| *** Prince781 has quit IRC | 03:51 | |
| *** toscalix has joined #buildstream | 07:41 | |
| *** Phil has joined #buildstream | 08:00 | |
| *** jonathanmaw has joined #buildstream | 08:41 | |
| gitlab-br-bot | buildstream: issue #307 ("Follow-up from "Add getting started section"") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/307 | 08:51 |
|---|---|---|
| *** Sebastian has joined #buildstream | 09:52 | |
| *** Phil has quit IRC | 09:59 | |
| *** Phil has joined #buildstream | 09:59 | |
| *** jonathanmaw has quit IRC | 10:02 | |
| *** jonathanmaw has joined #buildstream | 10:17 | |
| *** Sebastian has quit IRC | 10:22 | |
| gitlab-br-bot | buildstream: merge request (reduce_history_in_cache->master: Reduce history in cache) #482 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/482 | 10:32 |
| *** cs_shadow has joined #buildstream | 10:33 | |
| *** Sebastian_ has joined #buildstream | 10:34 | |
| *** Sebastian_ has quit IRC | 10:36 | |
| *** Sebastian_ has joined #buildstream | 10:38 | |
| gitlab-br-bot | buildstream: merge request (reduce_history_in_cache->master: Reduce history in cache) #482 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/482 | 10:38 |
| *** Phil has quit IRC | 10:40 | |
| *** Phil has joined #buildstream | 10:40 | |
| *** Sebastian_ has quit IRC | 10:47 | |
| *** Sebastian_ has joined #buildstream | 10:49 | |
| *** Sebastian_ has quit IRC | 10:51 | |
| *** Sebastian_ has joined #buildstream | 11:13 | |
| *** Sebastian_ has quit IRC | 11:16 | |
| *** Sebastian_ has joined #buildstream | 11:18 | |
| gitlab-br-bot | buildstream: merge request (380-untagged-bst->master: Resolve "Uncaught ValueError when tags are not pulled into local repository") #491 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/491 | 11:21 |
| *** Sebastian_ has quit IRC | 11:42 | |
| *** Sebastian has joined #buildstream | 11:42 | |
| *** Sebastian has quit IRC | 11:46 | |
| *** Sebastian has joined #buildstream | 11:47 | |
| gitlab-br-bot | buildstream: merge request (135-expire-artifacts-in-local-cache->master: WIP: Resolve "Expire artifacts in local cache") #347 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/347 | 12:21 |
| gitlab-br-bot | buildstream: merge request (135-expire-artifacts-in-local-cache->master: WIP: Resolve "Expire artifacts in local cache") #347 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/347 | 12:29 |
| *** bethw has joined #buildstream | 12:38 | |
| gitlab-br-bot | buildstream: merge request (135-expire-artifacts-in-local-cache->master: WIP: Resolve "Expire artifacts in local cache") #347 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/347 | 12:46 |
| gitlab-br-bot | buildstream: merge request (135-expire-artifacts-in-local-cache->master: WIP: Resolve "Expire artifacts in local cache") #347 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/347 | 12:50 |
| gitlab-br-bot | buildstream: merge request (328-support-for-downloading-sources-from-mirrors->master: WIP: Resolve "Support for downloading sources from mirrors") #404 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/404 | 13:01 |
| gitlab-br-bot | buildstream: merge request (valentindavid/331_include->master: WIP: Add support for include in project.conf) #471 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/471 | 13:14 |
| gitlab-br-bot | buildstream: merge request (valentindavid/331_include->master: WIP: Add support for include in project.conf) #471 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/471 | 13:28 |
| finn | Competition time | 13:31 |
| finn | What would you call the application side of BuildGrid i.e the cli | 13:31 |
| finn | At the moment it's just called BuildGrid App but that's a bit boring | 13:31 |
| * skullman would call it landis or gyr as a pun on the name of the company that makes my electricity meter and a pun on it being the power grid | 13:32 | |
| * tlater wonders if having more names just makes things confusing | 13:36 | |
| tlater | We already have at least 3 applications here, do we really need to name another thing? | 13:36 |
| *** xjuan has joined #buildstream | 13:39 | |
| *** Sebastian has quit IRC | 13:52 | |
| *** Sebastian has joined #buildstream | 13:53 | |
| gitlab-br-bot | buildstream: merge request (328-support-for-downloading-sources-from-mirrors->master: Resolve "Support for downloading sources from mirrors") #404 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/404 | 13:55 |
| *** Sebastian has quit IRC | 13:56 | |
| jonathanmaw | valentind: I've got my side of source mirroring in a complete solution (even if I doubt it'll be acceptable upstream in this state) now | 13:57 |
| jonathanmaw | https://gitlab.com/BuildStream/buildstream/merge_requests/404 | 13:57 |
| *** Sebastian has joined #buildstream | 13:57 | |
| jonathanmaw | valentind: have I accurately described your suggested API change to remove the need for hacks? | 13:58 |
| finn | tlater, I've made a small cli for the BuildGrid library. When it launches, it currently just says "BuildGrid App" which I found a little bland | 13:58 |
| finn | but I could just keep it as that | 13:58 |
| * tlater thinks "BuildGrid App" will be best, if a bit bland, at least it doesn't add to the potential confusion between BuildStream/BuildGrid/BuildWhateverOurClientWillBeCalled | 13:59 | |
| * finn wants something exciting like BuildGrid Ultimate Dyson Sphere | 14:02 | |
| skullman | bg-cli | 14:02 |
| valentind | jonathanmaw, OK. I will have a look. | 14:03 |
| finn | bg-dyson | 14:04 |
| jonathanmaw | ta valentind | 14:05 |
| tlater | finn: Are we starting a collectible trading card build system? | 14:10 |
| finn | tlater, adding it to BuildGrid issues now and assigning you | 14:11 |
| tlater | Hrm, my os.statvfs calls aren't returning the full struct they are supposed to according to the python documentation | 14:15 |
| tlater | They aren't even returning the struct `man` tells me they should | 14:15 |
| tlater | But *only* when I'm running the python test cases - running this in a shell manually on the same directory works just fine | 14:16 |
| tlater | Any idea what could be causing this? | 14:16 |
| *** Sebastian has quit IRC | 14:16 | |
| *** Sebastian has joined #buildstream | 14:18 | |
| *** Sebastian has quit IRC | 14:21 | |
| *** Sebastian has joined #buildstream | 14:23 | |
| *** Sebastian has quit IRC | 14:57 | |
| *** Phil has quit IRC | 14:58 | |
| *** Sebastian has joined #buildstream | 15:02 | |
| jonathanmaw | e]] | 15:16 |
| jonathanmaw | oops | 15:16 |
| * tlater really wonders how jonathanmaw produced that key stroke | 15:16 | |
| jonathanmaw | jennis: Can you try closing the merge request https://gitlab.com/BuildStream/bst-external/merge_requests/25 and try opening a new one? | 15:17 |
| jennis | sure | 15:17 |
| jonathanmaw | I was able to get the CI to work when I pushed a new branch https://gitlab.com/BuildStream/bst-external/commits/jonathan/test | 15:17 |
| tlater | Oh, if this sort of thing happens I suggest clearing the CI cache | 15:18 |
| jennis | jonathanmaw, done | 15:30 |
| * jonathanmaw mashes button | 15:31 | |
| *** jennis has left #buildstream | 15:31 | |
| *** jennis has joined #buildstream | 15:32 | |
| adds68 | juergbi, hey if the bst-artifact-server fails to start, where is best to look for the logs? | 15:39 |
| *** Prince781 has joined #buildstream | 15:39 | |
| juergbi | adds68: I think it's all in stderr | 15:41 |
| juergbi | only warnings/errors, though. there is no access log or anything like that at the moment | 15:42 |
| juergbi | but if it fails to start, I would expect a message on stderr | 15:42 |
| adds68 | juergbi, i am trying to start it via systemd, so don't get a very good error | 15:43 |
| adds68 | Just "Failed to determine user credentials: No such process" | 15:44 |
| juergbi | systemd should redirect stderr to the journal | 15:44 |
| juergbi | hm, that's a systemd error, might you be missing the user that you specified in the .service file? | 15:45 |
| juergbi | not sure why it says No such process, though | 15:45 |
| tlater | adds68: `journalctl -xe --unit <unit>.service` should help when debugging your unit | 15:45 |
| tlater | It should show exact errors | 15:46 |
| adds68 | tlater, awesome thank you and juergbi i will check that now | 15:46 |
| adds68 | juergbi, correct, type in the user name :) | 15:47 |
| adds68 | typo** haha! | 15:47 |
| tlater | ooi, is this a new fancy CAS server, or the old pushreceive script? | 15:48 |
| tlater | Do you do some cute systemd-socket based thing to run that script when someone opens an ssh connection? | 15:49 |
| adds68 | tlater, we did do that | 15:49 |
| adds68 | tlater, oh no we just ran the script when someone sshd in :) | 15:50 |
| adds68 | tlater, but this is the new shiny CAS | 15:50 |
| tlater | Ah, nice | 15:50 |
| * tlater hadn't realized it landed | 15:50 | |
| tlater | Running an actual daemon is so much nicer than that hack (: | 15:50 |
| adds68 | juergbi, it starts now, however it complains of a missing repo argument, even though i pass it a path to a repo ? | 15:50 |
| adds68 | tlater, ++ | 15:50 |
| tlater | adds68: Might be helpful to share the exact command systemd ends up running | 15:51 |
| adds68 | tlater, 2 ticks | 15:52 |
| adds68 | tlater, juergbi http://termbin.com/q5g8 | 15:54 |
| adds68 | That is my current systemd file | 15:54 |
| juergbi | you don't specify a repo arg... | 15:54 |
| adds68 | juergbi, i specify the repo as stated in the docs? https://gitlab.com/BuildStream/buildstream/blob/juerg/googlecas/doc/source/artifacts.rst | 15:55 |
| adds68 | "Serve cache over HTTPS" section | 15:55 |
| juergbi | the systemd file doesn't have that argument | 15:55 |
| juergbi | the last one is the server cert | 15:55 |
| tlater | adds68: Note the repo isn't an option but just a file path at the end | 15:56 |
| tlater | Might overlook it if you parse lines the way my mind does... | 15:57 |
| adds68 | juergbi, tlater *sigh* how did i not spot that | 15:57 |
| adds68 | I have 2 .service files and i'm comparing them both, one does have the option, the other doesn't | 15:57 |
| tlater | Come to think of it, actually, maybe we'd like to distribute one of these service files with the server, juergbi? | 15:57 |
| tlater | Some projects do, I always find it helpful when I try to run something without docker | 15:58 |
| juergbi | yes, this might indeed make sense given the prevalence of systemd | 15:58 |
| tlater | adds68: Can I create an issue and kindly ask you to contribute yours when you get a reasonably working one? | 15:58 |
| juergbi | that would be great | 15:58 |
| adds68 | tlater, yes certainly :) | 15:58 |
| *** Sebastian has quit IRC | 16:06 | |
| *** Sebastian has joined #buildstream | 16:09 | |
| gitlab-br-bot | buildstream: issue #419 ("Distribute a systemd `.service` file with the new CAS server") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/419 | 16:09 |
| tlater | adds68: Feel free to assign https://gitlab.com/BuildStream/buildstream/issues/419 to yourself | 16:10 |
| * tlater finds the issue template a little awkward | 16:10 | |
| *** Sebastian has quit IRC | 16:11 | |
| adds68 | ta tlater | 16:11 |
| tlater | adds68: Thank *you* for bringing this up ;) | 16:12 |
| *** Sebastian has joined #buildstream | 16:13 | |
| *** jennis has quit IRC | 16:18 | |
| adds68 | juergbi, i am now seeing permission error when the cache tried to read the cert.pm | 16:41 |
| adds68 | The user that the process is running under belongs to a group that has read permissions on that file | 16:41 |
| *** Sebastian has quit IRC | 16:42 | |
| adds68 | But it still sees permission denied | 16:42 |
| tlater | adds68: Long shot, but have you re-logged that user? | 16:42 |
| juergbi | adds68: maybe missing permission on containing directory? | 16:42 |
| adds68 | tlater, erm no, i could reboot the server | 16:42 |
| tlater | adds68: Only necessary if that user is signed in and you changed their groups | 16:43 |
| adds68 | juergbi, i set it to recurse and the cert.pem file has the correct group set | 16:43 |
| * tlater always forgets | 16:43 | |
| juergbi | for systemd started executable, relogin shouldn't be an issue | 16:44 |
| * tlater thinks so too :) | 16:44 | |
| *** Sebastian has joined #buildstream | 16:45 | |
| adds68 | juergbi, a paste for more context:https://paste.gnome.org/pxc7jcnwo | 16:46 |
| tlater | adds68: I think the directory has to be executable by the group | 16:46 |
| adds68 | tlater, oh really? | 16:46 |
| juergbi | yes | 16:47 |
| * tlater reflexively uses 755 when permissioning directories for some reason, and assumes that this is PTSD from fighting this issue before | 16:47 | |
| * adds68 tries that | 16:48 | |
| adds68 | No luck =/ | 16:50 |
| adds68 | It has been set, as i can see executable on the directory now | 16:50 |
| tlater | adds68: If you *really* want to debug this, use `su` to try and manually walk into that directory as that user | 16:50 |
| tlater | That should make it obvious which permission you don't have | 16:51 |
| adds68 | tlater, oo ok, good idea | 16:51 |
| tlater | Any parent directories will also need executable permissions and suchlike, so it's hard to tell from just an ls -l :) | 16:51 |
| adds68 | tlater, hm yes it seems artifacts gets permission denied at certs/ | 16:55 |
| tlater | Lemme guess, no g+x? | 16:55 |
| adds68 | tlater, ah no it seems the directory above it does not have the group set /o\ | 16:56 |
| tlater | Oh, haha | 16:56 |
| adds68 | tlater, \o/ :) | 16:57 |
| *** tristan has joined #buildstream | 17:00 | |
| tlater | Nexus: 2) Could you explain how exactly that error caused the hang? | 17:07 |
| tlater | Because we'd like to make sure this doesn't cause a situation in which debugging becomes difficult | 17:08 |
| tlater | Making a mistake and seeing an error is reasonable, not being able to see the error isn't. | 17:08 |
| Nexus | tlater: So it seems that, if you attempt to change directory, to one that later may not exist, the test tries to do something like "os.getcwd()" and hangs | 17:10 |
| Nexus | in my case, i was changing the directory that i was in, when making a repo inside the buildstream/tmp/ dir. And left it there, when i told it do go back to its previous directory, the hang stopped | 17:10 |
| tlater | Nexus: Pretty sure that os.getcwd() doesn't cause the hang | 17:10 |
| tlater | But the fact that os.getcwd() raises an exception in an unfortunate context | 17:10 |
| tlater | Which then causes buildstream to wait for a child process until the heat death of the universe | 17:10 |
| tlater | If you can determine the exact unfortunate context in which that happens, we can maybe do something about the hang | 17:11 |
| tlater | If not, there's not much we can do. | 17:11 |
| tlater | From my previous explorations into these test hangs, we can't do much but capture more and more corner cases around subprocesses | 17:11 |
| tlater | This isn't a front-end issue, luckily, so at worst causes a dev or two a bit of a headache for a while | 17:12 |
| tlater | And mind-share on how to debug hangs is becoming better, so it's not as critical anymore | 17:13 |
| * tlater wonders if there's a way to make pytest timeout on a test | 17:18 | |
| Nexus | that'd be nice | 17:19 |
| tlater | We might want to invest into this: https://pypi.org/project/pytest-timeout/ | 17:20 |
| tlater | It would make hangs debuggable | 17:20 |
| * tlater writes up another issue | 17:21 | |
| tlater | One of those days... | 17:21 |
| *** finn has quit IRC | 17:27 | |
| gitlab-br-bot | buildstream: issue #420 ("Make hanging pytest tests more debuggable") changed state ("opened") https://gitlab.com/BuildStream/buildstream/issues/420 | 17:29 |
| tlater | Nexus: ^^ will be of interest :) | 17:30 |
| *** bethw has quit IRC | 17:39 | |
| *** Sebastian has quit IRC | 17:54 | |
| *** Sebastian has joined #buildstream | 17:56 | |
| *** bethw has joined #buildstream | 18:08 | |
| gitlab-br-bot | buildstream: merge request (tristan/untagged-bst->master: Resolve "Uncaught ValueError when tags are not pulled into local repository") #496 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/496 | 18:12 |
| *** bethw has quit IRC | 18:12 | |
| gitlab-br-bot | buildstream: merge request (380-untagged-bst->master: Resolve "Uncaught ValueError when tags are not pulled into local repository") #491 changed state ("closed"): https://gitlab.com/BuildStream/buildstream/merge_requests/491 | 18:14 |
| *** toscalix has quit IRC | 18:16 | |
| *** toscalix has joined #buildstream | 18:16 | |
| gitlab-br-bot | buildstream: merge request (tristan/untagged-bst->master: Resolve "Uncaught ValueError when tags are not pulled into local repository") #496 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/496 | 18:17 |
| *** toscalix has quit IRC | 18:18 | |
| tristan | So I have a cute little way to run `bst` commands, capture their stdout/stderr with colors forced enabled, mark it up with html, and include it in the docs | 18:24 |
| tristan | Wonder if I should regenerate those html files every time we build the docs by running the said commands against the docs/examples projects | 18:24 |
| *** bethw has joined #buildstream | 18:25 | |
| albfan[m] | tristan: Nice | 18:27 |
| *** bethw has quit IRC | 18:29 | |
| tristan | albfan[m], I think the ToC is less noisy now too: http://buildstream.gitlab.io/buildstream/ | 18:40 |
| gitlab-br-bot | buildstream: issue #380 ("Uncaught ValueError when tags are not pulled into local repository") changed state ("closed") https://gitlab.com/BuildStream/buildstream/issues/380 | 18:51 |
| gitlab-br-bot | buildstream: merge request (tristan/untagged-bst->master: Resolve "Uncaught ValueError when tags are not pulled into local repository") #496 changed state ("merged"): https://gitlab.com/BuildStream/buildstream/merge_requests/496 | 18:51 |
| *** Prince781 has quit IRC | 18:55 | |
| *** Prince781 has joined #buildstream | 18:55 | |
| *** tristan has quit IRC | 18:58 | |
| *** Sebastian has quit IRC | 19:00 | |
| *** Sebastian has joined #buildstream | 19:02 | |
| *** Sebastian has quit IRC | 19:03 | |
| *** Sebastian has joined #buildstream | 19:04 | |
| *** tristan has joined #buildstream | 19:05 | |
| *** Sebastian has quit IRC | 19:08 | |
| *** Sebastian has joined #buildstream | 19:09 | |
| albfan[m] | tristan: Simple links, super! | 19:13 |
| *** Prince781 has quit IRC | 19:14 | |
| *** Prince781 has joined #buildstream | 19:15 | |
| *** Prince781 has quit IRC | 19:18 | |
| *** Prince781 has joined #buildstream | 19:19 | |
| *** Prince781 has joined #buildstream | 19:21 | |
| *** jsgrant has quit IRC | 19:24 | |
| *** Prince781 has quit IRC | 19:25 | |
| *** Prince781 has joined #buildstream | 19:26 | |
| *** xjuan_ has joined #buildstream | 19:27 | |
| *** xjuan has quit IRC | 19:29 | |
| *** toscalix has joined #buildstream | 19:40 | |
| *** toscalix has quit IRC | 19:41 | |
| *** jonathanmaw has quit IRC | 19:45 | |
| *** Sebastian has quit IRC | 19:53 | |
| *** Sebastian has joined #buildstream | 19:54 | |
| *** finn has joined #buildstream | 20:34 | |
| *** xjuan_ has quit IRC | 20:56 | |
| *** Sebastian has quit IRC | 20:58 | |
| *** Sebastian has joined #buildstream | 21:01 | |
| *** Sebastian has quit IRC | 21:04 | |
| *** Sebastian has joined #buildstream | 21:06 | |
| *** tristan has quit IRC | 21:08 | |
| *** Prince781 has quit IRC | 21:14 | |
| *** Sebastian has quit IRC | 21:18 | |
| *** Sebastian has joined #buildstream | 21:21 | |
| *** tristan has joined #buildstream | 21:29 | |
| *** Sebastian has quit IRC | 21:46 | |
| *** Sebastian has joined #buildstream | 22:05 | |
| *** Sebastian has joined #buildstream | 22:08 | |
| *** Sebastian has quit IRC | 22:10 | |
| *** Sebastian has joined #buildstream | 22:11 | |
| *** Sebastian has quit IRC | 22:14 | |
| *** Sebastian has joined #buildstream | 22:15 | |
| gitlab-br-bot | buildstream: merge request (jjardon/host_deps->master: Document Buildstream's plugins host packages dependencies) #495 changed state ("opened"): https://gitlab.com/BuildStream/buildstream/merge_requests/495 | 22:54 |
| *** Sebastian has quit IRC | 23:14 | |
| *** Sebastian has joined #buildstream | 23:16 | |
| *** Sebastian has quit IRC | 23:19 | |
| *** Sebastian has joined #buildstream | 23:20 | |
| *** tristan has quit IRC | 23:32 | |
| *** cs_shadow has quit IRC | 23:35 | |
| *** tristan has joined #buildstream | 23:38 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!