| *** cosm has quit IRC | 00:25 | |
| *** cosm has joined #baserock | 00:31 | |
| *** cosm has quit IRC | 01:36 | |
| *** cosm has joined #baserock | 01:50 | |
| *** cosm has quit IRC | 02:45 | |
| *** cosm has joined #baserock | 02:46 | |
| *** toscalix has joined #baserock | 03:44 | |
| *** cosm has quit IRC | 03:50 | |
| *** toscalix has quit IRC | 06:29 | |
| *** CTtpollard has joined #baserock | 06:59 | |
| *** faybrocklebank has quit IRC | 07:37 | |
| *** faybrocklebank has joined #baserock | 07:38 | |
| *** bashrc_ has joined #baserock | 08:00 | |
| *** rdale has joined #baserock | 08:13 | |
| *** ctbruce has joined #baserock | 08:21 | |
| *** anahuelamo has joined #baserock | 08:22 | |
| *** franred has joined #baserock | 08:30 | |
| *** tiagogomes has joined #baserock | 08:41 | |
| *** jonathanmaw has joined #baserock | 08:51 | |
| *** tiagogomes has quit IRC | 09:03 | |
| *** tiagogomes has joined #baserock | 09:05 | |
| *** ssam2 has joined #baserock | 09:11 | |
| *** ChanServ sets mode: +v ssam2 | 09:11 | |
| *** locallycompact has joined #baserock | 09:18 | |
| *** bwh has quit IRC | 10:01 | |
| *** benbrown_ has quit IRC | 10:01 | |
| *** locallycompact has quit IRC | 10:01 | |
| *** SotK has quit IRC | 10:01 | |
| *** paulsherwood has quit IRC | 10:02 | |
| *** paulsherwood has joined #baserock | 10:02 | |
| *** SotK has joined #baserock | 10:03 | |
| *** franred has quit IRC | 10:04 | |
| *** bwh has joined #baserock | 10:08 | |
| *** benbrown_ has joined #baserock | 10:08 | |
| *** locallycompact has joined #baserock | 10:11 | |
| *** franred has joined #baserock | 10:19 | |
| pedroalvarez | jjardon: sorry, I didn't spot this error when reviewing: https://gerrit.baserock.org/#/c/2049/ | 11:07 |
|---|---|---|
| pedroalvarez | red mason is red :) | 11:07 |
| franred | pedroalvarez, you can go ahead and merge it | 11:08 |
| jjardon | pedroalvarez: mmm, sorry; I think ybd builds that without complaining | 11:09 |
| pedroalvarez | no worries :) it's fixed now | 11:09 |
| *** cosm has joined #baserock | 11:27 | |
| *** cosm has quit IRC | 11:29 | |
| pedroalvarez | great! find the submit button here: https://gerrit.baserock.org/#/c/2048/2 | 12:00 |
| pedroalvarez | it has changed with latest release :) | 12:00 |
| franred | pedroalvarez, you can press it ;-) | 12:01 |
| pedroalvarez | the thing is that I can't, and that's what I mean. The button now says that you won't be able to when you hover over it | 12:03 |
| pedroalvarez | this is cool :) | 12:04 |
| pedroalvarez | franred: thanks for reviewing | 12:04 |
| *** edcragg has joined #baserock | 12:09 | |
| *** franred has quit IRC | 12:17 | |
| *** tiagogomes has quit IRC | 12:20 | |
| *** franred has joined #baserock | 12:29 | |
| *** tiagogomes has joined #baserock | 12:53 | |
| *** franred has quit IRC | 13:23 | |
| *** franred has joined #baserock | 13:23 | |
| *** franred has quit IRC | 13:53 | |
| *** franred has joined #baserock | 14:03 | |
| edcragg | i'm getting 'ERROR: Ref 48416261f38c2d1e4ee7a4482238565299c4a605 was not found in repo git://git.baserock.org/baserock/baserock/definitions.' when i try to run morph, i usually have changes in my local definitions that aren't on g.b.o, can't understand why this is happening | 14:47 |
| ssam2 | are you using --local-changes=ignore setting? | 14:58 |
| ssam2 | if so, all the commits need to be pushed, it actually ignores your local repo other than to find out what commit you want to build | 14:59 |
| ssam2 | the error could definitely be improved there (or the behaviour made less confusing) | 14:59 |
| ssam2 | i'm not sure there's any value in that behaviour beyond making it easy to know you're not accidentally building uncommitted stuff.. but there are other ways to check that | 15:00 |
| ssam2 | so morph should just read the files off disk where it can | 15:00 |
| *** ssam2 has quit IRC | 15:06 | |
| edcragg | i agree | 15:08 |
| edcragg | i'm not using --local-changes=ignore, though | 15:09 |
| *** ssam2 has joined #baserock | 15:22 | |
| *** ChanServ sets mode: +v ssam2 | 15:22 | |
| edcragg | thanks ssam2... seems to be avoided by changing the remote origin to my usual repo | 15:24 |
| edcragg | for the moment | 15:24 |
| *** cosm has joined #baserock | 15:25 | |
| *** cosm has quit IRC | 15:38 | |
| *** CTtpollard has quit IRC | 15:39 | |
| *** CTtpollard has joined #baserock | 15:40 | |
| *** ssam2 has quit IRC | 16:01 | |
| *** ssam2 has joined #baserock | 16:13 | |
| *** ChanServ sets mode: +v ssam2 | 16:14 | |
| jjardon | pedroalvarez: nice series! | 16:19 |
| pedroalvarez | jjardon: it's not complete, they are a lot... | 16:20 |
| pedroalvarez | sent them as separate commits to avoid merge conflicts when possible | 16:20 |
| pedroalvarez | worked on them this weekend, and I don't know when I'll have time to continue, that's why I sent it as it is | 16:23 |
| *** gtristan has joined #baserock | 16:34 | |
| edcragg | it appears ybd doesn't keep cache keys in its metadata format... anyone know otherwise? | 16:51 |
| paulsherwood | edcragg: what do you mean? | 16:53 |
| paulsherwood | iirc the meta file does not contain the cache-key, if that's your question | 16:54 |
| edcragg | paulsherwood: morph keeps record of the cache key in .meta files, i can't see any evidence of this in ybd (at least in built system images) | 16:54 |
| edcragg | yes, i think that's what i mean | 16:54 |
| paulsherwood | i think you're correct | 16:54 |
| paulsherwood | is there a benefit to including it? | 16:55 |
| *** ctbruce has quit IRC | 16:55 | |
| rdale | as the meta file is in the artifact's cached entry, you can't find it unless you know the cache key | 16:56 |
| richard_maw | or you could be scanning through the contents of /baserock | 16:56 |
| ssam2 | I like that Morph stores every factor that went into the cache key in the metadata | 16:57 |
| ssam2 | it means that there's a way to reconstruct the cache key from the contents of the file | 16:57 |
| ssam2 | which gives a way to validate it | 16:58 |
| ssam2 | i've never actually used it for that, but i like that it's there | 16:58 |
| richard_maw | also that for some systems you can get to something resembling the system you have by walking the metadata to get your build graph and run the commands yourself | 16:58 |
| edcragg | paulsherwood: it was previously used by the genivi manifest generation script, and included in the manifest output | 16:59 |
| edcragg | the more information the better, i think, for reproducibility | 16:59 |
| paulsherwood | https://github.com/devcurmudgeon/ybd/issues/202 | 17:00 |
| ssam2 | it does lead to long metadata files, because you have to list all deps | 17:00 |
| paulsherwood | patches welcome, or i'll get to it when i return from holiday :) | 17:00 |
| ssam2 | but not too long in practice -- morph's are not crazily long | 17:00 |
| *** bashrc_ has quit IRC | 17:01 | |
| *** jonathanmaw has quit IRC | 17:01 | |
| edcragg | the longest thing in most seems to be the list of files afaict | 17:02 |
| ssam2 | oh yeah, that's long too | 17:03 |
| edcragg | or at least, can be | 17:03 |
| ssam2 | but it's essential for providing 'what chunk contains this file' info | 17:03 |
| rdale | ybd meta files have the list of files for a chunk, or the list of components for a stratum and the git sha1s - i would expect that to be enough for ybd to construct the cache keys, but i might be wrong | 17:03 |
| ssam2 | oh, nice | 17:03 |
| *** toscalix has joined #baserock | 17:05 | |
| * edcragg assumed cache keys must depend on build commands too, but may very well be wrong | 17:05 | |
| rdale | ah, yes - it isn't just the git sha1s on the chunks, it is a reference the the sha1 of the definition perhaps that you need as well | 17:07 |
| *** ssam2 has quit IRC | 17:07 | |
| *** tiagogomes has quit IRC | 17:25 | |
| *** anahuelamo has quit IRC | 17:26 | |
| *** locallycompact has quit IRC | 17:34 | |
| *** toscalix has quit IRC | 17:43 | |
| *** edcragg has quit IRC | 17:51 | |
| *** rdale has quit IRC | 17:53 | |
| *** locallycompact has joined #baserock | 18:59 | |
| *** locallycompact has quit IRC | 19:39 | |
| *** locallycompact has joined #baserock | 20:11 | |
| *** gtristan has quit IRC | 21:48 | |
| *** cosm has joined #baserock | 21:56 | |
| *** edcragg has joined #baserock | 22:14 | |
| *** edcragg has quit IRC | 22:44 | |
| *** locallycompact has quit IRC | 23:01 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!