UbuntuIRC / 2014 /02 /23 /#ubuntu-ci-eng.txt
niansa
Initial commit
4aa5fce
[00:06] <rsalveti> ogra: hey
[00:06] <ogra> still building
[00:06] <ogra> :)
[00:06] <rsalveti> ogra: :-)
[00:06] <rsalveti> good
[00:06] <ogra> rsalveti, see -touch
[00:07] <ogra> xnox would like to rip ofono-scripts out
[00:07] <rsalveti> hm, why?
[00:07] <rsalveti> it's still used by people
[00:07] <rsalveti> let me look
[00:07] <ogra> because he is on a quest :)
[00:07] <ogra> to kill python2
[00:08] <rsalveti> right
=== rsalveti_ is now known as rsalveti
[00:13] <ogra> rsalveti, http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/20140222.4/
[00:14] <xnox_> ogra: i like that definition =)
[00:14] <ogra> :)
=== xnox_ is now known as xnox
[00:14] <xnox> and i got my name back.
[00:14] <xnox> apperantly i keep getting banned on channels. I'd like znc to autopart them...
[00:17] <rsalveti> ogra: lovely
[00:17] <ogra> system-image is still running
[00:18] <ogra> and cdimage will need some cleanup
[00:18] <ogra> the old goldfish stuff, the zips
[00:18] <ogra> (but nothing i will do tonight)
[00:19] <rsalveti> ogra: so do you want to try to add it to system-image or should we just ping stgraber?
[00:19] <rsalveti> ogra: yeah, that's fine
[00:19] <rsalveti> we can enable the system-image tomorrow as well
[00:19] <ogra> well, the known arches will just be imported
[00:20] <rsalveti> right
[00:20] <ogra> and the other stuff needs a certain sequence of commands
[00:20] <ogra> which i dont know
[00:20] <rsalveti> ogra: right, let me ping stgraber
[00:20] <ogra> he has some import toools
[00:21] <ogra> mako, manta and grouper should just work
[00:21] <rsalveti> yeah
[00:36] <rsalveti> ogra: how much time it usually takes to get the image published at system-image?
[00:36] <ogra> 20 min or so
[00:37] <rsalveti> alright, should be done any time I guess
[00:37] <ogra> 12316 ? R 15:19 /usr/bin/python /srv/system-image.ubuntu.com/bin/import-images
[00:37] <ogra> 15 done ...
[00:37] <ogra> shouldnt be long anymore
[00:43] <rsalveti> alright, cdimage looks good at least
[00:43] <ogra> hmm, did paul switch off testing on maguro
[00:56] <ogra> phew, takes pretty long today
[01:03] <ogra> === Image #203 DONE ===
[01:04] <ogra> ok, 2am ... bed time :)
[02:22] <rsalveti> ogra: thanks!!!
[10:36] <rsalveti> === Image 204 building ===
[10:38] <rsalveti> it's actually 205
[10:38] <rsalveti> === Image 205 building === :-)
[10:49] <popey> :D
[10:49] <popey> whats new?
[11:09] <rsalveti> just enabling mir for flo
=== popey_ is now known as popey
=== john-mca` is now known as john-mcaleely
=== lan3y is now known as Laneyu
=== Laneyu is now known as Laney
[18:18] <cjohnston> ogra: you around?
[18:20] <cjohnston> rsalveti: how about you?
[18:32] <cjohnston> I'm unable to boot using trusty-proposed on mako.. and also got the same network config error as in jenkins when trying to run the tests that way
[20:06] <ralsina_> So, I am guessing if I say cihelp that will alert anyone from CI that happens to be around on a sunday, right? ;-)
[20:11] <rsalveti> cjohnston: hey, what is the issue?
[20:11] <rsalveti> cjohnston: the network config issue was because I had to rebuild my silo
[20:12] <cjohnston> rsalveti: it gets stuck on the google logo
[20:13] <rsalveti> cjohnston: how did you update it?
[20:13] <rsalveti> cjohnston: can you reach it with adb?
[20:13] <cjohnston> phatlet-flash, yes.. im Not home now thi
[20:13] <cjohnston> tho
[20:13] <cjohnston> ralsina_: sup?
[20:14] <ralsina_> cjohnston: hi!
[20:14] <ralsina_> I need a silo reconfigured
[20:14] <cjohnston> ralsina_: ci team doesnt do that
[20:15] <ralsina_> cjohnston: hmmm ok, I may be confused about what team does what, then
[20:15] <ralsina_> rsalveti: you can give me a hand with reconfiguring silo 14, right?
[20:16] <ralsina_> cjohnston: sorry about the bad ping then!
[20:16] <cjohnston> ralsina_: the info is in the topic
[20:17] <rsalveti> cjohnston: hm, if you can try to reflash with 'ubuntu-device-flash -channel=trusty-proposed -bootstrap=true' later
[20:17] <rsalveti> or get adb to check what is going on
[20:17] <ralsina_> cjohnston: I just assumed CI Engineering team also handled CI Train. You know, because of all the CI in the names. So, sorry about that.
[20:17] <rsalveti> ralsina_: sure, what is the issue?
[20:18] <ralsina_> rsalveti: I changed the MP for silo 14
[20:18] <cjohnston> that'swhyits specified in the topic :)
[20:18] <rsalveti> ralsina_: let me check
[20:21] <rsalveti> ralsina_: https://code.launchpad.net/~diegosarmentero/unity-scope-click/open-after-install/+merge/207836, right?
[20:22] <ralsina_> rsalveti: exactly
[20:23] <rsalveti> hmmmm rsalveti is missing the Job/Build permission
[20:23] <rsalveti> would need to wait for robru or cyphermox then
[20:23] <rsalveti> let ask asac
[20:23] <rsalveti> asac`: http://162.213.34.102/job/prepare-silo/build
[20:24] <rsalveti> asac`: series: trusty
[20:24] <rsalveti> asac`: reconfigure_silo: landing-014
[20:24] <rsalveti> asac`: merge_proposals: https://code.launchpad.net/~diegosarmentero/unity-scope-click/open-after-install/+merge/207836
[20:25] <rsalveti> also no power
[20:25] <asac`> robru: there?
[20:25] <asac`> cyphermox: ^^
[20:26] <asac`> :)
[20:26] <asac`> wake up
[20:27] <rsalveti> ogra: plars: what is wrong with the dashboard? still not running tests for the newer images
[20:27] <rsalveti> mako specifically
[20:28] <asac`> cihelp ^^
[20:28] <cyphermox> what's going on?
[20:28] <asac`> cyphermox: silo reconfigure
[20:29] <asac`> 21:23 < rsalveti> asac`: http://162.213.34.102/job/prepare-silo/build
[20:29] <asac`> 21:24 < rsalveti> asac`: series: trusty
[20:29] <asac`> 21:24 < rsalveti> asac`: reconfigure_silo: landing-014
[20:29] <asac`> 21:24 < rsalveti> asac`: merge_proposals: https://code.launchpad.net/~diegosarmentero/unity-scope-click/open-after-install/+merge/207836
[20:29] <asac`> 21:23 < rsalveti> asac`: http://162.213.34.102/job/prepare-silo/build
[20:29] <asac`> 21:24 < rsalveti> asac`: series: trusty
[20:29] <asac`> 21:24 < rsalveti> asac`: reconfigure_silo: landing-014
[20:29] <asac`> 21:24 < rsalveti> asac`: merge_proposals: https://code.launchpad.net/~diegosarmentero/unity-scope-click/open-after-install/+merge/207836
[20:29] <asac`> ralsina_: whats this MP about?
[20:30] <asac`> cjohnston: maybe here rather
[20:31] <cyphermox> unless the MP is a different one altogether now, you shouldn't need to reconfigure for this
[20:31] <asac`> ralsina_: you surely should set a meaningful commit message for the MP
[20:31] <asac`> cyphermox: someone says its a new one
[20:32] <asac`> can you confirm?
[20:32] <ralsina_> asac`: sure, I'll add one
[20:33] <asac`> ralsina_: cool.always helpful to convey the purpose and context of what you are doing :)
[20:33] <cyphermox> indeed it's different
[20:33] <cyphermox> https://api.launchpad.net/devel/~dobey/unity-scope-click/uninstall/+merge/207557
[20:33] <asac`> right
[20:33] <asac`> so yeah we want a reconfigure
[20:33] <cyphermox> is that on purpose? should it be both or just hte new one?
[20:34] <asac`> cyphermox: just the new one... dobey wasnt around i figure
[20:34] <asac`> ralsina_: confirm that you just need the new one
[20:34] <asac`> and not the old one
[20:34] <cyphermox> this is messy
[20:35] <asac`> cyphermox: why?
[20:36] <ralsina_> asac`: yes, the new one includes the old one
[20:36] <asac`> k
[20:36] <asac`> cyphermox: so yeah. the new MP is marked as superseding the old one
[20:37] <cyphermox> should be able to build now
[20:38] <asac`> thx!
[20:38] <asac`> rsalveti: ralsina_: ^^
[20:38] <ralsina_> awesome, thanks asac`, cyphermox
[20:39] <rsalveti> great
[20:42] <rsalveti> plars: it seems we might have an issue with the wifi driver with the mako device we're using in CI, is there anyway to retrieve logs from it? I couldn't reproduce the issue with mine, so it'd be really nice to know what might be happening in there (syslog and logcat should be enough)
[20:44] <asac`> cihelp: ^^
[20:46] <cjohnston> rsalveti: that's What's happening with mine i assume. When i get home i Can be Avail to do What you need
[20:46] <asac`> plars: do you know why we fail hard if install-and-boot cannot setup wifi? thought we mangaed to get away from wifi being a critical resource for further testing
[20:46] <asac`> cjohnston: cool
[20:47] <rsalveti> cjohnston: alright, great
[21:15] <ev_> rsalveti: which mako device?
=== ev_ is now known as ev
[21:16] <rsalveti> let me get the link
[21:16] <ev> thanks
[21:16] <rsalveti> dashboard is super slow =\
[21:17] <rsalveti> ev: https://jenkins.qa.ubuntu.com/job/trusty-touch-mako-smoke-daily/77/console, mako-07
[21:17] <ev> really? It's pretty quick here.
[21:17] <ev> grabbing
[21:17] <rsalveti> ev: can you access it via adb?
[21:18] <ev> I can
[21:18] <ev> getting you syslog now
[21:19] <rsalveti> ev: thanks, syslog and logcat, please
[21:22] <ev> rsalveti: syslog: http://paste.ubuntu.com/6983989/ , logcat: http://paste.ubuntu.com/6983994/
[21:24] <rsalveti> ev: thanks, any way to have remote access to this device? just in case I get a kernel to test
[21:24] <ev> yes
[21:26] <ev> rsalveti: by the way, is this a critical issue. Should I be grabbing people to step in once I need to get to bed?
[21:27] <rsalveti> ev: well, the issue is critical, but it's more on my side at this point
[21:27] <rsalveti> there's an easy way to get this fixed, but then we might not necessarily be able to reproduce this anymroe
[21:27] <rsalveti> I'm unable to reproduce this issue with my device
[21:27] <rsalveti> that's why I'd only need access to see if I'm to get this fixed
[21:31] <rsalveti> brb, ~10min
[21:50] * rsalveti back
[21:54] <ev> cjohnston: if you're around can you give me admin access on q-jenkins' jenkins?
[21:54] <ev> I'd like to mark mako-07 offline to I can give rsalveti a shell without worrying about tests bursting into flames
[21:56] <cjohnston> ev: did you take it offline in nagios so that larry doesn't start getting paged?
[21:56] <cjohnston> rsalveti: I'm back if you want me to do stuff with mine
[21:56] <rsalveti> cjohnston: please also give your syslog and logcat (as root)
[21:56] <cjohnston> rsalveti: ack..
[21:57] <cjohnston> ev: which one are you
[21:57] <ev> cjohnston: I haven't yet, but I will do
[21:57] <ev> damned if I know. ev at the moment.
[21:57] <ev> but I appear to have evan or evand as well
[21:58] <ev> username squatting
[21:58] <cjohnston> ev: you should be good
[21:59] <ev> whoop
[22:01] <ev> cjohnston: is it your understanding as well that I can just mark mako-07 as offline - there's nothing more to it? The trusty-touch-mako-smoke-daily uses the daily-mako label, so it should grab one of the others.
[22:01] <cjohnston> ev: I'm not totally sure how it works anymore
[22:02] <rsalveti> cjohnston: how are the images actually upgraded when a new image is flashed?
[22:02] <rsalveti> just noticed they are still using the older kernel
[22:04] <rsalveti> using phablet-flash, alright
[22:04] <cjohnston> rsalveti: phablet-flash ubuntu-system -b --channel trusty-proposed
[22:04] <rsalveti> phablet-flash ubuntu-system -b --channel trusty-proposed
[22:04] <rsalveti> yeah
[22:04] <rsalveti> let me just try to update manually from system-settings and will give that a try
[22:05] <rsalveti> just flashed 202
[22:06] <rsalveti> I wonder if this command is actually flashing the boot partition
[22:06] <rsalveti> in theory we should all be using 'ubuntu-device-flash -channel=trusty-proposed -bootstrap=true' now
[22:07] <rsalveti> cjohnston: ev: can both of you get me the file /cache/recovery/last_log?
[22:07] <cjohnston> rsalveti: I don't believe that the install instructions are updated completely in the wiki, and I haven't seen any bugs or anything filed against the tests do have the switch made
[22:07] <cjohnston> rsalveti: I'm flashing now.. once I get done I will
[22:07] <rsalveti> yeah
[22:08] <rsalveti> ev: can you get this file from mako-07?
[22:09] <ev> rsalveti: http://paste.ubuntu.com/6984209/
[22:09] <rsalveti> awesome, thanks
[22:10] <rsalveti> yeah, for some reason it didn't flash either recovery nor boot
[22:10] <rsalveti> Flashing boot at /dev/block/platform/msm_sdcc.1/by-name/boot
[22:10] <rsalveti> Flashing recovery at /dev/block/platform/msm_sdcc.1/by-name/recovery
[22:10] <rsalveti> missing that
[22:15] <ev> rsalveti: if you need that shell it's rsalveti-tmp@ashes.ubuntu-ci with your public key
[22:15] <rsalveti> rvgreat, thanks
[22:15] <rsalveti> ev: ^
[22:15] <ev> adb -s 01e2f64788556934 shell
[22:15] <ev> is mako-07
[22:16] <rsalveti> ev: hm, asking passwd
[22:16] <rsalveti> did you use https://launchpad.net/~rsalveti/+sshkeys
[22:16] <rsalveti> ?
[22:16] <ev> I did
[22:16] * ev checks
[22:16] <rsalveti> hm
[22:16] <rsalveti> maybe permission?
[22:17] <rsalveti> .ssh/authorized_keys needs to be 700
[22:18] <cjohnston> http://paste.ubuntu.com/6984239/
[22:19] <ev> rsalveti: .ssh is 0700, authorized_keys is 0600
[22:19] <rsalveti> Jan 1 00:10:10 ubuntu-phablet kernel: [ 0.000000] Linux version 3.4.0-3-mako (buildd@kishi04) (gcc version 4.7.3 (Ubuntu/Linaro 4.7.3-7ubuntu3) ) #21-Ubuntu SMP PREEMPT Tue Oct 8 13:11:48 UTC 2013
[22:19] <rsalveti> cjohnston: same issue, kernel wasn't flashed
[22:19] <cjohnston> rsalveti: do you want me to try it with the new command?
[22:19] <rsalveti> cjohnston: yes, please
[22:20] <rsalveti> ev: still nothing :-(
[22:21] <cjohnston> rsalveti: remind me please what package the command comes form
[22:21] <cjohnston> from
[22:21] <cjohnston> ubuntu-devices-flash?
[22:21] <rsalveti> cjohnston: ubuntu-device-flash
[22:22] <cjohnston> ack
[22:23] <cjohnston> hrm
[22:25] <ev> rsalveti: try now
[22:25] <rsalveti> ev: yup, fine now, thanks!
[22:26] <cjohnston> rsalveti: when running the command, should I get any sort of visual feedback?
[22:27] <cjohnston> its been probably 3 or 4 minutes and I've seen nothing
[22:27] <rsalveti> cjohnston: oh, sorry, I hate that
[22:27] <rsalveti> cjohnston: please do adb reboot bootloader before
[22:27] <rsalveti> it expects the device to be at the bootloader already, but yeah, not a single message
[22:28] <cjohnston> that looks mo bettah
[22:29] <ev> for those of you following along at home, I've added this to the playbook: https://wiki.canonical.com/UbuntuEngineering/CI/Playbook#Providing_a_phone_for_remote_debugging
[22:30] <rsalveti> great
[22:30] <asac`> nice
[22:30] <cjohnston> rsalveti: I'll have to drop off for dinner in about 5 minutes but I'll be back after that
[22:30] <rsalveti> cjohnston: no worries, trying to reproduce the issue here
[22:31] <cjohnston> asac`: I'll look into the dashboard either later tonight or tomorrow..
[22:31] <asac`> sure
=== asac` is now known as asac
[22:34] <cjohnston> bbiab
[22:35] <ev> rsalveti: likewise I'm heading to bed soon. Do send me a PM when you're done with that shell.
[22:36] <rsalveti> ev: no worries, thanks
[22:47] <sergiusens> rsalveti, if I get a slot we can release more verbosity now
[22:47] <rsalveti> sergiusens: no worries, we can do this tomorrow
[22:47] <rsalveti> sergiusens: seems there's still something broken with 4.4.2's recovery
[22:48] <rsalveti> unable to flash boot and recovery somehow
[22:48] <rsalveti> checking now
[22:48] <rsalveti> wonder if we need to use unbuffered
[22:48] <sergiusens> rsalveti, worth trying
[22:48] <rsalveti> sergiusens: that might explain your issue with flo as well
[22:49] <sergiusens> rsalveti, did mako update fine?
[22:49] <rsalveti> sergiusens: update went ok, problem is flashing bootstrap
[22:49] <sergiusens> rsalveti, I guess this means that everyone in devel-proposed won't be able to update :-/
[22:49] <sergiusens> rsalveti, since the update uses recovery=current, next update will use recovery=4.4
[22:50] <rsalveti> right, but if you're coming from 4.2.2 and update using system-settings, you're fine
[22:50] <rsalveti> because the flash will use the older recovery image
[22:50] <sergiusens> rsalveti, yeah, but the next update will be broken
[22:50] <rsalveti> no, it's not broken
[22:50] <sergiusens> oh, really?
[22:50] <rsalveti> the only broken part is the one that looks for boot and recovery
[22:51] <rsalveti> I'm still checking
[22:52] <sergiusens> rsalveti, I didn't get that last statement; but if the previous is true; I should be able to update from 4.4's recovery; going to check
[22:52] <rsalveti> sergiusens: I mean, only the piece that looks for a newer kernel and recovery is broken with 4.4.2's recovery
[22:52] <rsalveti> at least this is what I saw
[22:53] <rsalveti> but it flashes the device and system image just fine
[22:53] <rsalveti> the only problem is that people who flashed devel-proposed might not be able to update to a newer recovery, or kernel
[22:53] <sergiusens> rsalveti, yeah
[22:53] <rsalveti> but we also need to fix that so system-image updates are always booting into the newer recovery instead of using the older one to do updates
[22:54] <rsalveti> anyway, let me check what is going on in here
[22:54] <sergiusens> +1
[23:00] <rsalveti> sergiusens: got it, fstab format is different now
[23:00] <rsalveti> http://paste.ubuntu.com/6984407/
[23:00] <rsalveti> old
[23:00] <rsalveti> http://paste.ubuntu.com/6984408/
[23:00] <rsalveti> new
[23:01] <rsalveti> and that's why we need to first update the recovery image
[23:01] <sergiusens> rsalveti, so it's standard now
[23:01] <sergiusens> yup
[23:02] <rsalveti> let me get this fixed
[23:03] <sergiusens> rsalveti, so this requires the image updater to see if a recovery image is bundled; add that to the ubuntu_command so it's the first thing it flashes and reboot into itself
[23:03] <sergiusens> and continue processig the commands
[23:03] <rsalveti> sergiusens: exactly
[23:11] <rsalveti> cool, new ubuntu logo
[23:12] <rsalveti> spinning
[23:12] <rsalveti> finally
[23:21] <rsalveti> sergiusens: https://code-review.phablet.ubuntu.com/188
[23:21] <rsalveti> flashed mako, worked fine
[23:21] <rsalveti> testing flo now
[23:22] <rsalveti> sergiusens: flo would only fail for you if you were coming from android
[23:40] <sergiusens> rsalveti, hmmm
[23:40] <sergiusens> rsalveti, so bootstrap should work?
[23:41] <rsalveti> sergiusens: with this recovery, yes
[23:41] <sergiusens> rsalveti, sorry, mother calls on sundays to talk :-P
[23:41] <sergiusens> now off the phone
[23:41] <rsalveti> haha, no worries
[23:41] <sergiusens> I can see there's a CR
[23:42] <rsalveti> sergiusens: yup
[23:42] <rsalveti> sergiusens: can you easily build it?
[23:42] <sergiusens> rsalveti, so I went to the doctor on Saturday; my head still hurts :-P
[23:42] <sergiusens> just commenting for the gossip factor :)
[23:42] <rsalveti> sergiusens: oh, still hurting?
[23:43] <sergiusens> yup
[23:43] <sergiusens> on pain killers now
[23:43] <sergiusens> didn't touch the computer at all yesterday
[23:43] <rsalveti> right, you might want to take a vacation
[23:43] <sergiusens> slept the whole day
[23:43] <sergiusens> I am :-)
[23:43] <rsalveti> just fly to floripa :-)
[23:43] <sergiusens> mini vacations for carnaval and then that mini eurotrip I commented on
[23:44] <sergiusens> that too
[23:44] <rsalveti> yeah, nice
[23:44] <sergiusens> thinking of april for a week
[23:44] <rsalveti> will take some days off after mwc as well
[23:44] <rsalveti> sergiusens: cool
[23:44] <cjohnston> sergiusens: did you get a chance to update the install wiki
[23:44] <rsalveti> cjohnston: phablet-flash can't flash flo, so I'd guess we also need to migrate to ubuntu-device-flash
[23:45] <cjohnston> rsalveti: the wiki is half updated.. it says to use u-d-f but doesn't tell you to install the package (or atleast didn't tell you to install the package)
[23:45] <cjohnston> rsalveti: using ubuntu-device-flash -channel=trusty-proposed -bootstrap=true it still didn't work.. same issues
[23:45] <cjohnston> need any logs from this attempt?
[23:45] <rsalveti> cjohnston: yeah, I found the issue
[23:45] <sergiusens> cjohnston, le me update
[23:45] <rsalveti> cjohnston: sergiusens is already reviewing it
[23:46] <rsalveti> reviewing the fix
[23:46] <sergiusens> cjohnston, while I test build the CR
[23:48] <cjohnston> sweet
[23:51] <sergiusens> hmm, my device crashed
[23:52] <sergiusens> darn, hate it when this happens, flashing manta recovery into flo :-P no checks in place
[23:53] <rsalveti> cjohnston: ev: done with mako-07
[23:53] <rsalveti> haha
[23:54] <cjohnston> ev: is all you did offline it and give rsalveti shell?
[23:54] <rsalveti> I believe so
[23:54] <rsalveti> https://wiki.canonical.com/UbuntuEngineering/CI/Playbook#Providing_a_phone_for_remote_debugging
[23:54] <rsalveti> guess basically what is described in there