| *** tristan_ has quit IRC | 00:06 | |
| *** tristan_ has joined #buildstream | 01:19 | |
| *** tristan_ has quit IRC | 02:08 | |
| *** tristan_ has joined #buildstream | 02:37 | |
| *** tristan_ has quit IRC | 02:38 | |
| *** tristan_ has joined #buildstream | 02:39 | |
| *** tristan_ has quit IRC | 02:45 | |
| *** tristan_ has joined #buildstream | 03:06 | |
| *** tristan_ has quit IRC | 03:08 | |
| *** tristan_ has joined #buildstream | 03:18 | |
| *** tristan_ has quit IRC | 04:09 | |
| *** tristan_ has joined #buildstream | 04:33 | |
| *** tristan_ has quit IRC | 04:36 | |
| *** tristan_ has joined #buildstream | 04:49 | |
| *** tristan_ has quit IRC | 05:12 | |
| *** tristan_ has joined #buildstream | 05:12 | |
| *** tristan_ has quit IRC | 05:51 | |
| *** tristan_ has joined #buildstream | 05:53 | |
| *** tristan_ has quit IRC | 06:49 | |
| *** tristan has quit IRC | 06:51 | |
| *** tristan has joined #buildstream | 07:48 | |
| *** tristan_ has joined #buildstream | 07:48 | |
| qinusty | Currently getting an interesting error message... https://paste.gnome.org/pghjqhuov/d1zmda/raw. empty bytes error failure :( | 08:00 |
|---|---|---|
| jennis | Yeah I've seen that before /o\ qinusty, not sure if you're aware of the docker compose that is using buildgrid in our test suite? | 08:02 |
| jennis | Might provide some inspiration | 08:02 |
| qinusty | oooo, could be worth looking at. Cheers jennis. | 08:02 |
| jennis | qinusty: https://gitlab.com/BuildStream/buildstream/blob/master/.gitlab-ci/buildgrid-compose.yml | 08:03 |
| jennis | But, that's basically running containers for each service you need, and you can use this as a client | 08:04 |
| jennis | Note that storage != artifact cache | 08:05 |
| jennis | qinusty, IIRC https://gitlab.com/BuildGrid/buildgrid.hub.docker.com/ contains a more generic manifest which I think you can just fire up and start using | 08:09 |
| jennis | There are some good instructions in the readme there | 08:09 |
| qinusty | Looks good, configuring it all for kubernetes at the minute but all of the launch options and configs will likely get me up and running | 08:10 |
| jennis | Kubernetes :O | 08:15 |
| qinusty | Whatever b'' is for, it appears to be originating from buildbox https://paste.gnome.org/pjnwdw40n | 08:35 |
| Kinnison | b'' is an empty bytestring | 08:39 |
| Kinnison | (says he probably unhelpfully now he's read backscroll) | 08:39 |
| qinusty | :D, it is also an unhelpful error message :P | 08:43 |
| qinusty | My guess is, somewhere buildstream takes an error response from buildbox and just prints it. And buildbox is being unhelpful | 08:43 |
| *** tristan_ has quit IRC | 08:50 | |
| *** tristan_ has joined #buildstream | 08:52 | |
| jennis | How do we tell App that something is a failure? | 08:55 |
| jennis | For artifact_push() in cli.py, we have a context manager: "with app.initialized(session_name="Push"):", which when finished, returns SUCCESS | 08:56 |
| jennis | Do I need to raise a specific error and special case it in app.initialize? :/ | 08:57 |
| tpollard | the scheduler should capture failure messages and send them to app | 08:59 |
| jennis | Yeah I tried a MessageType.ERROR, but this didn't work | 09:00 |
| jennis | Looks like raising a StreamError works | 09:00 |
| *** jonathanmaw has joined #buildstream | 09:01 | |
| *** phildawson has joined #buildstream | 09:03 | |
| *** lchlan has quit IRC | 09:14 | |
| *** lchlan has joined #buildstream | 09:14 | |
| *** tme5 has joined #buildstream | 09:26 | |
| *** lachlan has joined #buildstream | 09:28 | |
| juergbi | I'd appreciate a review of https://gitlab.com/BuildStream/buildstream-docker-images/merge_requests/125 (Add buildbox-casd to testsuite images) | 09:42 |
| *** tristan_ has quit IRC | 09:50 | |
| *** tristan_ has joined #buildstream | 09:51 | |
| gitlab-br-bot | jennis opened 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 | 10:05 |
| jennis | benschubert, ^ may be of interest to you, it addresses an issue you filed (#1095) | 10:07 |
| jennis | If anyone else could review, that'd be greatly appreciated :) | 10:07 |
| * tpollard looks | 10:12 | |
| *** lachlan has quit IRC | 10:14 | |
| *** lachlan has joined #buildstream | 10:30 | |
| *** tristan_ has quit IRC | 10:43 | |
| *** tristan_ has joined #buildstream | 10:48 | |
| *** tristan_ has quit IRC | 10:51 | |
| *** tristan_ has joined #buildstream | 10:52 | |
| benschubert | jennis: thanks! I added a comment there | 10:53 |
| *** lachlan has quit IRC | 10:56 | |
| *** phildawson_ has joined #buildstream | 10:56 | |
| *** phildawson has quit IRC | 10:58 | |
| *** tme5 has quit IRC | 11:00 | |
| *** tme5 has joined #buildstream | 11:00 | |
| *** tme5 has quit IRC | 11:15 | |
| *** tristan_ has quit IRC | 11:26 | |
| *** tristan_ has joined #buildstream | 11:26 | |
| *** tristan_ has quit IRC | 11:50 | |
| *** tristan_ has joined #buildstream | 11:52 | |
| jennis | How do we raise an "interactive" error? | 11:55 |
| jennis | One where we can choose to terminate/continue? | 11:55 |
| tpollard | check app | 11:56 |
| tpollard | _job_failed & _handle_failure | 11:57 |
| jennis | ah, the interrupt handler | 11:57 |
| jennis | thanks | 11:57 |
| jennis | Oh no, that's for ^C | 11:57 |
| jennis | but yeah app looks like the right place, thanks | 11:57 |
| tpollard | if the scheduler catches a job that errored, it should end up in that job_failed | 11:58 |
| *** tristan_ has quit IRC | 12:14 | |
| *** tristan_ has joined #buildstream | 12:15 | |
| *** phildawson_ has quit IRC | 12:20 | |
| gitlab-br-bot | juergbi opened MR !1525 (juerg/buildbox->master: Enable integration tests for buildbox sandboxing backend) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1525 | 12:27 |
| juergbi | WSalmon: fyi ^^ | 12:27 |
| gitlab-br-bot | willsalmon approved MR !1525 (juerg/buildbox->master: Enable integration tests for buildbox sandboxing backend) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1525 | 12:33 |
| *** toscalix has joined #buildstream | 12:33 | |
| juergbi | ta | 12:34 |
| gitlab-br-bot | marge-bot123 merged MR !1525 (juerg/buildbox->master: Enable integration tests for buildbox sandboxing backend) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1525 | 12:36 |
| juergbi | cs-shadow: if you have a bit of time, I'd appreciate a review of https://gitlab.com/BuildStream/buildstream-docker-images/merge_requests/125 (Add buildbox-casd to testsuite images) | 12:38 |
| juergbi | who is currently maintaining CI for WSL? maybe tpollard? what's the procedure to add a new dependency? | 12:42 |
| tpollard | it was jonathanmaw, I'm not sure who has the mantle now | 12:43 |
| tpollard | I'm sure I can get on the machine to add a package though if needed | 12:43 |
| juergbi | tpollard: for WSL CI to work with !1499 we'll have to install buildbox-common and buildbox-casd on WSL. i.e., the WSL equivalent of https://gitlab.com/BuildStream/buildstream-docker-images/merge_requests/125 | 12:45 |
| gitlab-br-bot | MR !1499: WIP: Use buildbox-casd for CAS access https://gitlab.com/BuildStream/buildstream/merge_requests/1499 | 12:45 |
| juergbi | maybe we could also do it in the CI job itself, but it would be better to align it with the other CI jobs. I'm also not sure how isolation currently works on WSL. is a fresh rootfs created for every job? | 12:47 |
| *** tristan_ has quit IRC | 12:50 | |
| *** tristan_ has joined #buildstream | 12:52 | |
| *** tme5 has joined #buildstream | 13:03 | |
| jennis | benschubert, cs-shadow: getting the push failure to continue on error has proved more difficult that I initially thought | 13:06 |
| jennis | We only have the interactive failures for when a completed job registers as failed | 13:07 |
| jennis | However, we need to check whether or not a cached artifact exists before we create the job | 13:08 |
| jennis | And it doesn't make sense to create a push job for an element we know isn't cached | 13:08 |
| jennis | Perhaps it makes more sense to handle this in stream and only enqueue elements (to the push queue) which we know are cached | 13:12 |
| *** phildawson_ has joined #buildstream | 13:13 | |
| tpollard | I wonder if you could have loader/load selection take an arg (assert_cached) or something | 13:13 |
| jennis | Mhmm, yeah, we should have support for this if we try to shell/checkout an uncached element too | 13:15 |
| jennis | It'd make more sense to be consistent with that | 13:15 |
| *** lachlan has joined #buildstream | 13:33 | |
| *** tristan_ has quit IRC | 13:50 | |
| gitlab-br-bot | coldtom opened (was WIP) MR !1520 (coldtom/move-cmake->master: Move cmake plugin to bst-plugins-experimental) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1520 | 14:07 |
| *** tristan_ has joined #buildstream | 14:08 | |
| *** tristan_ has quit IRC | 14:09 | |
| *** lachlan has quit IRC | 14:16 | |
| *** lachlan has joined #buildstream | 14:18 | |
| *** lachlan has quit IRC | 14:30 | |
| *** tristan_ has joined #buildstream | 14:32 | |
| cs-shadow | juergbi: hi, sorry, I just saw your message. Have left a couple comments on the MR | 14:37 |
| juergbi | ta | 14:37 |
| *** tiagogomes has quit IRC | 14:43 | |
| *** tiagogomes has joined #buildstream | 14:43 | |
| *** tristan_ has quit IRC | 14:50 | |
| *** tristan_ has joined #buildstream | 14:53 | |
| qinusty | Found the cause of b'' jennis https://gitlab.com/BuildGrid/buildgrid/issues/193 | 14:58 |
| *** lachlan has joined #buildstream | 15:05 | |
| *** bochecha has quit IRC | 15:09 | |
| *** lachlan has quit IRC | 15:11 | |
| tpollard | benschubert: I've either resolved all commented on all your comments for https://gitlab.com/BuildStream/buildstream/merge_requests/1500/ thanks for taking the time to raise them. Hopefully if all is ok it can get approval soon! | 15:14 |
| coldtom | is it possible to use two includes in the same yaml dict? I'm being told there's a repeated key "(@)" | 15:16 |
| benschubert | tpollard: I won't have time to have a look at it before tomorrow. If somebody approves, go ahead, otherwise I'll have a look tomorrow | 15:18 |
| tpollard | benschubert: np :) thanks | 15:20 |
| *** lachlan has joined #buildstream | 15:21 | |
| gitlab-br-bot | juergbi opened MR !1526 (juerg/expiry-test->master: tests/artifactcache/expiry.py: Limit to single fetcher) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1526 | 15:21 |
| juergbi | coldtom: you can't use multiple `(@)`, however, it accepts a list of strings, so you can still include multiple files | 15:24 |
| coldtom | ta juergbi :) | 15:24 |
| jennis | qinusty, ah, nice | 15:25 |
| qinusty | I don't imagine it requires that much work to fix, but bgd bot is on it's way out anyway. So I'm not sure it'll get fixed | 15:25 |
| *** lachlan has quit IRC | 15:27 | |
| *** phildawson has joined #buildstream | 15:27 | |
| *** phildawson_ has quit IRC | 15:28 | |
| *** lachlan has joined #buildstream | 15:30 | |
| *** lachlan has quit IRC | 15:33 | |
| *** tristan_ has quit IRC | 15:41 | |
| *** tristan_ has joined #buildstream | 15:52 | |
| gitlab-br-bot | juergbi closed MR !1526 (juerg/expiry-test->master: tests/artifactcache/expiry.py: Limit to single fetcher) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1526 | 15:54 |
| *** ChanServ sets mode: +o tristan_ | 15:59 | |
| *** lachlan has joined #buildstream | 16:31 | |
| *** tme5 has quit IRC | 16:44 | |
| *** tristan_ has quit IRC | 16:50 | |
| *** lachlan has quit IRC | 16:59 | |
| *** jonathanmaw has quit IRC | 17:03 | |
| *** lachlan has joined #buildstream | 17:04 | |
| *** tristan_ has joined #buildstream | 17:07 | |
| *** tristan_ has quit IRC | 17:09 | |
| *** tristan_ has joined #buildstream | 17:12 | |
| *** phildawson has quit IRC | 17:29 | |
| *** lachlan has quit IRC | 17:32 | |
| *** lachlan has joined #buildstream | 17:37 | |
| *** toscalix has quit IRC | 17:48 | |
| *** lachlan has quit IRC | 18:20 | |
| *** tristan_ has quit IRC | 18:50 | |
| *** lachlan has joined #buildstream | 18:52 | |
| *** lachlan has quit IRC | 19:00 | |
| *** lachlan has joined #buildstream | 19:14 | |
| *** lachlan has quit IRC | 19:22 | |
| *** lachlan has joined #buildstream | 19:33 | |
| *** lachlan has quit IRC | 19:44 | |
| *** lachlan has joined #buildstream | 19:58 | |
| *** lachlan has quit IRC | 19:59 | |
| *** tristan_ has joined #buildstream | 20:57 | |
| *** tristan_ has quit IRC | 21:51 | |
| *** tristan_ has joined #buildstream | 21:56 | |
| *** tristan_ has quit IRC | 22:45 | |
| *** tristan_ has joined #buildstream | 22:46 | |
| *** tristan_ has quit IRC | 23:15 | |
| *** tristan_ has joined #buildstream | 23:15 | |
| *** tristan_ has quit IRC | 23:24 | |
| *** tristan_ has joined #buildstream | 23:24 | |
| *** tristan_ has quit IRC | 23:25 | |
| *** tristan_ has joined #buildstream | 23:26 | |
| *** tristan_ has quit IRC | 23:30 | |
| *** tristan_ has joined #buildstream | 23:31 | |
| *** tristan_ has quit IRC | 23:50 | |
| *** tristan has quit IRC | 23:52 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!