| *** bochecha has quit IRC | 00:01 | |
| *** xjuan has quit IRC | 00:41 | |
| *** Prince781 has quit IRC | 01:47 | |
| *** Prince781 has joined #buildstream | 02:27 | |
| *** raoul has joined #buildstream | 03:59 | |
| *** coldtom has joined #buildstream | 04:56 | |
| *** Prince781 has quit IRC | 05:17 | |
| *** mohan43u has quit IRC | 07:05 | |
| *** mohan43u has joined #buildstream | 07:05 | |
| *** adds68 has joined #buildstream | 07:51 | |
| *** finn has joined #buildstream | 07:57 | |
| *** Nexus has joined #buildstream | 08:13 | |
| *** jennis has joined #buildstream | 08:24 | |
| *** paulsherwood has joined #buildstream | 08:31 | |
| gitlab-br-bot | phildawson closed MR !887 (aevri/prompt-config->master: Add 'prompt.*' config options to buildstream.conf) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/887 | 08:40 |
|---|---|---|
| gitlab-br-bot | phildawson reopened MR !887 (aevri/prompt-config->master: Add 'prompt.*' config options to buildstream.conf) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/887 | 08:40 |
| *** ikerperez has joined #buildstream | 08:52 | |
| *** jmac has joined #buildstream | 09:00 | |
| *** alatiera_ has joined #buildstream | 09:02 | |
| *** jonathanmaw has joined #buildstream | 09:14 | |
| gitlab-br-bot | jmacarthur approved MR !889 (valentindavid/ruamel_warnings->master: Fix some ruamel warnings during tests) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/889 | 09:19 |
| *** Nexus_ has joined #buildstream | 09:28 | |
| *** Nexus_ has joined #buildstream | 09:28 | |
| *** lachlan has joined #buildstream | 09:29 | |
| jennis | Where does our file that keeps track of the cache size get stored? | 09:30 |
| jennis | I'm guessing it's in ~/.cache/buildstream? | 09:30 |
| gitlab-br-bot | aevri opened MR !893 (aevri/unknown-uri-452->master: _downloadablefilesource: handle ValueError-s) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/893 | 09:31 |
| jennis | Ahh, ~/.cache/buildstream/artifacts | 09:31 |
| *** Nexus__ has joined #buildstream | 09:33 | |
| *** Nexus has quit IRC | 09:36 | |
| *** Nexus_ has quit IRC | 09:36 | |
| *** Nexus has joined #buildstream | 09:36 | |
| *** lachlan has quit IRC | 09:41 | |
| *** lachlan has joined #buildstream | 09:45 | |
| *** lachlan has quit IRC | 10:01 | |
| raoul | When terminating jobs I always get something along the lines of "Unknown child process pid 20093, will report returncode 255" which seems to be coming from the async library, this isn't supposed to be happening right? | 10:10 |
| raoul | Thought it was a change I made but it's happening on master | 10:11 |
| *** lachlan has joined #buildstream | 10:17 | |
| *** coldtom has quit IRC | 10:18 | |
| *** adds68 has quit IRC | 10:18 | |
| *** jennis has quit IRC | 10:18 | |
| *** ikerperez has quit IRC | 10:18 | |
| *** coldtom has joined #buildstream | 10:19 | |
| *** ikerperez has joined #buildstream | 10:21 | |
| *** jennis has joined #buildstream | 10:22 | |
| *** adds68 has joined #buildstream | 10:22 | |
| *** lachlan has quit IRC | 10:25 | |
| *** Prince781 has joined #buildstream | 10:29 | |
| *** Prince781 has quit IRC | 10:31 | |
| tlater[m] | raoul: This has been happening since my scheduler refactor, I believe. tristan thought he'd fixed it in the past... In any case, it's definitely not supposed to happen, and probably happens because BuildStream tries to kill a child process twice. | 10:35 |
| tlater[m] | Hm, I should really just make that the next issue I look at. | 10:35 |
| raoul | Yeah I saw some closed off issues, not sure if it's causing any issues but probably not good nonetheless | 10:37 |
| *** lachlan has joined #buildstream | 10:42 | |
| Kinnison | Hi all, I've got a fairly large MR in terms of commit count (!892) which is almost entirely lint cleanups. tlater[m] has already been over it, and I'd appreciate at least one more pair of eyes before I do as Tristan asked and rebase it into a smaller number of commits ready for merge. | 10:59 |
| Kinnison | Very few if any of the changes should be controversial | 10:59 |
| tlater[m] | Kinnison: As an aside, I think I would prefer a commit for every 'type', but that can be really annoying to orchestrate. file-granularity makes little sense, since the individual changes happen across files. Maybe one big commit is the way to go, except for certain kinds of errors that you think should be considered separate? | 11:01 |
| Kinnison | tlater[m]: Hopefully whoever can spare the time to go over the MR will also weigh in, I'm fine with any combination of rebase from "none" to "one giant commit of doom" but I'd be most comfortable with something along the lines of per-file, with the potential bug-fix split out | 11:02 |
| Kinnison | Though there would be something quite satisfying about merging a 43 commit sequence | 11:02 |
| Kinnison | Once this is merged, we can look to update pylint on our CI docker images | 11:03 |
| Kinnison | which will be nice | 11:03 |
| Kinnison | since the newer pylint definitely spotted a bug when I was doing this work | 11:03 |
| tlater[m] | Kinnison: Hah, well, I wouldn't be too opposed to the full 43 commit sequence. | 11:05 |
| * tlater[m] is really looking forward to this, his linter has been threatening him | 11:06 | |
| *** Prince781 has joined #buildstream | 11:17 | |
| juergbi | I'd definitely keep at least the actual bug fixes separate commits | 11:30 |
| juergbi | I like grouping by type as well, e.g., 'Remove unused imports' across the tree. however, I'd also be fine with other groupings or even no grouping, as long as the bug fixes are kept separate | 11:38 |
| Kinnison | juergbi: noted. | 12:04 |
| *** bochecha has joined #buildstream | 12:07 | |
| Kinnison | Oh for pity's sake | 12:13 |
| * Kinnison grumps at pylint | 12:14 | |
| tlater[m] | Haha | 12:25 |
| tlater[m] | #disable-line-too-long makes the line too long | 12:25 |
| tlater[m] | Adorable | 12:25 |
| *** tlaxkit has joined #buildstream | 12:58 | |
| * Kinnison wonders if juergbi or phildawson has had a chance to actually look through the changes | 13:11 | |
| * Kinnison trusts tlater[m] but would feel more comfortable with at least one other pair of eyes | 13:11 | |
| Kinnison | Then I'll make a decision on the commit sequence and whether I cba. to squish things | 13:11 |
| Kinnison | tlater[m]: Are you familiar with how gitlab.io instances are managed? | 13:12 |
| phildawson | Kinnison, not yet. I've got quarter of an hour before the next meeting so will have a quick look now :) | 13:13 |
| tlater[m] | Kinnison: Not particularly. jjardon set up the magic, I believe | 13:29 |
| *** lachlan has quit IRC | 13:33 | |
| *** lachlan has joined #buildstream | 13:34 | |
| *** catonano has joined #buildstream | 13:39 | |
| *** anahuelamo has quit IRC | 13:40 | |
| Kinnison | okies | 13:40 |
| Kinnison | lachlan: You may want to prod jjardon | 13:40 |
| *** anahuelamo has joined #buildstream | 13:41 | |
| lachlan | Kinnison: He has very kindly already provided some information to follow-up on. | 13:46 |
| Kinnison | cool | 13:47 |
| tlater[m] | Would someone mind taking a look at https://gitlab.com/BuildStream/buildstream/merge_requests/670 ? | 13:49 |
| tlater[m] | It's been reviewed a few times by tristan at this point, but I'd like to get someone to say it looks good before I prod him again ;) | 13:49 |
| Kinnison | tlater[m]: I can look, but I won't feel confident okaying it | 13:50 |
| tlater[m] | That's alright, I just want a "yup, this doesn't look horrifying" | 13:50 |
| * Kinnison skims | 13:51 | |
| Kinnison | tlater[m]: I don't see anything awful, and the general refactor looks nice. But I can't say if it's good or not | 14:00 |
| tlater[m] | Yes, that's fair | 14:01 |
| tlater[m] | There's also one little behavior detail that I want to discuss with tristan anyway | 14:01 |
| tlater[m] | Thanks for looking at it though :) | 14:01 |
| tlater[m] | Ah, tristan is tristan_ atm | 14:03 |
| tlater[m] | tristan: | 14:04 |
| tlater[m] | Hm, maybe neither | 14:04 |
| tlater[m] | Eh, I'll find him some other day. | 14:04 |
| Kinnison | I think he's at ELCE so probably not responding much | 14:04 |
| tlater[m] | Oh, right, I keep forgetting about ELCE | 14:05 |
| *** catonano has quit IRC | 14:27 | |
| cs-shadow | juergbi: hey! Would you consider taking a look at https://gitlab.com/BuildStream/buildstream/merge_requests/820 (bst source-checkout) when you get a chance | 14:31 |
| juergbi | sure | 14:32 |
| cs-shadow | thanks :) | 14:32 |
| Kinnison | juergbi: I've responded to your comments on my linting MR too when you have a moment :-D | 14:35 |
| *** catonano has joined #buildstream | 14:38 | |
| *** benschubert has joined #buildstream | 14:48 | |
| *** benschubert has joined #buildstream | 14:49 | |
| benschubert | Hey, I'd like to add a test to make sure that ensure we are not caching failures that are due to host tools. Where would be the best place to put the test? | 14:50 |
| Kinnison | If you actually need to run a proper sandbox then you'll need to put the test somewhere in tests/integration IIRC | 14:51 |
| benschubert | integration/cachedfail? | 14:52 |
| Kinnison | that sounds like a reasonable starting point at least | 14:52 |
| tpollard | +1 | 14:53 |
| benschubert | thanks I'll go for this! | 14:53 |
| *** WSalmon has joined #buildstream | 14:58 | |
| *** WSalmon_ has quit IRC | 14:58 | |
| *** lachlan has quit IRC | 15:05 | |
| gitlab-br-bot | juergbi opened issue #724 (Follow-up from "Fix lint issues": Avoid the need to disable `assignment-from-no-return` for abstract methods) on buildstream https://gitlab.com/BuildStream/buildstream/issues/724 | 15:10 |
| juergbi | Kinnison: commented | 15:13 |
| *** Prince781 has quit IRC | 15:23 | |
| *** catonano has quit IRC | 15:26 | |
| *** lachlan has joined #buildstream | 15:26 | |
| *** lachlan has quit IRC | 15:35 | |
| *** lachlan has joined #buildstream | 15:36 | |
| gitlab-br-bot | raoul.hidalgocharman opened issue #725 (Job cancellation on remote builds) on buildstream https://gitlab.com/BuildStream/buildstream/issues/725 | 15:41 |
| Kinnison | tlater[m]: Do you happen to know what triggers the new linting? pylint is locked at 2.1.1 which is what I have, and yet I get the lints where the current CI doesn't | 15:46 |
| raoul | juergbi, could I get dev access so I can assign an issue to myself? | 15:48 |
| tlater[m] | Kinnison: that's interesting. How do you run pylint? Through setup.py or manually? | 15:49 |
| Kinnison | setup.py | 15:49 |
| Kinnison | and I'm talking about what's in my .eggs/ | 15:49 |
| tlater[m] | It's possible setup.py doesn't actually run what's in there, but knowing you that's probably the only pylint on your system. | 15:50 |
| tlater[m] | My only suspicion is that you use he -m flag and that that picks up different modules than the default configuration. | 15:51 |
| tlater[m] | Have you tried updating the pylint version the CI uses, just in case? | 15:51 |
| Kinnison | I've not touched the CI | 15:51 |
| Kinnison | I was getting these lint warnings regardless of using "-m pylint" | 15:52 |
| tlater[m] | Hm, maybe we tagged the wrong version, but assert() was definitely part of the very first pylint version we used. I did find that odd. | 15:52 |
| tlater[m] | Let me have a look at the config... | 15:53 |
| * Kinnison thinks that when jonathanmaw tried to regenerate the CI worker images he encountered the lint issues | 15:55 | |
| Kinnison | jonathanmaw: ^^ is that true? | 15:55 |
| jonathanmaw | Kinnison: yep | 15:55 |
| Kinnison | So perhaps I don't need to touch dev-requirements.txt ? | 15:56 |
| Kinnison | jonathanmaw: related discussion: https://gitlab.com/BuildStream/buildstream/merge_requests/892#note_111509504 | 15:56 |
| jonathanmaw | Kinnison: linting runs at https://gitlab.com/BuildStream/buildstream/-/jobs/111173417 | 15:56 |
| Kinnison | That definitely looks like the pile'o'lints my MR fixes | 15:56 |
| juergbi | raoul: done | 15:57 |
| raoul | cheers juergbi | 15:57 |
| Kinnison | juergbi: Given ^^^ do you think I need to adjust dev-requirements.txt still? | 15:57 |
| jonathanmaw | Kinnison: there's a dev-requirements in the root of buildstream, which is used if you install by setup.py, and the CI workers have a dev-requirements which is used to populate the CI workers | 15:57 |
| Kinnison | Oh right | 15:57 |
| Kinnison | So juergbi might have meant the other dev-requirements.txt ? | 15:58 |
| * Kinnison is very confused | 15:58 | |
| juergbi | I meant both. aren't they kept in sync? | 15:59 |
| * Kinnison has literally no idea :( | 15:59 | |
| jonathanmaw | juergbi: they seem to have the same contents, but in a slightly different order | 16:00 |
| juergbi | Kinnison: my main point was that CI was apparently not catching these issues before. and whatever needs to be done to change this, should be in the branch (or in corresponding docker image branch) | 16:00 |
| Kinnison | juergbi: afaict what needs to be done is that the CI images need regenerating | 16:00 |
| Kinnison | juergbi: i.e. that if we regenerate them, these lints will hit us | 16:00 |
| tlater[m] | That can't be done in-branch | 16:00 |
| tlater[m] | Yes, that's annoying. | 16:01 |
| juergbi | well, you can generate new CI images and then change gitlab-ci in the main MR | 16:01 |
| juergbi | to use the new images | 16:01 |
| tlater[m] | Side effect of hosting the code separately from the testing environment. | 16:01 |
| tlater[m] | juergbi: The issue is that the image CI fails because it now suddenly picks up new errors | 16:01 |
| * Kinnison would definitely need help with that | 16:01 | |
| tlater[m] | So they can't update the images | 16:02 |
| Kinnison | Urgh | 16:02 |
| juergbi | ah, that was added, hm | 16:02 |
| tlater[m] | Iirc we'll first need to land the changes in master and then update the images. | 16:02 |
| juergbi | sure, not too much of an issue. let's just not forget it | 16:03 |
| juergbi | i.e., do it immediately | 16:03 |
| Kinnison | Would someone who knows what to do be prepared to merge the lint fixes MR and then immediately hit the image update? | 16:03 |
| tlater[m] | Sure - it seems jonathanmaw is already on it? | 16:04 |
| Kinnison | jonathanmaw: ^^ ? | 16:04 |
| tlater[m] | Otherwise I'm happy to | 16:04 |
| jonathanmaw | tlater[m]: I'm not currently working on that | 16:04 |
| jonathanmaw | I updated the image as part of adding pexpect (though it looks like I won't be using pexpect after all) | 16:05 |
| tlater[m] | Then I'll prepare the images :) | 16:05 |
| gitlab-br-bot | valentindavid opened MR !894 (valentindavid/sysroot_dependencies->master: Add support for importing build dependencies in sysroots.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/894 | 16:08 |
| tlater[m] | jonathanmaw: Ah, so you just updated the image ref, we already have a CI image that recognizes the new errors. | 16:10 |
| jonathanmaw | tlater[m]: yep, I needed a new image ref that included pexpect, and that's when I stumbled over a huge pile of linting errors | 16:11 |
| tlater[m] | Cool, well, that should be fixed with this branch. I'll just bump the version. | 16:12 |
| *** abderrahim3 has joined #buildstream | 16:14 | |
| *** abderrahim2 has quit IRC | 16:16 | |
| *** lachlan has quit IRC | 16:17 | |
| tlater[m] | Err, it looks like my push is not appearing in gitlab? | 16:17 |
| *** tlaxkit has quit IRC | 16:21 | |
| *** lachlan has joined #buildstream | 16:23 | |
| tlater[m] | Hm, no, just held back because CI takes a while. | 16:25 |
| laurence | jonathanclaw, jennis, I cannot remember, pls remind me - did we ever improve the filter documentation ? if so, can we close https://gitlab.com/BuildStream/buildstream/issues/278 ? | 16:27 |
| laurence | jonathanmaw, ^^ | 16:27 |
| jennis | I think phildawson ended up writing that | 16:29 |
| jonathanmaw | laurence: A MR was merged for it which improved the docstring | 16:30 |
| jonathanmaw | I think I paraphrased heavily what tristan outlined in the issue, so I wasn't confident to close the MR | 16:31 |
| jonathanmaw | I think it's been a while since people have been perplexed by filter elements at us, so maybe it's fine to close it now | 16:33 |
| *** lachlan has quit IRC | 16:33 | |
| phildawson | laurence, I started writing a filter example a while back but couldn't get it working at the time (people had their hands full at the time so I was struggling to get help to work it out). I think I've still got it sat on a branch somewhere. | 16:34 |
| laurence | jonathanmaw, thanks | 16:39 |
| laurence | phildawson, ah ok, i think that falls under examples in general rather than this specific issue | 16:39 |
| laurence | jonathanmaw, I suggest we close it for now and link to the MR tht was merged | 16:39 |
| jonathanmaw | laurence: it's already linked to the MR that was merged | 16:41 |
| jonathanmaw | ah, it was reopened as being related to https://gitlab.com/BuildStream/buildstream/issues/437 by the looks of it | 16:43 |
| laurence | dear me, right in that case sorry for the noise all | 16:43 |
| *** lachlan has joined #buildstream | 16:46 | |
| Kinnison | tlater[m]: Why is my MR stuck on pipeline pending again? | 16:57 |
| Kinnison | tlater[m]: oh no matter, I see your commit \o/ | 16:58 |
| * Kinnison decides to have some dinner | 16:58 | |
| tlater[m] | I had to add a commit for the image update. | 16:58 |
| tlater[m] | Now we need to wait for others' CI to finish... | 16:58 |
| Kinnison | :-D | 16:58 |
| tlater[m] | :) | 16:59 |
| Kinnison | Definitely dinner'o'clock then | 16:59 |
| * Kinnison waves | 16:59 | |
| Kinnison | I'll try and check in later | 16:59 |
| tlater[m] | o/ | 16:59 |
| gitlab-br-bot | aevri opened issue #726 ('Are you sure?' prompts don't suite all users) on buildstream https://gitlab.com/BuildStream/buildstream/issues/726 | 17:00 |
| *** lachlan has quit IRC | 17:05 | |
| *** lachlan has joined #buildstream | 17:10 | |
| laurence | WSalmon, is there a gitlab ticket raised for Workspaces - defaults? | 17:17 |
| *** lachlan has quit IRC | 17:25 | |
| *** raoul has quit IRC | 17:42 | |
| gitlab-br-bot | BenjaminSchubert opened MR !895 (bschubert/dont-cache-errors-from-host-tools->master: WIP: Don't cache host tools failures) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/895 | 17:47 |
| *** lachlan has joined #buildstream | 17:48 | |
| *** catonano has joined #buildstream | 17:50 | |
| gitlab-br-bot | BenjaminSchubert opened issue #727 (Don't cache host tools failures) on buildstream https://gitlab.com/BuildStream/buildstream/issues/727 | 17:50 |
| benschubert | Could I get your feedback on https://gitlab.com/BuildStream/buildstream/issues/727 ? Thanks! | 17:51 |
| benschubert | let me know if you want me to pass this to the ML | 17:51 |
| *** jonathanmaw has quit IRC | 17:59 | |
| *** finn has quit IRC | 18:04 | |
| *** finn has joined #buildstream | 18:06 | |
| *** finn has quit IRC | 18:09 | |
| *** xjuan has joined #buildstream | 18:26 | |
| *** lachlan has quit IRC | 18:34 | |
| *** xjuan has quit IRC | 19:00 | |
| *** alatiera_ has quit IRC | 19:04 | |
| *** mohan43u has quit IRC | 19:10 | |
| *** mohan43u has joined #buildstream | 19:13 | |
| *** catonano has quit IRC | 20:27 | |
| *** Prince781 has joined #buildstream | 20:42 | |
| *** catonano has joined #buildstream | 21:06 | |
| *** xjuan has joined #buildstream | 21:35 | |
| *** mohan43u has quit IRC | 21:40 | |
| *** mohan43u has joined #buildstream | 21:43 | |
| *** catonano has quit IRC | 21:58 | |
| *** Prince781 has quit IRC | 22:04 | |
| *** ssssam[m] has quit IRC | 22:44 | |
| gitlab-br-bot | cs-shadow opened MR !896 (chandan/source-doc-directory->master: plugins: Don't redundantly document common configurations) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/896 | 22:49 |
| *** alatiera has quit IRC | 22:59 | |
| *** dineshdb[m] has quit IRC | 22:59 | |
| *** inigomartinez has quit IRC | 22:59 | |
| gitlab-br-bot | cs-shadow approved MR !893 (aevri/unknown-uri-452->master: _downloadablefilesource: handle ValueError-s) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/893 | 23:07 |
| *** ssssam[m] has joined #buildstream | 23:18 | |
| *** Prince781 has joined #buildstream | 23:25 | |
| *** bochecha has quit IRC | 23:38 | |
| *** alatiera has joined #buildstream | 23:41 | |
| *** dineshdb[m] has joined #buildstream | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!