IRC logs for #buildstream for Wednesday, 2019-10-09

« Tuesday, 2019-10-08 Index Thursday, 2019-10-10 »
*** benschubert has quit IRC00:03
*** narispo has quit IRC05:30
*** narispo has joined #buildstream05:30
*** tejas has joined #buildstream06:46
*** slaf_ has joined #buildstream07:02
*** slaf_ has joined #buildstream07:02
*** slaf has quit IRC07:04
*** slaf_ is now known as slaf07:04
*** benschubert has joined #buildstream07:45
gitlab-br-botBenjaminSchubert opened (was WIP) MR !1631 (bschubert/fix-overnight->master: .gitlab-ci.yml: Bump versions for overnight tests) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163107:52
benschuberttlater[m]: we managed to fix !1142 right? I can close now?08:15
gitlab-br-botMR !1142: Increase read buffer size to improve performance https://gitlab.com/BuildStream/buildstream/merge_requests/114208:15
tlater[m]benschubert: I don't actually know - not seen the issue before08:44
benschubertoups not that one sorry08:44
benschubertI meant #114208:45
gitlab-br-botIssue #1142: Autocompletion of artifacts is throwing stack traces https://gitlab.com/BuildStream/buildstream/issues/114208:45
gitlab-br-botjjardon closed issue #1090 (overnigth tests are failing: 'CheckForbiddenElement' object has no attribute 'node_validate') on buildstream https://gitlab.com/BuildStream/buildstream/issues/109008:47
gitlab-br-botjjardon merged MR !1631 (bschubert/fix-overnight->master: .gitlab-ci.yml: Bump versions for overnight tests) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163108:47
benschubertjjardon: poor marge ;)08:48
gitlab-br-botBenjaminSchubert closed issue #1142 (Autocompletion of artifacts is throwing stack traces) on buildstream https://gitlab.com/BuildStream/buildstream/issues/114209:20
*** traveltissues has joined #buildstream09:21
*** traveltissues has quit IRC09:22
*** lachlan has joined #buildstream09:31
*** lachlan has quit IRC09:33
*** lachlan has joined #buildstream09:33
*** lachlan has quit IRC09:42
*** jonathanmaw has joined #buildstream09:45
*** lachlan has joined #buildstream09:46
jjardonbenschubert: apologies. You would think I should remember about her when we are using it in fdsdk since the beginning :)09:55
benschuberthahaha, no worries, when it's not busy it;s not a problem09:56
*** lachlan has quit IRC10:02
*** traveltissues has joined #buildstream10:03
*** tristan has joined #buildstream10:18
*** lachlan has joined #buildstream10:23
*** lachlan has quit IRC10:34
*** lachlan has joined #buildstream11:03
benschuberttraveltissues: could you double check that !1634 fixes #1158 for you too?11:19
gitlab-br-botIssue #1158: device failures do not cause buildstream failures https://gitlab.com/BuildStream/buildstream/issues/115811:19
traveltissuesi will have a look11:21
milloniwill buildstream ever be in ubuntu repos?11:23
millonioh it is in 19.0411:24
*** lachlan has quit IRC11:35
*** lachlan has joined #buildstream11:39
*** lachlan has quit IRC11:42
*** lachlan has joined #buildstream12:14
*** lachlan has quit IRC12:25
*** doras has quit IRC12:30
*** connorshea[m] has quit IRC12:31
*** kailueke[m] has quit IRC12:31
*** gimpnet-irc[m] has quit IRC12:31
*** dbuch has quit IRC12:31
*** dineshdb[m] has quit IRC12:31
*** cgmcintyre[m] has quit IRC12:31
*** krichter[m] has quit IRC12:31
*** tchaik[m] has quit IRC12:31
*** ssssam[m] has quit IRC12:31
*** mrmcq2u[m] has quit IRC12:31
*** rafaelff[m] has quit IRC12:31
*** theawless[m] has quit IRC12:32
*** verdre[m] has quit IRC12:32
*** skullone[m] has quit IRC12:32
*** jjardon[m] has quit IRC12:32
*** albfan[m] has quit IRC12:32
*** mattiasb has quit IRC12:32
*** waltervargas[m] has quit IRC12:32
*** pro[m] has quit IRC12:33
*** abderrahim[m] has quit IRC12:33
*** awacheux[m] has quit IRC12:34
*** m_22[m] has quit IRC12:35
*** Trevinho[m] has quit IRC12:35
*** nielsdg has quit IRC12:35
*** asingh_[m] has quit IRC12:35
*** connorshea[m] has joined #buildstream12:44
*** tlater[m] has quit IRC12:50
gitlab-br-bottraveltissues opened (was WIP) MR !1632 (traveltissues/strictxfail->master: setup.cfg: make xfails strict by default) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163212:55
*** reuben640[m] has quit IRC12:56
traveltissueslgtm benschubert13:03
benschubertawesome, to marge then :)13:04
gitlab-br-botmarge-bot123 closed issue #1158 (device failures do not cause buildstream failures) on buildstream https://gitlab.com/BuildStream/buildstream/issues/115813:05
gitlab-br-botmarge-bot123 merged MR !1634 (bschubert/fuse-permissions->master: _fuse/mount.py: Monitor the fuse process while waiting for the mount) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163413:05
*** dineshdb[m] has joined #buildstream13:05
*** narispo has quit IRC13:05
*** narispo has joined #buildstream13:05
*** cgmcintyre[m] has joined #buildstream13:14
*** krichter[m] has joined #buildstream13:19
*** lachlan has joined #buildstream13:31
*** lachlan has quit IRC13:39
*** narispo has quit IRC13:41
*** narispo has joined #buildstream13:41
*** narispo has quit IRC14:00
*** narispo has joined #buildstream14:00
gitlab-br-botBenjaminSchubert approved MR !1632 (traveltissues/strictxfail->master: setup.cfg: make xfails strict by default) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163214:13
*** pro[m] has joined #buildstream14:15
*** lachlan has joined #buildstream14:18
*** awacheux[m] has joined #buildstream14:34
*** waltervargas[m] has joined #buildstream14:43
gitlab-br-botBenjaminSchubert opened MR !1635 (bschubert/partial-source-cache->master: Gracefully fallback to fetching source if remote doesn't have every blob cached) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163514:49
benschubertjuergbi: ^ I think you might have the most context around that if you have time for a review14:57
juergbiwill take a look14:58
*** bochecha_ has joined #buildstream15:00
*** Trevinho[m] has joined #buildstream15:01
*** bochecha has quit IRC15:01
*** bochecha_ is now known as bochecha15:01
*** narispo has quit IRC15:09
*** narispo has joined #buildstream15:13
*** narispo has quit IRC15:28
*** narispo has joined #buildstream15:28
*** theawless[m] has joined #buildstream15:28
gitlab-br-botjuergbi opened issue #1166 (Consolidate code duplicated in ArtifactCache and SourceCache) on buildstream https://gitlab.com/BuildStream/buildstream/issues/116615:37
*** verdre[m] has joined #buildstream15:38
gitlab-br-botjuergbi approved MR !1635 (bschubert/partial-source-cache->master: Gracefully fallback to fetching source if remote doesn't have every blob cached) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163515:38
benschubertthanks juergbi !15:38
*** bochecha_ has joined #buildstream15:56
*** skullone[m] has joined #buildstream15:56
*** bochecha has quit IRC15:59
*** bochecha_ is now known as bochecha15:59
benschubertjuergbi: is that expected: https://gitlab.com/BuildStream/buildstream/blob/master/src/buildstream/_sourcecache.py#L143-163 and https://gitlab.com/BuildStream/buildstream/blob/master/src/buildstream/_artifactcache.py#L162-176 ? I would think the source cache implementation is the correct one, right?16:14
juergbibenschubert: the deep check appears to indeed be required for sources given how it's used16:18
juergbifor artifacts it's currently only used for dead code (incremental workspace build), afaict16:18
juergbimaking it consistent certainly sounds sensible, though16:19
benschubertSo taking the source implementation and moving it to _basecache?16:19
juergbibenschubert: more or less, yes (proto handling appears to be different)16:22
benschubertthat's something I'd like to fix too :)16:22
juergbibenschubert: there are some conceptual differences, though16:24
juergbie.g., while source is essentially just a directory (although wrapped in a proto), artifacts have multiple parts, and not all have to exist16:25
benschubertgood point16:25
benschubertyeah might not make sense for this one then16:25
juergbinot sure whether we'll keep the source proto long term in light of the Fetch API proposal16:26
*** jjardon[m] has joined #buildstream16:26
gitlab-br-botBenjaminSchubert opened MR !1636 (bschubert/unify-caches->master: Unify part of the artifact/source caches) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163616:31
benschubertjuergbi: ^ a first draft of unification if you are interested. I think doing more than that is more involved, we can do it but I'm not sure whether it's worth the effort16:32
juergbibenschubert: you consider the directory fetch part to be outside the scope?16:33
benschubertoh right forgot that one16:34
benschubertlet me add this16:34
gitlab-br-bottraveltissues opened (was WIP) MR !1632 (traveltissues/strictxfail->master: setup.cfg: make xfails strict by default) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163216:39
*** lachlan has quit IRC16:42
*** albfan[m] has joined #buildstream16:43
benschubertjuergbi: https://gitlab.com/BuildStream/buildstream/blob/master/src/buildstream/_artifactcache.py#L652 is that the other part of 'fetch' you had in mind? If we move it to be the same as the one in source cache, it will be incredibly verbose. I'm actually not sure it's worth consolidating16:52
*** mattiasb has joined #buildstream16:52
ironfootcan I patch the source of an element in a junction?16:54
benschubertironfoot: you can patch a junction, so you can add a patch to a junction in order to patch an element in the junction (not sure if that's clear)16:57
ironfootyes, exactly what I'm doing currently16:57
ironfootbenschubert: thanks for confirming though :)16:57
benschubertno problem. I don't have any simpler way but I do use it extensively :)16:58
ironfootsound like a common usecase for you then, maybe possible to add support for it?16:58
benschubertnot sure we could do much in BuildStream itself "easily" and not sure it's a very common use case either17:00
*** lachlan has joined #buildstream17:00
juergbibenschubert: we should at least consolidate the part that fetches directory with blobs (i.e., equivalent to the nested __pull_digest) and ensures it will always raise BlobNotFound if blob is not found, not RpcError with NOT_FOUND17:04
juergbi(and that should be in CASCache)17:05
benschubertCASCache raises BlobNotFound already17:05
juergbiright, that may be the case now. the NOT_FOUND check is probably a left-over from pre-casd where it was inconsistent (batch vs. non-batch)17:06
*** lachlan has quit IRC17:06
juergbiso I guess we can also defer consolidation until we switch to casd FetchTree17:06
benschubertagreed17:06
juergbibenschubert: however, we're still missing except BlobNotFound in artifactcache, aren't we?17:07
benschubertah that's possible17:07
benschubertII'll write a test for it tomorrow17:07
benschubertand fix it if that's the case :)17:07
*** doras has joined #buildstream17:26
*** m_22[m] has joined #buildstream17:32
*** tiagogomes has quit IRC17:39
gitlab-br-botmarge-bot123 merged MR !1632 (traveltissues/strictxfail->master: setup.cfg: make xfails strict by default) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/163217:40
*** traveltissues has quit IRC17:46
*** jonathanmaw has quit IRC17:52
*** abderrahim[m] has joined #buildstream17:53
*** gimpnet-irc[m] has joined #buildstream17:56
*** kailueke[m] has joined #buildstream18:02
*** dbuch has joined #buildstream18:02
*** raoul has quit IRC18:14
*** narispo has quit IRC18:26
*** narispo has joined #buildstream18:27
*** mohan43u has quit IRC18:53
*** tchaik[m] has joined #buildstream19:07
*** lachlan has joined #buildstream19:10
*** nielsdg has joined #buildstream19:26
*** asingh_[m] has joined #buildstream19:26
*** ssssam[m] has joined #buildstream19:41
*** mrmcq2u[m] has joined #buildstream19:57
*** rafaelff[m] has joined #buildstream20:14
*** tlater[m] has joined #buildstream20:26
*** lachlan has quit IRC20:28
*** reuben640[m] has joined #buildstream20:44
*** narispo has quit IRC20:51
*** narispo has joined #buildstream20:51
*** narispo has quit IRC20:55
*** narispo has joined #buildstream20:55
*** narispo has quit IRC21:10
*** narispo has joined #buildstream21:11
jjardonWould anyone mind if we change the name of bst-plugins-experimental repo to bst-plugins-external or similar? Not sure It is a good idea to mix the status of the repo with its name22:29
*** tristan has quit IRC23:15
« Tuesday, 2019-10-08 Index Thursday, 2019-10-10 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!