[00:01] trainguards could someone delete ubuntu-system-settings out of the vivid silo 0 ppa ?? we've reconfig'd that out of there, so no longer needed [00:01] kgunn: sure, one sec [00:02] kgunn: alright, i hit the button, that should disappear shortly. [00:02] ta [00:02] yw === _salem is now known as salem_ [01:17] Grr, what happened to rtm build 241? It still hasn't finished... === salem_ is now known as _salem [01:33] ToyKeeper: I can try to look it up [01:35] started building 4 hours ago or so? [01:36] Yes, about 4.5 hours. [01:36] yeah, looks right [01:36] so image 20150219 [01:37] some part of it got built at least: http://cdimage.ubuntu.com/ubuntu-touch/ubuntu-rtm/14.09/daily-preinstalled/20150219/ [01:44] ToyKeeper: does this look right? seems like the right changes to me at least: http://people.canonical.com/~ogra/touch-image-stats/rtm/20150219.changes [01:44] I think so, but I don't see 241 on the image servers. [01:45] yeah, there's some kind of second step that may be still running I guess? [01:45] rsalveti: still around to look into this? ^^ [01:45] Just confused, since usually it takes about 90 minutes but something seems to have failed this time. [01:48] cyphermox: the import happens in the same machine [01:48] but yeah, 4 hours is too much [01:48] usually if it's showing under ogra's website it means it got published in system-image [01:48] let me check [01:49] right [01:49] the dated image is there but not the numbered one [01:49] I just know that UDF query is showing me 240 still. [01:49] so it goes beyond what I know of that build process [01:50] yeah, no image for today in system-image [01:50] https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu-rtm/14.09/ubuntu-touch/ [01:50] rootfs was built 4 hours ago [01:51] # System image imports [01:51] MAILTO=stgraber@ubuntu.com [01:51] #*/5 * * * * TMPDIR=/srv/system-image.ubuntu.com/tmp /srv/system-image.ubuntu.com/bin/import-images [01:51] cyphermox: from nusakan [01:51] best to ask stgraber [01:52] stgraber might not be very reachable, afaik he's away [01:52] at least not in this channel :) [01:53] right, pinged him in release [01:53] ok [01:53] wonder if that was because of snappy or whatever [01:53] hard to know [01:53] sergiusens: do you know? [01:54] ToyKeeper: it seems this is critical enough to call someone, right? [01:54] as QA would need to validate that image [01:54] Possibly, since this is the promotion candidate we're hoping to get out the door by the end of the week. [01:55] In practical terms, getting it fixed now would only save about ~6 hours though. [01:58] ToyKeeper: cyphermox: running importer by hand now [01:58] cross your fingers [01:59] ok [02:05] === IMAGE 106 building (started: 20150220-02:05) === [02:17] should take ~30min I guess [02:20] === IMAGE RTM 241 DONE (finished: 20150220-02:20) === [02:21] s/30/3/ ... heh, what's an order of magnitude between friends? :) [02:21] === changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/241.changes === [02:21] rsalveti: Thanks! I'm not sure what broke, but I think this unblocks the image promotion process. :) [02:21] yeah, there you go [02:48] trainguards, dumb request, could I get a link to the spreadsheet please? :-) [02:48] veebers: wiki.ubuntu.com/citrain :) [02:49] Ursinha: awesome thanks. Thats the link I was looking for [02:49] you're welcome! [02:49] (and should have probably already known about :-P) [02:50] It's a little scary how much I rely on my browsers history and bookmarks, means when I'm using a backup I've lost a bunch of info [03:08] Ursinha: whoa, you're up late! [03:08] veebers: you got silo 9 [03:08] robru: awesome, thank you [03:08] robru: yeah, am here by accident and there was a fire on [03:08] veebers: you're welcome [03:09] veebers: also, the channel topic links to the dashboard, which has a link to the spreadsheet, just in case you ever lose it again [03:09] robru: ah right, now I feel _really_ blind :-) Cheers [03:09] veebers: hehe [03:10] oh, the jenkins looks quite different too [03:28] veebers: oh yeah, we did a big rollout recently. migrated from precise to trusty, got a new version of jenkins, etc. [03:29] robru: nice. I understand that it's improved performance a bit? [03:29] veebers: oh I hadn't noticed a change in performance... I think that email was talking about some other jenkins. [03:30] ah ok :-p [03:30] === IMAGE 106 DONE (finished: 20150220-03:30) === [03:30] === changelog: http://people.canonical.com/~ogra/touch-image-stats/106.changes === [03:32] veebers: yeah, ted' email is congratulating people for s-jenkins speeding up. That's the one that does CI on MPs among other things. [03:32] ah ok, my bad. I take back the nice things I said ;-) [03:45] veebers: damn right you will! :-P [03:45] veebers: https://ci-train.ubuntu.com/job/ubuntu-landing-009-1-build/lastBuild/ one nice thing is that diffs are now produced at build time, so you can see those if you want [03:45] robru: oh that is cool, I'll check it out [05:30] g morning [08:28] Is anyone around yet who can investigate what happened to image 241? [08:30] ogra_, ^ is it something you can investigate or someone else is needed? [08:30] imgbot, status 241 [08:30] Status: succeeded, Started: 2015-02-19 20:54:30 UTC, Finished: 2015-02-19 21:43:00 UTC [08:30] Build URL: https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu-rtm/14.09/ubuntu-touch/+build/20307 [08:30] Changelog: http://people.canonical.com/~ogra/touch-image-stats/rtm/241.changes [08:31] looks fine to me [08:31] ogra_, apparently the build never completed on its own, and the image doesn't show up for OTA. I don't know much I just woke up [08:33] jibel, the bot reads from the server, so it must be there :) [08:33] i think the image importer on system-image was disabled for a few hours yesterday [08:33] ogra_, I'm on 240, go to update and don't see the image [08:36] the image is properly shown at http://system-image.ubuntu.com/ubuntu-touch/ubuntu-rtm/14.09-proposed/krillin/index.json [08:36] server side everything is fine [08:37] it would be a client side issue? [08:37] i see nothing wrong on the server [08:43] ToyKeeper, did you check on the client if anything was wrong? [08:44] imgbot is confused. [08:44] === IMAGE RTM 241 building (started: 20150219-21:00) === [08:44] === IMAGE RTM 241 DONE (finished: 20150220-02:20) === [08:45] hm? [08:45] What's up? [08:46] sil2100: The build/publish process failed for image 241, and we're not sure why. [08:46] So we don't have image 241? [08:46] ogra_: ^ ? [08:46] We do, sort of, but it's invisible for OTA updates. [08:46] It was invisible to UDF and imgbot too, until rsalveti kicked something manually. [08:47] huh, now that's strange [08:47] "running importer by hand", he said. [08:47] Aah, damn, yeah, maybe the release team switched the importer off for FF or related [08:47] Or the .2 release even [08:48] sil2100, ogra_ system-image crashed on the client [08:48] systemimage.gpg.SignatureError: [08:48] sig path : b23d5b2a14027f6775f4012123f5e8c1 [08:48] But strange that it's not visible as an OTA, shouldn't matter [08:48] /tmp/system-image-bk33lz3z/channels.json.asc [08:49] full error: http://paste.ubuntu.com/10321090/ [08:49] sil2100, slangasek worked on the channel config for snappy afaik ... [08:49] importer was off til ~3aam UTC [08:49] Ok, so maybe there's some issue with channels right now? [08:49] looking on the server now it appears everything is fine [08:50] * sil2100 looks at available channels [08:50] http://system-image.ubuntu.com/ubuntu-touch/ubuntu-rtm/14.09-proposed/krillin/index.json [08:50] Looks ok here [08:50] right [08:50] but jibel's error above is worrying [08:51] jibel: does that happen constantly? Let me fetch my phone, I should be upgradable as well [08:51] sil2100, yes [08:51] ToyKeeper, can you check on your device in /var/log/system-image/client.log [08:52] jibel: I've reflashed too many times to still have that log. [08:52] ToyKeeper, then flash 240 and try to upgrade [08:52] Already started. ;) [08:52] thanks [08:53] after 222 and 234, 241 was not a nice build number for a release anyway [08:59] sil2100, bug 1423837 [08:59] bug 1423837 in system-image (Ubuntu) "OTA from 240 to 241 fails with systemimage.gpg.SignatureError" [Undecided,New] https://launchpad.net/bugs/1423837 [09:07] I really should make it beep when it finishes flashing. [09:08] So... yeah, same error jibel got. [09:09] I get the impression that it might just fix itself on a new build, assuming the build process simply didn't sign the image correctly. [09:10] well, its not the image, its the channels that have a bad sig [09:10] and i fear we need stgraber or slangasek for this ... i doubt a re-build will fix it [09:12] Hmm, okay. Not sure how that part of the pipeline works. [09:12] it means not before 4 hours min. [09:13] yeah ... but i'm in the same camp as ToyKeeper here [09:13] not sure if mvo has more insight into channel handlin [09:13] g [09:13] I would say 241 can be tested anyway, but not the OTA parts [09:14] I started on that, at least. It fixed the wizard bug in 240. Woot! [09:14] ogra_: not much more insight, no, I get the same error on snappy fwiw [09:14] right, that what i assumed [09:15] i suspect the /var/lib/system-image/keyrings/image-signing.tar.xz doesnt have the matching content [09:15] or some such [09:15] I started pretty late though, since the build was delayed, and got sidetracked a lot by the weirdness which probably won't exist in 242. [10:32] Mirv: greyback: ↑ ? [10:33] tsdgeos: check the option like "don't check for twins" [10:33] tsdgeos: greyback: ignoring the twin package for now, doing another build === ev changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping ev | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: ubuntu-rtm landing gates now closed [10:47] Mirv: would you throw me a silo please? [10:48] * Mirv throws a heavy silo in the general direction of bzoltan_. it has "0 2 4" carved on the side of it. [10:49] * bzoltan_ cathces it like if it were a beer can [10:49] kalikiana: tsdgeos: ^ u1db-qt is ready to be tested in vivid silo 19 [11:01] sil2100, jibel, i think we have the channes.json fixed [11:01] my phone just started an OTA ... [11:01] might take a while for you, i had to kill system-image-dbus to trigger it [11:03] * ogra_ hits the install button [11:09] \o/ [11:10] greyback: tsdgeos: ^ qtmir built [11:10] davmor2: ↑ can you try this one (not sure if you were one of the ones seeing the slowlyness) [11:11] tsdgeos: nope regression testing ota1 latter I can [11:11] ok [11:11] i [11:17] I think I have to play more Robots vs Robots to be sure about the performance fix .... [11:17] I'm sure I'd suddenly find a horde of QA volunteers to help, though [11:28] Mirv: I'm on hard level 6 it's enough to make you weep I tell you [11:29] I'm enough weeping on level 1, but I'm also quite confident the silo 021 is very helpful for perf [11:47] cihelp: can someone help me understand why https://code.launchpad.net/~flscogna/ubuntu-terminal-app/json-keyboard-profiles/+merge/249754 is failing? specifically "ubuntuuitoolkit._custom_proxy_objects._common.ToolkitException: No actions in overflow" from http://91.189.93.70:8080/job/generic-mediumtests-utopic/2059/consoleText === _salem is now known as salem_ [12:12] popey, The problem appears to be unrelated to that MP, those tests are also failing on trunk and on the dashboard , http://ci.ubuntu.com/smokeng/vivid/touch/mako/106:20150220:20150210/12348/ubuntu_terminal_app/ === MacSlow is now known as MacSlow|lunch [12:23] fginther: its running the wrong tests [12:24] popey, the dashboard is running the wrong tests? or jenkins too? [12:24] well the dashboard seems to be [12:25] the new terminal is https://code.launchpad.net/~ubuntu-terminal-dev/ubuntu-terminal-app/reboot [12:25] which doesn't have the circle menu which the old terminal had [12:25] so perhaps it's pulling tests from the wrong place? [12:29] popey, ack, I see now that the branch was switched out. [12:31] k [12:41] popey, have the terminal app tests been updated for the reboot branch? It appears that the tests directory is identical between the lp:ubuntu-terminal-app/reboot and lp:ubuntu-terminal-app/trunk branches [12:41] fginther: I'll file a bug to get them updated [12:49] sil2100, do we have 241? [12:49] pmcgowan: yes! [12:49] pmcgowan: but... [12:49] Mirv: silo 19 looks to be fine as far as u1db using apps goes, no explosions etc. [12:49] pmcgowan: we had issues with making it OTAable, not sure if it's fixed already [12:49] pmcgowan: but all in all QA is testing it already [12:50] sil2100, ok great, for some reason I didnt get a notification for it [12:50] Mirv: as we talked about this I'm actually wondering if it'd be good if I could easily run unit tests against the silo… [12:50] but that's pretty much the issue you were already describing… it's not [12:51] pmcgowan: yeah, OTA might be still b0rken [12:51] still, didnt know it was [12:52] Mirv: (well, I did it by hand in addition to manual testing, just is a bit awkward) [12:52] kalikiana: ok, thanks! did you test it on device, and if so, which device + image? [12:52] kalikiana: it'd be nice indeed [12:55] Mirv: d'oh good you're asking about what image… my testing it on rtm was kind of pointless… [12:55] I'll re-do that on vivid :-] === alan_g is now known as alan_g|lunch [12:56] kalikiana: hah! [13:00] sil2100, I didnt know u1-db was broken on rtm, only on vivid with qt5.4? Mirv ? [13:02] davmor2, jibel: how's testing going so far? [13:02] pmcgowan: I think it's only on vivid [13:03] sil2100, it is looking good so far. Still have to run a bunch of telephony tests and upgrade tests [13:04] sil2100, I am confusing myself looking at both dashboards :) [13:04] sil2100, I could upgrade frmo 240 to 241 after ogra fixed channels.ini but didn't receive the notification [13:04] plars or cihelp, can I remove the "backwards" compatibility from u-d-f already? [13:04] s/.ini/.json/ [13:04] whatever he fixed [13:04] pmcgowan: only vivid and the silo fixes it [13:04] ack [13:05] pmcgowan: telegram works again at least [13:05] jibel, I wonder if he can manually trigger the notification [13:05] sergiusens, excellent [13:08] arg [13:08] it's really annoying that udf just ignores typos and wrongcommands [13:08] I want those 10min of my life back [13:10] ogra_: what happened with the channels that made ota to fail? [13:11] I had to run the importer by hand yesterday as it was disabled [13:11] slangasek enabled it after that [13:19] rsalveti, cahennels.json apparently got edisted by hand instead of via s-i-shell [13:19] rsalveti, s-i-shell gpg signs it after every change ... editing by hand doesnt :) [13:20] so we had to re-generate the signature and all was fine (simply calling hide_channel and sho_channel for a random channel in si-shell did that) [13:27] ogra_: oh, looks like someone didn't know what he was doing :-) [13:28] but cool that you could easily fix it [13:28] and it worries me that it's so easy to break it [13:28] rsalveti, well, and someone else who didnt know what he was doing fixed it by pressing random buttons ;) [13:28] this is how system-image maintenance works :P [13:28] * Mirv blames someone [13:29] ogra_: is there a todo item about making that process more.. solid/less errorprone? [13:29] why ? it worked fine :P [13:30] (i dont think there is a todo ... but by the wiki documentation it was guess-able what was needed) [13:33] trainguards: rtm silo 11 failed to build because it says there are some symbols changed, but by running strings on the rtm library I see the symbols were already there [13:34] trainguards: any idea? === MacSlow|lunch is now known as MacSlow [13:35] boiko: let me take a look at the error [13:37] sil2100: thanks [13:41] Mirv: seems good. now also vivid on the device :-P [13:42] ok, so #image 106 mako? or krillin? [13:50] Mirv: 106, mako [13:50] (and on vivid desktop) [13:51] kalikiana: thank you, marked so [13:52] boiko: looking [13:52] boiko: hmmm [13:52] boiko: how are you making those symbols files btw.? [13:53] sil2100: so, it is my first time updating telepathy-qt on rtm, I actually didn't even know I was supposed to update those symbol files [13:53] rsalveti: do you have an idea when/how the symbol files were updated? [13:53] boiko: ok, looking at the source now === alan_g|lunch is now known as alan_g [13:55] boiko: you usually know by building and checking the diff (after it fails to build) [13:55] let me check the build log [13:55] boiko: sil2100: rsalveti: ^ looking already [13:55] and updating symbols [13:55] oh, great then [13:55] yeah, Mirv knows it all [13:55] Mirv: so, any idea how the .so had the symbols but the deb didn't have them listed? [13:56] boiko: oh god, those are raw symbols, I would simply copy the version returned from dpkg-gensymols and just use it instead [13:56] boiko: problem with C++ symbols files is that the linker exports everything by default [13:57] sil2100: ah ok [13:57] boiko: so without proper linker handling the symbols that are leaked have all the inner symbols like Qt, C++ STL and similar [13:57] sil2100: got it [13:57] But anyway, I guess this can be solved easily, Mirv will certainly fix it in a moment ;) [14:04] sergiusens: hi, which backwards compatability do you mean? the assumed "touch" mode in the command line? [14:06] boiko: what sil said, c++ :( uploaded another build and monitoring how it goes [14:06] * Mirv more than familiar with C++ symbols files via Qt [14:18] boiko ^ watch that space, x86 built already so unless armhf has anything special it should also succeed [14:24] plars: yes, that one [14:30] sergiusens: I believe so, let me check a couple of things first [14:32] plars: sure, not going to do today or anything like that; I'll add you as a reviewer to the MP when I get to it [14:33] sergiusens: it looks safe [14:34] sergiusens: ah, the gatekeeper job for autopilot at least still uses the old method, no idea why [14:34] sergiusens: will try to talk to veebers about it [14:42] Mirv: thanks! [14:52] hmmmm ^^ Mirv what does this mean? [14:57] kalikiana: something wrong with the branches. I don't see anything wrong, but lately there has been more and more of this. looking. [15:03] eh... bzr: ERROR: Cannot lock LockDir(chroot-79771920:///~uonedb-qt/u1db-qt/trunk/.bzr/branch/lock): Transport operation not possible: readonly transport [15:03] sil2100: could you look at wtf is https://ci-train.ubuntu.com/job/ubuntu-landing-019-3-merge-clean/37/console for kalikiana, or further delegate to robru when he's online? [15:04] I'm obliged to start doing other "important stuff" (cleaning the house...) [15:04] Mirv: looking [15:05] Mirv: ok ;) Good luck! [15:06] sil2100: something else too on the air, looking at boiko's silo: https://ci-train.ubuntu.com/job/ubuntu-rtm-landing-011-1-build/104/console "cupstream2distro.errors.CITrainError: dpkg-source: error: unpack target exists" [15:06] thanks :) [15:06] the build itself with updated symbols was fine, so I'm happy [15:06] Silo 11 looks like we're missing some clean-up ;/ [15:30] jibel: how much more testing left? [15:31] sil2100, we are done. I'm reviewing the bugs and writing the report. [15:31] \o/ === chihchun_afk is now known as chihchun [16:44] jibel, sent you my two cents on silos, but I am weak :) [16:50] sil2100, is there anything important for me in the landing meeting ? i'm in another one, but if it is important that i'm there i can drop off there for a bit, to join LT [16:54] ogra_: I think we're fine, just a 'thank you' for fixing the OTA's ;) [16:54] np [17:05] sil2100: so, any update on what is going on now on rtm silo 11? [17:24] ev: hi, looks like autopkgtest some host or other is out of space: https://jenkins.qa.ubuntu.com/view/Vivid/view/AutoPkgTest/job/vivid-adt-cairo/lastBuild/ARCH=amd64,label=adt/console [17:25] boiko: it's strange, robru is looking into that as well as it might be some strange condition in the new code [17:25] ev: ah, wazn [17:26] sil2100: I have a branch for that first log you showed me. were there any other problems? [17:26] Mirv: readonly transport means the bot doesn't have permission to merge that branch. you gotta ask the team admins to add the bot if they want to use the train for landings. [17:27] Mirv: https://code.launchpad.net/~uonedb-qt/u1db-qt/trunk if you look here, the last release was done by the old bot creds, so they never added the new bot when we switched over, and now it's their first time trying to use the train with the new creds. [17:28] Mirv: and apparently you are an admin ;-) [17:29] sil2100: oups, ok [17:34] jibel: just double confirming - 241 gets a +1 on promotion? ;) [17:37] sil2100, yes, krillin 241, mako 199 and x86 191 [17:38] x86 191? Not 193? [17:39] jibel: ^ ? [17:39] sil2100, you're right, 193 [17:39] 191 is flo [17:41] == IMAGE RTM 19 promoted == [17:42] (for other platforms it's mako 17 and x86 16) [17:42] john-mcaleely: ^ [17:42] It's in the rc channel [17:42] ubuntu-touch/rc/bq-aquaris.en [17:42] \o/ [17:43] Ok, I need to drive home now, I'll write up the e-mail later on [17:43] o/ [20:16] Mirv: https://ci-train.ubuntu.com/view/1.%20Build/job/ubuntu-rtm-landing-011-1-build/106/console ok, fixed that one for you, sorry about that. [20:23] boiko: ah, looks like you need to rebuild and retest! [20:26] robru: ah oups, forgot to rebuild telepathy-ofono in there, thanks for the heads up [20:26] boiko: you're welcome! === sil2100 changed the topic of #ubuntu-ci-eng to: Need a silo or CI Train support? ping trainguards | Need help with something else? ping ev | Train Dashboard: http://bit.ly/1mDv1FS | QA Signoffs: http://bit.ly/1qMAKYd | Known Issues: - [21:26] trainguards i got something i've never seen [21:26] https://ci-train.ubuntu.com/job/ubuntu-landing-000-1-build/337/console [21:26] any help ? [21:27] kgunn: looking [21:27] thanks [21:27] strange one [21:27] to me at least [21:27] kgunn: oh great [21:27] kgunn: yeah unicode handling is a bit weird, I've been working on it, looks like I made a mistake [21:28] kgunn: let me just revert that so you're not blocked... [21:28] thanks man [21:32] kgunn: alright, retrying: https://ci-train.ubuntu.com/job/ubuntu-landing-000-1-build/338/console [21:32] cool [21:41] robru: silo 25 rebuilt and retested, all good [21:41] boiko: great, thanks === chihchun is now known as chihchun_afk [21:42] boiko: just need a packaging ack ;-) [21:42] kenvandine: mterry: anybody around for a packaging ack? https://ci-train.ubuntu.com/job/ubuntu-landing-025-2-publish/21/ [21:44] kenvandine: can you take it a look? === salem_ is now known as _salem === _salem is now known as salem_ === salem_ is now known as _salem [21:46] robru, seems fine [21:53] mterry: thanks! [21:54] boiko, sure [21:54] oh [21:54] mterry beat me :) [21:54] mterry, thx :) [21:54] kenvandine: thanks anyway! [21:54] mterry: kenvandine: thanks [23:00] W. T. F