[00:57] <robru> jhodapp: sorry for the spam, was using your silo to debug an issue in production.
[00:57] <robru> i think it's fixed now
[00:57] <jhodapp> robru, dang
[00:58] <jhodapp> robru, did I move silos now?
[00:59] <robru> jhodapp: oh yeah, sorry, that was a separate issue, had to free you and reassign it. you're in 5 now
[00:59] <robru> rtm5
[00:59] <jhodapp> robru, ok
[01:00] <robru> jhodapp: ok, there's a package in your PPA, so the issue I was debugging is fixed, *PHEW* ;-)
[01:00] <jhodapp> robru: good news :)
[01:00] <jhodapp> robru, hopefully after I get a review by ricmm this weekend, I'll be landing that
[01:01] <robru> jhodapp: My bug rendered citrain completely incapable of building *anything* in RTM ;-)
[01:01] <jhodapp> robru, oh man, no good
[01:01] <robru> jhodapp: at 6PM on a friday ;-)
[01:01] <robru> jhodapp: but I fixed it! Now it's time for dinner...
[01:02] <jhodapp> robru, so when I land in Ubuntu and want to then land in the ubuntu-rtm, what do I need to do?
[01:02] <jhodapp> robru, there were so many emails sent about that I got so confused
[01:02] <robru> jhodapp: oh yeah it's rubbish
[01:03] <robru> jhodapp: basically now you have to make two requests, the utopic one as normal, and then the rtm one is just a copypaste, but change your MP to empty and your source packages can just say 'sync:' (we'll fill in what to sync after the utopic silo lands).
[01:03] <robru> jhodapp: when in doubt, copy what other people are doing. there's lots of examples in the spreadsheet
[01:04] <jhodapp> robru, so then that gets built for the ubuntu-rtm branch, and then both need to be manually landed by the CI team?
[01:05] <cwayne1> did we do the rtm fork a bit early?
[01:05] <cwayne1> just based on the fact that like more than 50% of the MPs then include a sync:
[01:06] <jhodapp> cwayne1, yeah, when would an MP only be landed in one anyway?
[01:07] <robru> jhodapp: utopic needs to publish before rtm starts to build
[01:07] <robru> bbl, dinner for real
[01:07] <jhodapp> robru, ok cool
[01:07] <jhodapp> enjoy
=== robru changed the topic of #ubuntu-ci-eng to: Train support: trainguards | Vanguard: cihelp | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: robru is tearing this train apart. Ping him if any jenkins jobs have unexpected results.
[01:10] <cwayne1> jhodapp: right? no idea
[01:30]  * ToyKeeper wonders what's with everyone trying to land huge silos or core app changes at the last minute on a Friday
[02:05] <imgbot> === trainguards: IMAGE 229 building (started: 20140906 02:05) ===
[03:05] <imgbot> === trainguards: RTM IMAGE 21 building (started: 20140906 03:05) ===
[03:23] <bzoltan> cjwatson: OK, i have approved the MR
[03:50] <imgbot> === trainguards: IMAGE 229 DONE (finished: 20140906 03:50) ===
[03:50] <imgbot> === changelog: http://people.canonical.com/~ogra/touch-image-stats/229.changes ===
[04:05] <tedg> robru, K, done
[04:20] <imgbot> === trainguards: RTM IMAGE 21 DONE (finished: 20140906 04:20) ===
[04:20] <imgbot> === changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/21.changes ===
[08:07] <popey> upgraded my krillin to 21 OTA and unity8 is in an apport/crash loop. going back to #20
[08:07] <popey> same happened when i went 16->17
[08:09] <asac> hmm. for me it was a bit flaky ... e.g. hart to convince to do the upgrade (had to go to settings a few times)
[08:09] <asac> but 21 is running here
[08:19] <popey> went back to 20
[08:19] <popey> i am wondering if there's some issue with download/checksum here
[08:20] <cjwatson> bzoltan: thanks
[08:20] <popey> because when i went 16->17 the download took forever
[08:20] <bzoltan> cjwatson: Just ping me when this MR lands so I can port it to Trusty and release in the SDK PPA.
[08:23] <cjwatson> ok
[08:27] <cjwatson> sigh, mvo is offline and I can't set a commit message there
[08:28] <cjwatson> bzoltan: would you mind setting the commit message on that MP to "Run pkcon with --allow-untrusted (LP: #1360582)." ?
[08:28] <ubot5> Launchpad bug 1360582 in PackageKit "Can't manually install clicks "Signature verification error" since #205" [Medium,Confirmed] https://launchpad.net/bugs/1360582
[08:28] <cjwatson> then I can get on with building it
[09:46] <bzoltan> cjwatson:  done
[09:46] <bzoltan> Sorry, my response time is not the best on Saturday. I had to suspend two girls.
[09:48] <bzoltan> ogra_: sergiusens: It seems that the `phablet-config edges-intro --disable` does not do what used to do. The device boots to the edges intro.
[09:49] <ogra_> bzoltan, yes, thats a bug in cgroups, mterry and hallyn were working on it all day yesterday
[09:49] <ogra_> or rather in the combo of cgroups and lightdm as i understood
[09:50] <bzoltan> ogra_:  OK, good to know. It means I can not reboot between AP tests.
[09:51] <ogra_> you could hack something together that runs as root ... and upstart job with the dbus call from "phablet-config edges-intro" for example
[09:52] <ogra_> (the prob is that the user has no permissions to talk to the accountservice via dbus)
[09:52] <bzoltan> ogra_: that sounds doable
[09:52] <bzoltan> ogra_:  ehh
[10:05] <bzoltan> ogra_:  is there any way to prevent screen lock?
[10:14] <popey> bzoltan: powerd-cli active
[10:14] <popey> dunno if that's the "official" way but that's how I do it.
[10:14] <bzoltan> popey:  thanks
[10:14] <popey> keeps the screen on
[10:23] <ogra_> there is some new unlock code as well in the unity8-autopilot package iirc
[11:04] <asac> ogra_: do you know if johns tarball made it into rtm branch yesterday?
[11:04]  * asac didnt get an escalation so he assumes it is in
[11:06] <ogra_> in image 20
[11:06] <asac> nice
[11:07]  * asac crosses finger that this powerd issue is really not comgin back
[11:07] <asac> so far the phone is still up
[11:08] <ogra_> i havent seen it in more than a week here
[11:09] <asac> i saw it every day
[11:09] <asac> ogra_: guess you are not running around with your phone much :)
[11:09] <asac> and reboot regularly?
[11:09] <ogra_> no, i dont reboot unless something is seriously broken
[11:10] <ogra_> but i charge over night so it only has long sleep phases during the day
[11:10] <asac> interesting
[11:10] <asac> ok
[11:10] <ogra_> and these are rarely more than 2h
[11:10] <asac> i dont know what long is though
[11:10] <asac> think long is long enough for deep sleep
[11:10] <asac> doesnt matter how long in that sleep mode i would think
[11:10] <ogra_> probably we have multiple stages of deep sleep in the HW
[11:13]  * asac happy that gmail webapp seems to not forget his cookies anymore
[11:13]  * asac hugs dbarth if he would be here
[11:18] <bzoltan> ogra_:  I am running tests on #229. the good news is that actually some of the tests pass
[11:18] <ogra_> cool
[11:20] <asac> some :)
[11:25] <bzoltan> even better... all sudoku app tests pass :) the first fully validated app
[11:27] <ogra_> are you testing rtm or utopic ?
[11:30] <ogra_> asac, seeing your "word suggestion" mail ... try a sudo command in the terminal-app with that on ;)
[11:31] <ogra_> (and enjoy having your sudo pasword from now on in the word suggestion cache :P
[11:31] <ogra_> )
[11:35] <bzoltan> ogra_:  utopic #229
[11:35] <ogra_> k
[11:50] <asac> u know what i hate
[11:50] <asac> auto correction
[11:50] <asac> thats the most useless thing in the world
[11:50] <asac> really
[11:51] <asac> whoever invented that had no clue how folks dont want to type proper german or english
[11:52] <asac> ogra_: it doesnt remember my pin
[11:52] <asac> if i dont tell it to
[11:52] <asac> what it does though is post it first thing on terminal after i enter it to get acecess to terminal
[11:53] <asac> guess thats a bug though
[11:53] <asac> auto suggestion doesnt auto remember
[11:53] <asac> you have to click the word after typing so it adds it to its dictionary
[11:53] <ogra_> it will show your pin as first suggestion in the wordlist if you enter the first char of it for the duration of the terminal session
[11:53] <asac> at least thats how i learned it and how i see it here
[11:53] <asac> ogra_: no it doesnt
[11:53] <ogra_> it does here
[11:54] <asac> not for me
[11:54] <ogra_> if my pin is 1234 and i type 1 the first suggestion on the left is always 1234
[11:54] <asac> i think you accidentially added it to your dict
[11:54] <asac> by clicking on it when it was in the suggestion bar while typing
[11:54] <ogra_> how wouold i do that ??
[11:54] <ogra_> hmm
[11:54] <asac> ogra_: so when you type you see the suggestions and on the left you see what you type
[11:54] <asac> if you click on that it will add that to dict
[11:54] <asac> try with brbriwer
[11:55] <asac> it will do a whitespace after you hit it and add to dict
[11:55] <asac> and its the ONLY way to actually leave a word that isnt suggested like it is if you have  auto correction
[11:55] <asac> so if i want to write with autocorrection "mennomeno" i have to click it ohterwise it gets transformed into something else
[11:55] <asac> :(
[11:55] <asac> and then its in dict forever
[11:56] <asac> anyway, in general apps should be able to set modes
[11:56] <asac> that dont have auto suggestions
[11:56] <asac> like password
[11:56] <asac> otoh, in browser you can also remember you pass :)
[11:56] <asac> but thats different
[11:56] <asac> ogra_: i dont know where the ddictionary is kept
[11:56] <asac> but i am sure you will find it in there
[11:57] <asac> oh ... there is another bug
[11:57] <asac> 1. after i unlock terminal it prints the pin in the first line
[11:57] <ogra_> it is gone when i close the terminal
[11:57] <asac> 2. and of course at that point it shows the 3050
[11:57] <asac> but only for the first line
[11:57] <asac> after that
[11:57] <asac> weird
[11:57] <ogra_> only happens within one session
[11:57] <asac> but thats a bug
[11:57] <asac> right, but for me it doesnt happen after the second line
[11:57]  * asac tries again
[11:58]  * ogra_ sighs ... why is krillin always behavinng different :(
[11:58] <asac> 1. kill terminal
[11:58] <asac> 2. open terminal enter pin
[11:58] <asac> 3. click in terminal pastes my pin :)
[11:58] <asac> 4. /me uses backspace to delete
[11:58] <asac> 5. type ls
[11:59] <asac> 6. type first digit of pin -> does not suggest my pin
[11:59] <cjwatson> bzoltan: thanks.  response time entirely understandable, I don't expect people to be around on Saturday :)
[11:59] <asac> thats what i experience
[11:59] <asac> ogra_: try that :)
[11:59] <cjwatson> asac: autocorrect is an utter pain when I'm trying to type Irish ...
[11:59] <asac> you just not hit enter after the pin gets pasted on first line i guess
[12:00] <ogra_> hmm, right
[12:00] <ogra_> doesnt happen anymore
[12:00] <asac> cjwatson: hey :) yes, i hate it a lot for whatever. i use my own shorthands and special english-germano self inventions and it ONLY gets in my way :)
[12:00] <asac> and the bad is you dont notice that it changed your word until after you send the text :P
[12:00] <asac> i think i have it never do the right correction for me in android
[12:01] <asac> cant imagine it helps anyone else :P
[12:01] <cjwatson> I'm fine with autocorrect being offered (autosuggest or whatever); the annoying bit is when it's automatically applied when you go to the next word by entering space or whatever
[12:01] <cjwatson> that just gets in the way
[12:01] <asac> yes thats autocorrection
[12:01] <asac> suggestion is fine, just dont bloody do it :)
[12:01] <cjwatson> amen
[12:01] <asac> see the mail thread :)
[12:01] <asac> i think we should turn on suggestion, but nothing else by default
[12:02] <asac> ogra_: its weird... i think the big problem is the initial paste
[12:02] <asac> which is super buggy
[12:02] <asac> ogra_: a) why does it paste anything when i first click in the terminal
[12:03] <asac> b) why is my pin in the clipboard
[12:03] <asac> -> that certainly is a security bug too
[12:03] <asac> ogra_: who would know? mterry? where do i file this bug against?
[12:04] <mterry> asac, for Terminal?
[12:04] <ogra_> terminal-app i would guess
[12:04] <asac> mterry: yeah see the steps here:
[12:04] <asac> 13:57 < asac> 1. kill terminal
[12:04] <asac> 13:57 < asac> 2. open terminal enter pin
[12:04] <asac> 13:57 < asac> 3. click in terminal pastes my pin :)
[12:04] <asac> 13:58 < asac> 4. /me uses backspace to delete
[12:04] <asac> 13:58 < asac> 5. type ls
[12:04] <mterry> asac, I'm guessing they forgot to turn off the predictive hint for their keyboard entry
[12:04] <asac> 13:58 < asac> 6. type first digit of pin -> does not suggest my pin
[12:04] <mterry> asac, I think the lockscreen also does that, shame on me
[12:04] <asac> look at step 3
[12:05] <mterry> asac, I don't know why it's in the clipboard
[12:05] <asac> i enter pin and in the unlock screen and its in the clipboard it seems
[12:05] <asac> i dont know why terminal pastes when i click in it :)
[12:05] <asac> but it must come from somewhere
[12:05] <asac> hehe
[12:05] <mterry> asac, but I remember hearing someone blame the OSK / predictive text for that
[12:05]  * asac doesnt know if its in the clipboard
[12:05] <asac> just assumes
[12:05] <asac> mterry: for stuff being in the clipboard?
[12:06] <mterry> asac, well, for the text showing up like that, wherever it came from
[12:06] <asac> mterry: read the instructions
[12:06] <asac> it doesnt show up :)
[12:06] <mterry> asac, well it's pasted in?
[12:06] <asac> (as suggestion)
[12:06] <asac> yeah
[12:06] <asac> but not as suggestion
[12:06] <asac> just as real content
[12:06] <asac> it is not suggested after that (unless you accept it of course)
[12:06] <mterry> asac, I think someone mentioned the same problem with the lockscreen and pasting text into app
[12:07] <asac> intersting
[12:07] <asac> mterry: where shall i file this against?
[12:07] <mterry> asac, and at the time it was theorized it was due to the OSK's treatment of predictive text
[12:07] <asac> so we dont forget to look into it
[12:07] <asac> predictive text == suggestion?
[12:07] <mterry> asac, with the suggestion that the lockscreen should really disable predictive text for that field (or really, maybe the password hint should automatically disable it)
[12:07] <mterry> asac, I think so yes
[12:07]  * asac turns all that stuff off and tries again
[12:08] <mterry> asac, well I have that off and I still see it
[12:08] <mterry> asac, I'd guess OSK for now, to file the bug
[12:08] <asac> mterry: what do you have off? all options? or just auto correction?
[12:08] <asac> this time with all options turned off it didnt happen
[12:08] <mterry> asac, uh... I think I have the default set of options
[12:08]  * asac tries more options
[12:08] <asac> mterry: can you check in settings -> language?
[12:08]  * asac doesnt know what the defaults are
[12:09] <asac> yeah so just word suggestion withtout anything causes this
[12:09] <asac> so indeed
[12:09] <mterry> asac, uh no, I can't check right now actually -- but I remember that I saw this bug with a test device, not my dogfood one, so it was with the new default of predicitive=on
[12:09] <asac> guess auto suggestion uses clipboard to get the current text to then look in dictionary
[12:09] <asac> mterry: dont worry
[12:10] <asac> have confirmed its the suggestion feature
[12:10]  * asac will file bug
[12:10] <asac> whats the package?
[12:10] <ogra_> ubuntu-keyboard ?
[12:10] <asac> really?
[12:11]  * asac tries
[12:11] <ogra_> well, that handles all auto suggestion
[12:16] <asac> ok bug 1366314
[12:16] <ubot5> bug 1366314 in ubuntu-keyboard (Ubuntu) "security issue? auto suggest seems to copy credentials into clipboard" [Critical,New] https://launchpad.net/bugs/1366314
[12:17] <asac> subscribed jamie and marc
[12:17] <asac> will shoot them a mal so they keep that on their radar
[12:20] <ogra_> is it my fingers or is the krillin touchscreen a lot more sensitive recently ? +
[12:21] <ogra_> i can hardly scrolll in webapps without tapping a link
[12:23] <cjwatson> whoops, I forgot citrain inserts bug refs
[12:23] <cjwatson> oh well, just a slightly noisy qtcreator-plugin-ubuntu changelog
[12:24] <ogra_> heh, i gave up wondering about weird changelogs
[12:24] <cjwatson> things I love: ppc64el (and nowadays powerpc too) build times.  we should just ram power8s into all our phones
[12:25] <ogra_> haha
[12:25] <ogra_> its getting winter soon ... i could need a pocket oven :)
[12:25] <asac> power is cool and powerful :P
[12:25] <asac> they  dont waste energy on heat... just pure computing :P
[12:25] <asac> lol
[12:26] <ogra_> yeah, i guess sparc would be the better oven here
[12:26]  * cjwatson sits back and contemplates the laws of thermodynamics
[12:27] <asac> heh
[12:29] <asac> i heard go will reduce build times a lot for us :P
[12:29] <ogra_> lol, go ...
[12:29] <asac> no more C++ energy waste
[12:29] <ogra_> go away
[12:30] <cjwatson> it had better, since once we have go shared libraries it sounds like we'll be rebuilding the universe frequently :)
[12:31] <asac> cjwatson: you say with shared library we will do more rebuilding?
[12:31] <asac> or was that a typo?
[12:31] <cjwatson> see the thread on -tech
[12:31] <asac> hmm
[12:31] <cjwatson> the current upstream strategy for go shared libraries will basically bump the soname equiv on every source change
[12:31] <cjwatson> at least to start with
[12:32] <cjwatson> I'm ok with it as long as it improves later :)
[12:32] <ogra_> another regular haskell bomb for -changes
[12:32] <ogra_> :)
[12:32] <asac> heh
[12:32] <cjwatson> exactly, except with runtimes as well
[12:32] <cjwatson> closer to current ocaml
[12:32] <cjwatson> so ... just as well we know how to deal with that kind of thing now ;)
[12:36] <asac> what was the motivation driver for folks doing a shared library approach like they did now?
[12:38] <cjwatson> I think it's effectively the simplest possible strategy
[12:38] <cjwatson> they did flag it as something they might well refine later
[12:38] <asac> right. but did they do this effort because of us (aka distros) asking for it or because they found their own reason to need this?
[12:39] <cjwatson> oh, right, not totally sure but we have definitely been asking about it, and they noted that part of their proposal was explicitly targeted at distribution builders
[12:39] <cjwatson> the proposal is a bit broader than just shared libraries for distros; I only read parts of it
[12:45] <bzoltan> as expected the unity8 tests are not really happy with the lockings
[12:49] <cjwatson> how do I get click and packagekit into the silo config for ubuntu-rtm/landing-003?
[12:50] <cjwatson> is the spreadsheet syntax wrong there?  hmm, maybe it needs to lose the :version
[12:50]  * cjwatson tries that
[12:51]  * ogra_ notes that lxc-android-config made it into the archive and triggers a utopic image build
[12:53] <cjwatson> much better
[12:56] <cjwatson> asac: I've mailed mvo to let him know that's done, but not expecting him to test over the weekend
[12:58] <asac> cjwatson: yes perfect.
[12:58] <asac> thanks!
[12:58] <asac> cjwatson: if its in rtm silo qa will also help testing and give more assurance; they just need goo guidance what type of stuff could happen
[12:59] <asac> anyway...
[12:59]  * asac steps out for couple hours
[12:59] <asac> hope you guys are having weekend then
[12:59]  * asac waves
[12:59] <cjwatson> yeah, will send it off to QA on Monday
[12:59] <cjwatson> I'm off too, I smell lunch
[13:00] <imgbot> === trainguards: IMAGE 230 building (started: 20140906 13:00) ===
[14:35] <imgbot> === trainguards: IMAGE 230 DONE (finished: 20140906 14:35) ===
[14:35] <imgbot> === changelog: http://people.canonical.com/~ogra/touch-image-stats/230.changes ===