| *** barlas has joined #buildstream | 00:55 | |
| *** bochecha has quit IRC | 01:03 | |
| *** leopi has joined #buildstream | 03:04 | |
| *** leopi has quit IRC | 03:25 | |
| *** codex2064 has joined #buildstream | 03:55 | |
| *** leopi has joined #buildstream | 04:33 | |
| *** mohan43u has joined #buildstream | 05:11 | |
| *** leopi has quit IRC | 05:48 | |
| *** noisecell has joined #buildstream | 05:56 | |
| *** Peetz0r has joined #buildstream | 06:00 | |
| *** tristan has joined #buildstream | 06:05 | |
| *** leopi has joined #buildstream | 06:31 | |
| *** toscalix has joined #buildstream | 07:28 | |
| gitlab-br-bot | buildstream: issue #331 ("Project configuration required is too much") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/331 | 07:59 |
|---|---|---|
| gitlab-br-bot | buildstream: issue #195 ("Errors are needed when referring to local things outside of the project") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/195 | 07:59 |
| *** tiagogomes_ has joined #buildstream | 08:01 | |
| *** mohan43u has quit IRC | 08:01 | |
| toscalix | LInk to the board we are working with in the meeting: https://gitlab.com/BuildStream/buildstream/boards/580466?milestone_title=BuildStream_v1.2&=&label_name[]=Blocker | 08:05 |
| *** coldtom has joined #buildstream | 08:06 | |
| toscalix | feel free to join: https://appear.in/toscalix | 08:06 |
| toscalix | we are going to discuss how the tasks related with the freeze are going | 08:07 |
| gitlab-br-bot | buildstream: issue #195 ("Errors are needed when referring to local things outside of the project") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/195 | 08:14 |
| coldtom | is it possible to connect to the network through bst shell? | 08:18 |
| toscalix | board with Urgent tickets https://gitlab.com/groups/BuildStream/-/boards?=&label_name[]=Bug | 08:19 |
| jmac | As far as I know, the sandbox is meant to isolate that, coldtom | 08:19 |
| coldtom | something similar to flatpak run --share=[substystem] would be very handy for troubleshooting | 08:20 |
| coldtom | currently i'm waiting ~3 hours for llvm6 to build to test an issue with openssl in freedesktop-sdk | 08:20 |
| Kinnison | coldtom: Would weak rebuilds help your cycle speed? | 08:21 |
| coldtom | Kinnison: that doesn't rebuild elements if dependencies change right? | 08:22 |
| coldtom | because if so, yes | 08:22 |
| Kinnison | coldtom: It only rebuilds elements if their sources, or their dependency chain *shape* changes afaict | 08:23 |
| Kinnison | coldtom: not if a dependency rebuilds | 08:23 |
| Kinnison | coldtom: --no-strict might be your friend | 08:23 |
| coldtom | Kinnison: this probably solves my issue, ty | 08:23 |
| *** leopi has quit IRC | 08:28 | |
| *** finn_ has joined #buildstream | 08:36 | |
| *** leopi has joined #buildstream | 08:38 | |
| *** finn has quit IRC | 08:38 | |
| * qinusty is looking for issue suggestions | 08:38 | |
| *** mohan43u has joined #buildstream | 08:46 | |
| gitlab-br-bot | buildstream: issue #550 ("pulling from the cache doesn't honor the fetchers setting") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/550 | 08:54 |
| *** tiagogomes_ has quit IRC | 09:06 | |
| *** tiagogomes_ has joined #buildstream | 09:08 | |
| gitlab-br-bot | buildstream: merge request (dp0/513/cas-cache-client-certs->master: WIP: Support dynamic client certificates for CAS cache) #594 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/594 | 09:12 |
| gitlab-br-bot | buildstream: merge request (dp0/513/cas-cache-client-certs->master: WIP: Support dynamic client certificates for CAS cache) #594 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/594 | 09:12 |
| laurence | I think I will send a mail to the ML re back-porting to 1.2. | 09:26 |
| qinusty | I'm not too familiar with the process of backporting. I assume you're just rebasing a bunch of commits from an MR onto the 1.2 branch and adding the necessary commits that're missing? | 09:32 |
| tiagogomes_ | qinusty: yes. You rebase, fix eventual conflicts and then create another merge request but this time targeting the bst-1.2 branch | 09:38 |
| *** mohan43u has quit IRC | 09:50 | |
| laurence | Re notification emails from gitlab, there's always been a 'view it on gitlab' link on the ones i receive | 09:55 |
| laurence | but it does not seem to be there anymore | 09:55 |
| laurence | anyone else? | 09:56 |
| jmac | Hmm, you're right | 09:56 |
| jmac | But I have emails from yesterday which do have the link | 09:57 |
| jmac | And one today which doesn't | 09:57 |
| gitlab-br-bot | buildstream: merge request (tiagogomes/issue-500->master: Handle gracefully files of type socket) #570 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/570 | 09:58 |
| laurence | ah well at least it's not just me - will see if it's been reported to gitlab upstream | 09:58 |
| tristan | laurence, as I receive those in gmail, which I dont know if it's munging the messages in some way; I have always had to look at the first email in the thread which I received from gitlab to see that link | 10:00 |
| tristan | it's become a habit for me to click the first email in the thread to get at that link | 10:00 |
| tristan | not sure if that has regressed / changed | 10:00 |
| laurence | I use thunderbird and | 10:00 |
| laurence | woops - and they're usually on every mail | 10:01 |
| tristan | coldtom, regarding bst shell and network; in fact we reversed the default of that quite a while back, and added the `--isolate` option to differentiate | 10:02 |
| tristan | coldtom, in a build, network is always disabled, but for `bst shell`, it seems more popularly used to test a built application, as such we default to allowing networking there | 10:03 |
| jmac | Yeah, GitLab have broken the HTML version of their email | 10:03 |
| tristan | interestingly also, running `make distcheck` in a `bst shell --build` normally breaks without network access | 10:04 |
| jmac | `Reply to this email directly or <a href="https://gitlab.com/BuildGrid/buildgrid/merge_requests/34#note_92306915">view it on GitLab</a>` | 10:04 |
| tristan | jmac, indeed, I'm seeing the yuck right now | 10:05 |
| tristan | in my inbox :-S | 10:06 |
| *** mohan43u has joined #buildstream | 10:07 | |
| laurence | I'll raise an issue on gitlab/production | 10:24 |
| laurence | https://gitlab.com/gitlab-com/production/issues/389 | 10:28 |
| *** coldtom has quit IRC | 10:30 | |
| toscalix | tristan https://gitlab.com/BuildStream/nosoftware/communication/issues/12 | 10:33 |
| toscalix | should I add anything else? | 10:33 |
| toscalix | hummm, maybe the link to the previous communication, for reference | 10:34 |
| jjardon | tristan: would it be possible to tag when we know there is a possibility we are breaking stuff? It would make our work easier to know what version to use and prepare for possible breackage | 10:36 |
| jjardon | for example, the patch that now makes bst to fail when there are stuff outside the project folder, or when there is a change in the cache-key calculation | 10:37 |
| tristan | jjardon, I think there is an issue blocking such things | 10:40 |
| tristan | jjardon, i.e. I am perfectly fine with making symbolic tags, I think we need that | 10:40 |
| tristan | jjardon, but we need to make versioneer cope with it first, that was fallout from the last time we tried it | 10:40 |
| jjardon | tristan: issue where? in buildstream? | 10:41 |
| tristan | yes | 10:41 |
| tristan | jjardon, remember you asked for a pre-cas tag, and then it started failing ? | 10:41 |
| tristan | that resulted in an issue, I dont know off hand where it's filed; we aught to make sure it's filed and being addressed, though | 10:41 |
| jjardon | ah ok, so this was already requested | 10:42 |
| jjardon | I will search the buildstream issue tracker | 10:42 |
| jjardon | tristan: what started failing exactly? | 10:42 |
| tristan | Well, the specific tag you mean is not requested, but I have always presumed the ability to make symbolic tags, not all tags are release tags after all | 10:42 |
| tristan | jjardon, bst fails to derive it's version when the latest tag in history is not a numeric version tag | 10:43 |
| tristan | which... sucks | 10:43 |
| tristan | and causes pretty much any bst invocation to break | 10:43 |
| jjardon | tristan: oops, well a (numeric) tag with a little description about why is done would help a lot. Only to have a reference | 10:44 |
| tristan | jjardon, that would not be ideal, because numeric tags, like 1.2.0 for instance, are release tags; if we tagged an arbitrary number and it caused BuildStream to not "fail", it would still cause BuildStream to report an incorrect version | 10:45 |
| tristan | also, it would not convey the meaning of the symbolic tag in the tag name, which also sucks | 10:45 |
| tristan | found it | 10:46 |
| tristan | jjardon, https://gitlab.com/BuildStream/buildstream/issues/516 | 10:46 |
| paulsherwood | not sure if it's relevant, but ybd calculates its version using some git magic https://gitlab.com/baserock/ybd/blob/master/ybd/repos.py#L75 | 10:47 |
| paulsherwood | (but ybd expects always to be run from git) | 10:47 |
| jjardon | tristan: I only suggesting to tag more often: 1.1.3, 1.1.4, 1.1.5 and specially when we know there is a possibility to break stuff (the explanation can go in the tag body message) | 10:47 |
| tristan | paulsherwood, what we do is similar, but there is a bug in that code | 10:47 |
| paulsherwood | there is? | 10:48 |
| tristan | we use this versioneer thing to do that, it has a bug | 10:48 |
| paulsherwood | hmmm... i thought you meant a bug in the ybd code. it's been working for several years | 10:48 |
| tristan | jjardon, understood, but I still think we should fix 516, it's probably not *that* hard to fix, probably easier than sitting around and discussing the issue :) | 10:48 |
| tristan | paulsherwood, yeah it's not a bug in YBD, what we do which is similar (but also encodes the correct version in the package info when generating a dist tarball), has a bug | 10:49 |
| paulsherwood | ack... drop versioneer, then :) | 10:52 |
| paulsherwood | "you had one job..." | 10:52 |
| tristan | paulsherwood, I dropped the last thing because it didnt do it's job properly (it needed an additional ./setup.py install --user -e invocation to update it's version) | 10:53 |
| tristan | Fixing it is not a huge deal; this code which is in our repo, is doing a list of the tags in chronological order, it just needs to skip tags it doesnt understand as a release tag | 10:54 |
| tristan | That is probably also easier than "dropping it" | 10:54 |
| tristan | it also serves an important purpose; whether or not humans consume release tarballs, package maintainers will either consume the tarballs, or need to rely on that so that BuildStream knows it's correct version when deployed via a distro | 10:55 |
| tristan | toscalix, reading through https://gitlab.com/BuildStream/nosoftware/communication/issues/12, I'm not sure what the purpose of all of that is. I was most certainly already going to mail the list and announce feature freeze along with the 1.1.5 beta, and have additionally confirmed that I will announce it in our meeting this morning | 11:01 |
| tristan | toscalix, maybe this issue is leading towards some documentation about our process, rather than a big issue about a mail I have to send out on monday ? | 11:02 |
| tristan | if the latter, I understand the purpose of it | 11:02 |
| tristan | toscalix, there is also some other material I have regarding the routine involved in making releases which could go into some useful documentation somewhere | 11:03 |
| *** wizonesolutions12 has joined #buildstream | 11:05 | |
| tristan | qinusty, do I have another moderated message from you ? or is that another notification from the message you posted before having joined the list ? | 11:12 |
| tristan | ah, this appears to be the original message, since you have already subscribed and sent it, I will delete it from the moderation queue | 11:13 |
| WSalmon | if i checkout a non numeric tag then i get a error saying that it cant turn the tag name in to a int | 11:15 |
| WSalmon | i was gona make a issue but do we want it to suport this or just have a more graceful error message | 11:15 |
| tristan | WSalmon, I posted the exact issue for that above, recently | 11:20 |
| tristan | WSalmon, it needs to be fixed, plain and simple | 11:20 |
| noisecell | WSalmon, tristan, there is an issue open already for this: https://gitlab.com/BuildStream/buildstream/issues/516 | 11:27 |
| tristan | noisecell, that is the one I'm referring to yes | 11:28 |
| *** mohan43u has quit IRC | 11:30 | |
| *** mohan43u has joined #buildstream | 11:34 | |
| *** tristan has quit IRC | 11:40 | |
| WSalmon | so i can wrap the int('s in try and have a nice message saying please version me better? or do we want to beable to suport these tags some how? | 12:07 |
| gitlab-br-bot | buildstream: merge request (tiagogomes/includes-1.12-backport->bst-1.2: Add support for include in project.conf) #599 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/599 | 12:08 |
| noisecell | WSalmon, for my discussion with tristan, we would like to make versioner to understand release and not release tags, and if it is a no release tag (as the non numerical are at the moment) continue using the previous tag | 12:12 |
| noisecell | at least that is what I understood of what it is written in #516 | 12:12 |
| *** toscalix has quit IRC | 12:16 | |
| *** toscalix has joined #buildstream | 12:16 | |
| toscalix | tristan beta is august 6th | 12:46 |
| toscalix | together with freeze | 12:47 |
| toscalix | I am just documenting the release | 12:47 |
| toscalix | assuming freeze and beta does not need to be the same day | 12:47 |
| toscalix | so for the next release I just need to copy the tickets | 12:47 |
| gitlab-br-bot | buildstream: merge request (tiagogomes/includes-1.12-backport->bst-1.2: Add support for include in project.conf) #599 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/599 | 12:48 |
| toscalix | in the original plan freeze and beta were not the same day. It is do to a delay. I will make a note in the comments about this once I finish the creation of the beta communication ticket | 12:49 |
| qinusty | toscalix, tristan isn't connected via bouncer. He'll miss that message | 12:53 |
| *** toscalix has quit IRC | 12:54 | |
| *** toscalix has joined #buildstream | 12:54 | |
| tiagogomes_ | If one issue is blocked by another issue, how that should be marked. Through "Related Issues" or is there a better way? | 12:57 |
| qinusty | I'd argue related with a comment in the description saying something about how #<blocker> needs to be addressed first | 12:58 |
| laurence | Related Issues and also a note in the description is the only thing you can do, sadly | 12:58 |
| *** bochecha has joined #buildstream | 13:03 | |
| toscalix | tristan https://gitlab.com/BuildStream/nosoftware/communication/issues/13 | 13:07 |
| jjardon | Hey, only coming here to say we see massive improvements with the new CAS server (much less CPU usage, and most important, no timeout failures anymore). Congrats everyone juergbi and everyone involved! :) | 13:08 |
| qinusty | toscalix, tristan is not currently connected and won't see backscroll | 13:08 |
| toscalix | qinusty: I know, but he will probably read the logs on Monday | 13:09 |
| juergbi | jjardon: good to hear, thanks for the feedback | 13:09 |
| toscalix | I will send him a mail though, with the full explanation. Thanks for poointing | 13:09 |
| toscalix | great to hear | 13:09 |
| toscalix | good news on friday afternoon.... | 13:10 |
| noisecell | qinusty, the channel is logged, so everyone can read the backscroll, though -- https://irclogs.baserock.org/buildstream | 13:11 |
| noisecell | jjardon, these are great news | 13:12 |
| qinusty | TIL | 13:12 |
| qinusty | ty noisecell | 13:12 |
| bochecha | jjardon: I had a timeout failure today | 13:13 |
| qinusty | How do people raise warnings to the interface when the message handler isn't initialized? | 13:19 |
| tiagogomes_ | sys.stderr.write maybe | 13:20 |
| WSalmon | noisecell tristan, i can fix this by changing line 242 in _version.py but it looks like the file is auto generated or has something to do with versioner. do i need to try and resolve this upstream in versioner so we can continue to upgrade versioner or can i just submit a change to bst with this small change to bst's copy of _version.py? | 13:23 |
| tiagogomes_ | I am trying to build freedesktop sdk, and there are a lot of these messages@ | 13:23 |
| tiagogomes_ | [00:00:00][][] SUCCESS cache_size.15157.log | 13:23 |
| tiagogomes_ | [--:--:--][][] START cache_size.15159.log | 13:23 |
| noisecell | WSalmon, regarding ttps://gitlab.com/BuildStream/buildstream/issues/516 (last paragraph) -- we would like to solve that in versioneer if possible | 13:29 |
| noisecell | and if it does involve big effort, open an issue and create a workaround in our code | 13:30 |
| jjardon | bochecha: do you have a link to the logs? | 13:34 |
| bochecha | jjardon: https://gitlab.com/freedesktop-sdk/freedesktop-sdk/pipelines/27064115/builds | 13:35 |
| bochecha | jjardon: 3 out of 4 timed out | 13:35 |
| bochecha | jjardon: it's building a lot straight from bootstrap though, so now that I think about it, it probably had nothing to do with the new cache | 13:35 |
| jjardon | bochecha: no related with CAS really | 13:35 |
| jjardon | simply the build took more than 20h, which is the current limit | 13:35 |
| jjardon | maybe we should increase it | 13:36 |
| bochecha | I've relaunched it, let's see if this time it hits the cache and builds less, and thus faster :) | 13:36 |
| *** xjuan has joined #buildstream | 14:00 | |
| gitlab-br-bot | buildstream: merge request (relative_workspaces->master: WIP: Patch for issue #191 support relative workspaces) #504 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/504 | 14:02 |
| WSalmon | Could some one have a quick look at https://gitlab.com/BuildStream/buildstream/merge_requests/595 at some point, thanks | 14:21 |
| *** jennis2 is now known as jennis | 14:22 | |
| *** coldtom has joined #buildstream | 14:23 | |
| *** toscalix has quit IRC | 14:38 | |
| *** toscalix has joined #buildstream | 14:39 | |
| *** noisecell has quit IRC | 14:53 | |
| *** leopi has quit IRC | 14:54 | |
| gitlab-br-bot | buildstream: merge request (tiagogomes/issue-520->master: Fix race condition when calculating disk usage) #600 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/600 | 15:04 |
| gitlab-br-bot | buildstream: merge request (willsalmon/versionTagRegrex->master: Search for tags with the *.*.* patten for version) #601 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/601 | 15:10 |
| tiagogomes_ | WSalmon: commented on the PR | 15:14 |
| gitlab-br-bot | buildstream: merge request (Qinusty/491->master: Cache quota is now restricted to available disk space) #563 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/563 | 15:20 |
| *** Ragnor has joined #buildstream | 15:20 | |
| WSalmon | So following up on #516 I have made !601 noisecell tristan | 15:20 |
| gitlab-br-bot | buildstream: merge request (Qinusty/502->master: cascache.py: Raise ArtifactError on grpc error) #576 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/576 | 15:21 |
| gitlab-br-bot | buildstream: merge request (Qinusty/275->master: Indicate where artifacts are going to and coming from in the log) #553 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/553 | 15:22 |
| gitlab-br-bot | buildstream: merge request (relative_workspaces->master: WIP: Patch for issue #191 support relative workspaces) #504 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/504 | 15:23 |
| qinusty | tiagogomes_, If I'm looking to rebase for a backport. I assume I'm checking out 1.2 and cherry picking my commits from the relevant MR? | 15:26 |
| WSalmon | you may consider "git rebase --onto 1.2.x commitBefore commitTo" if you just want a few and dont want to cherry pick each one | 15:31 |
| tiagogomes_ | I normally cherry-pick it is only one or two commits. Otherwise I do a git rebase instead | 15:33 |
| *** toscalix has quit IRC | 15:33 | |
| gitlab-br-bot | buildstream: merge request (Qinusty/backport-576->bst-1.2: Backport unhandled exception fix in cascache.py) #602 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/602 | 15:34 |
| *** leopi has joined #buildstream | 15:35 | |
| *** leopi has quit IRC | 15:43 | |
| qinusty | Alright, I've got a few I'll do monday as rebases | 15:43 |
| qinusty | What is the merge policy onto 1.2? I've made a MR https://gitlab.com/BuildStream/buildstream/merge_requests/602 | 15:44 |
| gitlab-br-bot | buildstream: merge request (willsalmon/versionTagRegrex->master: Search for tags with the *.*.* patten for version) #601 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/601 | 15:45 |
| tiagogomes_ | qinusty: I think the unofficial policy is that if was approved in master and it makes sense to backport it to 1.2, you can create a merge request and set it to merge automatically when the CI passes | 15:51 |
| gitlab-br-bot | buildstream: merge request (willsalmon/trackWarning->master: Add warning to git track if track and ref are not present) #580 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/580 | 15:58 |
| *** rdale has quit IRC | 16:02 | |
| phildawson | The aarch64 runnner that bst-docker-images uses seems to be down. jjardon do you know anything about this? | 16:11 |
| *** tiagogomes_ has quit IRC | 16:22 | |
| *** tiagogomes_ has joined #buildstream | 16:22 | |
| *** tiagogomes_ has quit IRC | 16:28 | |
| gitlab-br-bot | buildstream: merge request (Qinusty/backport-576->bst-1.2: Backport unhandled exception fix in cascache.py) #602 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/602 | 16:30 |
| gitlab-br-bot | buildstream: merge request (jmac/cas_virtual_directory->master: WIP: CAS-backed virtual directory implementation) #481 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/481 | 16:37 |
| *** rdale has joined #buildstream | 16:41 | |
| gitlab-br-bot | buildstream: merge request (jmac/cas_virtual_directory->master: WIP: CAS-backed virtual directory implementation) #481 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/481 | 16:50 |
| gitlab-br-bot | buildstream: merge request (jmac/cas_virtual_directory->master: WIP: CAS-backed virtual directory implementation) #481 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/481 | 16:54 |
| gitlab-br-bot | buildstream: merge request (jmac/cas_virtual_directory->master: WIP: CAS-backed virtual directory implementation) #481 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/481 | 16:57 |
| *** mohan43u has quit IRC | 17:06 | |
| *** Pici has joined #buildstream | 17:10 | |
| *** mohan43u has joined #buildstream | 17:10 | |
| *** rdale has quit IRC | 17:11 | |
| gitlab-br-bot | buildstream: merge request (toscalix->master: removed space in task template so checkboxes work as expected) #588 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/588 | 17:29 |
| *** xjuan has quit IRC | 18:54 | |
| *** leopi has joined #buildstream | 19:16 | |
| *** lantw44 has quit IRC | 19:55 | |
| *** cs_shadow has quit IRC | 20:23 | |
| *** beaver has joined #buildstream | 21:50 | |
| *** bochecha has quit IRC | 21:59 | |
| *** Dan_au has joined #buildstream | 22:05 | |
| *** Hello71 has joined #buildstream | 22:10 | |
| *** tristan has joined #buildstream | 22:13 | |
| *** Welcome has joined #buildstream | 22:35 | |
| *** leopi has quit IRC | 22:50 | |
| *** tristan has quit IRC | 22:56 | |
| *** pelegreno______ has joined #buildstream | 23:30 | |
| *** justyns has joined #buildstream | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!