| gitlab-br-bot | buildstream: issue #644 ("Allow BuildStream to be installed in a remote-only mode?") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/644 | 00:16 |
|---|---|---|
| *** rdale has quit IRC | 02:05 | |
| *** tristan has joined #buildstream | 03:48 | |
| *** ChanServ sets mode: +o tristan | 03:48 | |
| gitlab-br-bot | buildstream: issue #632 ("CAS server: Implement BatchReadBlobs") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/632 | 04:23 |
| *** leopi has joined #buildstream | 06:04 | |
| *** leopi has quit IRC | 07:06 | |
| *** leopi has joined #buildstream | 07:07 | |
| *** leopi has quit IRC | 07:11 | |
| *** leopi has joined #buildstream | 07:12 | |
| *** toscalix has joined #buildstream | 07:52 | |
| *** tristan has quit IRC | 08:07 | |
| *** tristan has joined #buildstream | 08:20 | |
| *** ChanServ sets mode: +o tristan | 08:21 | |
| toscalix | tristan: reading your mail, are you planning to turn all those regressions into tickets or are you expecting developers to do it? | 08:23 |
| *** rdale has joined #buildstream | 08:30 | |
| juergbi | jmac: is https://gitlab.com/BuildStream/buildbox/issues/2 still an issue? finn or mablanch, have you seen this or does it work for you? | 08:33 |
| mablanch | juergbi: Nope, I don't remember seeing it. | 08:34 |
| juergbi | ta, closing it as needinfo for now | 08:46 |
| tiagogomes | tristan we do it as well from the main process | 08:50 |
| jmac | I haven't tried a large enough build to reproduce it recently - happy to close it for now | 09:02 |
| tiagogomes | I wonder if before commit/pull/clean we should remove the cache_size, and rewrite it again when these operations finish, so that if something crashes there isn't a stale cache_size file | 09:03 |
| *** tpollard has joined #buildstream | 09:04 | |
| *** bochecha has joined #buildstream | 09:06 | |
| gitlab-br-bot | buildstream: merge request (richardmaw/element-cache-state-simplify->master: Simplify element state by removing `__cached`) #784 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/784 | 09:07 |
| *** abderrahim has quit IRC | 09:10 | |
| juergbi | any opinions on what to call the repo for a buildstream project that builds buildbox? buildbox-integration? | 09:11 |
| *** abderrahim has joined #buildstream | 09:11 | |
| Kinnison | That would follow skullman's suggestion for buildstream-integration being a buildstream project for building buildstream | 09:11 |
| Kinnison | skullman: Did you stick with that in the end? | 09:11 |
| skullman | I did, yes. | 09:11 |
| juergbi | ok, let's go with that for now, ta | 09:12 |
| *** jonathanmaw has joined #buildstream | 09:22 | |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: WIP: out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 09:23 |
| *** lachlan has joined #buildstream | 09:27 | |
| flatmush | I'm looking at the patch documentation here: https://buildstream.gitlab.io/buildstream/sources/patch.html | 09:53 |
| flatmush | And it's not immediately obvious to me how I'd use that to apply a patch to a bst element. | 09:53 |
| flatmush | My specific case is that I want to add the OpenSSL 1.1 patch to OpenSSH as documented here: http://www.linuxfromscratch.org/blfs/view/svn/postlfs/openssh.html | 09:53 |
| flatmush | Can anyone point me in the right direction? | 09:53 |
| gitlab-br-bot | buildstream: merge request (richardmaw/subprocess-PWD->master: WIP: Address post-merge review of Ensure PWD is set in process environment) #788 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/788 | 09:53 |
| *** lachlan has quit IRC | 09:54 | |
| gitlab-br-bot | buildstream: merge request (richardmaw/subprocess-PWD->master: WIP: Address post-merge review of Ensure PWD is set in process environment) #788 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/788 | 09:54 |
| tpollard | flatmush: freedesktop use patch with openssl here https://gitlab.com/freedesktop-sdk/freedesktop-sdk/blob/18.08/elements/base/openssl.bst | 09:56 |
| flatmush | tpollard: cool, thanks | 10:00 |
| *** cs-shadow has joined #buildstream | 10:04 | |
| gitlab-br-bot | buildstream: merge request (richardmaw/builddir-sockets->master: Fix: While caching build artifact: "Cannot extract [path to socket file] into staging-area. Unsupported type.") #783 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/783 | 10:07 |
| *** lachlan has joined #buildstream | 10:10 | |
| *** lachlan has quit IRC | 10:19 | |
| *** lachlan has joined #buildstream | 10:29 | |
| toscalix | tiagogomes: can I add a License file in the website repo or that will impact the web? | 10:33 |
| tiagogomes | It won't change anything regarding the website deployed | 10:34 |
| *** lachlan has quit IRC | 10:34 | |
| toscalix | ok, thanks | 10:34 |
| *** lachlan has joined #buildstream | 10:39 | |
| *** tristan has quit IRC | 10:44 | |
| *** lachlan has quit IRC | 11:11 | |
| *** jonathanmaw_ has joined #buildstream | 11:18 | |
| *** lachlan has joined #buildstream | 11:18 | |
| *** jonathanmaw has quit IRC | 11:18 | |
| *** lachlan has quit IRC | 11:21 | |
| *** lachlan has joined #buildstream | 11:25 | |
| *** lachlan has quit IRC | 11:29 | |
| *** alatiera_ has joined #buildstream | 11:37 | |
| *** alatiera_ has quit IRC | 11:39 | |
| *** lachlan has joined #buildstream | 11:41 | |
| *** leopi has quit IRC | 11:42 | |
| *** lachlan has quit IRC | 11:45 | |
| *** mohan43u has quit IRC | 11:47 | |
| *** mohan43u has joined #buildstream | 11:48 | |
| *** alatiera_ has joined #buildstream | 11:55 | |
| *** lachlan has joined #buildstream | 12:12 | |
| *** lachlan has quit IRC | 12:19 | |
| 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 | 12:28 |
| WSalmon | any one else getting: | 12:46 |
| WSalmon | Please use node.get_closest_marker(name) or node.iter_markers(name). | 12:46 |
| WSalmon | Docs: https://docs.pytest.org/en/latest/mark.html#updating-code | 12:46 |
| WSalmon | content = request.keywords.get('datafiles').args | 12:46 |
| WSalmon | Please use node.get_closest_marker(name) or node.iter_markers(name). | 12:46 |
| WSalmon | Docs: https://docs.pytest.org/en/latest/mark.html#updating-code | 12:46 |
| WSalmon | options.update(request.keywords.get('datafiles').kwargs) | 12:46 |
| WSalmon | when running ```python3 setup.py test --addopts --integration``` | 12:46 |
| WSalmon | locally | 12:47 |
| WSalmon | Please use node.get_closest_marker(name) or node.iter_markers(name). | 12:47 |
| WSalmon | Docs: https://docs.pytest.org/en/latest/mark.html#updating-code | 12:47 |
| WSalmon | content = request.keywords.get('datafiles').args | 12:47 |
| WSalmon | Please use node.get_closest_marker(name) or node.iter_markers(name). | 12:47 |
| WSalmon | Docs: https://docs.pytest.org/en/latest/mark.html#updating-code | 12:47 |
| WSalmon | options.update(request.keywords.get('datafiles').kwargs) | 12:47 |
| jmac | Sounds familiar | 12:48 |
| WSalmon | irc is magling my message, i will paste bin | 12:48 |
| WSalmon | https://hastebin.com/nijiwikigu.sql | 12:48 |
| jmac | Yep | 12:49 |
| WSalmon | oh, i have only just started seeing it | 12:49 |
| jmac | `pip3 install pytest-pylint==0.12.1` will get rid of it temporarily | 12:49 |
| WSalmon | like maybe a week | 12:49 |
| WSalmon | ta | 12:49 |
| jmac | They've started appearing since pytest-pylint 0.12.2 appeared | 12:49 |
| *** alatiera_ has quit IRC | 12:50 | |
| WSalmon | its one thing to want but i get it like 1000 times | 12:50 |
| WSalmon | maybe not tha much but like 90% of my terminal | 12:51 |
| WSalmon | scrollback | 12:51 |
| WSalmon | *its one thing to warn | 12:51 |
| *** alatiera_ has joined #buildstream | 12:52 | |
| WSalmon | interesting that the ci is still using 0.12.1 i wonder if its pegged in there.. | 12:53 |
| gitlab-br-bot | buildstream: merge request (richardmaw/subprocess-PWD->master: Address post-merge review of Ensure PWD is set in process environment) #788 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/788 | 12:55 |
| *** alatiera_ has quit IRC | 12:55 | |
| *** alatiera_ has joined #buildstream | 13:03 | |
| *** lachlan has joined #buildstream | 13:34 | |
| ikerperez | I had some error when building a project with Buildstream in my Debian laptop that doesn't happen when I try it in ubuntu. When I build the project in Debian at some point it start to create directories inside ~/.cache/buildstream without never stopping and finally leads into a linking error | 14:31 |
| *** lachlan has quit IRC | 14:31 | |
| ikerperez | The results is a directory like ~/.cache/buildstream/A/B/A/B/A/B/A/B | 14:32 |
| ikerperez | I haven't been the only one that had this error | 14:32 |
| tlater[m] | ikerperez: Most devs use debian, so I think it's unlikely to be distro specific. That said, do you have logs for this you could share, perhaps open an issue? | 14:33 |
| *** lachlan has joined #buildstream | 14:35 | |
| flatmush | The issue I was having a week or two ago where there seemed to be a recursive variable issue was also on a standard debian install | 14:35 |
| flatmush | the issue iker describes on debian is very similar | 14:36 |
| ikerperez | I shared them on Monday. I don't have them now but I will produce them again | 14:36 |
| flatmush | it's possible that there's an undocumented dependency that's not there by default on debian? | 14:36 |
| toscalix | if that is the case, it would be good to add it to the feature page: https://buildstream.build/feature.html | 14:37 |
| * tlater[m] thinks it's unlikely, variables are resolved by buildstream itself | 14:38 | |
| tlater[m] | But weirder things have happened | 14:38 |
| toscalix | tiagogomes: when you can, I hopefully addressed your concerns: https://gitlab.com/BuildStream/website/merge_requests/77 | 14:38 |
| gitlab-br-bot | buildstream: merge request (mac_fixes->master: WIP: Implement compatibility fixes for MacOSX and WSL Blocks #411 and #412") #726 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/726 | 14:39 |
| gitlab-br-bot | buildstream: merge request (mac_fixes->master: WIP: Implement compatibility fixes for MacOSX and WSL Blocks #411 and #412") #726 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/726 | 14:42 |
| *** hergertme_ has joined #buildstream | 14:44 | |
| *** lachlan has quit IRC | 14:44 | |
| *** hergertme has quit IRC | 14:44 | |
| *** lachlan has joined #buildstream | 14:45 | |
| flatmush | tlater[m]: I'm not sure the issue is to do with variables | 14:46 |
| tlater[m] | I think it's more likely to be a regression from some CAS changes that seem to have happened recently | 14:47 |
| tlater[m] | Probably a slight mistake in an algorithm somewhere | 14:47 |
| flatmush | All I know is that every time we've tried to run with debian, it has failed with similar looking errors. The quickest way to resolve this has been to install Ubuntu. Might be worth having some CI to test this kind of thing? | 14:47 |
| tlater[m] | flatmush: We do, we test on debian, ubuntu and fedora to my knowledge | 14:48 |
| tlater[m] | Hence I sort of believe that the distro is just a coincidence. But it's possible that some sort of dependency mess happens - in any case we'll need logs to debug, this is all conjecture :) | 14:50 |
| flatmush | this is the log for my issue: https://gitlab.com/trustable/distros/minimal-distro/-/jobs/93411861 | 14:56 |
| flatmush | the same code worked fine when I installed Ubuntu on the CI runner | 14:56 |
| tlater[m] | This is definitely interesting... | 15:01 |
| tlater[m] | flatmush: Is there an issue for this yet? I don't have the time to debug it right now, unfortunately, and it looks like nobody else is chiming in either. | 15:03 |
| flatmush | I could never really characterize the issue well enough to write a detailed bug report, and I wondered if it was just an odd issue with my install | 15:04 |
| flatmush | the only reason I'd add weight to this is that iker seems to be having a very similar issue | 15:04 |
| tlater[m] | flatmush: That log and the exact circumstances in which you trigger it is enough detail IMO. More important that we know it's out there than that it's easy to debug. | 15:06 |
| ikerperez | tlater[m], this are the logs https://paste.gnome.org/pqfh04sb2 | 15:07 |
| tlater[m] | Thanks ikerperez | 15:07 |
| tlater[m] | Interesting that it;s /lib/debug/tools | 15:08 |
| tlater[m] | iirc that touches a patch Sam Thursfield wrote not too long ago. | 15:08 |
| *** lachlan has quit IRC | 15:08 | |
| * tlater[m] will write an issue and have a look at it some other day | 15:08 | |
| mablanch | juergbi: For BuildBox pre-fetch support, you are implementing and using BatchReadBlobs() right? Not GetTree()? | 15:13 |
| gitlab-br-bot | buildstream: issue #645 ("Obscure Bug when using Buildstream on Debian") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/645 | 15:13 |
| *** robjh has joined #buildstream | 15:14 | |
| flatmush | ikerperez: Can you attach your logs and a description to this: https://gitlab.com/BuildStream/buildstream/issues/645 | 15:14 |
| ikerperez | sure | 15:16 |
| gitlab-br-bot | buildstream: issue #646 ("BuildStream attempts to create very long filenames after finisihing a build") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/646 | 15:17 |
| gitlab-br-bot | buildstream: issue #646 ("BuildStream attempts to create very long filenames after finisihing a build") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/646 | 15:20 |
| robjh | Hi all. I'm getting this error when trying to build base.bst | 15:23 |
| robjh | [00:00:03][866680d4][fetch:base.bst ] FAILURE 403 Client Error: Forbidden for url: https://production.cloudflare.docker.com/registry-v2/docker/registry/v2/blobs/sha256/5e/5e4eaac9a3437e1b74ba3d71d11b6dbd3822a11eab91049fb185e0ce3ef28fd3/data?verify=1536681998-ChxlDoJ%2BmY9gcpGfNRXmdRaOzdc%3 | 15:23 |
| robjh | The weird thing being, i can access that url using a browser, but curl gives me this: {"status":403,"message":"Error: invalid URL signature"} | 15:23 |
| *** lachlan has joined #buildstream | 15:24 | |
| persia | robjh: BuildStream performs builds in a sandbox without network access, so you'll need to pre-download things and make sure that the build script can use the pre-downloaded copies. | 15:33 |
| persia | (unless I misunderstand, in which case someone else csn likely provide better support) | 15:33 |
| juergbi | mablanch: yes, only BatchReadBlobs() for now. GetTree() might actually be slower in the case where most of the tree is already in the local cache. also, not sure whether BatchReadBlobs() alone is ever substantially slower than using GetTree() | 15:34 |
| juergbi | persia: this doesn't seem to be in the sandbox | 15:36 |
| mablanch | juergbi: Make sense for BuildBox. When you've got nothing locally GetTree() is nice in that you don't have to fetch sub-Directories one by one for full recursive download. | 15:36 |
| robjh | persia, i didnt write this module, but it's bst file doesnt reference that url directly | 15:36 |
| juergbi | robjh: I get 403 here in the browser as well. does this URL require credentials that your browser has stored? | 15:37 |
| * persia is now curious what the problem ends up being | 15:37 | |
| juergbi | it might be a redirect | 15:37 |
| juergbi | what URL is in that base.bst then? | 15:37 |
| robjh | juergbi, to my knowledge, ive not logged into cloudflare.docker.com before | 15:37 |
| juergbi | might be regular docker.com login | 15:38 |
| *** lachlan has quit IRC | 15:39 | |
| robjh | the git log shows that tlater[m] wrote this bst file. so he might be able to weigh in. the files contests are as follows; | 15:40 |
| robjh | kind: import | 15:40 |
| robjh | sources: | 15:40 |
| robjh | - kind: docker | 15:40 |
| robjh | image: buildstream/image-tools | 15:40 |
| robjh | track: latest | 15:40 |
| robjh | ref: 2b959153ad739754b1882cfe3445fe22b3a6e72099101f30c6ea76240c92d76 | 15:40 |
| tlater[m] | Hmm | 15:41 |
| tlater[m] | It's been at least a month since I last tried our docker plugin | 15:41 |
| tlater[m] | It's possible dockerhub changed | 15:41 |
| * tlater[m] thinks cs-shadow would know best | 15:41 | |
| tlater[m] | I have not seen that particular error myself, at any rate | 15:42 |
| robjh | tlater[m], when you access that url do you get a byte stream or a 403 error? | 15:43 |
| tlater[m] | 403 on my end | 15:43 |
| tlater[m] | That is manually accessing it with a browser however. It's possible dockerhub refuses access if you send headers that it doesn't expect. | 15:44 |
| robjh | it works for me from a browser (from the codethink office) but not from curl, and not from the `bst build base` command | 15:45 |
| tpollard | does not work for me on a browser in said office | 15:46 |
| tlater[m] | That *does* sound like they changed the requirements for what clients can access it. robjh, tpollard out of interest, which browser/version? | 15:47 |
| robjh | lol. im now getting {"status":403,"message":"TypeError: Invalid URL string."} in my browser | 15:48 |
| robjh | Anyway; Firefox ESR 52.8.0 | 15:48 |
| juergbi | robjh: the URL indeed looks wrong | 15:49 |
| juergbi | it ends with %3, i.e., incomplete escaped character | 15:49 |
| juergbi | don't know whether that's just cut off in the console or whether that's related to the actual issue | 15:50 |
| robjh | ahhh... maybe i copied it wrong | 15:50 |
| robjh | The last character is D | 15:51 |
| tlater[m] | With that I do get a byte stream | 15:52 |
| robjh | okay, good. and now bst has stopped complaining too | 15:53 |
| gitlab-br-bot | buildstream: merge request (willsalmon/outOfSourecBuild->master: WIP: out of source builds) #776 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/776 | 15:53 |
| robjh | so something weird was happening and now it isnt. im going to go ahead and blame CDN pixies | 15:53 |
| robjh | sorry for the bother everyone :) | 15:54 |
| tlater[m] | Haha, damn gremlins | 15:54 |
| robjh | i'll be back if it happens again ;) | 15:54 |
| *** lachlan has joined #buildstream | 15:57 | |
| *** lachlan has quit IRC | 16:28 | |
| *** lachlan has joined #buildstream | 16:38 | |
| *** toscalix has quit IRC | 16:39 | |
| cs-shadow | tlater[m]: robjh: Sorry I missed your ping earlier but happy that it works now | 16:45 |
| robjh | cs-shadow, no worries. we're blaming CGN gremlins (but it was probably me being stupid) :) | 16:46 |
| *** lachlan has quit IRC | 16:47 | |
| *** lachlan has joined #buildstream | 16:56 | |
| *** ikerperez has quit IRC | 16:57 | |
| *** jonathanmaw_ has quit IRC | 17:02 | |
| *** alatiera__ has joined #buildstream | 17:31 | |
| *** lachlan has quit IRC | 17:32 | |
| *** alatiera_ has quit IRC | 17:33 | |
| robjh | so, these buildstream is working, how would i make it give me an image for an alternate architecture? Or would i need to run it on that architecture? | 17:57 |
| gitlab-br-bot | buildstream: issue #647 ("BuildStream occasionally tries to copy files in alphabetical order rather than safe linking order") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/647 | 18:04 |
| juergbi | robjh: a buildstream project may support cross compilation, in which case there should be a project option for the target architecture | 18:11 |
| cs-shadow | robjh: You can just specify the architecture and os like so: https://gitlab.com/BuildStream/bst-external/blob/master/bst_external/sources/docker.py#L46 | 18:11 |
| * cs-shadow will try to find a better documentation link | 18:12 | |
| juergbi | however, I expect most buildstream projects not to support cross compilation, which means you actually need to run bst on another architecture | 18:12 |
| juergbi | (with remote execution things will be more flexible) | 18:12 |
| juergbi | cs-shadow: this will select the docker image but it still requires a compatible host (or an emulator) | 18:14 |
| robjh | juergbi, cs-shadow, thanks for your respective inputs | 18:15 |
| cs-shadow | juergbi: That's right. My point is only valid if one wants to download the docker image for another arhictecture | 18:15 |
| cs-shadow | but yeah, the build might not work after that | 18:15 |
| robjh | i guess the easiest thing to do would be to bring a raspberry pi into work tomorrow :) | 18:16 |
| *** rdale has quit IRC | 19:42 | |
| *** xjuan has joined #buildstream | 21:54 | |
| *** bochecha has quit IRC | 22:16 | |
| *** mohan43u has quit IRC | 22:36 | |
| *** mohan43u has joined #buildstream | 22:39 | |
| *** alatiera__ has quit IRC | 23:21 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!