**** BEGIN LOGGING AT Thu Nov 29 02:59:59 2018 Nov 29 07:30:39 Morning! Nov 29 07:30:54 Tofe: Merged all and kicked off a testing build, let's see where we get :) Nov 29 07:35:03 dkirker: YGM Nov 29 07:43:48 Morning ! Nov 29 07:44:05 Herrie|Laptop: thanks ! Nov 29 07:44:17 expect a mistake or two though :) Nov 29 07:51:20 Tofe: Well of course... Seems that most of the work I did wasn't too bad after all on migration. Seems 70-80% was pretty OK ;) Nov 29 07:55:53 Herrie|Laptop: yes, it helped a lot as a starting ground Nov 29 08:09:17 Herrie: first mistake: "ERROR: luna-sysmgr-3.0.0-3+gitAUTOINC+86cba93fb5-r0 do_fetch: Fetcher failure: Unable to find revision 86cba93fb57f27f102d93bd51fb6bb82ab72b715 in branch webosose even from upstream" Nov 29 08:18:27 Tofe: That one should be fixed now Nov 29 08:19:48 app-services is another one Nov 29 08:20:07 Let me cherry pick the commit from webos-ports/master in webosose and fix that one as well Nov 29 08:30:52 Herrie|Laptop: shouldn't we point to the webos-ports/master branch instead? Nov 29 08:31:12 Tofe: Well your PR went to webosose ;) Nov 29 08:31:18 oops Nov 29 08:31:43 But yeah you can PR to master Nov 29 08:31:47 ah maybe it didn't want to go to webos-ports/master because of a completely different history Nov 29 08:31:57 Since the chances I did in OSE are gone becuase I dropped the whole mojomail folder Nov 29 08:32:07 In master Nov 29 08:32:21 webos-ports/master I mean Nov 29 08:33:13 mojomail which is in https://github.com/webOS-ports/mojomail was duplicated into app-services Nov 29 08:33:19 So I removed it from app-services Nov 29 08:33:25 And the duplicate bits from mojomail itself Nov 29 08:34:42 ah ok, so that's why there was a conflict Nov 29 08:34:55 Yup Nov 29 08:38:15 Tofe: I cherry picked your commit into webos-ports/master Nov 29 08:38:17 So we should be good Nov 29 08:39:43 ok thanks Nov 29 08:40:05 I'll update the recipe Nov 29 08:41:43 Let's restart a build Nov 29 08:44:38 Biggest thing missing now are the nyx hybris modules I guess? Nov 29 08:50:30 I don't think there's anything left to do on that front -- but maybe I missed a point here Nov 29 08:57:14 Tofe: Could be just this needs to be enabled again: https://github.com/Tofee/meta-webos-ports/blob/0d744a3f170416916cc61c1de8f5576980837b89/meta-luneos/recipes-core/packagegroups/packagegroup-luneos-extended.bb#L133 Nov 29 08:58:41 oh, I didn't see that one... Nov 29 08:59:00 Tofe: I'm sure those need work Nov 29 08:59:03 I got stuck there Nov 29 08:59:12 ok, I'll have a look Nov 29 08:59:15 I did the API migration, but the rest was over my head and the code could use some cleaning up too Nov 29 08:59:29 Lots of warnings with compiling due to code style it seems Nov 29 08:59:48 yes probably Nov 29 09:25:00 Herrie|Laptop: the latest work was https://github.com/webOS-ports/nyx-modules-hybris/commits/webosose ? Nov 29 09:25:51 I think so Nov 29 09:25:53 It's been a while Nov 29 09:26:03 There's also https://github.com/webOS-ports/nyx-modules-hybris/tree/webosose-2 Nov 29 09:26:08 They should be similar-ish Nov 29 09:26:15 But I don't remember which was latest Nov 29 09:26:24 I would say the webosose since it has more commits Nov 29 09:26:32 Which would need to be squashed but I really got stuck there Nov 29 09:27:19 I'll manage from there, you already did a nice job Nov 29 09:27:42 When it gets too much c++ it's over my head sorry ;) Nov 29 09:28:03 I can follow instructions and change some things, but anything more then that I'm more likely to break v.s. fix ;) Nov 29 09:29:17 hehe :) Nov 29 09:43:27 Hmmz Hammerhead failed with "| /home/jenkins/workspace/luneos-testing/webos-ports/tmp-glibc/deploy/images/hammerhead/zImage-dtb: No such file or directory' Nov 29 09:44:37 That's weird Nov 29 09:44:43 out of space? Nov 29 09:46:40 nope; must be something else. And luneos-dev-package succeeded... Maybe it's a dependency issue that has always been there Nov 29 10:05:19 Tofe: Yeah could be Nov 29 10:28:43 Seems we might be running out of space though.... NOTE: Running task 7093 of 7184 (/home/jenkins/workspace/luneos-testing/webos-ports/meta-webos-ports/meta-luneos/recipes-core/images/luneos-dev-package.bb:do_install) is taking forever to complete Nov 29 10:29:39 mmh Nov 29 10:39:26 Will need to poke novaldex or JaMa I guess Nov 29 10:39:31 novaldex|away: ^ Nov 29 10:40:10 is that my cue? morning.. Nov 29 10:41:37 novaldex: ^ yeah ;) Nov 29 10:42:17 Seems we might be running out of space again somewhere :P Nov 29 10:43:16 I.e. http://jenkins.nas-admin.org/job/luneos-testing_hammerhead/lastBuild/console seems stuck Nov 29 10:43:30 Normally this step finishes in a minute or 2, now been stuck here for hr+ Nov 29 10:43:36 So suspect we're out of disk space again somewhere Nov 29 10:43:49 Morning BTW;) Nov 29 10:46:22 i'm thinking it's the builder, as there's tons of space on Kaylee Nov 29 10:46:38 '/dev/xvdc1 102718924 32137132 70057528 32% /mnt/jenkins Nov 29 10:48:31 Should be on Bonaire I guess Nov 29 10:48:47 I thought disk space would be increased there or we're still waiting for those disks from LG :P ? Nov 29 10:49:22 yes, bonaire: /dev/xvdb1 515928504 483120236 9221200 99% /home Nov 29 10:49:46 I can't remember how big it was before, but that's a 500GB volume? Nov 29 10:52:04 I think it was 500 before, but not sure Nov 29 10:53:02 i'll check with ka6sox later today about it Nov 29 10:53:10 can we clean anything up do you know? Nov 29 11:02:03 Well 500 is already a lot, there's not a whole lot we can clean in terms of images Nov 29 11:02:10 You could empty the testing in general Nov 29 11:02:15 Since we just pushed out a release Nov 29 11:02:34 http://build.webos-ports.org/luneos-testing/images/ Nov 29 11:08:26 This could go to: http://build.webos-ports.org/luneos-stable/images/ Nov 29 11:15:42 the images for the build website are stored on a different server, we've got a good amount of space left on that one still Nov 29 11:15:53 Ah ok Nov 29 11:16:02 Seems that Mako is running now though :S Nov 29 11:16:09 Not sure what's up really Nov 29 11:16:10 '/dev/xvdf1 1048510004 790217684 258292320 76% /home/jenkins (for the build website) Nov 29 11:16:22 Ah OK good Nov 29 11:16:24 there's a tiny bit of room just now, about 8G Nov 29 11:16:44 I guess need to ping JaMa when he's around Nov 29 11:16:51 I don't know much about how & what on the builder Nov 29 11:17:19 i'm looking in the workspace dir just now, I think that's where most files end up Nov 29 11:17:31 problem is they get so darn big it's hard to tell! Nov 29 11:18:17 I think that's 10 compiler threads running simultaneously on bonaire? Nov 29 11:18:19 Well there's a workspace clean task Nov 29 11:18:34 That we run in between various targets Nov 29 11:18:35 12 maybe? more? Nov 29 11:18:38 That might clean things up Nov 29 11:18:40 Yeah could be :) Nov 29 11:18:47 busy little box! Nov 29 11:19:07 long as it's running just now, we'll keep an eye on it & run workspace clean up soon as we can Nov 29 11:21:10 We'll see Nov 29 11:23:28 JaMa: Morning ;) Nov 29 11:23:37 We were just talking about you, so you come at the right time ;) Nov 29 11:23:46 Seems we're runnign out of diskspace on Bonaire Nov 29 11:23:52 You might have an idea what to clean Nov 29 11:24:24 We're trying to build OSE-ified LuneOS ;) Nov 29 13:19:59 Herrie|Laptop: will check in few minutes Nov 29 13:20:33 JaMa: Thnx Nov 29 13:40:24 218G luneos-shared Nov 29 13:40:25 4.6G luneos-stable Nov 29 13:40:25 33G luneos-testing Nov 29 13:40:36 139G luneos-unstable Nov 29 13:40:41 bye-bye unstable Nov 29 13:41:19 BTW: I've already deleted OSE a while ago and stopped running the builds, because of that question from skype.. Nov 29 13:41:47 ka6sox: did you receive the hard disks promised in Paris for building webOS OSE on LuneOS servers? Nov 29 13:47:13 Herrie|Laptop: I'll kill whole local sstate cache for testing and stable builds (161G luneos-shared/sstate-cache/), but let me finish building all targets once (with manually tmpfs cleanup between) without running rsync job Nov 29 13:47:35 because rsync job would remove the sstate from the file server as well -> causing next builds to take significantly longer Nov 29 13:58:35 JaMa: Thnx Nov 29 14:13:50 JaMa: oops seems I killed your rsync Nov 29 14:13:58 I thought it was started by mistake so I killed it Nov 29 14:20:42 no problem, will run it again :) Nov 29 14:21:02 just don't touch it, until I'm finished :P Nov 29 14:26:14 Herrie|Laptop: I'll remove 2nd luneos-testing_workspace-compare-signatures from http://jenkins.nas-admin.org/view/luneos-testing/job/luneos-testing_build-all/ IIRC it does exactly the same check just twice Nov 29 14:26:55 onyx isn't aarch64? Nov 29 14:34:59 JaMa: No Snapdragon 801, 32 bits Nov 29 14:37:29 ok, ty Nov 29 16:16:38 Herrie|Laptop: Tofe: damn qtwebengine isn't reused from sstate so this rebuild will take a while Nov 29 16:16:53 can I break emulator a bit and merge my pending changes for VirtualBox? Nov 29 16:17:06 so that we save one bigger rebuild? Nov 29 16:22:03 Herrie|Laptop: check your OSE change in meta-luneos/recipes-webos/nyx-modules/nyx-modules.bb looks like you've accidentally removed PV variable Nov 29 16:22:28 nyx-modules went backwards which would break package feeds from (0:6.1.0-94+git0+bd739fa608-r0 to 0:1.0-r0) [version-going-backwards] Nov 29 16:28:00 heh found an excuse to push it now :) Nov 29 16:28:02 Nov 24 16:58:36 JaMa: it should be fine to try some things on pyro Nov 29 16:32:26 JaMa: Good Nov 29 16:32:33 Not sure what happened with nyx-modules, checking Nov 29 16:32:41 Been a while since I touched the OSE bits Nov 29 16:32:47 Tofe picked up where I left ;) Nov 29 16:33:20 JaMa: Seems mistake when I moved the SRCREV around Nov 29 16:34:19 LG's equivalent of PV = WEBOS_VERSION right? Nov 29 16:35:06 Herrie|Laptop: https://github.com/webOS-ports/meta-webos-ports/commit/9d14ac00cd8fe8be23d640e79ae7562c3fb184ec Nov 29 16:35:12 yes Nov 29 16:35:25 JaMa: Yes found it, putting it back now Nov 29 16:36:03 WEBOS_VERSION is PV + SRCREV + check in webos_enhanced_submissions that PV contains submission number and corresponding submission tag is annotated and matches with SRCREV Nov 29 16:37:10 JaMa: This should do it: https://github.com/webOS-ports/meta-webos-ports/commit/a7a653fb5d09265677532fc8f9046bca34eabe61 Nov 29 16:37:20 I think I saw something else in logs with same error Nov 29 16:37:27 akc Nov 29 16:37:28 ack Nov 29 16:38:13 all 1.0 versions are suspicious Nov 29 16:40:29 I also see: WARNING: activitymanager-3.0.0-1+gitAUTOINC+6e38fc5791-r0 do_packagedata: QA Issue: Package version for package activitymanager-lic went backwards which would break package feeds from (0:3.0.0-123+git5+1e3e1f2a94-r0.66 to 0:3.0.0-1+git0+6e38fc5791-r0.0) [version-going-backwards] Nov 29 16:40:52 And WARNING: configurator-3.0.0-1+gitAUTOINC+10464dbc52-r0 do_packagedata: QA Issue: Package version for package configurator-lic went backwards which would break package feeds from (0:3.0.0-59+git9+86ca6b12e2-r0.11 to 0:3.0.0-1+git0+10464dbc52-r0.0) [version-going-backwards] Nov 29 16:49:22 did you switch to their submission tags? if it's from our forked repo with different SRCREV then I would keep the old submission number (or higher) Nov 29 16:49:38 everything in OSE reset submission tags to 1 :/ Nov 29 16:52:40 JaMa: I thought since we took their repo as source, I'd also switch the submission tag? Nov 29 16:52:42 Or shouldn't? Nov 29 16:53:10 Basically imported their OSE repo into a new branch at our end Nov 29 16:53:16 And patched where needed Nov 29 16:53:25 If Nov 29 16:53:30 if you switched to their repo then it's fine (or rather "their" fault) :) Nov 29 16:53:59 Well at some point when we're happy with all, we should cut the links with OWO upstream and switch to OSE I guess Nov 29 16:54:02 But that's for future Nov 29 17:44:30 JaMa: yes, feel free to experiment a bit, we can live without qemu target for some days (even more *these* days, when I probably broke pyro myself...) Nov 29 17:48:06 hopefully I didn't break other targets as well :) Nov 29 17:48:49 but I changed qtbase a bit, so better to do bigger rebuild just once, the fixes should be smaller and restricted to qemu Nov 29 17:50:46 ok Nov 29 18:13:41 Herrie: you were actually one little modification away from building nyx-modules-hybris successfully :p Nov 29 18:23:55 ok, let's go and rebuild qt... Nov 29 19:00:35 http://forum.webosose.org/t/sharing-news-for-collaboration-with-luneos/713/2 heh Nov 29 19:01:11 I wonder what it means as well :) Nov 29 19:53:22 Tofe: Great! Will merge! Nov 29 20:54:36 JaMa: it's a bit awkward n-jay guesses are both incorrect :p Nov 29 20:55:20 (or, do I misunderstand the work accomodate... that could be... I'll check) Nov 29 20:55:24 word* Nov 29 20:56:33 (no, it does mean "integrate", so it's incorrect :) ) Nov 29 21:03:54 JaMa: I have issues building e2fsprogs-native: is that to be expected ? Nov 29 21:35:59 Tofe: not expected, what kind of issues? Are you using 16.04? Nov 29 21:36:15 it shouldn't even need rebuild based on recent changes Nov 29 22:00:42 * JaMa needs a drink, virtual toast anyone? Nov 29 22:01:01 Cheers! Nov 29 22:01:47 Cheers! Nov 29 22:17:24 Cheers! Nov 29 22:18:28 JaMa: on 18.04, something like https://bpaste.net/show/41acd7b632ef Nov 29 22:18:49 but I must have done something wrong, I guess Nov 29 22:21:23 Tofe: do you have latest oe-core/pyro branch? Nov 29 22:21:47 looks like you're missing all recent fixes Nov 29 22:22:06 "Your branch is behind 'origin/pyro' by 237 commits, and can be fast-forwarded." huh....... Nov 29 22:22:08 thanks! Nov 29 22:22:08 Tofe: http://git.openembedded.org/openembedded-core/commit/?h=pyro&id=350f2a4ad6e21acf1d357a90ba37b2c149ec7864 Nov 29 22:22:31 don't forget to update other layers and bitbake as well Nov 29 22:22:56 I thought I did a global update, but it must have been stuck without me noticing Nov 29 22:23:05 and expect rebuilt from scratch Nov 29 22:23:11 oh, that I do :) Nov 29 22:23:15 you can pretty much delete your old sstate-cache at this point Nov 29 22:23:48 good idea, that'll clean up a bit Nov 29 22:24:03 I'll build the image tomorrow then Nov 29 22:24:51 Herrie: are you also using 18.04? if you both use 18.04 (or want to use 18.04) we should upgrade jenkins servers as well to generate usable sstate-cache for you Nov 29 22:25:04 or set uninative for LuneOS Nov 29 22:25:14 but that will need to wait after OE upgrade Nov 29 23:28:32 http://jenkins.nas-admin.org/view/luneos-testing/job/luneos-testing_mido/16/console failed with: Nov 29 23:28:35 nyx-modules-hybris/0.1.0-1+gitAUTOINC+e363872871-r0/git/src/haptics/haptics.c:41:41: error: macro "nyx_info" requires 3 arguments, but only 1 given Nov 29 23:28:43 but that might be before the latest fix Nov 29 23:30:53 looks like nyx-modules-hybris was returned to image, but SRCREV bump is missing, will fix Nov 30 00:29:47 looks like every MACHINE is building its own qtwebengine, that's why the builds are so slow now Nov 30 01:00:44 Herrie: the db8 change from commit 9d14ac00cd8fe8be23d640e79ae7562c3fb184ec which adds dependency on VIRTUAL-RUNTIME_stat breaks build with newer Yocto, you forgot to import this part: meta-webos/conf/distro/include/webos-preferred-providers.inc:VIRTUAL-RUNTIME_stat = "busybox" Nov 30 01:00:50 Herrie: will fix that **** ENDING LOGGING AT Fri Nov 30 03:00:00 2018