| *** tristan_ has joined #buildstream | 01:17 | |
| *** tristan_ has quit IRC | 01:38 | |
| *** tristan_ has joined #buildstream | 02:31 | |
| *** tristan_ has quit IRC | 02:38 | |
| *** admin has quit IRC | 03:07 | |
| *** swick has joined #buildstream | 03:08 | |
| *** tristan_ has joined #buildstream | 03:34 | |
| *** tristan_ has quit IRC | 03:39 | |
| *** tristan_ has joined #buildstream | 03:58 | |
| *** tristan_ has quit IRC | 03:59 | |
| *** tristan_ has joined #buildstream | 04:00 | |
| *** tristan_ has quit IRC | 04:02 | |
| *** tristan_ has joined #buildstream | 04:05 | |
| *** tristan_ has joined #buildstream | 04:30 | |
| *** tristan_ has quit IRC | 04:31 | |
| *** tristan_ has joined #buildstream | 04:37 | |
| *** tristan_ has quit IRC | 04:39 | |
| *** tristan_ has joined #buildstream | 04:44 | |
| *** tristan_ has quit IRC | 04:49 | |
| *** tristan_ has joined #buildstream | 04:49 | |
| *** tristan_ has quit IRC | 04:50 | |
| *** tristan_ has joined #buildstream | 04:51 | |
| *** tristan_ has quit IRC | 04:53 | |
| *** tristan_ has joined #buildstream | 05:08 | |
| *** tristan_ has quit IRC | 05:09 | |
| *** tristan_ has joined #buildstream | 05:10 | |
| *** tristan_ has quit IRC | 05:11 | |
| *** tristan_ has joined #buildstream | 05:12 | |
| *** tristan_ has quit IRC | 05:13 | |
| *** tristan_ has joined #buildstream | 05:59 | |
| *** tristan_ has quit IRC | 06:01 | |
| *** tristan_ has joined #buildstream | 06:01 | |
| *** tristan_ has quit IRC | 06:04 | |
| *** tristan_ has joined #buildstream | 06:05 | |
| *** tristan_ has quit IRC | 06:07 | |
| *** tristan_ has joined #buildstream | 06:12 | |
| *** tristan_ has quit IRC | 06:14 | |
| *** tristan_ has joined #buildstream | 06:19 | |
| *** tristan_ has joined #buildstream | 06:21 | |
| *** tristan_ has quit IRC | 06:23 | |
| *** tristan_ has joined #buildstream | 06:34 | |
| *** tristan_ has quit IRC | 06:36 | |
| *** tristan_ has joined #buildstream | 06:57 | |
| gitlab-br-bot | danielsilverstone-ct opened MR !1543 (danielsilverstone-ct/missing-sh-fix->master: Cache failed builds caused by missing-command) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1543 | 07:15 |
|---|---|---|
| *** tristan_ has joined #buildstream | 07:22 | |
| *** tristan_ has quit IRC | 07:25 | |
| *** narispo has quit IRC | 07:31 | |
| *** narispo has joined #buildstream | 07:31 | |
| *** narispo has quit IRC | 07:36 | |
| Kinnison | Have I recently been granted more access than I previously had? I now seem to have "merge when pipeline succeeds" buttons where I didn't have them before | 07:40 |
| qinusty | Doesn't buildstream require atleast 1 approval? | 07:40 |
| * qinusty shrugs | 07:41 | |
| qinusty | apparently it's changed, who knows | 07:41 |
| juergbi | is there a sensible place where I can upload a ~2MB static buildbox casd binary (maybe digitalocean server)? it will be used by WSL CI (as it doesn't use docker images) and by users who don't want to build it on their own | 07:44 |
| *** narispo has joined #buildstream | 07:44 | |
| juergbi | would initially only be linux x86-64, but we could add binaries for other platforms | 07:45 |
| juergbi | built using buildstream but not yet [buildbox] CI | 07:46 |
| jennis | juergbi, on digital ocean we have the bastion and the tutorial cache server which are permanent | 08:04 |
| jennis | I don't think the tutorial cache server is being used though, perhaps we can take that down and spin up another one for this? | 08:05 |
| juergbi | well, I don't think we need an extra VM for this | 08:06 |
| juergbi | don't we already store e.g. the alpine base image on digital ocean? | 08:07 |
| juergbi | can we easily store additional files like that? | 08:07 |
| alexandrufazakas | + | 08:07 |
| alexandrufazakas | oops | 08:07 |
| jennis | mhmm yeah, let me see if I can find it (assuming it's on the BuildStream account...) | 08:08 |
| juergbi | thanks | 08:08 |
| jennis | juergbi, it says you're on the digital ocean account, can you log on? | 08:17 |
| jennis | I've created a "space" called buildbox-casd-linux-x86-64, you can upload the file here | 08:17 |
| jennis | Just go to spaces on the LHS | 08:18 |
| jennis | In fact, if the spaces will take additional binaries, I'll change the name to buildbox-casd-binaries | 08:20 |
| juergbi | yes, I think that works. didn't realize I already had access | 08:22 |
| juergbi | thanks | 08:22 |
| jennis | juergbi, looks like your branch is somehow appending a new line to the summary each time we write output | 08:29 |
| *** rdale has joined #buildstream | 08:41 | |
| gitlab-br-bot | coldtom opened issue #1104 (Filter elements don't work on stacks) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1104 | 08:43 |
| *** phildawson has joined #buildstream | 09:04 | |
| *** jonathanmaw has joined #buildstream | 09:07 | |
| *** tme5 has joined #buildstream | 09:17 | |
| gitlab-br-bot | tlater approved MR !1524 (jennis/push_unbuilt_artifact->master: Ensure push fails when trying to push an unbuilt element) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1524 | 09:19 |
| *** lachlan has joined #buildstream | 09:28 | |
| *** phildawson_ has joined #buildstream | 09:36 | |
| *** phildawson has quit IRC | 09:36 | |
| gitlab-br-bot | tlater approved MR !1543 (danielsilverstone-ct/missing-sh-fix->master: Cache failed builds caused by missing-command) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1543 | 09:41 |
| *** traveltissues has joined #buildstream | 09:42 | |
| jennis | Kinnison, do you mind if I cherry-pick Ben's commit and repush your branch? | 10:11 |
| jennis | This commit was the test Ben added which made me realise this error in the first place | 10:11 |
| Kinnison | I do not object | 10:11 |
| jennis | Ok, thanks | 10:11 |
| gitlab-br-bot | jennis approved MR !1543 (danielsilverstone-ct/missing-sh-fix->master: Cache failed builds caused by missing-command) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1543 | 10:15 |
| jennis | Kinnison, done, thanks for fixing that | 10:17 |
| Kinnison | no probs, took me a bit of time to track down, and then nearly no time to fix :D | 10:17 |
| tpollard | \o/ | 10:17 |
| jennis | gitlab is being slow | 10:18 |
| tme5 | my MR is ready for a look also :) https://gitlab.com/BuildStream/buildstream/merge_requests/1541 | 10:18 |
| tme5 | it's sooo slow isn't it | 10:18 |
| jennis | Also the pipeline are "pending" for quite a while before they begin, but I've checked digital ocean, we have space | 10:19 |
| jennis | So that might be related to the general gitlab slowness | 10:19 |
| Kinnison | they're claiming that they *had* an incident earlier today but that it's now resolved | 10:19 |
| Kinnison | this might just be residual slowdown as it recovers | 10:19 |
| gitlab-br-bot | tmewett opened issue #1105 (Dependency selection is inconsistent across commands) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1105 | 10:44 |
| tme5 | RFC on above | 10:46 |
| tlater[m] | tme5: ooi, are you sure PLAN includes the runtime dependencies? | 10:57 |
| tlater[m] | iirc it only includes runtime dependencies of build dependencies | 10:58 |
| jennis | tlater[m], why did you unassign marge to Kinnison's MR? | 10:58 |
| jennis | s/to/from | 10:58 |
| tlater[m] | jennis: Did I? | 10:58 |
| * tlater[m] thought he assigned her | 10:58 | |
| jennis | Tristan Maat @tlater unassigned @marge-bot123 21 minutes ago | 10:58 |
| tlater[m] | :| | 10:58 |
| tlater[m] | gitlab! | 10:58 |
| jennis | haha | 10:58 |
| tlater[m] | tme5: coming back to that point... | 10:59 |
| tlater[m] | We obviously need runtime dependencies of build dependencies because, well, we need to run build dependencies to build our element. | 10:59 |
| tlater[m] | So it's not as arbitrary as you might think :) | 10:59 |
| tlater[m] | Provided that it doesn't include runtime dependencies of our element, but I suspect you mis-wrote/read that. | 11:00 |
| tme5 | tlater[m], I am sure | 11:00 |
| tme5 | look in Planner.plan_element | 11:00 |
| * tme5 really hopes his python reading was correct | 11:00 | |
| tlater[m] | tme5: I think it might not have been | 11:02 |
| tlater[m] | Oh, hang on | 11:03 |
| tlater[m] | No, you're right | 11:03 |
| * tlater[m] sent a long message: < https://gnome.modular.im/_matrix/media/v1/download/gnome.org/rNIQggUVhyNgHeUgssfutfet > | 11:03 | |
| tme5 | jennis, Kinnison and I looked into this yesterday, and I'm fairly sure that's what we found empirically too | 11:04 |
| tlater[m] | Ok, cool, I said nothing :) | 11:05 |
| *** lachlan has quit IRC | 11:34 | |
| *** lachlan has joined #buildstream | 11:47 | |
| tlater[m] | jennis: Could you put marge back onto that MR? | 12:01 |
| tlater[m] | She's not appearing in my assignment list | 12:01 |
| *** lachlan has quit IRC | 12:04 | |
| tpollard | can't until the pipeline is fixed or she can push new commits | 12:04 |
| jennis | tlater[m], ahh workspaces with buildbox need to be marked with xfail | 12:04 |
| jennis | I'll sort it | 12:04 |
| tlater[m] | Can you not assign marge to anything that has a failing pipeline? | 12:05 |
| Kinnison | she'll just hand it back because it fails | 12:06 |
| jennis | mhmm failures on both MRs for the tests-buildbox | 12:06 |
| tlater[m] | Kinnison: Well, she doesn't show up in the list of possible assignees in the first place for me, I wanted to try if she'd restart the pipeline for me :) | 12:16 |
| Kinnison | I don't think she'll rerun pipelines, no | 12:16 |
| adds68 | thanks for the review cs-shadow | 12:29 |
| *** lachlan has joined #buildstream | 12:30 | |
| *** alexandrufazakas has quit IRC | 12:35 | |
| gitlab-br-bot | marge-bot123 closed issue #1101 (Sometimes failed builds are not cached as failed) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1101 | 12:43 |
| gitlab-br-bot | marge-bot123 merged MR !1543 (danielsilverstone-ct/missing-sh-fix->master: Cache failed builds caused by missing-command) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1543 | 12:43 |
| * tlater[m] is annoyed there is no `bst source delete` | 13:18 | |
| tlater[m] | I think we need to do something to make artifacts and sources equivalent to the front-end | 13:19 |
| tlater[m] | Because under the hood they're all just CAS keys anyway | 13:19 |
| tlater[m] | Would keep us from duplicating so damn much for the two types of CAS content. | 13:20 |
| qinusty | Without the support of seperate endpoints for CAS/Artifact server does buildstream currently work with buildgrid (with remote caching support)? Perhaps tlater[m]? | 13:38 |
| tlater[m] | qinusty: It should do | 13:51 |
| tlater[m] | To my knowledge | 13:51 |
| juergbi | traveltissues: thinking about it more, I expect it will be somewhat cumbersome and also not very efficient to maintain a separate file for timestamps of the whole tree | 14:17 |
| traveltissues | i can't immediately think of an alternative if we're ruling out metadata changes | 14:18 |
| juergbi | extending the Directory proto would likely be much simpler and more efficient. the main concern there is compatibility with other servers | 14:19 |
| Kinnison | tlater[m]: sources and artifacts are different in terms of where they live and what their semantics are in terms of sharing | 14:19 |
| Kinnison | tlater[m]: It'd not be complex to add `bst source {list,delete}` | 14:19 |
| juergbi | traveltissues: however, it's also not trivial to achieve compatibility with a separate file (would have to apply timestamps at the beginning via shell script and then create that timestamp file via shell script at the end) | 14:19 |
| tlater[m] | No, but the underlying code is extremely similar | 14:19 |
| tlater[m] | Kinnison: I'm worried that we're just writing the same bits of boilerplate for the two over and over again. | 14:20 |
| traveltissues | juergbi, yes, aiui this compatibility issue was a blocker | 14:20 |
| tlater[m] | (Having gone through the effort of implementing source {push, pull}) | 14:20 |
| juergbi | traveltissues: so I'm wondering whether we should rather say that (incremental) workspace builds are restricted to REAPI servers that support that extension (+ local builds, of course). | 14:20 |
| juergbi | might be acceptable as long as we don't break compatibility for non-workspace builds | 14:21 |
| Kinnison | tlater[m]: I suppose the underlying stuff might be unifiable, though I wouldn't be sad if it weren't short-term | 14:21 |
| tlater[m] | Yeah, it's a long-term thought | 14:21 |
| tlater[m] | Since I'm sitting here getting annoyed over how many Queues I need to implement that check the same 3 variables with slightly different names ;) | 14:22 |
| juergbi | traveltissues: this might need some further/broader discussion | 14:22 |
| Kinnison | hehe | 14:23 |
| juergbi | #38 is also somewhat related to this | 14:23 |
| gitlab-br-bot | Issue #38: Lost file metadata in artifacts and images https://gitlab.com/BuildStream/buildstream/issues/38 | 14:23 |
| traveltissues | juergbi, we can definitely/should support that locally. but this would mean having to maintain a separate proto. I can't really say what the performance hit from a data file would be atm | 14:24 |
| juergbi | maybe we could standardize an extension mechanism | 14:25 |
| traveltissues | yes, that's related at some point in this effort. bb would probably need some work to support this aiui | 14:25 |
| juergbi | possibly key-value pairs for extended attributes | 14:25 |
| traveltissues | in bb fuse? | 14:26 |
| juergbi | I mean standardize in REAPI upstream | 14:26 |
| juergbi | and then support it in bb, of course | 14:26 |
| juergbi | afaik, other REAPI users are not currently interested in this, but maybe we could still get agreement on at least reserving a field that would be suitable for extensions | 14:27 |
| juergbi | I think we should at least going down that route as it would be much nicer in many aspects | 14:28 |
| juergbi | *at least try | 14:28 |
| juergbi | or do you see this differently? | 14:28 |
| traveltissues | no, i'd prefer to avoid using some sort of index file if possible | 14:31 |
| juergbi | we should send a mail to the remote-execution-apis list. or possibly first discuss it more broadly among BuildStream/BuildGrid/BuildBox devs and then send a mail there | 14:33 |
| gitlab-br-bot | marge-bot123 merged MR !1539 (coldtom/move-modulebuild->master: Move modulebuild plugin to bst-plugins-experimental) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1539 | 14:35 |
| gitlab-br-bot | coldtom opened MR !1545 (coldtom/move-qmake->master: Move qmake plugin to bst-plugins-experimental) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1545 | 14:44 |
| *** bochecha_ has joined #buildstream | 15:01 | |
| *** lachlan has quit IRC | 15:03 | |
| *** bochecha has quit IRC | 15:03 | |
| *** bochecha_ is now known as bochecha | 15:03 | |
| gitlab-br-bot | marge-bot123 closed issue #1095 (BuildStream crashes when trying to push a non-build artifact) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1095 | 15:18 |
| gitlab-br-bot | marge-bot123 merged MR !1524 (jennis/push_unbuilt_artifact->master: Ensure push fails when trying to push an unbuilt element) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1524 | 15:18 |
| Kinnison | tme5: I'm not sure about !1541 | 15:30 |
| gitlab-br-bot | MR !1541: Clarify bst artifact subcommand help text https://gitlab.com/BuildStream/buildstream/merge_requests/1541 | 15:30 |
| * tlater[m] takes a peak while waiting for his test suite | 15:40 | |
| tlater[m] | Grmbl grmbl pytest | 15:40 |
| adds68 | cs-shadow, hey sorry for the delay: https://gitlab.com/BuildStream/buildstream-docker-images/merge_requests/129 is that ok? | 15:42 |
| tlater[m] | Kinnison: isn't that what the doc says? | 15:42 |
| adds68 | Gitlab will squash those commits also | 15:42 |
| tlater[m] | Kinnison: > Element | 15:43 |
| tlater[m] | paths may be given instead, in which case the artifact ref is calculated | 15:43 |
| tlater[m] | from the element. | 15:43 |
| tlater[m] | Or are you unsure whether we can mix them? | 15:43 |
| *** bochecha_ has joined #buildstream | 15:53 | |
| tpollard | is the buildbox pipeline known to be broken/ | 15:55 |
| tpollard | ? | 15:55 |
| *** bochecha has quit IRC | 15:56 | |
| *** bochecha_ is now known as bochecha | 15:56 | |
| tme5 | as far as I understand (which may not be far at all), elements give rise to artifact refs | 15:59 |
| tme5 | so i wrote it in the way I did to clarify that; the commands operate on artifacts, but you can give element paths to use their innate artifact ref | 16:00 |
| tme5 | tlater[m], on the subject of mixing them, i've seen jennis write in multiple places that bst artifact checkout/pull/push are not actually supposed to take refs yet. So this documents current behaviour, maybe not intended :) | 16:03 |
| *** phil has joined #buildstream | 16:40 | |
| *** phildawson_ has quit IRC | 16:41 | |
| *** toscalix has joined #buildstream | 16:41 | |
| *** lachlan has joined #buildstream | 16:46 | |
| *** toscalix has quit IRC | 16:51 | |
| *** lachlan has quit IRC | 16:53 | |
| *** jonathanmaw has quit IRC | 17:03 | |
| *** lachlan has joined #buildstream | 17:09 | |
| *** lachlan has quit IRC | 17:15 | |
| *** tme5 has quit IRC | 17:17 | |
| *** lachlan has joined #buildstream | 17:30 | |
| *** phil has quit IRC | 17:43 | |
| *** bochecha has quit IRC | 17:49 | |
| Kinnison | tlater[m]: when I read it, it felt like the emphasis had shifted enough to confuse me, I'll re read it tomorrow | 17:49 |
| *** traveltissues has quit IRC | 17:55 | |
| *** lachlan has quit IRC | 17:56 | |
| *** lachlan has joined #buildstream | 18:17 | |
| *** tristan_ has joined #buildstream | 18:22 | |
| *** lachlan has quit IRC | 18:22 | |
| *** lachlan has joined #buildstream | 18:23 | |
| *** tristan_ has quit IRC | 18:38 | |
| *** tristan_ has joined #buildstream | 18:39 | |
| *** tristan_ has quit IRC | 18:41 | |
| *** tristan_ has joined #buildstream | 18:51 | |
| *** tristan_ has quit IRC | 18:53 | |
| *** tristan_ has joined #buildstream | 18:59 | |
| *** tristan_ has quit IRC | 19:02 | |
| *** tristan_ has joined #buildstream | 19:02 | |
| *** tristan_ has joined #buildstream | 19:04 | |
| *** tristan_ has quit IRC | 19:05 | |
| *** tristan_ has joined #buildstream | 19:05 | |
| *** tristan_ has quit IRC | 19:10 | |
| *** tristan_ has joined #buildstream | 19:11 | |
| *** tristan_ has quit IRC | 19:13 | |
| *** narispo has quit IRC | 19:16 | |
| *** narispo has joined #buildstream | 19:18 | |
| *** tristan_ has joined #buildstream | 19:20 | |
| *** tristan_ has quit IRC | 19:24 | |
| *** lachlan has quit IRC | 20:04 | |
| *** cs-shadow has quit IRC | 20:57 | |
| *** narispo has quit IRC | 22:01 | |
| *** narispo has joined #buildstream | 22:01 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!