[00:06] when does it come out [00:07] so I have a toro version of a galaxy nexus, and I want to try to install the takju galaxy nexus builds of ubuntu touch on it to see what happens [00:07] phablet-flash -b -d takju doesn't work [00:08] anyone have any advice? [00:08] when does it come out [00:09] I just want to know how to tell phablet-flash "just pretend install the builds for device y, ignore that you don't recignize the device" [00:09] just install* [00:12] nvm, i'm dumb [00:12] -d maguro [00:20] hi === chriadam|away is now known as chriadam [00:33] when does it come out === chihchun_afk is now known as chihchun === chriadam is now known as chriadam|away === VDVsx_ is now known as VDVSx [04:56] clear [04:56] does Ubuntu Touch use wpa_supplicant for encrypted wifi? I cant find /data/ubuntu/etc/wpa_supplicant.conf [04:58] I found /data/ubuntu/etc/wpa_supplicant/ but it just has 3 .sh scripts in it [04:59] ps aux | grep wpa shows... [04:59] /sbin/wpa_supplicant -B -P /run/sendsigs.omit.d/wpasupplicant.pid -u -s -O /var/run/wpa_supplicant [04:59] looks like its running but not sure where the conf file is :( [05:02] oh snap............ theres a command line interface with wpa_cli === forcev is now known as Funkypenguin [06:54] Hi, is there an arm-eabi-gdb executable that works under ubuntu touch, and if so, how do I find it? [07:40] ok who is the wise one to break wpa_supplicant [07:41] just did apt-get dist-upgrade rebooted and now its not running anymore [07:45] is there any where do download ubuntu for android source code? is it released for a phone, or is ubuntu edge the first phone to get it? [07:48] reffering to this http://www.ubuntu.com/phone/ubuntu-for-android [07:48] effbiai, neither binary nor source of UfA are public [07:48] if the edge happens it will be released alongside though [07:49] DJJeff, use nm-cli, dont use wpa_supplicant directly, this will most likely break stuff [07:51] -bash: nm-cli: command not found === hikiko is now known as hik|afk [08:00] DJJeff, ah, sorry, no dash === chihchun is now known as chihchun_afk [08:00] if you already fiddled with the setup it might indeed now work though [08:01] *not [08:01] DJJeff, is there any reaso why you dont use the UI btw ? [08:03] (phablet-network-setup will help too in case your host PC has wlan) [08:08] my UI shows the networks but wont let me type a password [08:08] will only put a check mark [08:08] even if I select more then one network they all get check marks [08:09] ran tail -f /var/log/syslog in my ssh session and got nothing === chihchun_afk is now known as chihchun [08:10] DJJeff, does your host PC have a wlan connection by chance ? [08:10] nope [08:10] ogra_, do you know if it is easy to compile an android binary? Just a small c file that will open an android library and call a function in there. [08:10] ah, well, then nmcli it is [08:10] ogra_, using hybris in that scenario would be overkill perhaps [08:11] diwic, xnox did roll a bionic/android toolchain we build the android packages with ... i need to look up the package name ... [08:11] gcc-arm-linux-androideabi [08:11] you should be able to cross build for android with it [08:11] ah, thanks [08:12] cjwatson, is that the package name? [08:13] yes [08:13] cjwatson, aha, it is only available on amd64, not on armhf itself [08:15] diwic, for cross building, yes [08:16] I only wish my WLAN0 TXPOWER was 1496 dBm lol [08:16] http://i.imgur.com/0bjS6JM.png [08:16] peed my pants from laughing so hard [08:17] well, again, dont use wpa_supplicant directly ... [08:18] nmcli d help [08:18] start from there [08:18] just googled this [08:19] http://pravin.paratey.com/posts/manage-wireless-networks-with-nmcli [08:19] nmcli -p dev wifi list [08:19] returned nothing for me :( [08:20] what phone is that ? [08:20] rfkill list shows that nothing is blocked [08:21] and it worked just fine before I did apt-get dist-upgrade [08:21] with nmcli ? [08:21] editing wpa configs doesnt count [08:21] I was using wpa_supplicant before with wpa_cli [08:21] right ... [08:22] root@ubuntu-phablet:~# iw wlan0 scan | grep SSID [08:22] shows my network just fine [08:23] well, so should: nmcli d wifi list [08:24] it does not :( === hik|afk is now known as hikiko [08:25] well, the good question would be now ... did it before you tinkered with the wpa settings ... but i guess you didnt try [08:27] http://puu.sh/3Ubc6.png === schwuk_away is now known as schwuk [08:30] DJJeff, well, its hard to tell if it isnt whon because you changed the wpa defaults or if it has to do anything with your disct upgrade [08:30] *shown [08:31] * ogra_ would just flash freshly to make sure nothing is messed up, and then use nmcli on a fresh system .... [08:32] if you hack around it now by not using teh default tools, it will likely just break again on a subsequent upgrade [08:33] seems that each time I set something up and get it working it breaks when I upgrade :( [08:33] I really need to stop upgrading [08:33] or just wait for this to become more stable before diving into the settings [08:33] what device are you on ? [08:34] Samsung GT 10.1 Wifi (p4wifi) [08:34] (also be extremely careful with upgrading, there are shared pieces in android that arent packaged so you need to re-flash a freshly built zip alongside on upgrades) [08:35] (wont influence wlan though ... ) [08:35] past upgrades have been ok (for the last 3 weeks) [08:36] untill today where I noticed network-manager got updated [08:36] and same with wpa_supplicant [08:38] been using open non encrpyted networks in the past with just (iwconfig wlan0 essid "networkname") and (dhclient wlan0) [08:43] asac, looking at the dashboard it smells like another week without release ... [08:43] all new tests seem to fail === dholbach_ is now known as dholbach [08:44] (we should probably rename the "dailys" to "weeklys" :P ) [08:44] gema, ^^^ [08:48] ogra_: see, and you want minutely! :p [08:48] ogra_: daily releases are being to move on a 3 hours schedule probably FYI [08:49] didrocks, well, minutely would indeed assume thests are working :P [08:49] ogra_: right, welcome to my world of pain ;) [08:49] (and that we invent some kind fo time warp mechanism to make them finish in under a minute) [08:49] ogra_: let me make some "research" playing chrono cross :) [08:49] hehe [08:50] welll, i think multiple dailies would be a good start already [08:50] won't fix the "flacky/failing tests" though [08:50] but yeah, you will get quicker to the image [08:50] no, indeed [08:50] thats up to the devs of the apps [08:51] (or devs of the tests ... depending where what fails) [08:51] I need to get some locking system in place though to prevent if we can't make the 3 hours target because of timing out tests [08:51] that should be easy enough, but need some thoughts to be scalable [08:52] yeah, could get hairy if they dleay your stuff more than once ... like ten times in a row .... and suddenly you test two days old stuff from a queue [08:52] *delay [08:52] ogra_: right, hence the easiest model is "if you miss the tick, you'll be picked in the next one" [08:53] yup [08:53] (but also, it expects pushing stuff in coherence every 3 hours) [08:53] let's see how it goes, pushing stuff in coherence was already hard to get ;) [09:10] Good morning all, happy Traffic Light Day! :-D === psivaa_ is now known as psivaa [10:04] tsdgeos, Saviq, mzanetti: that's a plain Jenkins-failure, right -> https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-saucy/1596/console ? [10:04] MacSlow, yes [10:04] yep [10:04] just retrigger it [10:04] ok [10:04] hmpf, unity8 eats constantly ~40% CPU here [10:04] my load is around 3.6 [10:05] ogra_, it's a feature, it's to keep your pocket warm in winter [10:05] heh, yeah [10:05] ogra_: dednick is working on trying to find out what that happes [10:05] surfaceflinger seems busy as well ... not as much as unity8 though [10:06] tsdgeos, ah, awesome [10:06] it sucks that is not easily/trivially reproducible [10:07] i see it on a parted device more often than on a nexus [10:07] *ported [10:09] my wpa_supplicant process quite often goes to +95% usage and stays there... [10:11] making reproducing a pain. [10:11] dednick, yeah, we have rasec all over the place [10:11] *races [10:12] jodh's upstart bridge into android will help a lot with that (hopefully) [10:12] ogra_: yeah - death to sleeps! :) [10:13] !! === zsombi1 is now known as zsombi [10:24] gusch, i think ricmm might be able to help you, he reworked the PID handling when we flipped ... [10:24] ogra_: thx [10:29] Morning all [10:29] ogra_: how are the images coming along? [10:29] 0804 looks fine [10:30] waiting for a burp from cdimage for 05 [10:30] the new tests mostly fail though [10:30] ogra_: yeah I have 04 [10:30] so i wouldnt expect a release soon [10:30] ogra_: fair enough [10:31] ogra_: I noticed that the images test rates were all over the place [10:31] all over the place ? [10:33] ogra_: 30-odd % saturday 70-odd% sunday [10:33] ah, yeah, that perscentage computing is totally weird [10:34] but 100% are 100% at least :) [10:35] ogra_: 100% of the test are definitely tests [10:35] well, i meant success rate indeed :) [10:38] im happy with my load / cpu usage hehehe http://puu.sh/3UekO.png [10:38] 03:36:58 up 4 min, 1 user, load average: 1.16, 0.84, 0.38 [10:39] above 1 doesnt look so good for an idling device though [10:39] here is a screenshot of my HTOP http://puu.sh/3Ueoh.png [10:42] when I first flashed ubuntu on my tablet network manager was constantly @ 100% cpu till I stopped the ofono service === MacSlow is now known as MacSlow|lunch [11:08] hi [11:13] root@ubuntu-phablet:/tmp# nmcli dev list [11:13] GENERAL.NM-MANAGED: no [11:14] if you had re-flashed when i told you, you might have a working system since 1h :) [11:20] oh I think I solved it.... :() had to edit /etc/NetworkManager/NetworkManager.conf [11:20] took out ofono and put managed=true [11:21] nmcli con now shows [11:21] Ifupdown (wlan0) 5391eba4-6426-faca-338e-5828034ff9d1 802-3-ethernet never [11:22] I choose reflash/reinstall as a last resort as you loose your work and dont learn anything [11:24] yup problem solved http://puu.sh/3Ufye.png [11:24] congrats [11:26] oh and now wpa_cli decided to start working [11:26] so I guess doing apt-get dist-upgrade wiped out my .conf files and that was the issue [11:26] right, and it would have done the same on next upgrade === alan_g is now known as alan_g|lunch [11:34] ogra_: moin moin [11:34] moin [11:35] dashboard doesnt look so good with the new tests === chihchun is now known as chihchun_afk [11:35] doanac: gema: plars: dashboard needs help i guess [11:35] maguro seems to hang [11:35] ogra_: yesterday or so it was still ok (no big regressions on the existing ones) [11:35] yeah, i was referring to the new ones [11:35] they have a lot failures [11:36] ogra_: balloons is the guy [11:36] (not unexpected) [11:36] he promissed quick fix [11:36] balloons, ^^ [11:36] :) [11:37] balloons: how is the test cleanup going? [11:37] balloons: ogra has high hopes in you [11:37] me ?!? [11:37] you said he's the man :) [11:37] ogra_: i guess today will be calm... lots of folks shaking out their pain [11:37] yeah ... i'm melting anyway [11:37] :) [11:38] ogra_: move in a modern flat ... here in the north room i dont really get much heat [11:38] though i at least want to have the livecd-rootfs android implementation done today [11:38] ogra_: is everything in archive now? [11:38] my basement is cold too [11:38] asac, yes, but we still need a plan for doign automated daily builds for the package [11:39] currently xnox does them manually on request [11:39] ogra_: what packages? [11:39] android [11:39] ogra_: so phablet ppa is dead? [11:39] can we kill it"? [11:39] ah [11:39] phablet PPA waits for ofono to be done afaik [11:39] ogra_: just stuff it in [11:39] instead of the other stuff [11:39] awe and rsalveti are on it [11:39] :P [11:39] asac, thats my plan, but we still need an automated rebuild strategy [11:40] ogra_: so what i want is that all that OEM has to touch is in container [11:40] else hybris can go out of sync and you stay unbootable [11:40] ogra_: so we might want to move more in there [11:40] and maybe the android container should be a separate image like thing [11:40] thing [11:40] it is [11:40] that we distribute as a separate baseline [11:40] ogra_: good. then just live-build it [11:41] done [11:41] we cant distribute it separately [11:41] ogra_: we already do, no? [11:41] it always needs the matching hybris ABI [11:41] e.g. +maguro [11:41] why do we change hybris api? [11:41] yes, but the build is 100% bound to the content of the armhf zip [11:41] what changes hybris api? i would think if we dont move to a new android [11:41] ABI :) [11:41] we dont change that api [11:41] not api [11:42] abi changes, but api not? [11:42] you need binary compatibility on both sides [11:42] that feels odd [11:42] i know [11:42] i dotn care about api at all :P [11:42] but... the hybris api/abi doesnt change much, i would think [11:42] i do care about failing boots :) [11:43] if hybris on the android side changed you need to rebuild android and need to re-flash along with updating the hybris deb on the ubuntu side [11:45] and since hybris and platform-api can hook into HW specific bits you cant just have a generic cross built hybris package that you could dump into the container [11:46] ogra_: when does hybris change? isnt that just when new android comes out? [11:58] asac, it changes all the time [12:00] ogra_: i can build 4.3 android emulator from aosp and it does run (that's target arch arm, running under qemu). Do you think it's sensible to try, ubuntu touch port against aosp builds? for some reason the cyanogenmod emulator builds are borked, and i don't feel like fixing that. [12:02] xnox, why not, if you can get it to work === MacSlow|lunch is now known as MacSlow === _xtr3m3 is now known as xtr3m3 === LarrySteeze|Away is now known as LarrySteeze === zsombi is now known as zsombi|afk === _salem is now known as salem_ [12:53] renato_: hi, can you look at https://bugs.launchpad.net/address-book-app/+bug/1208343 which blocks phone stack publishing [12:54] Ubuntu bug 1208343 in address-book-app "AP test address_book_app.tests.test_contactlist.TestContactList.test_contact_list failing" [Undecided,New] [12:54] rsalveti, btw, you said sensorservice was moved to late_start to fix a race ... was that before or after the flip ? :) [12:54] ogra_: after flip [12:55] ah, k [12:55] sad, i had hopes === alan_g|lunch is now known as alan_g === bregma_ is now known as bregma [13:25] gusch, hey, are you working on the gallery-app (seems so from the commit history)? [13:26] ogra_: hmmm that will be an interesting bug then, on 05 disable wifi to allow 3g to be tested enable wifi refuses to connect correctly 3 then crashes the entire device instant reboot nice :) [13:26] seb128: yep - although currently busy with other stuff ;) [13:27] gusch, I've some questions for you if you have a bit of spare time :-) [13:27] gusch, 1- can I --pick-photo on a specific album (if not, is that planned)? [13:28] ogra_: do we at all have: unflipped userdata as .img ? [13:28] seb128: no, not planned at the moment [13:29] hum, ok [13:29] gusch, next question "is there a way to call that picker from qml"? [13:30] seb128: Kaleo is working on that picker API for QML, but the picker is not yet functional :( [13:31] gusch, ok, I'm asking for https://wiki.ubuntu.com/Appearance#Phone (to give some context) [13:32] gusch, I think I'm just going to make one entry calling gallery-app --pick-photo meanwhile [13:33] seb128: I'd say that's a good idea [13:34] Kaleo, hey, where can I get details on the content picker that you are working on and what it's going to do/support? [13:34] gusch, thanks for the replies ;-) [13:35] seb128: https://docs.google.com/a/canonical.com/document/d/1j7LsTDHLlYODMxXkz6PI-wqgDMvDuwlz8NWzSyYF9Zg/edit#heading=h.nm4hvq9ryltd [13:35] Kaleo, thanks [13:36] seb128: not working on it right now [13:36] Kaleo, do you have any idea when that's going to be available/usable? [13:37] seb128: not before I'v done the backend for it - don't hold your breath :( [13:38] hum [13:38] are you guys saying we should better build a custom image grid/picker for the background selection? [13:38] that google doc has 13.08 as target for a qml api in the sdk [13:38] I guess that needs to be revisited? [13:41] seb128: probably it has; maybe; maybe we manage to get something for end of month [13:41] seb128: can you work on something else in the meantime, [13:41] ? [13:41] Kaleo, yes [13:41] I'm just going to put that on hold with a note to revisit in septembre [13:42] if by then the content picker is still not there we might need a plan B [13:42] ok [13:42] gusch, Kaleo: thanks [13:43] gush: if I read that google doc correctly the picker is going to allow important photos from apps (gallery and camera are the ones we need) and from specific folders? === dednick is now known as dednick|lunch [13:45] seb128: not sure what you mean, you'll be able to import the photos, that are available in gallery-app [13:46] gusch, what about pointing to a directory? like /usr/share/backgrounds [13:46] gusch, those are not user photos/in the gallery, but we are going to need to pick files from there for the background [13:46] gusch, e.g is the picker going to support an "import from " ? [13:46] xnox, no, we dont have userdata as img at all [13:46] (i mean we dont use it) [13:47] davmor2, wait for awe ... he was working on a fix for all this (or at least coordinating the work) [13:47] ogra_: i'll see if i can fake one, by repackaging unflipped .zip roughly following what our deploy script does. Emulator runs off .imges only. [13:47] seb128: no - no plan for that [13:48] balloons: good morning [13:48] xnox, well, the build surely produces an img ... you can use sim2img to make it mountable, install the tarball in it and re pack it with img2simg [13:48] seb128: interessting requirement ... [13:48] balloons: ubuntu-clock-autopilot and ubuntu-clock-app-autopilot were both on our list of requested app tests to add. Are they really different from one another though? [13:48] xnox, what android produces is otherwise just the directory structure inside the img [13:49] gusch, where would be the right place to open a wishlist? [13:50] seb128: gusch: add /usr/share/backgrounds to a default album in the gallery app? [13:50] and done? [13:50] seb128, we ship ubuntu-bug now :) [13:51] Kaleo, that would work for me, we might not want to list backgrounds there though [13:51] mmmm bugs [13:51] seb128: sounds like a design question [13:52] ogra_: will do ta [13:52] mpt, ^ opinion/can you check with other designers? [13:52] seb128: mpt: especially Rachel and Oren (who take care of the picker) === jhodapp is now known as jhodapp|brb [13:59] hello [14:01] rsalveti: sergiusens: hey, how's it going? [14:01] btw I still have https://code.launchpad.net/~mathieu-tl/phablet-extras/brcm-rename/+merge/176270 waiting for review ;) [14:02] cyphermox: hey, welcome back [14:03] yeah, had some comments but forgot to update that, just a sec [14:07] cyphermox: just updated it [14:07] thanks! [14:08] cyphermox: just a better check for getprop, as this will be installed by default [14:08] aye [14:09] cyphermox: the mako one is missing 'stop on' as well [14:09] yeah, that's on purpose [14:09] great [14:09] there is not really a stop case for mako [14:09] yeah, just thought about that [14:10] cyphermox: why do we need to export a PATH in there? [14:10] probably not, I'll re-check [14:10] cyphermox: cool, and sleep 15 is really a long sleep, do we need a high value like that? [14:10] sleep: yeah, seems like it does, but if you give me a few minutes I'll triple-check [14:11] but brcm-patchram takes a little bit to get done [14:11] cyphermox: perfect [14:13] I'm reflashing my nexus 4 right now, I'll verify all of this [14:14] cyphermox, davmor2 also had some weird behavior with wlan interacting with 3G (might be UI related though, he can surely explain better) [14:14] ok [14:14] let's look into it [14:15] davmor2: ? [14:16] ogra_: hmmm that will be an interesting bug then, on 05 disable wifi to allow 3g to be tested enable wifi refuses to connect correctly 3 then crashes the entire device instant reboot nice :) [14:16] from the backlog [14:17] cyphermox: so for me I wanted to test the 3g connection, so I opened the NM-indicator deselected wifi. Tested the 3g connection worked and then went to re-enable the wifi. The wifi instantly flicked back to X in nm-indicator so I tried it 2 more time then the device crashed and rebooted [14:17] ok [14:17] what device was that? [14:17] cyphermox: maguro [14:18] cyphermox: that is on 05 [14:19] ok [14:21] heh, the indicator is as broken as ever [14:21] hello, is there a blog or any site I can go to see the progress on ubuntu touch? [14:21] it restarts so often it's not impossible that NM got confused by the state changes if it changed states a lot [14:22] davmor2: ^ [14:23] cyphermox: this was disable, open the webbrowser, wait for the first page to load click on the indigogo link wait for that, close browser enable wifi, so a minute or more maybe? === jhodapp|brb is now known as jhodapp [14:24] davmor2: yeah, but I meant something else actually [14:24] when you enable or disable, the indicator itself crashes and restarts [14:24] that's why you see the app flashing and stuff [14:24] cyphermox: ah right [14:24] cyphermox: yes that I see [14:25] I started looking into that, with help from desrt, but I didn't manage to fix that one issue yet [14:25] not sure whether it's worth spending too much time on it though since tedg is working on the indicator... [14:25] Anyone know where I can find a log for the Porting Clinic discussion? Been on vacation and missed it. [14:26] dejello, irclogs.ubuntu.com has all ubuntu related irc logs [14:26] (including this channel) [14:27] Thank you [14:28] cyphermox: the other thing as well is I believe there is a new indicator hopefully landing soon so I don't know how much effort you will want to put into it, ogra_ that was the outcome from tedg right? [14:28] yeah [14:29] davmor2: yeah that's what I was just saying :) [14:29] not sure of that also is true for NM though [14:29] cyphermox: ah sorry :) [14:29] yes [14:29] ah, good === alan_g is now known as alan_g|tea [14:29] it's the network indicator, for driving wifi and 3G [14:30] cyphermox: I thought it also included signal strength for telecoms too [14:30] yes [14:30] or you know well, [14:30] possibly, I don't know ;) [14:30] haha [14:31] cyphermox: I think you missed a maybe and a perhaps out of that sentence :) === dednick|lunch is now known as dednick [14:31] indeed [14:44] plars, I'm confused by clock-app and ubuntu-clock.. probably a misprint. There's only one clock app [14:45] balloons: but there are two test packages [14:45] balloons: ubuntu-clock-autopilot and ubuntu-clock-app-autopilot [14:49] balloons: there is the clock indicator too or is that in an other part? [14:51] hi everyone [14:51] balloons: The libpoppler-qt5 as just been released in saucy, I think we will be able to work on pdf tests soon :) [14:51] Chocanto, :-) [14:51] wohoo [14:52] * ogra_ finally wants to be able to read his ebooks on ubuntu touch [14:52] plars, so the ubuntu-clock-app is using ubuntu-clock-app-autopilot [14:52] balloons: but are the tests doing something different? [14:53] ogra_: It would be awesome ! :) [14:54] ++ === alan_g|tea is now known as alan_g [15:00] plars, as far as I can tell ubuntu-clock-autopilot isn't doing anything and isn't needed [15:01] seb128, Kaleo: That's why I wrote "items for “Ubuntu Art” ... *and* an item for each album in the gallery" [15:01] seb128, Kaleo: So it's a variation of the image content picker specifically in that here it shows the Ubuntu Art collection whereas everywhere else it does not. === Namidairo is now known as Namidairo`bnc [15:03] mpt, would "Ubuntu Art" be part of the gallery albums? [15:03] mpt, seems like the picker as designed atm doesn't support what you want, at least it doesn't have a "pick from album"... [15:03] seb128, no, that's why I wrote "*and*" there [15:03] hum [15:04] Should I change it to "plus"? :-) [15:04] mpt, I guess I should open a bug because there might be a disconnect between the design and what the picker is going to allow us to do [15:04] mpt, no, the design is fine, but it seems like the picker is not going to let pick from something which is not a gallery album [15:05] seb128, probably. I did raise this with Rachel earlier. I haven't seen the design for the content picker though. [15:06] mpt: https://docs.google.com/a/canonical.com/document/d/1j7LsTDHLlYODMxXkz6PI-wqgDMvDuwlz8NWzSyYF9Zg/edit#heading=h.nm4hvq9ryltd [15:06] kenvandine: hi! Please don't forget about https://code.launchpad.net/~mardy/ubuntu-system-settings-online-accounts/client/+merge/178038 (the diff is huge, but mostly because of the docs) [15:08] balloons: ok, it will probably go away when we move to the click packages then [15:09] mardy, i won't :) [15:09] plars, lol.. well that's one way of looking at :-) I was trying to track down where it was coming from [15:09] plars, you mean when we have to rewrite all tests from scratch anyway ? :) [15:09] seb128, I just checked with rachelliu. The design of the content picker is being changed at the moment, but she will raise this issue of extensibility with tvoss_ on Wednesday. [15:10] mpt, ok, thanks [15:10] gusch, Kaleo: ^ jfyi [15:11]