| *** gtristan has quit IRC | 02:07 | |
| *** gtristan has joined #baserock | 03:13 | |
| *** paulwaters_ has joined #baserock | 07:51 | |
| *** gtristan has quit IRC | 08:37 | |
| *** jonathanmaw has joined #baserock | 08:38 | |
| noisecell | jjardon, do you know what could be happening: https://gitlab.com/baserock/definitions/-/jobs/76768144 ? | 08:53 |
|---|---|---|
| jjardon | noisecell: did you check the logs? | 08:54 |
| noisecell | jjardon, yeah, but it is giving me: "Job's log exceeded limit of 4194304 bytes." | 08:54 |
| jjardon | oh, seems logs are not being stored | 08:54 |
| jjardon | yeah, there is a limit of the amount of logs you can show there | 08:54 |
| jjardon | 4MB is currenttly that limit | 08:55 |
| jjardon | noisecell: maybe you can try the build again so no so many logs are being generated? | 08:55 |
| jjardon | as everything should be cached? | 08:55 |
| noisecell | jjardon, ok, although I have tried at least 2 times and Im running the pipeline manually because if I run all the "Build-2" jobs all the builds fail retrieving repositories :/ | 08:56 |
| * noisecell presses the button | 08:57 | |
| jjardon1 | noisecell: so the build artifact are not being cached? | 08:57 |
| jjardon1 | artifacts* | 08:57 |
| noisecell | jjardon, https://gitlab.com/baserock/definitions/-/jobs/76768144 -- this is the pipeline (well you can see so far that all the jobs are green) | 08:57 |
| noisecell | jjardon, not sure if intermediate artifacts are cached, I will check in my next build | 08:58 |
| jjardon | noisecell: Are you building exactly the same 2 time and the second time you need to build everything again? | 08:58 |
| noisecell | I think there is a patch in mainline to cache artifacts even if the final build fails | 08:58 |
| noisecell | jjardon, Im pressing the "Retry" button, yes | 08:59 |
| jjardon | noisecell: ybd should be filled up when something is build, doesnt matter if the final job fails or not | 08:59 |
| jjardon | noisecell: can you file a bug report, please? | 09:00 |
| jjardon | noisecell: let me increase the maximum amount of logs for now | 09:00 |
| noisecell | jjardon, is this ybd or buildstream? | 09:00 |
| noisecell | jjardon, cheers | 09:00 |
| jjardon | oh sorry | 09:00 |
| jjardon | this is buildstream XD | 09:00 |
| jjardon | but I think it works in the same way; not sure though | 09:01 |
| jjardon | noisecell: limit increased to 16MB, hope is enough! | 09:04 |
| noisecell | jjardon, cheers, Im going to launch 2 or 3 jobs and see what is going on | 09:04 |
| *** paulwaters_ has joined #baserock | 09:36 | |
| noisecell | jjardon, how do you feel to move out of foundation btrfs-progs.bst and tbdiff.bst ? is this required for any project? | 10:52 |
| * noisecell thinks we are not using btrfs nowadays as filesystem in any image, and it could be added in another element if required, but not as foundation | 10:53 | |
| noisecell | this was a hard requirement for morph/baserock deployments but not sure if that is still true for ybd or buildstream | 10:54 |
| noisecell | ironfoot, ^^ | 10:54 |
| jjardon | noisecell: I do not mind, open a MR if you want | 10:57 |
| jjardon | ybd doesnt support bst format anyway | 10:57 |
| * paulsherwood wonders if it should | 10:58 | |
| ironfoot | tbdiff was needed for upgrades, btrfs too | 10:59 |
| ironfoot | if we want to re-implement that, we should keep it | 10:59 |
| ironfoot | but IMO, using ostree to implement this would be easier, now that we have it | 10:59 |
| noisecell | ironfoot, wasn't tbdiff hardlinked to btrfs? | 10:59 |
| paulsherwood | yes | 10:59 |
| ironfoot | noisecell: I can't remember, not sure if it assumes some volumes already mounted somewhere, being not btrfs dependant | 11:00 |
| noisecell | ok, I will check it | 11:00 |
| ironfoot | I don't mind removing it at all, seriously, upgrades need to be rethought | 11:01 |
| noisecell | as I can see definitions is still using some stacks which Im not sure are longer fine | 11:02 |
| noisecell | given that we can use elements, we might want to reconfigure these stacks/old strata in something else | 11:02 |
| noisecell | and add runtime/build dependencies | 11:02 |
| * noisecell is not happy that changing a link in systemd makes openstack to build almost fully :/ | 11:03 | |
| noisecell | s/openstack/openstack-system/ | 11:03 |
| ironfoot | bst has some features to avoid that during development | 11:04 |
| ironfoot | in reality, i don't know if changing systemd could trigger failures when building any of the components of openstack | 11:04 |
| noisecell | ironfoot, surely not this one, it is a symlink on installation, as liking /sbin/init to lib/systemd/systemd | 11:05 |
| ironfoot | but i feel like some of the elements will depend on systemd being installed | 11:05 |
| noisecell | yes, but at runtime, not buildtime possibly? | 11:05 |
| noisecell | ironfoot, ^^ | 11:06 |
| ironfoot | i might be wrong, but most of the software that installs services check for the version of systemd at build time | 11:07 |
| * noisecell doesn't recall that dependency, but it was long time ago | 11:08 | |
| ironfoot | I might be mixing up my memory with genivi elements | 11:09 |
| ironfoot | and we could try to remove the dependency and see what happens | 11:10 |
| noisecell | ironfoot, yeah, I would like to have an image that boots and then I will like to make it as small as I can. Probably, reducing the stratas/actual stacks to maybe use only a list of elements | 11:11 |
| ironfoot | nice | 11:11 |
| * noisecell is trying to make openstack one-node image come back from the graveyard | 11:12 | |
| ironfoot | deRIP :D | 11:13 |
| *** paulwaters_ has joined #baserock | 12:49 | |
| * noisecell makes openstack image to boot ... without any openstack and services files | 13:09 | |
| ironfoot | :) | 13:09 |
| * noisecell notes that any image that uses systemd for booting will be broken until I submit this branch for review | 13:09 | |
| noisecell | and to submit this branch I need to fix some other oddities that happens when replacing STAGE2_SYSROOT to "/" | 13:10 |
| noisecell | s/happens/happen/ | 13:10 |
| ironfoot | eww | 13:47 |
| ironfoot | :( | 13:47 |
| ironfoot | sounds like definitions is a bit more broken than what we thought | 13:47 |
| jjardon1 | It's what happen with nobody is actually maintaining the stuff :( | 13:49 |
| ironfoot | yep | 13:50 |
| ironfoot | we need someone to buy us some "beers" | 13:50 |
| paulsherwood | o/ | 13:51 |
| gitlab-br-bot | infrastructure: merge request (pedro/update-certs-jun-2018->master: SSL: Update LetsEncrypt certs Jun 2018) #27 changed state ("opened"): https://gitlab.com/baserock/infrastructure/merge_requests/27 | 14:28 |
| ironfoot | Let's Ecrypt gave me a 1 day warning :/ | 14:29 |
| * noisecell doesn't have permissions to merge | 14:32 | |
| gitlab-br-bot | infrastructure: merge request (pedro/update-certs-jun-2018->master: SSL: Update LetsEncrypt certs Jun 2018) #27 changed state ("opened"): https://gitlab.com/baserock/infrastructure/merge_requests/27 | 14:33 |
| gitlab-br-bot | infrastructure: merge request (pedro/update-certs-jun-2018->master: SSL: Update LetsEncrypt certs Jun 2018) #27 changed state ("merged"): https://gitlab.com/baserock/infrastructure/merge_requests/27 | 14:33 |
| *** paulwaters_ has joined #baserock | 15:03 | |
| *** paulwaters_ has quit IRC | 15:04 | |
| *** gtristan has joined #baserock | 15:23 | |
| *** jonathanmaw has quit IRC | 17:20 | |
| *** jonathanmaw has joined #baserock | 17:34 | |
| *** jonathanmaw has quit IRC | 17:38 | |
| *** gtristan has quit IRC | 18:40 | |
| *** gtristan has joined #baserock | 21:23 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!