| *** nimish has joined #buildstream | 02:06 | |
| *** mohan43u has quit IRC | 02:31 | |
| *** mohan43u has joined #buildstream | 02:35 | |
| *** nimish has quit IRC | 03:18 | |
| *** tpollard has joined #buildstream | 08:40 | |
| doras[m] | When is 1.2.4 expected? | 09:02 |
|---|---|---|
| *** toscalix has joined #buildstream | 09:43 | |
| *** kapil___ has joined #buildstream | 09:48 | |
| *** toscalix has quit IRC | 10:06 | |
| *** toscalix has joined #buildstream | 10:07 | |
| *** toscalix has quit IRC | 10:15 | |
| *** jonathanmaw has joined #buildstream | 10:17 | |
| *** toscalix has joined #buildstream | 10:19 | |
| *** jmac has joined #buildstream | 10:25 | |
| *** toscalix has joined #buildstream | 10:29 | |
| *** lachlan has joined #buildstream | 10:42 | |
| * laurence has only just learnt that you can filter MRs using WIP yes/no | 10:46 | |
| jjardon | It's a recent new feature (2 or 3 releases ago I think) | 10:48 |
| *** raoul has joined #buildstream | 10:50 | |
| adds68 | Hi all, i have sent out the final email for the logo, there are now 2 choices, which ever gets the most votes by the end of today (can be longer), i will send over the designer for finalisation | 10:51 |
| *** toscalix has quit IRC | 10:55 | |
| laurence | adds68, thanks again for pushing this - I hope my intervention wasn't too awkward, but I felt the wheel needed an airing | 10:59 |
| laurence | and some key contributors seem to agree, which is nice (i think their vote counts for more than folks who don't even have a patch) | 10:59 |
| *** raoul has quit IRC | 11:00 | |
| adds68 | laurence, no problem, it's hard to try and please everyone with things like this and i also don't want people to feel left out. I also have to keep the designer happy, so once we have a final version, we can always ask for small revisions/colours as he is happy to do things like that | 11:00 |
| adds68 | laurence, and yes agreed :) | 11:00 |
| *** alatiera has joined #buildstream | 11:01 | |
| doras[m] | Should "cleanup" take... minutes? | 11:04 |
| *** lachlan has quit IRC | 11:05 | |
| doras[m] | Looks like "~/.cache/buildstream/artifacts" is 57G, while I set its quota to be 50G. So it does make sense that a cleanup would occur, but it also seems to remove about 4 kilobytes per 5 seconds. That's awfully slow. | 11:09 |
| *** lachlan has joined #buildstream | 11:11 | |
| gitlab-br-bot | jmacarthur approved MR !1129 (snakeviz->master: contributing: snakeviz replaces pyflame+flamegraph) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1129 | 11:11 |
| doras[m] | Is there a way to trigger cleanup on-demand? | 11:14 |
| *** lachlan has quit IRC | 11:18 | |
| doras[m] | Trying to build again still leaves me with an extremely slow cleanup :\ | 11:23 |
| *** lachlan has joined #buildstream | 11:28 | |
| jjardon | doras[m]: 1.2.4 was expected before christmas I think; I guess it will be released soon | 11:29 |
| jjardon | doras[m]: about the cleanup: https://gitlab.com/BuildStream/buildstream/issues/734 | 11:29 |
| *** toscalix has joined #buildstream | 11:37 | |
| doras[m] | Thanks for the reference, jjardon. Is there a workaround for this? Can I also delete objects directly somehow to skip this process? | 11:39 |
| jjardon | doras[m]: I'm not aware of any, sorry | 11:39 |
| doras[m] | What I currently end up doing is clean up my entire cache with rm -rf every week or so. This isn't what everybody else does, is it? | 11:43 |
| jjardon | doras[m]: that is what I do as well | 11:43 |
| doras[m] | There must be some way to selectively remove what I know isn't needed anymore. | 11:43 |
| doras[m] | Well, bummer. | 11:44 |
| juergbi | doras[m]: we have plans to improve this (switching to object-based expiry) | 11:44 |
| jjardon | doras[m]: if you find other solution, please comment in the issue :) | 11:44 |
| tpollard | I believe there's a plan to add a command for delete element artifacts | 11:44 |
| juergbi | right now a very expensive part of the cleanup process is to prune objects, i.e., figure out which object files are still needed | 11:45 |
| juergbi | with object-based expiry, we can skip that, however, it requires a few changes in BuildStream to accommodate incomplete/partial artifacts | 11:45 |
| juergbi | (on the server side, we already follow this approach, as it can deal with partial artifacts) | 11:45 |
| jjardon | would that be available in the 1.2 branch, or is too big change? | 11:46 |
| juergbi | I think it's rather unlikely | 11:46 |
| juergbi | however, hopefully, 1.90.x will be acceptable for general use | 11:46 |
| juergbi | at least the snapshots | 11:46 |
| doras[m] | If I remove refs manually in bulk, can the garbage collector figure out rather quickly which objects are no longer needed? | 11:48 |
| juergbi | if you still have many refs left, it might still take a while as it has to enumerate all reachable objects | 11:49 |
| doras[m] | I see. Looking at what is actually deleted, it appears that it only deletes refs. It doesn't even reach the part of deleting the actual objects. | 11:54 |
| *** phildawson has quit IRC | 12:03 | |
| *** phildawson has joined #buildstream | 12:03 | |
| doras[m] | I wonder when it actually removes the objects, then. | 12:10 |
| doras[m] | Or otherwise how I can figure out which objects can be deleted by myself. | 12:15 |
| gitlab-br-bot | jonathanmaw approved MR !1130 (are_you_sure2->master: userconfig: rm really-workspace-close-project-inaccessible) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1130 | 13:17 |
| *** phildawson has quit IRC | 13:19 | |
| gitlab-br-bot | tpollard opened (was WIP) MR !1135 (tpollard/896->master: Optional creation of buildtrees) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1135 | 13:28 |
| tpollard | tests-fedora-update-deps just took 2 hours to complete a test run | 13:33 |
| tpollard | and counting | 13:33 |
| *** raoul has joined #buildstream | 13:47 | |
| *** kapil___ has quit IRC | 13:48 | |
| *** raoul has quit IRC | 13:52 | |
| gitlab-br-bot | jennis approved MR !1129 (snakeviz->master: contributing: snakeviz replaces pyflame+flamegraph) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1129 | 13:55 |
| gitlab-br-bot | jennis approved MR !1130 (are_you_sure2->master: userconfig: rm really-workspace-close-project-inaccessible) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1130 | 14:00 |
| laurence | jennis, earlier this week, Kinnison, benshubert and I had a chat and wrt benchmarks, we mentioned having an easy way to compare a local checkout against master | 14:00 |
| *** nimish has joined #buildstream | 14:00 | |
| laurence | jennis, is this feasible with the current set up? what do you think? | 14:00 |
| *** phildawson has joined #buildstream | 14:01 | |
| *** tpollard has quit IRC | 14:04 | |
| jennis | AFAIK, you can run the benchmarking locally on your laptop, so I don't think it will be too difficult to add a helper in where we run this for master and the branch in concern | 14:04 |
| jennis | jonathanmaw, looks like we're having problems with the windows runner: https://gitlab.com/BuildStream/buildstream/-/jobs/158499449 | 14:11 |
| jonathanmaw | hmm, wifi problems on that computer | 14:12 |
| jennis | :( | 14:12 |
| Kinnison | phildawson: Good news - the experiment from this morning, assuming we manage to cache usually, is ca. 2x the speed | 14:13 |
| Kinnison | phildawson: it's *marginally* slower if we *always* have to re-parse | 14:13 |
| * Kinnison will now try and plumb it alongside _variables.py to do a proper profile | 14:14 | |
| jonathanmaw | jennis: there we go, I've moved it to wired internet | 14:14 |
| phildawson | Nice Kinnison :) | 14:15 |
| Kinnison | phildawson: looks like the number of times we call len() it's worth caching | 14:47 |
| gitlab-br-bot | jennis approved MR !1075 (phil/plugin-testing-api->master: Expose basic api for testing external plugins.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1075 | 14:48 |
| *** lachlan has quit IRC | 15:16 | |
| *** lachlan has joined #buildstream | 15:23 | |
| *** toscalix has quit IRC | 15:33 | |
| WSalmon | hi all | 15:35 |
| WSalmon | i just tried to follow https://buildstream.build/source_install.html#installing_dependencies | 15:35 |
| WSalmon | for fedora but then installed master | 15:35 |
| WSalmon | it failed | 15:35 |
| WSalmon | is this guild only for 1.2? | 15:35 |
| WSalmon | or should this work? | 15:35 |
| WSalmon | should i be issuing a bug agenst the website/ | 15:36 |
| tlater[m] | WSalmon: If the docker image builds, I find it unlikely that this fails | 15:42 |
| tlater[m] | But yes, I'd file a bug against the website | 15:42 |
| tlater[m] | If you'd like a bit more debugging help, feel free to paste errors and suchlike here :) | 15:42 |
| WSalmon | so the website is not detailing master so all is fine in that regard | 15:48 |
| WSalmon | but the git clone bit dosent seem to be rendering well so if some one else could check its not just me | 15:49 |
| WSalmon | then i could put a issue up for that | 15:49 |
| doras[m] | I see: git checkout <desired release tag> | 15:58 |
| WSalmon | not just me then, thanks doras[m] | 16:01 |
| WSalmon | https://gitlab.com/BuildStream/website/issues/28 | 16:08 |
| gitlab-br-bot | phildawson closed issue #847 (Expose testing api for use by external plugins) on buildstream https://gitlab.com/BuildStream/buildstream/issues/847 | 16:28 |
| gitlab-br-bot | phildawson merged MR !1075 (phil/plugin-testing-api->master: Expose basic api for testing external plugins.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1075 | 16:28 |
| gitlab-br-bot | valentindavid approved MR !1056 (jmac/make_marked_directories->master: _sandboxremote.py: Create marked directories before running) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1056 | 16:37 |
| gitlab-br-bot | jjardon opened MR !1136 (jjardon/fedora_27->master: .gitlab-ci.yml: Remove testing on Fedora 27) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1136 | 16:49 |
| *** bilelmoussaoui has joined #buildstream | 17:00 | |
| *** bilelmoussaoui has quit IRC | 17:26 | |
| adds68 | any last votes for the buildstream logo: http://www.strawpoll.me/17389951 ? | 17:30 |
| adds68 | 7.9% of people have voted | 17:30 |
| adds68 | See the mailing list for the #1/#2 images also | 17:31 |
| gitlab-br-bot | Issue #1: sphinx docs fail on python2-default systems https://gitlab.com/BuildStream/buildstream/issues/1 | 17:31 |
| adds68 | haha silly gitlab-br-bot :) | 17:32 |
| WSalmon | dose it do that for any #22 | 17:33 |
| gitlab-br-bot | Issue #22: failed to upgrade buildstream at 640a734ec1db923b5 https://gitlab.com/BuildStream/buildstream/issues/22 | 17:33 |
| WSalmon | haha thats cool | 17:33 |
| gitlab-br-bot | jonathanmaw opened (was WIP) MR !1134 (jonathan/junction-no-tmpdir->master: Stage junctions into .bst instead of a tmpdir) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1134 | 17:34 |
| adds68 | WSalmon, also !1111 | 17:36 |
| gitlab-br-bot | MR !1111: Run tests using tox https://gitlab.com/BuildStream/buildstream/merge_requests/1111 | 17:36 |
| adds68 | ironfoot is a wizz kid | 17:36 |
| ironfoot | ;) | 17:37 |
| ironfoot | You are all gitlabIRCed | 17:37 |
| *** nimish has quit IRC | 17:38 | |
| *** bilelmoussaoui has joined #buildstream | 17:45 | |
| *** bilelmoussaoui has joined #buildstream | 17:46 | |
| *** lachlan has quit IRC | 18:03 | |
| *** jonathanmaw has quit IRC | 18:05 | |
| *** lachlan has joined #buildstream | 18:06 | |
| *** bilelmoussaoui has quit IRC | 18:20 | |
| jjardon | valentind: number of concurrent jobs were limited to 20; I have increased it to 50 | 18:20 |
| valentind | ok | 18:20 |
| jjardon | seems the pipeline has grow a lot of parallel jobs lately | 18:21 |
| jjardon | Can I have reviews of https://gitlab.com/BuildStream/buildstream/merge_requests/1136 , please? | 18:55 |
| gitlab-br-bot | jjardon closed issue #618 (BuildStream pulls from remote cache, even when configuration removed) on buildstream https://gitlab.com/BuildStream/buildstream/issues/618 | 20:29 |
| gitlab-br-bot | jjardon closed issue #401 (If a project B use project A as a junction, B should not try to use the bst remote cache of A (at least by default)) on buildstream https://gitlab.com/BuildStream/buildstream/issues/401 | 20:29 |
| gitlab-br-bot | jjardon merged MR !1113 (abderrahim/artifact-cache-junction->master: Use artifact cache specs from the parent project before those defined in junctions) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1113 | 20:29 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!