| *** xjuan has quit IRC | 02:14 | |
| *** catonano has joined #buildstream | 02:57 | |
| *** catonano has quit IRC | 03:29 | |
| *** catonano has joined #buildstream | 05:28 | |
| *** alatiera_ has joined #buildstream | 05:57 | |
| *** ahmed89 has quit IRC | 06:07 | |
| *** ahmed89 has joined #buildstream | 06:08 | |
| gitlab-br-bot | buildstream: merge request (juerg/dummy-sandbox->master: _platform/linux.py: Accept all configs for dummy sandbox) #843 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/843 | 06:27 |
|---|---|---|
| gitlab-br-bot | buildstream: merge request (juerg/cas-batch-1.2->bst-1.2: CAS: Implement BatchUpdateBlobs support) #844 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/844 | 06:31 |
| gitlab-br-bot | buildstream: merge request (juerg/sandbox->master: sandbox/_sandboxremote.py: Use helpers to set up cwd and env) #841 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/841 | 06:34 |
| *** ahmed89 has quit IRC | 07:02 | |
| *** ahmed89 has joined #buildstream | 07:02 | |
| *** catonano has quit IRC | 07:02 | |
| *** catonano has joined #buildstream | 07:07 | |
| gitlab-br-bot | buildstream: merge request (juerg/sandbox->master: sandbox/_sandboxremote.py: Use helpers to set up cwd and env) #841 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/841 | 07:08 |
| *** toscalix has joined #buildstream | 07:40 | |
| valentind | The gnome wiki page for the buildstream events is locked. | 08:36 |
| valentind | How can I edit it? | 08:36 |
| *** jonathanmaw has joined #buildstream | 08:47 | |
| juergbi | valentind: I guess you're not in https://wiki.gnome.org/TrustedEditorGroup yet. what's your wiki username? | 08:49 |
| valentind | juergbi, ValentinDavid | 08:50 |
| juergbi | valentind: added, should work now | 08:50 |
| valentind | juergbi, thank you | 08:50 |
| tiagogomes | Why GitLab devs thought it would be better to show next to "Changes" the number of files changed, instead of the number of lines changed | 09:06 |
| *** abderrahim has quit IRC | 09:07 | |
| *** abderrahim has joined #buildstream | 09:08 | |
| Kinnison | IIRC it's the number of diff hunks | 09:08 |
| Kinnison | no? | 09:08 |
| tiagogomes | The "7 changed files" here https://gitlab.com/BuildStream/buildstream/merge_requests/836/diffs leads me to believe it is not | 09:16 |
| Kinnison | boo :( | 09:16 |
| Kinnison | You are quite correct | 09:16 |
| *** tristan has joined #buildstream | 09:16 | |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 09:20 |
| *** lachlan has joined #buildstream | 09:28 | |
| *** bochecha has joined #buildstream | 09:30 | |
| tpollard | is anyone able to run 'test_invalid_cache_quota' from artifactcache/expiry.py locally without it failing? | 09:44 |
| tiagogomes | tpollard master? | 09:45 |
| tpollard | I'm running the tests with --addopts --integration, yet afaict the test isn't run in the CI at all | 09:45 |
| tpollard | yep | 09:45 |
| tiagogomes | it works for me | 09:47 |
| tpollard | sorry it is triggered in CI | 09:47 |
| tpollard | mine is claiming that 50% is not a valid cache quota | 09:47 |
| tpollard | 'Error instantiating artifact cache: Your system does not have enough available space to support the cache quota specified' with 210G available.... | 09:48 |
| tiagogomes | tpollard yeah, you will have to free some space on your test machine ¬_¬ | 09:48 |
| tiagogomes | https://gitlab.com/BuildStream/buildstream/issues/530 | 09:49 |
| tpollard | phew, not just me then | 09:49 |
| *** ChanServ sets mode: +o tristan | 09:50 | |
| tristan | tpollard, you need an env var | 09:50 |
| tristan | tpollard, if I recall, I only get that error when I run a bst invocation which ran in a test to test it out myself | 09:51 |
| tristan | and running ./setup.py test will automatically set the BST_TEST_SUITE env var | 09:52 |
| tristan | Or, maybe I'm thinking of a different check | 09:52 |
| tristan | right | 09:53 |
| tpollard | I can ignore it locally, just wanted to confirm I wasn't needing to go down a rabbit hole (especially as my changes are not in its scope) | 09:54 |
| tristan | Commented on the issue just now | 09:59 |
| tristan | tpollard, If I recall correctly, we use some big numbers for the cache expiry tests, however they are just numbers really - no reason why the same tests shouldnt work if we just run them with a very small cache quota | 10:01 |
| tpollard | 50 & 200% currently | 10:02 |
| *** solid_black has joined #buildstream | 10:03 | |
| tristan | Ahhhh, it's testing 50% and 200% | 10:03 |
| tpollard | and pony :D | 10:03 |
| tristan | I see, that is tricky, all of this to ensure that we parse the % properly ? | 10:03 |
| tiagogomes | So can also create a loopbacked filesystem and point the cache location to there :) But that might be considered over-engineering | 10:07 |
| Kinnison | tristan, juergbi: I was looking at what it'd take to get the Linux platform's sandbox code to cleanly report if bwrap is too old (i.e. not at least 0.1.2) and I am confused by _local_sandbox_available() | 10:15 |
| Kinnison | it seems redundant to call os.path.exists() around a utils.get_host_tool() and also would it not be of more use to call _site.check_bwrap_version(0, 1, 2) instead? | 10:15 |
| juergbi | Kinnison: yes, that should be used there | 10:16 |
| tristan | Kinnison, I think the main check is done at installation time | 10:16 |
| Kinnison | tristan: Yeah, but that's a problem for Linux systems which will only be used for remote execution | 10:16 |
| Kinnison | tristan: I have a use-case where a Linux system doesn't have bwrap or FUSE, but can contact a RE service | 10:17 |
| * Kinnison was hoping to prepare an MR which removed the install-time check for bwrap, but only if there's a good runtime check which gives the user useful feedback | 10:17 | |
| tristan | Right, so all of that will become optional | 10:17 |
| juergbi | yes, we most likely need to improve the user feedback on the runtime check | 10:17 |
| juergbi | providing details why the local sandbox is not available | 10:18 |
| juergbi | otherwise it will get confusing for the common case of local builds | 10:18 |
| Kinnison | Perhaps SandboxDummy should be constructed with a reason why it was selected | 10:18 |
| Kinnison | Then if it tries to .run() it'd have a good message for the exception | 10:19 |
| juergbi | could be an option | 10:19 |
| tristan | tiagogomes, I think a fuse mount would be good for those tests really; it's hard to test without a controlled, sized volume | 10:19 |
| tristan | tiagogomes, And really, monkey patching system calls to work around that and "fake it" is error prone anyway | 10:19 |
| * tiagogomes nods | 10:20 | |
| tristan | I would steer away from a loopback cause I wouldnt want to require root for running those tests (and we require fuse anyway) | 10:20 |
| tiagogomes | mount on fedora has the setuid bit, but I guess that's not the same for other distros | 10:21 |
| juergbi | tiagogomes: that's common but that doesn't allow loop mount, afaik | 10:22 |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: Out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 10:22 |
| juergbi | (setuid mount doesn't allow arbitrary mounts) | 10:22 |
| Kinnison | setuid mount is for entries in fstab which have the 'user' option listed | 10:22 |
| Kinnison | fusermount exists explicitly to circumvent this for fuse filesystems | 10:23 |
| tiagogomes | aah "mount: only root can use "--options" option" | 10:24 |
| tristan | And this is why we use `mkfs -d` when generating system images :) | 10:24 |
| gitlab-br-bot | buildstream: merge request (jonathan/pickle-yaml->master: WIP: Add a cache of parsed and provenanced yaml) #787 changed state ("closed"): https://gitlab.com/BuildStream/buildstream/merge_requests/787 | 10:33 |
| *** WSalmon_ has joined #buildstream | 10:38 | |
| *** WSalmon has quit IRC | 10:39 | |
| tristan | juergbi, So mr !844 does the client side of batch blobs correct ? | 10:47 |
| juergbi | tristan: that's both client and server side, but for uploads / artifact push | 10:48 |
| juergbi | only recently implemented in master | 10:48 |
| tristan | juergbi, I'm curious why it modifies _artifactcache/casserver.py... oh you beat me to it | 10:48 |
| tristan | I see | 10:48 |
| tristan | juergbi, Did we merge the client side for pull ? | 10:48 |
| juergbi | batch download support is already in both master and 1.2 | 10:48 |
| juergbi | 1.2.1/1.2.2 has the server side but not the client side yet | 10:49 |
| tristan | Ok, that's what I wanted to know :) | 10:49 |
| tristan | Nice | 10:49 |
| juergbi | abderrahim did some testing and was seeing large speedups | 10:49 |
| juergbi | fdo cache server already is on 1.2.1, iirc | 10:49 |
| tristan | great, I will try it :) | 10:50 |
| tpollard | I've not taken metrics but my scenario seems faster too | 10:50 |
| juergbi | <abderrahim> 5 min vs 50+min | 10:51 |
| juergbi | <abderrahim> (for bst build with everything in the cache) | 10:51 |
| juergbi | <abderrahim> the above is for the flatpak runtimes, for the full gnome I get 13min vs 1h50m | 10:51 |
| tristan | Sweet | 10:51 |
| coldtom | valentind: upstreaming the fdo strip commands fails CI because the tests expect the debug symbols to be included, but the new commands remove them entirely | 10:52 |
| valentind | Why does it remove them? | 10:52 |
| valentind | Are you sure it is just not a different path? | 10:52 |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 10:53 |
| coldtom | valentind: looks like it should be the right directory :/ | 11:06 |
| valentind | coldtom, I am running it locally to understand what is happening. | 11:07 |
| jjardon | Hi, now that buildstream is in https://pypi.org/project/BuildStream/, do you plan to upload bst-external there as well? | 11:08 |
| jjardon | juergbi: fdo is updated, correct | 11:09 |
| lachlan | tristan: I have updated !836 with my thoughts - will rework asap | 11:11 |
| *** lachlan has quit IRC | 11:14 | |
| tristan | jjardon, I hadn't considered it really, but I don't see an issue with that; unlike distro packaging, the expectations for API stability are different on PyPI | 11:15 |
| tristan | One can always install the exact version they want from PyPI, too | 11:16 |
| jjardon | tristan: BTW, there were some discussion about this; what is your vision about bst-external? place for plugins that have to mature to be included in buildstream repo or actually they will always be there? People seems to think is the former | 11:25 |
| tiagogomes | I would prefer the latter even if complicates a bit using buildstream | 11:26 |
| tiagogomes | I don't like the package name though | 11:26 |
| jjardon | OK, wherever is It would be good It's documented somewhere | 11:26 |
| tristan | jjardon, The former (almost), maybe not well named for communicating that, I might have preferred bst-experimental | 11:34 |
| tristan | jjardon, However, when I say (almost), what I mean is that it is a place for plugins to mature, not necessarily to land in BuildStream | 11:34 |
| tristan | jjardon, I.e. I would expect that more fringe plugins will mature and graduate to a more stable repo | 11:35 |
| tristan | Some of those plugins will probably land in core BuildStream, some of them might land elsewhere in more use-case specific repos | 11:36 |
| tristan | Should probably be documented in the bst-external repo's README | 11:36 |
| juergbi | staging might be a suitable name | 11:37 |
| tristan | jjardon, I should note too that it is also a place for us to get our plugin story straight and improve on that :D | 11:38 |
| valentind | coldtom, "sh: read: line 34: illegal option -d" | 11:45 |
| valentind | coldtom, I suppose alpine uses busybox | 11:45 |
| tristan | valentind, Yes, that is true | 11:47 |
| tristan | sounds familiar | 11:47 |
| valentind | read command is not documented in busybox. | 11:51 |
| tristan | valentind, frankly, we might end up removing the default strip commands from BuildStream proper some day | 11:55 |
| *** ahmed89 has quit IRC | 11:56 | |
| tristan | It would be a breaking change, but it's very OS/Platform specific, as such doesnt really belong in the defaults | 11:56 |
| valentind | tristan, I agree. | 11:56 |
| valentind | Though we should keep some recipes somewhere. | 11:56 |
| valentind | It is difficult to make it correct. | 11:56 |
| tristan | yes | 11:56 |
| coldtom | it could be kept as an include somewhere | 11:56 |
| tristan | maybe it can be a non-breaking change if we had some kind of platform specific defaults, not sure | 11:57 |
| valentind | tristan, for example in freedesktop sdk we use debugedit to extract necessary source files. | 11:57 |
| valentind | And also change the path. | 11:57 |
| valentind | Also some people might want to use build id for the path of debug. | 11:57 |
| *** catonano has quit IRC | 11:59 | |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: Out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 12:06 |
| *** catonano has joined #buildstream | 12:07 | |
| gitlab-br-bot | buildstream: merge request (tristan/fix-status-messages->master: fix status messages) #845 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/845 | 12:15 |
| gitlab-br-bot | buildstream: merge request (tristan/fix-status-messages-1.2->bst-1.2: fix status messages (1.2)) #846 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/846 | 12:16 |
| * tristan found the source of the obnoxious status messages which came back to haunt us | 12:19 | |
| tristan | git cat-file etc | 12:19 |
| *** mohan43u has quit IRC | 12:20 | |
| tristan | juergbi, llvm6 = 10 min 44 seconds | 12:38 |
| tristan | As opposed to 1.5 hours :D | 12:38 |
| juergbi | :) | 12:39 |
| Kinnison | juergbi, tristan: Following on from our chat this morning, I have a WIP branch at https://gitlab.com/BuildStream/buildstream/tree/danielsilverstone-ct/bwrap-check-runtime-only (not filed as an MR yet, tests underway) | 12:42 |
| juergbi | tristan: do you remember how long this took with OSTree? | 12:42 |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: Out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 12:42 |
| tristan | juergbi, I don't, but I suspect we are now faster | 12:43 |
| tristan | juergbi, I also recall there was an issue with freedesktop-sdk with many concurrent pulls, something to do with too many open connections | 12:43 |
| tristan | That is probably much improved with these changes | 12:44 |
| tristan | We only really optimized push, because ostree-push was severely non performant | 12:44 |
| juergbi | yes, we don't actually do any parallel requests for single pull job, so it should be light on connections | 12:44 |
| Kinnison | juergbi, tristan: If either of you peek at that branch and have ideas for good ways to test the changes, I'd appreciate some hints, I'm a bit lost in the test suite currently | 12:45 |
| juergbi | Kinnison: branch makes sense at first glance. I guess testing is somewhat tricky as this checks host requirements | 12:46 |
| Kinnison | juergbi: yeah, I've been testing locally by temporarily renaming the bwrap binary, or replacing it with a shell script claiming different versions | 12:47 |
| juergbi | yes, manual testing like that might be sufficient for this, imo | 12:48 |
| juergbi | Kinnison: one initial comment, I would still keep _local_sandbox_available() around (could also be a variables instead of a function) to avoid code duplication | 12:49 |
| tristan | Kinnison, juergbi, One approach is to have a separate image for testing a bwrap-less environment | 12:49 |
| juergbi | Kinnison: with https://gitlab.com/BuildStream/buildstream/merge_requests/843 we'd otherwise have three times fuse+bwrap check | 12:49 |
| Kinnison | juergbi: I was thinking at some point that the logic might want to be opened up a little bit (e.g. for fuse/no-bwrap -> SandboxUnix) | 12:50 |
| tristan | Then we mark tests to skip if bwrap exists | 12:50 |
| Kinnison | tristan: Interesting idea, so that'd be another docker container and branch on the .gitlab-ci.yml ? | 12:51 |
| tristan | Kinnison, exactly | 12:51 |
| Kinnison | juergbi: Hmm I see, yes that might be a problem | 12:51 |
| Kinnison | tristan: that'd be well beyond me at the moment to set up I fear :( | 12:51 |
| tristan | Well, it's the way things are intended to work - test things which are testable in the environment | 12:52 |
| tristan | And skip tests which would not pass in the test environment | 12:52 |
| tristan | We have tests/testutils/site.py where we declare HAVE_FOO for a bunch of things | 12:52 |
| juergbi | it should actually not be too bad with regards to CI time given that we'd skip all integration tests | 12:52 |
| Kinnison | I meant the docker stuff, I could probably manage a test of some kind :-) | 12:52 |
| tristan | That is in buildstream-docker-images, which Chandan maintains | 12:53 |
| tristan | So yeah, we would need to spin up a new image for that first | 12:53 |
| juergbi | ideally, we'd even test on WSL but that's probably not straight forward to set up | 12:53 |
| Kinnison | Don't we have an image for testing the SandboxUnix stuff? | 12:53 |
| Kinnison | Or does that still have bwrap on, just with the backend forced? | 12:54 |
| tristan | I don't know, honestly I'm not extremely picky about this case, but setting up that image seems like it will be worthwhile for more tests than just this one | 12:54 |
| juergbi | Kinnison: it uses the same fedora 27 image with bwrap | 12:54 |
| Kinnison | oh :/ | 12:54 |
| tristan | Kinnison, No, we unfortunately use the same image and force the unix backend :( | 12:54 |
| juergbi | but if we create a bwrap-less image, we could use that for unix as well | 12:55 |
| gitlab-br-bot | buildstream: merge request (mablanch/630-remote-execution-reconn->master: Handle connection losses during remote build execution) #806 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/806 | 12:55 |
| Kinnison | juergbi: aye, though it'd still need forcing because otherwise it'll pick the linux platform | 12:55 |
| Kinnison | and the linux platform doesn't do unix sandbox | 12:55 |
| tristan | Yeah, I'd like it if we were testing the unix backend on, say... a bsd machine | 12:55 |
| juergbi | yes, that would be great | 12:55 |
| tristan | It's supposed to be "fallback", we don't really know that it works on non-linux or not | 12:55 |
| persia | I remember someone trying to get it working on BSD: did that effort succeed? | 12:56 |
| tristan | persia, No follow up really, someone provided a first iteration patch | 12:56 |
| tristan | It should be very doable too | 12:56 |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: Out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 13:00 |
| *** catonano has quit IRC | 13:04 | |
| *** catonano has joined #buildstream | 13:05 | |
| tpollard | tristan: would it be possible for you to give a once over on this? There's some comments left dangling https://gitlab.com/BuildStream/buildstream/merge_requests/786/ | 13:05 |
| *** lachlan has joined #buildstream | 13:11 | |
| tristan | tpollard, w00t ! I will - but I don't think I can do a review today of that | 13:11 |
| tpollard | tristan: no worries, there's just a couple of comments opens on some specific points | 13:12 |
| tpollard | mainly the type of configuration it should be set at | 13:12 |
| tpollard | (not at project.conf :)) | 13:13 |
| tristan | yeah I recall that :) | 13:13 |
| tristan | Commented on that :) | 13:18 |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 13:18 |
| gitlab-br-bot | buildstream: merge request (tristan/fix-status-messages-1.2->bst-1.2: fix status messages (1.2)) #846 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/846 | 13:19 |
| tpollard | tristan: cheers! | 13:19 |
| *** lachlan has quit IRC | 13:19 | |
| *** lachlan has joined #buildstream | 13:20 | |
| gitlab-br-bot | buildstream: merge request (tristan/fix-status-messages->master: fix status messages) #845 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/845 | 13:20 |
| tristan | Someone care to take a look at https://gitlab.com/BuildStream/buildstream/merge_requests/845 ? | 13:21 |
| tristan | It adds a very *weird* loop construct, but solves the issue of irrelevant status messages getting printed in the UI, as a result of consuming generator items from Source.get_source_fetchers() | 13:22 |
| tristan | Another thought, is instead of the weird loop, maybe I could subclass `iter` with an object that overrides the __next__ implementation | 13:27 |
| finn | Hey - anyone seen this error before?: File "/home/buildgrid/buildstream/buildstream/_ostree.py", line 66, in ensure | 13:27 |
| finn | repo.reload_config() | 13:27 |
| finn | AttributeError: 'Repo' object has no attribute 'reload_config' | 13:27 |
| *** tristan has quit IRC | 13:33 | |
| finn | I'm getting this when running the flatpak-autotools example | 13:37 |
| finn | on master | 13:39 |
| gitlab-br-bot | buildstream: issue #690 ("flatpak-autotools example broken") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/690 | 13:42 |
| gitlab-br-bot | buildstream: issue #690 ("flatpak-autotools example broken") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/690 | 13:57 |
| *** tristan has joined #buildstream | 13:58 | |
| *** lachlan has quit IRC | 14:12 | |
| *** lachlan has joined #buildstream | 14:17 | |
| gitlab-br-bot | buildstream: merge request (danielsilverstone-ct/bwrap-check-runtime-only->master: WIP: Make bwrap check runtime only) #847 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/847 | 14:27 |
| gitlab-br-bot | buildstream: merge request (danielsilverstone-ct/bwrap-check-runtime-only->master: WIP: Make bwrap check runtime only) #847 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/847 | 14:28 |
| tiagogomes | Something that is missing from the documentation: BuildStream for previous Yocto users | 14:30 |
| tiagogomes | With some pointers for porting | 14:36 |
| tpollard | one of the first questions will be, where is my poky equivalent | 14:50 |
| tiagogomes | hmm, there isn't one, but we can point to freedestop-sdk, gnome-meta and minimal distro projects | 14:53 |
| jonathanmaw | tristan: good news, I've managed to replicate your error case in the wild. Now I need to figure out how it's differing from my test cases. | 14:56 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 15:05 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 15:08 |
| valentind | adds68, Do you know when you will be testing !830 on Freedeesktop SDK infra? | 15:13 |
| adds68 | valentind, i'm currently just rebasing the branch now, i will then run the release and open an MR | 15:13 |
| adds68 | Probably ~1 hour | 15:13 |
| valentind | OK, great. | 15:14 |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 15:14 |
| adds68 | valentind, lots of changes since switching to CAS | 15:14 |
| *** lachlan has quit IRC | 15:15 | |
| gitlab-br-bot | buildstream: merge request (tpollard/494->master: WIP: Don't pull artifact buildtrees by default) #786 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/786 | 15:18 |
| *** bochecha has quit IRC | 15:36 | |
| *** lachlan has joined #buildstream | 15:52 | |
| tiagogomes | hmm, bitbake has download progress bars | 15:59 |
| Nexus | juergbi around? | 15:59 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 16:00 |
| gitlab-br-bot | buildstream: merge request (lachlan/pickle-yaml-test-list-composite->master: Add yaml cache testing to yaml tests) #836 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/836 | 16:00 |
| *** lachlan has quit IRC | 16:15 | |
| *** bochecha has joined #buildstream | 16:15 | |
| juergbi | Nexus: yes | 16:20 |
| Nexus | juergbi: could you have a look at https://gitlab.com/BuildStream/buildstream/merge_requests/848 for me pls? | 16:21 |
| *** lachlan has joined #buildstream | 16:23 | |
| tpollard | will be good to have search working again :) | 16:23 |
| Nexus | +1 | 16:25 |
| Nexus | how do i set up a project with a remote artifact cache in pull only mode? | 16:28 |
| tpollard | if you don't add push:true it won't attempt to push | 16:29 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 16:29 |
| tpollard | or are you asking about running the artifact server itself? | 16:29 |
| Nexus | yup, that one :) | 16:29 |
| tpollard | https://buildstream.gitlab.io/buildstream/install_artifacts.html ? | 16:31 |
| tpollard | just don't include --enable-push | 16:31 |
| juergbi | Nexus: one comment, otherwise looks fine | 16:31 |
| juergbi | Nexus: oh, and let's get used to the filename prefix in the commit summaries. i.e., .gitlab-ci.yml: | 16:32 |
| jonathanmaw | Of course those tests failed! I go down an entirely different code path for the git source! (╯°□°）╯︵ ┻━┻ | 16:35 |
| tiagogomes | I don't use filename prefix, but rather subsystem prefix | 16:35 |
| Nexus | no need to get angry jonathanmaw! ┬──┬︵ſ(°□°ſ) | 16:35 |
| tiagogomes | core, scheduler, artifactcache, ci, docs | 16:35 |
| *** lachlan has quit IRC | 16:35 | |
| tiagogomes | Reason being, sometimes a commit spreads across more than one file if you want it to be in a working state | 16:37 |
| jonathanmaw | I spent ages on this because I failed to notice an obvious fact! (╯°Д°）╯︵/(.□ . \) | 16:37 |
| jonathanmaw | (ranting over, just wanted an excuse to use that emoticon) | 16:37 |
| Nexus | lol | 16:38 |
| tiagogomes | Also the recommended line length for the title in the commit message is too short to write "buildstream/_artifactcache/cascache:" for example | 16:38 |
| Nexus | ノ┬─┬ノ ︵ ( \o°o)\ | 16:38 |
| jonathanmaw | lol | 16:38 |
| gitlab-br-bot | buildstream: merge request (jonathan/source-mirror-project-refs->master: tests: Add regression test for mirroring with project.refs) #823 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/823 | 16:39 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 16:40 |
| *** toscalix has quit IRC | 16:42 | |
| *** lachlan has joined #buildstream | 16:42 | |
| *** phildawson has quit IRC | 16:42 | |
| gitlab-br-bot | buildstream: issue #691 ("BuildStream does not automatically fetch a junction within a junction / displays an inaccurate helper") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/691 | 16:45 |
| *** solid_black has quit IRC | 16:50 | |
| *** bochecha has quit IRC | 16:50 | |
| *** bochecha has joined #buildstream | 16:51 | |
| *** bochecha has quit IRC | 16:55 | |
| *** bochecha has joined #buildstream | 16:55 | |
| Nexus | juergbi: Now using 1.7.9, have confirmed it works | 16:55 |
| Nexus | setting it to merge once pipeline succeeds | 16:55 |
| juergbi | ok, ta | 16:56 |
| Nexus | search will work again \o/ | 16:56 |
| *** jonathanmaw has quit IRC | 17:04 | |
| * laurence is seeing a different coloured fox favicon in buildstream gitlab tabs | 17:10 | |
| * tiagogomes takes credit on that one | 17:10 | |
| tiagogomes | But it is a beaver, not a fox | 17:11 |
| tiagogomes | Ah wait, I didn't do anything on the buildstream gitlab, just the website | 17:12 |
| laurence | tiagogomes, nice | 17:12 |
| laurence | it's odd because i'm only seeing it on certain tabs... | 17:12 |
| gitlab-br-bot | buildstream: issue #681 ("Search functionality in the Documentation has stopped working") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/681 | 17:14 |
| gitlab-br-bot | buildstream: merge request (docs_Search_not_working->master: Fixing: Search functionality in the Documentation has stopped working) #848 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/848 | 17:14 |
| *** benbrown_ has quit IRC | 17:22 | |
| *** Nexus has quit IRC | 17:22 | |
| *** aiden has quit IRC | 17:23 | |
| *** jmac has quit IRC | 17:23 | |
| *** paulsher1ood has quit IRC | 17:23 | |
| *** milloni has quit IRC | 17:23 | |
| *** coldtom has quit IRC | 17:23 | |
| *** adds68 has quit IRC | 17:23 | |
| *** catonano has quit IRC | 17:23 | |
| *** johnward has quit IRC | 17:24 | |
| *** jennis has quit IRC | 17:24 | |
| *** mablanch has quit IRC | 17:24 | |
| *** valentind has quit IRC | 17:24 | |
| *** johanneshilling has quit IRC | 17:24 | |
| *** bethw has quit IRC | 17:24 | |
| *** laurence has quit IRC | 17:24 | |
| *** Nexus has joined #buildstream | 17:24 | |
| *** jennis has joined #buildstream | 17:24 | |
| *** laurence has joined #buildstream | 17:24 | |
| *** benbrown has joined #buildstream | 17:24 | |
| *** mablanch has joined #buildstream | 17:24 | |
| *** johanneshilling has joined #buildstream | 17:24 | |
| *** coldtom has joined #buildstream | 17:24 | |
| *** adds68 has joined #buildstream | 17:24 | |
| *** bethw has joined #buildstream | 17:24 | |
| *** johnward has joined #buildstream | 17:25 | |
| *** paulsherwood has joined #buildstream | 17:25 | |
| *** valentind has joined #buildstream | 17:25 | |
| *** jmac has joined #buildstream | 17:25 | |
| *** aiden has joined #buildstream | 17:28 | |
| *** milloni has joined #buildstream | 17:28 | |
| gitlab-br-bot | buildstream: issue #692 ("SSL Root cettificate error when configuration a cache") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/692 | 17:39 |
| *** lachlan has quit IRC | 17:42 | |
| *** solid_black has joined #buildstream | 17:45 | |
| *** raoul has quit IRC | 17:49 | |
| *** johanneshilling has quit IRC | 17:58 | |
| *** jennis has quit IRC | 17:58 | |
| *** adds68 has quit IRC | 17:58 | |
| *** coldtom has quit IRC | 17:58 | |
| *** mablanch has quit IRC | 18:15 | |
| *** milloni has quit IRC | 18:15 | |
| *** bethw has quit IRC | 18:15 | |
| *** johnward has quit IRC | 18:15 | |
| *** laurence has quit IRC | 18:15 | |
| *** valentind has quit IRC | 18:15 | |
| *** aiden has quit IRC | 18:15 | |
| *** aiden has joined #buildstream | 18:18 | |
| *** laurence has joined #buildstream | 18:19 | |
| *** valentind has joined #buildstream | 18:20 | |
| *** bethw has joined #buildstream | 18:21 | |
| *** johnward has joined #buildstream | 18:21 | |
| *** milloni has joined #buildstream | 18:22 | |
| *** mablanch has joined #buildstream | 18:23 | |
| *** benbrown has quit IRC | 18:24 | |
| *** Nexus has quit IRC | 18:24 | |
| *** jmac has quit IRC | 18:24 | |
| *** paulsherwood has quit IRC | 18:24 | |
| *** benbrown has joined #buildstream | 18:29 | |
| *** coldtom has joined #buildstream | 18:34 | |
| *** toscalix has joined #buildstream | 18:38 | |
| *** toscalix has quit IRC | 18:39 | |
| *** toscalix has joined #buildstream | 18:54 | |
| *** toscalix has quit IRC | 18:54 | |
| *** catonano has joined #buildstream | 19:03 | |
| *** tristan has quit IRC | 19:41 | |
| *** adds68 has joined #buildstream | 20:05 | |
| gitlab-br-bot | buildstream: merge request (valentindavid/rmtree_oserror->master: Catch correct exception from shutil.rmtree) #849 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/849 | 21:25 |
| *** xjuan has joined #buildstream | 21:41 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!