the release [13:57] fginther, 'stood [13:57] boiko: hm, is the request line 58 a critical fix requiring a OTA-3.5? I see the bug is only high [13:57] OTA-3.5?? [13:57] boiko: was there a client's request to spin an rtm image for that? [13:57] sil2100: so, bfiller asked me to get rtm silos and pmcgowan will evaluate the need for a hotfix [13:58] boiko: ok then [13:58] sil2100: same thing for the rtm silo 003 I already have [13:58] Saviq: OTA-3.5, so a hotfix release ;) [13:58] sil2100, should we think about targeting the scope update bug for this too? [13:58] boiko: I thought so, but the severity of the bug got me confused [13:59] Saviq: maybe, although I'm not sure if there are real plans for the hotfix release [13:59] ok ;) [13:59] I'll know more after the RTM status meeting later today [13:59] if there are, we should, if there are none, we shouldn't :) [13:59] rvr, I’ve seen that you’re testing silo 7, thanks! I just added a corresponding manual test to https://wiki.ubuntu.com/Process/Merges/TestPlan/webbrowser-app [14:00] Mirv: silo 27 (ubuntu) and silo 5 (rtm) are the silos we are testing [14:00] rvr, note that most of the functionality is unit-tested already [14:00] oSoMoN: Yeah, I'll begin soon, just had lunch [14:00] Mirv: for sycnevo [14:00] Mirv: not sure what the other lines are [14:00] oSoMoN: Ack [14:01] bfiller: yes, those are lines 54/55, you have also line 27 "Fix for lp:1438662 - Calendar sync not working" for vivid and line 33 for syncing to rtm :) I'll remove those [14:01] Mirv: ack, thanks [14:05] sil2100: I've gone through the spreadsheet now again, finding many "already landed" and such, and now hitting the archive button [14:05] Mirv: thanks for the clean up! :) [14:14] o/ line 44 for an oxide packages copy [14:15] as a binary copy please, to minimize the build time and land in time for the image [14:35] sil2100: can you please free vivid silo 004? we are done testing already, I will remove the row from the spreadsheet [14:36] sil2100: actually quick question: can I just remove the row from the spreadsheet, or should I just clear its contents? [14:36] boiko: hey! [14:36] boiko: let me take care of that - we need to remove the silo as well :) [14:36] sil2100: ok, thanks, it's silo 004 and row 20 [14:37] sil2100: maybe you can reuse that silo for line 45? ;) [14:37] boiko: ;) [14:37] boiko: assigning those in a moment, had to finish up a meeting [14:37] sil2100: no hurry, take your time [14:40] === IMAGE 180 DONE (finished: 20150421-14:40) === [14:40] === changelog: http://people.canonical.com/~ogra/touch-image-stats/180.changes === [14:40] dbarth: regarding the oxide request - I suppose this is supposed to land to the main vivid archive, right? [14:41] ogra_: ^ [14:41] nice [14:41] just hope that removing libdbusmenu-qt5 doesn't break anything [14:41] :-) [14:42] rsalveti: it doesn't have any touch rdeps, so I think we're safe ;) [14:42] sil2100: hmm, considering the final freeze, i would land in the overlay ppa [14:42] yeah if it breaks something it needs a dep somewhere [14:43] sil2100: it will be available in the security updates anyway [14:43] but should be good [14:43] ogra_: how much did we save with such changes? [14:43] down to 462M from 486M for the cdimage tarball :) [14:43] i havent summed up the unpacked size ... but likely around 40-50M [14:44] dbarth: so it will be released to vivid later after final-freeze then? [14:45] yes, as an official security update [14:45] sil2100: chris is copying 1.6.4 there as well anyway [14:46] Ok then [14:46] Oh, actually I thought the next security update will have 1.7 [14:47] dbarth, sil2100 if it lands in security pocket we pick it up from there [14:47] and wouldnt need it in the ppa [14:48] Right, that will be after final-freeze though [14:48] dbarth: do you need to have it ASAP? [14:49] That's why I didn't assign a silo yet, need to clear out all the doubts [14:50] We have plenty of free vivid silos, but oxide is very storage exhaustive [14:51] sil2100, we do need it asap yes [14:51] we need it last week :) [14:51] hah, you don't like the crashing T&C ;) ? [14:53] oh right thats still in there [14:53] I guess the earlier it's on our images the better, more dogfooding time - although it's faring pretty well on 14.09 [14:57] storage> I'm tempted to set the size of all ci-train-ppa-service landing-* archives to 20480 across the board [14:58] They're a bit of a random mix right now [14:58] sil2100: yes, please [14:58] sil2100: mind reconfiguring ubuntu silo 10? I added another project to the list of MR's [14:59] There are two in excess of that, but neither is using all that space right now, and I suspect that they're only set that large because of the lack of garbage collection we had a while back [15:06] bfiller: done [15:06] sil2100: thanks [15:13] sil2100, can i have a silo for row #45? [15:15] pstolowski: on it :) [15:15] awesome, thanks :) [15:23] oSoMoN: Approving silo 7 [15:24] rvr, awesome, thanks! [15:31] dbarth: oxide-qt copied to the oxide silo ;) [15:32] jibel, robru, davmor2, popey, rvr, ogra_: do you want to discuss anything on the landing meeting today? [15:32] sil2100, no [15:32] nope, i think the disaster recovery this morning was fine :) [15:32] nothing to discuss [15:33] sil2100: Is the last episode of Games of Thrones a valid topic? ;) [15:33] rvr: oh shit [15:33] sil2100: it's okay you can say you hate us we won't take it personally :P [15:33] davmor2: ;) === 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: n/a [15:34] davmor2: I'm just a good listener, and robru mentioned multiple times he wants to have time to wake up properly in the mornings [15:34] ;) [15:34] sil2100: pfffff [15:35] * ogra_ hands dbarth bucket and mop for silo 22 [15:37] oops [15:37] :) [15:37] yes, let me check [16:23] ping cihelp, the license check in Jenkins is rejecting https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/generatedApDocs/+merge/256913 the files in question are MIT licensed generated files; adding _build to the excluded folders as can be seen in the branch would resolve it, I don't know where that script is maintained, though [16:27] trainguards: silo 7 is landing, and I’m seeing that webbrowser-app is in the proposed pocket. I thought it would go directly into the overlay PPA, what’s up with that? [16:27] oSoMoN: let me take a look [16:28] oSoMoN: ouch, ok, someone set the column name but didn't reconfigure [16:28] Ok, let me reconfigure and re-publish - thanks for noticing! [16:29] sil2100, I did change the destination PPA column after reading slangasek’s announcement, sorry I didn’t know it required a reconfigure [16:30] oSoMoN: no worries, I should have double checked when publishing, I usually do but it was probably published in a hurry ;p [16:31] sil2100, alright. I guess the package can be easily rejected from proposed so as to not affect the archive, right? [16:31] sil2100: can you copy the package from -proposed back to the overlay ppa? then I can delete it from -proposed [16:32] slangasek: sure thing, on it now [16:36] slangasek: ok, package copied (should be in the overlay PPA) [16:37] * sil2100 switches context back to UITK [16:37] sil2100, oSoMoN: ok. In thinking about it, I realize that removing the package means it won't be re-copiable back to w when it opens without a new upload; so I think I'm going to leave the package in -proposed for now, which is valid [16:38] slangasek: makes sense, indeed [16:39] slangasek, sil2100: thanks [16:39] sil2100, can I merge and clean the silo now? [16:40] oSoMoN: yes :) [16:40] Just use the force flag if anything [16:40] ok [16:48] sil2100: don't forget the dashboard is authoritative; check for 'stable-phone-overlay' there before clicking publish [16:50] robru: right'o! Forgot about that, was actually checking in the JSON configs [16:51] sil2100: yeah, sorry I should have added that info to the dashboard sooner. Didn't think of it until yesterday morning. [16:51] robru: I read that you added that, but somehow with all the stuff happening it just jumped out of my head [16:52] kalikiana: Looking... [16:58] dbarth: Approving silo 6 [17:00] o/ [17:01] mardy: dbarth: https://ci-train.ubuntu.com/job/ubuntu-landing-006-2-publish/88/console need some MPs approved [17:01] sil2100: good evening! [17:06] sil2100: I think you need to fix the version of the gles [17:07] bzoltan_: eek, will do that, maybe some typo [17:07] In one moment, OTP now [17:10] Ok, on it now [17:13] bzoltan_: hmmm, the version looks fine to me, I can see it in the PPA [17:13] Did I do something wrong? [17:13] bzoltan_: argh, and I see another problem ;/ [17:15] bzoltan_: why do I see other changes in the PPA's changelog? [17:15] Isn't trunk = what's in the archive? [17:16] Ah! [17:16] Ok, it's all ok [17:16] It's in the overlay PPA and the diff is made against the archive [17:16] nvm [17:24] bzoltan_: could you check the PPA and merge and tell me what I'm doing wrong? I see that CI Train builds the package correctly, so the versionin needs to be ok, but nothing gets uploaded to the PPA [17:30] sil2100: ask an archive admin to check why the upload was rejected. I don't see any obvious reason for it (eg, that version doesn't already exist in the ppa as far as I can see...) [17:31] robru: I poked cjwatson, but maybe I could poke slangasek as well [17:31] sure [17:32] slangasek: hey, you have a moment? [17:34] slangasek: I handled sil2100's request on #ubuntu-touch [17:34] sil2100: slangasek doesn't have LP log access AFAIK [17:34] slangasek: ...unping [17:34] cjwatson: thanks :) [17:34] cjwatson: I think I'll try to fetch the source and re-upload myself with debuild -S -sa to make sure we upload the tarball [17:34] sure [17:34] I have no idea why it didn't include it in the upload [17:37] plars: any idea at all? [17:40] kalikiana: we were discussing how to handle it, fginther has an MP for it in flight [17:52] ogra_: btw. smoketesting is busted for 180... [17:52] sigh [17:54] sil2100, thats your falt though ... [17:54] *fault [17:54] package ubuntu-ui-toolkit, version 1.2.1485+15.04.20150417.1-0ubuntu1 not found in https://api.launchpad.net/devel/ubuntu/vivid [17:54] Oh uh! [17:54] i guess you want to re-roll the immage after that landed :) [17:54] to have the versions match [17:55] ogra_: not particulary my fault, more like 'our fault' and pointing at the LT ;p [17:55] No no, it's because it's in the overlay PPA [17:55] ah [17:55] A re-spin won't fix it [17:55] right [17:55] * sil2100 sighs ;) [17:55] Actually, damn, I might need to change my commitlog generation scripts too ;/ [17:55] robru, will the citrain script take the overlay ppa into acocunt somehow ... [17:55] (and does it need to ?) [17:55] robru: on it [17:56] or do we just assume everything yu need for testin is actually in the image or silo [17:56] dbarth: thanks [17:56] ogra_: it shouldn't need to, as those packages get rolled into image builds === chihchun is now known as chihchun_afk [17:57] robru, right, but if the image is a day behind and the silo got built against the overlay PPA your silo packages might be uninstallable [17:57] until the image has the overlay packages available [17:57] thats surely only a corner case though [17:57] ogra_: the silo ppas now depend on the overlay ppa so it should take care of itself shouldn't it? [17:58] not with the citrain script where we /dev/null all sources.lis entries except the silo [17:59] the overlay package would not be available ... but the silo has a dep against it [17:59] as i said, likely just a corner case ... but it could happen [17:59] ogra_: i don't really understand how ski dependencies work [17:59] * sil2100 really needs to go AFK now [17:59] o/ [17:59] Be back later [18:00] ogra_: "silo dependencies". Anyway if it happens you'll just have to install the overlay ppa and dist upgrade [18:00] robru, well, the image has foo-123 ... i upload foo-124 to the overlay ppa ... now you push package bar to a silo that depends on foo ... and gets a binary dep via shlibdeps against foo-124 ... [18:01] bar will be uninstallable in the image unless i enable the overlay ppa or until the foo package hits the image [18:01] right, if thats our answer that is fine [18:02] (it might taint your image and the test might be moot though ) [18:03] ogra_: well, I've seen in some cases already that necessary deps were copied from distro into a silo ppa, specifically because citrain tool doesn't add distro updates on purpose. So if such a case arose, people should just copy the dep from the overlay ppa back to the silo. [18:04] cool, thats fine then [18:05] ogra_: i think it's a rare case. If it happens a lot we can consider changing the tool, but i believe qa specifically requested the tool only install silo contents, and nothing else. [18:05] yeah [18:20] kalikiana: bzoltan_ ping? Need merges approved https://ci-train.ubuntu.com/job/ubuntu-landing-021-2-publish/48/console [18:59] dbarth: uh, so empathy was in your silo and that got copied into the overlay ppa, was that on purpose? should we delete that? [19:00] dbarth: I mean it was in your silo PPA but not in your silo config, seems like a mistake. [19:01] Laney: ^ you uploaded this empathy into silo 6 PPA, did you intend for that to wind up in the overlay PPA? [19:03] Laney: I guess if you wanted that uploaded to distro you'll have to copy it from the overlay ppa yourself. https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay [19:50] robru: do you still need my approval? As I see the MRs are landing [19:51] bzoltan_: not sure where you see that? the link says "ERROR" and https://code.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/generatedApDocs/+merge/256970 is still unapproved [19:51] (I approved one because I could, I don't have permission on the other) [19:52] robru: ohh.. the main. I approved [19:52] robru: yes, the gles you could, thanks for that [19:52] bzoltan_: great, thanks [20:09] robru: it does seem like a mistake indeed [20:10] robru: i wonder why it got pulled in, except that for qa purposes you needed to make sure that empathy was not listed in teh FB/MSN account settings [20:48] plars, Hi! [20:48] plars, the dashboard is showing multiple runs as 'Running' wonder whats causing that ? [20:51] om26er: it looks like phablet-click-test-setup blew up quite badly [20:51] it seems to be causing us all sorts of grief lately === salem_ is now known as _salem [20:53] plars, can you report bugs for phablet-click-test-setup, I can look into fixing those ? [20:56] om26er: fginther has a patch for the one we hit last week, not sure what this one is yet [20:59] om26er, plars, that's an error I had not seen before [20:59] fginther: yeah, it seems to have started in the latest image [21:01] plars, can you share the stacktrace ? [21:02] om26er: https://jenkins.qa.ubuntu.com/job/vivid-touch-mako-smoke-daily/544/console [21:03] package ubuntu-ui-toolkit, version 1.2.1485+15.04.20150417.1-0ubuntu1 not found in https://api.launchpad.net/devel/ubuntu/vivid [21:03] ti seems something change in uitk? [21:04] there is a new uitk uploaded today, yes. [21:17] plars: om26er: the new one from today is version ...0421, and it's sitting in the UNAPPROVED queue (eg not in the archive at all). shouldn't have anything to do with this ...0417.1 issue you're seeing [21:21] indeed, ...0417.1 doesn't exist: https://launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+publishinghistory [21:23] robru: om26er: this ought to be simple to reproduce locally by doing a fresh install and running phablet-click-test-setup [21:23] plars: fresh install of what? you mean the latest image on a krillin? [21:24] robru: yes, or even on mako [21:25] plars: is it picking up 0417.1 from a ppa or something? [21:25] plars, robru, om26er, indeed I can reproduce it locally. The problem appears to be that 1.2.1485+15.04.20150417.1-0ubuntu1 is coming from a PPA and not the archvie [21:26] phablet-click-test-setup does not know how to extract sources from PPAs [21:26] heh [21:26] robru: right [21:26] so I guess this is the overlay ppa because vivid is frozen [21:26] oh duh, 0417.1 is in the overlay ppa [21:26] plars: yeah [21:27] fginther: plars: so I guess p-c-t-s needs to grow awareness of the overlay ppa [21:27] om26er: ^ [21:27] robru, yep... trying to think of how this might be done [21:28] fginther: catch that exception and fallback to downloading the package from the ppa? [21:32] robru, ideally it could just find the PPA from looking at the attached phone [21:33] fginther: but the phone doesn't have the ppa enabled; the packages from the ppa are rolled right into the images. [21:33] as far as I know [21:34] robru, my mako does indicate that uitk came from a PPA, is this correct? [21:34] $ apt-cache policy qtdeclarative5-ubuntu-ui-toolkit-plugin [21:34] qtdeclarative5-ubuntu-ui-toolkit-plugin: [21:34] Installed: 1.2.1485+15.04.20150417.1-0ubuntu1 [21:34] Candidate: 1.2.1485+15.04.20150417.1-0ubuntu1 [21:34] Version table: [21:34] *** 1.2.1485+15.04.20150417.1-0ubuntu1 0 [21:34] 500 http://ppa.launchpad.net/ci-train-ppa-service/stable-phone-overlay/ubuntu/ vivid/main armhf Packages [21:35] fginther: oh, ok, great. I wasn't aware that it worked that way. looks right [21:55] robru, plars, I can find the uitk source urls with the follow two queries: https://api.launchpad.net/devel/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay?ws.op=getPublishedSources&source_name=ubuntu-ui-toolkit&source_package_version=1.2.1485+15.04.20150417.1-0ubuntu1 [21:55] then https://api.launchpad.net/devel/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay/+sourcepub/4955514?ws.op=sourceFileUrls [21:55] boiko: ping [21:56] I'm trying to hack in some changes, but I'll have to jump offline for a few hours soon [21:56] boiko: You sent another silo to ubuntu-rtm [22:13] plars, om26er, robru, this appears to work (it's a huge hack) [22:13] https://code.launchpad.net/~fginther/phablet-tools/skip-missing-click-tests/+merge/256931 [22:13] * fginther -> away for a while [22:40] plars, can you please deploy that ? [22:43] * om26er >> out