Friday, 2017-12-08

*** gmann_afk is now known as gmann00:01
*** threestrands has joined #openstack-dev00:01
*** yamamoto has quit IRC00:03
*** charcol has quit IRC00:03
*** achampion has quit IRC00:04
*** nplanel has joined #openstack-dev00:07
*** TxGirlGeek has quit IRC00:11
*** baoli has quit IRC00:13
*** jmlim has quit IRC00:21
*** slaweq has joined #openstack-dev00:22
*** rcernin has quit IRC00:25
*** rcernin has joined #openstack-dev00:25
*** gouthamr has quit IRC00:25
*** slaweq has quit IRC00:26
*** sdague has quit IRC00:29
*** kumarmn has joined #openstack-dev00:31
*** achampion has joined #openstack-dev00:33
*** kumarmn has quit IRC00:35
*** sasheto has quit IRC00:39
*** vnogin has joined #openstack-dev00:45
*** masayukig[m] has joined #openstack-dev00:46
*** vnogin has quit IRC00:49
*** SumitNaiksatam has quit IRC00:51
*** tovin07_ has joined #openstack-dev00:55
*** kumarmn has joined #openstack-dev00:56
*** slaweq has joined #openstack-dev00:57
*** yamamoto has joined #openstack-dev00:59
*** kumarmn has quit IRC01:01
*** gyee has quit IRC01:01
*** slaweq has quit IRC01:02
*** achampion has quit IRC01:02
*** yamamoto has quit IRC01:04
*** scanepa[m] has joined #openstack-dev01:05
*** csmart has joined #openstack-dev01:05
*** retr0h has joined #openstack-dev01:05
*** slaweq_ has joined #openstack-dev01:12
*** kumarmn has joined #openstack-dev01:12
*** sasheto has joined #openstack-dev01:12
*** achampion has joined #openstack-dev01:14
*** esberglu has joined #openstack-dev01:16
*** slaweq_ has quit IRC01:16
*** kumarmn has quit IRC01:16
*** david-lyle has joined #openstack-dev01:18
*** achampion has quit IRC01:19
*** esberglu has quit IRC01:21
*** ajmaidak has joined #openstack-dev01:23
*** mikal has quit IRC01:24
*** Dinesh_Bhor has joined #openstack-dev01:28
*** mikal has joined #openstack-dev01:30
*** slaweq has joined #openstack-dev01:32
*** fragatina has quit IRC01:33
*** david-lyle has quit IRC01:36
*** slaweq has quit IRC01:37
*** tserong has quit IRC01:41
*** vnogin has joined #openstack-dev01:45
*** achampion has joined #openstack-dev01:45
*** fragatina has joined #openstack-dev01:46
*** jmlim has joined #openstack-dev01:49
*** vnogin has quit IRC01:50
*** fragatina has quit IRC01:51
*** dalgaaf has quit IRC01:57
*** dalgaaf has joined #openstack-dev01:59
*** yamamoto has joined #openstack-dev02:00
*** harlowja has quit IRC02:03
*** yamamoto has quit IRC02:04
*** ldl has joined #openstack-dev02:06
*** slaweq has joined #openstack-dev02:12
*** slaweq has quit IRC02:17
*** qlixed has joined #openstack-dev02:28
*** pbourke has quit IRC02:29
*** pbourke has joined #openstack-dev02:31
*** achampio1 has joined #openstack-dev02:45
*** vnogin has joined #openstack-dev02:46
*** zhurong has joined #openstack-dev02:47
*** slaweq has joined #openstack-dev02:48
*** jjmojojjmojo has quit IRC02:50
*** vnogin has quit IRC02:51
*** jjmojojjmojo has joined #openstack-dev02:51
*** slaweq has quit IRC02:52
*** fragatina has joined #openstack-dev03:01
*** yamamoto has joined #openstack-dev03:01
*** yamamoto has quit IRC03:06
*** ekcs has quit IRC03:09
*** MattMan has quit IRC03:11
*** MattMan has joined #openstack-dev03:12
*** slaweq has joined #openstack-dev03:19
*** fragatina has quit IRC03:19
*** crtcji has joined #openstack-dev03:19
*** slaweq has quit IRC03:24
*** slaweq has joined #openstack-dev03:25
*** Dinesh_Bhor has quit IRC03:25
*** SumitNaiksatam has joined #openstack-dev03:29
*** slaweq has quit IRC03:30
*** SumitNaiksatam_ has joined #openstack-dev03:32
*** SumitNaiksatam has quit IRC03:33
*** SumitNaiksatam_ is now known as SumitNaiksatam03:33
*** mriedem has quit IRC03:36
*** bobh has quit IRC03:38
*** vnogin has joined #openstack-dev03:47
*** bobh has joined #openstack-dev03:49
*** vnogin has quit IRC03:51
*** bobh has quit IRC03:53
*** slaweq has joined #openstack-dev03:57
*** slaweq has quit IRC04:02
*** yamamoto has joined #openstack-dev04:03
*** rosmaita has quit IRC04:03
*** Dinesh_Bhor has joined #openstack-dev04:06
*** yamamoto has quit IRC04:06
*** armax has quit IRC04:06
*** yamamoto has joined #openstack-dev04:06
*** armax has joined #openstack-dev04:07
*** armax has quit IRC04:07
*** TuanLA has joined #openstack-dev04:13
*** dhajare has joined #openstack-dev04:14
*** Dinesh_Bhor has quit IRC04:19
*** Dinesh_Bhor has joined #openstack-dev04:19
*** kumarmn has joined #openstack-dev04:26
*** zhurong has quit IRC04:26
*** kumarmn has quit IRC04:30
*** kumarmn has joined #openstack-dev04:36
*** slaweq has joined #openstack-dev04:36
*** qlixed has quit IRC04:39
*** slaweq has quit IRC04:41
*** bobh has joined #openstack-dev04:43
*** kumarmn has quit IRC04:46
*** achampio1 has quit IRC04:46
*** vnogin has joined #openstack-dev04:48
*** ldl has quit IRC04:48
*** bobh has quit IRC04:51
*** vnogin has quit IRC04:53
*** dhajare has quit IRC04:53
*** adreznec has joined #openstack-dev04:54
*** adreznec has quit IRC04:55
*** karthiks has joined #openstack-dev04:56
*** adreznec has joined #openstack-dev04:57
*** sree has joined #openstack-dev05:00
*** dpawar has joined #openstack-dev05:06
*** slaweq has joined #openstack-dev05:07
*** rcernin has quit IRC05:11
*** slaweq has quit IRC05:11
*** slaweq_ has joined #openstack-dev05:13
*** ratailor has joined #openstack-dev05:13
*** slaweq_ has quit IRC05:18
*** dhajare has joined #openstack-dev05:22
*** threestrands has quit IRC05:24
*** ldl has joined #openstack-dev05:25
*** harlowja has joined #openstack-dev05:37
*** threestrands has joined #openstack-dev05:39
*** threestrands has quit IRC05:39
*** threestrands has joined #openstack-dev05:39
*** radhikam has quit IRC05:39
*** slaweq has joined #openstack-dev05:40
*** zhurong has joined #openstack-dev05:40
*** slaweq has quit IRC05:45
*** armax has joined #openstack-dev05:45
*** fragatina has joined #openstack-dev05:52
*** armax has quit IRC05:55
*** armax has joined #openstack-dev05:55
*** armax has quit IRC05:55
*** armax has joined #openstack-dev05:56
*** armax has quit IRC05:56
*** fragatina has quit IRC05:57
*** mikal has quit IRC05:57
*** mikal has joined #openstack-dev05:59
*** armaan has joined #openstack-dev06:01
*** armaan has quit IRC06:02
*** rcernin has joined #openstack-dev06:11
*** sree has quit IRC06:12
*** sree has joined #openstack-dev06:13
*** harlowja has quit IRC06:14
*** slaweq has joined #openstack-dev06:16
*** sree has quit IRC06:17
*** slaweq has quit IRC06:21
*** arcolife has joined #openstack-dev06:22
*** threestrands has quit IRC06:28
*** fragatina has joined #openstack-dev06:29
*** fragatina has quit IRC06:30
*** fragatina has joined #openstack-dev06:30
*** Hien has quit IRC06:39
*** Hien has joined #openstack-dev06:39
*** sree has joined #openstack-dev06:43
*** zhurong has quit IRC06:47
*** marios has joined #openstack-dev06:52
*** slaweq has joined #openstack-dev06:55
*** arcolife has quit IRC06:57
*** sree has quit IRC06:57
*** sree has joined #openstack-dev06:58
*** cyberworm54__ has joined #openstack-dev07:00
*** slaweq has quit IRC07:00
-openstackstatus- NOTICE: Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning07:01
*** ChanServ changes topic to "Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning"07:01
*** sree has quit IRC07:03
*** cyberworm54_ has quit IRC07:03
*** serverascode has quit IRC07:07
*** Ng has quit IRC07:07
*** Ng has joined #openstack-dev07:07
*** mugsie has quit IRC07:07
*** bgmccollum has quit IRC07:07
*** persia has quit IRC07:07
*** serverascode has joined #openstack-dev07:08
*** onovy has quit IRC07:08
*** Dmitrii-Sh has quit IRC07:08
*** kencjohnston has quit IRC07:08
*** odyssey4me has quit IRC07:08
*** tdasilva has quit IRC07:08
*** vkmc has quit IRC07:08
*** melwitt has quit IRC07:08
*** mugsie has joined #openstack-dev07:08
*** mugsie has quit IRC07:08
*** mugsie has joined #openstack-dev07:08
*** persia has joined #openstack-dev07:08
*** cmurphy has quit IRC07:08
*** nguyentrihai has joined #openstack-dev07:09
*** arcolife has joined #openstack-dev07:09
*** cmurphy has joined #openstack-dev07:09
*** sree has joined #openstack-dev07:10
*** melwitt has joined #openstack-dev07:10
*** vkmc has joined #openstack-dev07:10
*** vkmc has quit IRC07:10
*** vkmc has joined #openstack-dev07:10
*** melwitt is now known as Guest905407:11
*** onovy has joined #openstack-dev07:11
*** nguyentrihai has quit IRC07:11
*** igarcia__ has joined #openstack-dev07:11
*** bgmccollum has joined #openstack-dev07:12
*** odyssey4me has joined #openstack-dev07:12
*** tdasilva has joined #openstack-dev07:13
*** kencjohnston has joined #openstack-dev07:13
*** csmart has quit IRC07:14
*** arcolife has quit IRC07:14
*** masayukig[m] has quit IRC07:14
*** retr0h has quit IRC07:14
*** cyberworm54__ has quit IRC07:15
*** scanepa[m] has quit IRC07:15
*** sree has quit IRC07:17
*** armaan has joined #openstack-dev07:17
*** bandini has quit IRC07:17
*** sree has joined #openstack-dev07:17
*** ajmaidak has quit IRC07:17
*** bandini has joined #openstack-dev07:19
*** ajmaidak has joined #openstack-dev07:20
*** armaan has quit IRC07:21
*** sree has quit IRC07:25
*** arcolife has joined #openstack-dev07:26
*** paramite has joined #openstack-dev07:30
*** slaweq has joined #openstack-dev07:34
*** slaweq has quit IRC07:39
*** makowals has quit IRC07:45
*** crtcji has quit IRC07:45
*** Dmitrii-Sh has joined #openstack-dev07:46
*** slaweq has joined #openstack-dev07:46
*** makowals has joined #openstack-dev07:47
*** sree has joined #openstack-dev07:47
*** cablekevin has joined #openstack-dev07:48
*** slaweq has quit IRC07:50
*** slaweq has joined #openstack-dev07:50
*** slaweq has quit IRC07:51
*** david-lyle has joined #openstack-dev07:52
*** hseipp has joined #openstack-dev07:53
*** slaweq has joined #openstack-dev07:53
*** vnogin has joined #openstack-dev07:54
*** shardy has joined #openstack-dev07:58
*** kobis has joined #openstack-dev08:02
*** slaweq has quit IRC08:12
*** slaweq has joined #openstack-dev08:12
*** zhurong has joined #openstack-dev08:16
*** tesseract has joined #openstack-dev08:20
*** kobis has quit IRC08:23
*** mgoddard has joined #openstack-dev08:26
*** shardy is now known as shardy_afk08:31
*** adreznec has quit IRC08:32
*** adreznec has joined #openstack-dev08:32
*** etingof has quit IRC08:34
*** jpich has joined #openstack-dev08:36
*** armaan has joined #openstack-dev08:45
*** mgoddard has quit IRC08:46
*** shardy_afk is now known as shardy08:48
*** dalgaaf has quit IRC08:51
*** armaan has quit IRC08:53
*** armaan has joined #openstack-dev08:55
*** lucas-afk is now known as lucasagomes08:57
*** nplanel has quit IRC08:57
*** armaan has quit IRC08:57
*** dalgaaf has joined #openstack-dev08:59
*** gszasz has joined #openstack-dev09:00
*** dalgaaf has quit IRC09:04
*** hseipp has quit IRC09:05
*** crtcji has joined #openstack-dev09:06
*** owalsh has quit IRC09:10
*** owalsh has joined #openstack-dev09:12
*** jbadiapa has joined #openstack-dev09:14
*** athomas has joined #openstack-dev09:15
*** dalgaaf has joined #openstack-dev09:15
*** masayukig[m] has joined #openstack-dev09:17
cmurphyNobodyCam: stable/pike is the last version with v209:18
*** mgoddard has joined #openstack-dev09:18
*** etingof has joined #openstack-dev09:19
*** karthiks is now known as karthiks_afk09:20
*** dalgaaf has quit IRC09:20
*** jbadiapa has quit IRC09:21
*** dalgaaf has joined #openstack-dev09:24
*** e0ne has joined #openstack-dev09:27
*** csmart has joined #openstack-dev09:36
*** scanepa[m] has joined #openstack-dev09:36
*** retr0h has joined #openstack-dev09:36
*** Dinesh_Bhor has quit IRC09:37
*** yamamoto has quit IRC09:41
*** vnogin has quit IRC09:41
*** yamamoto has joined #openstack-dev09:43
*** hseipp has joined #openstack-dev09:47
*** gmann is now known as gmann_afk09:48
*** vnogin has joined #openstack-dev09:50
*** tserong has joined #openstack-dev09:56
*** Alex_Staf has joined #openstack-dev09:56
*** jimbaker_ has quit IRC10:01
*** saggi has joined #openstack-dev10:03
*** jimbaker_ has joined #openstack-dev10:04
*** vnogin has quit IRC10:09
*** brault has quit IRC10:11
*** brault has joined #openstack-dev10:11
*** gszasz has quit IRC10:12
*** brault has quit IRC10:12
*** dtantsur|afk is now known as dtantsur10:16
*** Alex_Staf has quit IRC10:16
*** tovin07_ has quit IRC10:23
*** rraja has joined #openstack-dev10:25
*** vnogin has joined #openstack-dev10:25
*** sdague has joined #openstack-dev10:28
*** electrofelix has joined #openstack-dev10:30
*** gszasz has joined #openstack-dev10:31
*** akrivoka has joined #openstack-dev10:40
*** zhurong has quit IRC10:40
*** ldl has quit IRC10:44
*** gszasz has quit IRC10:45
*** ioggstream has joined #openstack-dev10:50
*** achampion has quit IRC10:51
*** tserong has quit IRC10:53
*** achampion has joined #openstack-dev10:55
*** tserong has joined #openstack-dev10:57
*** ldl has joined #openstack-dev11:00
*** karthiks_afk is now known as karthiks11:08
*** yamamoto has quit IRC11:21
*** sree has quit IRC11:25
*** claudiub has joined #openstack-dev11:28
*** jkilpatr has quit IRC11:28
*** tobiajo has quit IRC11:29
*** claudiub2 has joined #openstack-dev11:30
*** gfidente has joined #openstack-dev11:33
*** ioggstream has quit IRC11:35
*** dhajare has quit IRC11:36
*** brault has joined #openstack-dev11:38
*** charcol has joined #openstack-dev11:40
*** tobiajo has joined #openstack-dev11:41
*** ldl has quit IRC11:41
*** _crtcji has joined #openstack-dev11:41
*** crtcji has quit IRC11:44
*** gszasz has joined #openstack-dev11:47
*** yamamoto has joined #openstack-dev11:49
*** tesseract has quit IRC11:50
*** tesseract has joined #openstack-dev11:51
*** vnogin has quit IRC11:52
*** __crtcji has joined #openstack-dev11:56
*** _crtcji has quit IRC11:58
*** cdent has joined #openstack-dev11:59
*** nplanel has joined #openstack-dev12:06
*** ansmith has joined #openstack-dev12:07
*** Exhar has quit IRC12:09
*** claudiub2 has quit IRC12:12
*** claudiub has quit IRC12:14
*** vnogin has joined #openstack-dev12:14
*** _crtcji has joined #openstack-dev12:16
*** Exhar has joined #openstack-dev12:17
*** vnogin has quit IRC12:19
*** __crtcji has quit IRC12:19
*** _crtcji has quit IRC12:20
*** _crtcji has joined #openstack-dev12:22
*** _crtcji has quit IRC12:22
*** jkilpatr has joined #openstack-dev12:23
*** vnogin has joined #openstack-dev12:26
*** TuanLA has quit IRC12:26
*** vnogin has quit IRC12:27
*** efried is now known as fried_rice12:29
*** ratailor has quit IRC12:32
*** yikun_ has joined #openstack-dev12:32
*** bobh has joined #openstack-dev12:34
*** salv-orlando has joined #openstack-dev12:42
*** vnogin has joined #openstack-dev12:48
*** rosmaita has joined #openstack-dev12:51
*** yamamoto has quit IRC12:51
*** makowals has quit IRC12:57
*** ikhan_ has joined #openstack-dev12:58
*** makowals has joined #openstack-dev12:59
*** vnogin has quit IRC13:02
*** ikhan_ has quit IRC13:02
*** ikhan_ has joined #openstack-dev13:02
*** gordc has joined #openstack-dev13:17
*** kumarmn has joined #openstack-dev13:20
*** claudiub has joined #openstack-dev13:20
*** nguyentrihai has joined #openstack-dev13:26
*** jaypipes has joined #openstack-dev13:28
*** salv-orlando has quit IRC13:29
*** jaypipes is now known as leakypipes13:29
*** salv-orlando has joined #openstack-dev13:30
*** salv-orlando has quit IRC13:34
*** edleafe is now known as figleaf13:36
*** nplanel has quit IRC13:36
*** lpetrut has joined #openstack-dev13:39
*** markvoelker has quit IRC13:40
*** arcolife has quit IRC13:41
*** markvoelker has joined #openstack-dev13:43
*** tesseract has quit IRC13:47
*** mriedem has joined #openstack-dev13:49
*** tesseract has joined #openstack-dev13:50
*** yamamoto has joined #openstack-dev13:52
*** jkilpatr has quit IRC13:53
*** achampion has quit IRC13:54
*** yamahata has joined #openstack-dev13:56
*** morazi has joined #openstack-dev13:56
*** yamamoto has quit IRC13:58
*** ChanServ changes topic to "OpenStack development || Support is in #openstack || channel logs http://eavesdrop.openstack.org/irclogs/%23openstack-dev/"13:58
-openstackstatus- NOTICE: The issues have been fixed, Zuul is operating fine again but has a large backlog. You can recheck jobs that failed.13:58
*** makowals has quit IRC13:58
*** pcregut has joined #openstack-dev13:59
*** yamamoto has joined #openstack-dev14:01
*** baoli has joined #openstack-dev14:04
*** makowals has joined #openstack-dev14:04
*** yehowyada has joined #openstack-dev14:05
*** kgiusti has joined #openstack-dev14:09
*** pcregut has left #openstack-dev14:13
*** vnogin has joined #openstack-dev14:23
*** esberglu has joined #openstack-dev14:24
*** dansmith is now known as superdan14:25
*** awaugama has joined #openstack-dev14:41
*** gimps has quit IRC14:42
*** yamahata has quit IRC14:47
*** pbourke has quit IRC14:50
*** Guest9054 is now known as melwitt14:52
*** ihrachys has joined #openstack-dev14:54
*** pbourke has joined #openstack-dev14:55
*** achampion has joined #openstack-dev14:55
*** rcernin has quit IRC14:55
lbragstadcmurphy: so another question i've thought about; is how much does limit validation matter regardless of the model?15:00
lbragstadlike - regardless of the enforcement model, should i be able to set x to 5 on project A and 10 on project B15:01
cmurphylbragstad: you mean if B is a child of A?15:01
lbragstadyeah15:01
lbragstadi guess is there a set of validation rules that are independent of the model15:02
*** slaweq has quit IRC15:02
* lbragstad pulls up https://review.openstack.org/#/c/441203/15:03
*** slaweq has joined #openstack-dev15:03
*** salv-orlando has joined #openstack-dev15:04
lbragstadhttp://logs.openstack.org/03/441203/18/check/build-openstack-sphinx-docs/2d514db/html/specs/keystone/backlog/hierarchical-quota-scenarios.html is easier to read15:04
cmurphylbragstad: if there is no hierarchical enforcement model, then I think there is not enough information to say whether UPDATE B SET limit = 10 is a violation or not15:04
cmurphybecause what happens when we have project C that is a child of B15:05
cmurphyand we try to set its limit to 1015:05
cmurphydoes it fail because its parent is not high enough?15:05
cmurphydoes it fail because its grandparent + its parent is not high enough?15:05
cmurphywe don't have the rules yet to make these decisioins15:06
cmurphythat was my comment on ps16 on https://review.openstack.org/#/c/455709/1615:06
lbragstadcorrect me if i'm wrong, but the only one we seem to have the rules for would be a flat structure http://logs.openstack.org/03/441203/18/check/build-openstack-sphinx-docs/2d514db/html/specs/keystone/backlog/hierarchical-quota-scenarios.html#flatland15:07
lbragstadbecause it literally doesn't have anything to do with a hierarchy15:07
cmurphylbragstad: yes. so i guess the question is whether we can apply a flatland model to a hierarchical project structure and just not worry about the hierarchy for the sake of limits, or do we forbid applying limits to child projects because we don't have a model to support it yet15:09
* lbragstad leans towards the latter15:10
*** signed8bit has joined #openstack-dev15:10
lbragstadif we apply any sort of model to a hierarchy, we've going to be at risk of backwards incompatible changes if we decide to change the validation of the limits in that model, yeah?15:11
lbragstadwe're*15:11
cmurphyi think i agree15:11
*** rraja has quit IRC15:11
lbragstadso - next question, is that direction still useful?15:11
lbragstadsans hierarchies15:12
lbragstadas something that can be accomplished in queens, while trying to nail down an enforcement model15:13
cmurphyyes, it's still useful even for flat structures15:13
lbragstadi agree15:13
cmurphybecause keystone can validate projects15:13
cmurphywhereas nova can't15:13
lbragstadyeah...15:13
lbragstadthe projects consuming this should be checking the limits information for the project in question15:14
cmurphynod15:14
lbragstadwhich should hopefully stay consistent regardless of the project being part of a flat structure or a hierarchy15:14
lbragstadi'm trying to think if services will have additional work to do and how much once the hierarchy is introduced15:15
lbragstador can they implement the usage check + enforcement and just consuming a limit hierarchy once we implement it in keystone15:15
* lbragstad is curious about sdague's take on ^15:16
cmurphywell we'll have that "what information do i need to file a ticket" question15:17
cmurphyhttps://review.openstack.org/#/c/455709/17/specs/keystone/queens/limits-api.rst@78915:18
*** tssurya has joined #openstack-dev15:20
lbragstadmmm15:20
lbragstadat line 80615:20
lbragstadso nova would have to relay some information to the user based on the hierarchy15:21
*** ioggstream has joined #openstack-dev15:22
*** tssurya has quit IRC15:23
lbragstadi guess that doesn't change building out the enforcement in each service thoguh15:23
lbragstadwhich is a good thing15:23
*** armax has joined #openstack-dev15:24
*** gfidente has quit IRC15:25
*** radhikam has joined #openstack-dev15:25
*** rraja has joined #openstack-dev15:27
*** claudiub has quit IRC15:28
sdaguelbragstad: trying to play catch up here15:28
*** vnogin has quit IRC15:28
sdaguelbragstad: if I understand the question correctly it is:15:29
sdagueWill projects need to understand the hierarchical model to do quota enforcement?15:29
sdagueAnd the answer as best I could work out, was "it depends on the model, but probably yes"15:29
sdagueWhich is why there was a suggestion in the spec that hierarchical models, limit changes, and quota enforcement, are implemented in a common library, and referenced symbolically, so that logic is handled outside of the project15:31
lbragstadsdague: ack15:31
sdagueAnd, depending on the model, you might need the topology of the hierarchy. Especially on any of that "overcommit" approaches15:31
lbragstadahh15:32
lbragstadi forgot about that one, and the overcommit approach makes it pretty clear that the service needs to know more about the hierarchy i suppose15:32
lbragstadsdague: the tl;dr of what we're running into with the spec is that we're not sure how much validation of limits in a hierarchy we can do15:33
lbragstadcc kmalloc cmurphy ^15:33
*** makowals has quit IRC15:34
*** salv-orlando has quit IRC15:34
*** salv-orlando has joined #openstack-dev15:35
*** vnogin has joined #openstack-dev15:35
lbragstadi was thinking we could try and implement validation of project limits regardless of the project being in a hierarchy, but cmurphy raised a couple concerns there15:35
cmurphywithout having a defined a model yet, there is no rule that even says child B can't have limit greater than parent A15:36
sdaguevalidating changing the limits?15:36
cmurphyregardless of how unintuitive it is to think about it15:36
cmurphysdague: right15:36
sdaguecmurphy: right, for sure, you need to actually have a model and the rules for it defined15:36
sdaguethe scenarios was like a 10% beginning of that work15:37
cmurphyright15:37
cmurphybaby steps15:37
lbragstadso - before we can implement limits in a hierarchy, we'll need to really flesh those models out15:37
sdaguewell, I would pick one hierarchical model and do an implementation on it15:38
sdagueeven if it turns out to be one no one really will use15:38
sdaguethat will give you something to plumb the limits setting code against15:38
lbragstadok - so keystone's limit validation will be dependent on that model, and service usage will be dependent on that model, right?15:39
sdagueThe Stanek model15:39
sdagueis probably a good one to flesh out lots of things15:39
sdaguelbragstad: right, it was envisioned that you basically set the model name for the deployment in keystone (config) and it's the source of truth for all services as well15:39
cmurphysdague: you think that implementation belongs in this first spec?15:39
sdaguebecause they all have to be operating under the same assumptions15:39
*** salv-orlando has quit IRC15:40
lbragstadsdague: right - otherwise i can't imagine the confusion you'd end up with15:40
sdaguecmurphy: I don't know. I'm torn on whether the limits in keystone API is useful enough without any hierarchy models existing15:40
*** hseipp has quit IRC15:40
lbragstadi was under the assumption that the services would also have a configuration option that deployers set that specifies enforcement15:41
sdaguebut, you all know the code and architecture better, so you probably have better thoughts15:41
cmurphyI think it is still useful15:41
sdaguelbragstad: it could, but if you desynchronize for any reason, everything goes crazy15:41
cmurphybut there's also the risk that we'll implement somethign and then realize it will have to be totally different once there's hierarchy15:41
lbragstadsdague: right - that's seems scary, i like you're idea better15:41
lbragstads/you're/your/15:41
sdagueso, it honestly might make more sense to make it so that services in their process of querying keystone for limit info, get the model info as well15:42
cmurphyyes15:42
lbragstad++15:42
sdagueso they can't desynchronize15:42
lbragstadbased on whatever that is, the service does enforcement15:42
sdagueyep15:42
lbragstadthat'd make things a lot easier for deployers, too15:42
cmurphythe model definitely needs to be discoverable, otherwise we have interop problems15:42
sdagueyep15:43
*** charcol has quit IRC15:43
lbragstadyeah15:43
*** jloehel_ has quit IRC15:44
lbragstadso the question now become, what model do we implement first? or do we implement things for a flat structure and push a model implementation to Rocky?15:44
lbragstadbecomes*15:45
*** hseipp has joined #openstack-dev15:45
*** paramite has quit IRC15:45
cmurphyif we stick to our spec deadline we won't be able to flesh out the hierarchy stuff in time15:46
lbragstadright - not without an exception15:46
lbragstadi think it'd be pretty easy to rework what is there to fit a flat structure15:46
lbragstadand target it to queens15:46
lbragstadthe implementation would introduce registered limits and project limits for projects that don't have parents or children15:47
*** shardy has quit IRC15:47
*** felipemonteiro_ has joined #openstack-dev15:47
NobodyCamcmurphy: Awesome, Perfect, Thank you :)15:48
cmurphylbragstad: what happens if someone tries to define a limit on a project in a tree?15:48
cmurphyNobodyCam: np :)15:48
NobodyCam:)15:48
lbragstadcmurphy: validation error?15:48
lbragstadcmurphy: say we don't support limits in a project hierarchy yet?15:49
cmurphylbragstad: so then what happens when a hierarchical model is introduced and its not invalid any more?15:49
lbragstadyeah... that'd be a 4XX -> 2XX15:50
* lbragstad sigh15:50
cmurphythe user would have to opt in to the new behavior somehow15:50
kmalloclbragstad: 500 error! then it can be 500->200 /s15:51
*** paramite has joined #openstack-dev15:51
cmurphylol15:51
*** felipemonteiro__ has joined #openstack-dev15:51
lbragstad:)15:52
kmallocso... we can abuse http a little here15:52
kmallocwe could 501 (don't hurt me, i know it's the wrong "not implemented")15:52
kmallochonestly, i think we have to hit the model at the same time =/15:53
kmallocbecause... keystone has no other way to handle w/o breaking API contracts15:53
lbragstadwhat if switching from model to model also breaks API contracts?15:53
cmurphywe kind of need like microversions-lite just for /limits :P15:53
*** slaweq has quit IRC15:53
kmallocwe could do the same thing as a versionless auth (similar to microversions lite)15:54
kmallocjust make the user ask for the model they want.15:55
kmallocpick a sane-ish default [or don't]15:55
cmurphyuser can't choose the model15:55
cmurphyadmin chooses the model15:55
*** felipemonteiro_ has quit IRC15:55
cmurphybut we can make it discoverable with like GET /limits/model15:55
sdagueright, so I think the previous cheat was pick "flatland" as the default model15:55
sdaguewhich is today's behavior15:56
kmallocor....15:56
sdaguefor everyone except cinder15:56
*** openstackgerrit has joined #openstack-dev15:56
openstackgerritMonty Taylor proposed openstack/governance master: Shift tripleo-ci from infra to tripleo  https://review.openstack.org/52671515:56
kmallochm.15:56
*** gimps has joined #openstack-dev15:56
*** gimps has joined #openstack-dev15:56
kmallocgod this feels like such a bad API design.15:56
lbragstadok - so if we implement flatland15:56
lbragstadthen 4XX is acceptable for trying to set a limit on a project in a hierarchy15:57
lbragstadbut if we come out with a model that allows that later15:57
lbragstadthat's fine?15:57
sdaguelbragstad: well flatland just means that models ignore parents15:57
kmallocyeah we never 400.15:57
kmallocin flat-model15:57
lbragstadoh...15:57
kmallocany limit set is ok15:57
sdagueyou can still set it on parents, but models are only evaluated in the context of their own project15:57
lbragstadi misunderstood that then15:57
sdaguelbragstad: yeh, I might have been unclear, or maybe thinking evolved15:58
cmurphylbragstad: haha i was right :P15:58
kmallocok15:58
lbragstad:)15:59
kmallocthe real solution here is simple, i don't think this matters.15:59
sdaguekmalloc: I don't think it's a bad API, it's just a hard problem, and assuming it's going to be super crisp on attempt 1 I think will mostly mean attempt 1 never happens15:59
kmallocthis is a fairly narrow API change based upon the model selected. the API definition should say "keystone will evaluate the limits, and will 400 if the limits are invalid, with a proper error code"15:59
cmurphyokay, I don't think it's an API break if we have GET /limits-model which for now will only report "flat" and we say up front that different operations will have different responses based on /GET /limits-model15:59
kmallocopening us up for multiple models.16:00
sdaguecmurphy: ++ agreed16:00
kmallocdefine the API loosely enough in the documentation16:00
kmallocthis is really not bad16:00
kmallockeystone will evaluate, based upon the model, and it doesn't really break interop, because we are emitting errors that make sense and the API itself doesn't change16:00
sdagueI think the important thing is that when calls fail, their is enough error information to explain why16:00
kmallocwe can make the model discoverable16:00
kmallocbut the error needs to be clear16:00
sdagueso the admin can consider what their next step is16:01
kmalloci am 100% ok with this after thinking it thorugh and don't feel dirty for the design16:01
cmurphycool16:01
kmallocsdague: ++16:02
lbragstadwhat happens if a deployment wants to switch from "flat" to "something-else" later on?16:02
kmalloclbragstad: they can, old limits set will remain16:02
lbragstadis that a possibility?16:02
kmallocwe don't "fix" limits in a model change16:02
kmallocdon't even try to.16:02
kmallocit is on the admin to say "we'16:02
lbragstadright - but how painful will that be?16:02
kmallocre changing models and need to deal with it"16:02
kmallocpainful.16:02
kmalloci don't want to have to assume we know what the limit usage is downstream16:03
kmallocwe *Cant* afaict update limits.16:03
kmallocit really is an admin action and admin update.16:03
kmallocit is a "going forward we adhere to the model" if you make a change.16:04
kmallocbecause it's only invoked on "set"16:04
cmurphyit would only start applying when users want to update limits, if they find the tree is out of whack becasue the model changed the admin will have to go fix it16:04
lbragstadko16:04
lbragstadok*16:04
kmallocit explicitly grandfathers any limits into the old model on the change, new limits are new model, and it might mean you're over limit / cant set limits in some cases until the tree is fixed16:05
cmurphyyep16:05
kmallocbut again, we can't claim to be smart enough to handle the downstream effects (what nova is doing)16:05
lbragstadyeah - i think that's the right thing to do, but it'll just mean operators really have to think about how to move from one model to the other16:06
lbragstadif they want to16:06
kmallocgood. changes on this level should require real thought not just willy-nilly changes16:06
kmalloc;)16:06
lbragstadi just expect to have people coming to us when they do16:06
kmallocwe are explicitly defining the API to support multiple modes without breaking interop/contracts with minor behavior differences.16:07
lbragstadtry and do something like this16:07
kmallocdocumentation documentation documentation16:07
*** vhosakot has joined #openstack-dev16:07
kmallocnow, we should consider a bulk limit update API with this as well.16:07
kmalloc(future?)16:07
sdagueagree with cmurphy / kmalloc16:08
sdaguekeystone only does validation on an update call16:08
lbragstad++16:08
kmalloci think a "get me all limits" and "bulk update via json" might be very valid for assisting in model switches16:09
*** vnogin has quit IRC16:09
sdagueif someone decided to change their model, then presumably they also wrote themselves a batch script to go through their project tree and set a new world16:09
lbragstadright16:09
sdaguekmalloc: maybe, bulk update actually makes for a harder validation problem16:09
sdaguevs. making them make all the calls from the root down to create sanity16:09
kmallocsdague: perhaps.16:09
*** paramite has quit IRC16:10
kmalloci am happy to punt on bulk update16:10
kmallocand just back burner it to see if it makes sense16:10
sdagueyeh, future possible optimization16:10
kmallocit should be something we hold onto in case we want it, even if we don't implement it right away16:10
sdaguethat would be an api change to add it, but that's fine. may never actually need it16:10
sdagueok, got to drop off for a bit, catch you all later16:11
kmallocadding /limits/change_all_the_limits_I_know_what_im_doing is easy :)16:11
kmallocsdague: cheers and thanks!16:11
lbragstadsdague: thanks for the time!16:11
lbragstadi think this cleared a lot of things up...16:12
*** ioggstream has quit IRC16:12
cmurphy++16:13
lbragstadi think targetting the flat model first makes sense16:13
cmurphy+116:13
lbragstadmy concerns about the API things were addressed16:14
lbragstadit gives people a way to start using this without having the development burden of hammering out all the details exactly right on the first try16:14
*** xarses has quit IRC16:14
*** weshay|ruck is now known as weshay|ruck|MOD16:15
*** rraja has quit IRC16:16
*** yehowyada has quit IRC16:17
*** vhosakot has quit IRC16:25
*** baoli has quit IRC16:27
*** felipemonteiro__ has quit IRC16:28
*** felipemonteiro__ has joined #openstack-dev16:28
*** kashyap has left #openstack-dev16:30
*** gszasz has quit IRC16:31
*** gszasz has joined #openstack-dev16:31
*** zaitcev has quit IRC16:33
*** baoli has joined #openstack-dev16:38
*** ioggstream has joined #openstack-dev16:43
*** dpawar has quit IRC16:44
*** gszasz has quit IRC16:46
*** zaitcev has joined #openstack-dev16:46
*** radhikam has joined #openstack-dev16:49
*** d0ugal has quit IRC16:52
*** d0ugal has joined #openstack-dev16:54
*** d0ugal has quit IRC16:54
*** d0ugal has joined #openstack-dev16:54
*** fragatina has quit IRC16:57
*** david-lyle has quit IRC16:57
*** ioggstream has quit IRC16:58
*** marios has quit IRC16:58
*** baoli_ has joined #openstack-dev16:59
*** fragatina has joined #openstack-dev17:00
*** baoli has quit IRC17:01
*** aludwar1 has joined #openstack-dev17:02
*** aludwar1 has quit IRC17:03
*** slaweq has joined #openstack-dev17:03
*** kobis has joined #openstack-dev17:03
*** fragatina has quit IRC17:05
*** xarses has joined #openstack-dev17:07
*** felipemonteiro_ has joined #openstack-dev17:07
*** rosmaita has quit IRC17:09
*** vnogin has joined #openstack-dev17:09
*** slaweq_ has joined #openstack-dev17:09
*** slaweq has quit IRC17:10
*** slaweq_ has quit IRC17:10
*** rosmaita has joined #openstack-dev17:11
*** felipemonteiro__ has quit IRC17:11
*** slaweq has joined #openstack-dev17:11
*** lucasagomes is now known as lucas-afk17:12
*** jpich has quit IRC17:12
*** vnogin has quit IRC17:14
*** gyee has joined #openstack-dev17:14
*** panda is now known as panda|off17:18
*** karthiks has quit IRC17:19
*** baoli_ has quit IRC17:20
*** schmots has joined #openstack-dev17:20
*** dtantsur is now known as dtantsur|afk17:24
*** tobiajo has quit IRC17:24
*** hseipp has quit IRC17:27
*** baoli has joined #openstack-dev17:27
*** tesseract has quit IRC17:29
*** lpetrut has quit IRC17:30
*** jkilpatr has joined #openstack-dev17:34
*** david-lyle has joined #openstack-dev17:36
*** kobis has quit IRC17:39
*** jkilpatr has quit IRC17:39
*** kobis has joined #openstack-dev17:40
*** fried_rice is now known as fried_rolls17:41
*** fragatina has joined #openstack-dev17:43
*** signed8bit is now known as signed8bit_Zzz17:45
*** signed8bit_Zzz has quit IRC17:45
*** salv-orlando has joined #openstack-dev17:50
*** claudiub has joined #openstack-dev17:51
*** vnogin has joined #openstack-dev17:51
*** kbringard1 has quit IRC17:53
*** jkilpatr has joined #openstack-dev17:53
*** kbringard has joined #openstack-dev17:53
*** kobis has quit IRC17:54
*** morazi has quit IRC17:59
*** salv-orlando has quit IRC18:01
*** e0ne has quit IRC18:01
*** xarses has quit IRC18:02
*** yamamoto has quit IRC18:02
*** mdorman has joined #openstack-dev18:03
*** mgoddard has quit IRC18:03
*** openstackgerrit has quit IRC18:03
*** jkilpatr has quit IRC18:06
*** thorre has quit IRC18:11
*** thorre has joined #openstack-dev18:12
*** vnogin has quit IRC18:14
*** sdague has quit IRC18:16
*** yamamoto has joined #openstack-dev18:16
*** ekcs has joined #openstack-dev18:19
*** david-lyle has quit IRC18:21
*** yamamoto has quit IRC18:21
*** slaweq has quit IRC18:22
*** jkilpatr has joined #openstack-dev18:23
*** slaweq has joined #openstack-dev18:25
*** electrofelix has quit IRC18:28
*** slaweq has quit IRC18:29
*** yamamoto has joined #openstack-dev18:32
*** penick has joined #openstack-dev18:35
*** yamamoto has quit IRC18:36
*** signed8bit has joined #openstack-dev18:40
*** sdague has joined #openstack-dev18:40
*** yamamoto has joined #openstack-dev18:44
*** yamamoto has quit IRC18:44
*** yamamoto has joined #openstack-dev18:44
*** yamamoto has quit IRC18:45
*** mgoddard has joined #openstack-dev18:52
*** penick has quit IRC18:55
*** kobis has joined #openstack-dev18:57
*** vnogin has joined #openstack-dev18:58
*** vnogin has quit IRC18:58
*** penick has joined #openstack-dev18:58
*** kobis has quit IRC19:02
*** Guest99196 has quit IRC19:04
*** zaneb has quit IRC19:05
*** kbringard1 has joined #openstack-dev19:07
*** kbringard has quit IRC19:08
*** vnogin has joined #openstack-dev19:11
*** zaneb has joined #openstack-dev19:19
*** harlowja has joined #openstack-dev19:21
*** schmots has quit IRC19:23
*** slaweq has joined #openstack-dev19:24
*** morazi has joined #openstack-dev19:28
*** vnogin has quit IRC19:28
*** mgoddard has quit IRC19:29
*** e0ne has joined #openstack-dev19:31
*** vnogin has joined #openstack-dev19:34
*** crtcji has joined #openstack-dev19:36
*** vnogin has quit IRC19:38
*** slaweq has quit IRC19:39
*** kobis has joined #openstack-dev19:45
*** yamamoto has joined #openstack-dev19:46
*** yamamoto has quit IRC19:52
*** fried_rolls is now known as fried_rice19:53
*** kobis has quit IRC19:55
*** kobis has joined #openstack-dev19:55
*** kobis has quit IRC19:56
*** kobis has joined #openstack-dev19:56
*** kobis has quit IRC19:57
*** gouthamr has joined #openstack-dev20:05
*** lpetrut has joined #openstack-dev20:07
*** penick has quit IRC20:08
*** david-lyle has joined #openstack-dev20:16
*** ChanServ has quit IRC20:17
*** lpetrut has quit IRC20:18
*** slaweq has joined #openstack-dev20:20
*** penick has joined #openstack-dev20:22
*** ChanServ has joined #openstack-dev20:24
*** barjavel.freenode.net sets mode: +o ChanServ20:24
*** penick has quit IRC20:25
*** sridharg has quit IRC20:28
*** david-lyle has quit IRC20:32
*** Odd_Bloke has quit IRC20:37
*** Odd_Bloke has joined #openstack-dev20:37
*** penick has joined #openstack-dev20:39
*** penick has quit IRC20:42
*** penick has joined #openstack-dev20:42
*** penick has quit IRC20:45
*** jkilpatr has quit IRC20:47
*** yamamoto has joined #openstack-dev20:48
*** yamamoto has quit IRC20:52
*** visbits has quit IRC20:54
*** kbringard1 has quit IRC20:57
*** penick has joined #openstack-dev20:57
*** ttx has quit IRC20:57
*** ttx has joined #openstack-dev20:57
*** xarses has joined #openstack-dev21:01
*** gordc has quit IRC21:03
*** gouthamr has quit IRC21:07
*** penick has quit IRC21:09
*** fried_rice is now known as efried_cya_jan21:12
*** penick has joined #openstack-dev21:12
*** cdent has quit IRC21:19
*** achampion has quit IRC21:20
*** etingof has quit IRC21:22
*** kbringard has joined #openstack-dev21:23
*** diablo_rojo has joined #openstack-dev21:23
*** david-lyle has joined #openstack-dev21:24
*** e0ne has quit IRC21:26
*** david-lyle has quit IRC21:29
*** akrivoka has quit IRC21:29
*** e0ne has joined #openstack-dev21:33
*** leakypipes has quit IRC21:33
*** kgiusti has quit IRC21:34
*** e0ne has quit IRC21:35
*** kgiusti has joined #openstack-dev21:35
*** etingof has joined #openstack-dev21:37
*** rcernin has joined #openstack-dev21:44
*** yamamoto has joined #openstack-dev21:49
*** yamamoto has quit IRC21:54
*** edmondsw has joined #openstack-dev21:55
*** edmondsw has quit IRC21:56
*** rcernin has quit IRC22:03
*** vnogin has joined #openstack-dev22:03
*** rcernin has joined #openstack-dev22:03
*** awaugama has quit IRC22:10
*** felipemonteiro_ has quit IRC22:11
*** gouthamr has joined #openstack-dev22:12
*** openstackgerrit has joined #openstack-dev22:15
openstackgerritDoug Hellmann proposed openstack/governance master: stop linking to documentation from governance  https://review.openstack.org/52319522:15
*** ianychoi_ has joined #openstack-dev22:17
*** ianychoi has quit IRC22:19
*** gmann_afk is now known as gmann22:21
openstackgerritMerged openstack/governance master: Tags are applied to deliverables, or teams  https://review.openstack.org/52388622:21
openstackgerritMerged openstack/governance master: Officialize election organization working group  https://review.openstack.org/52106222:30
openstackgerritMerged openstack/governance master: Remove redundant links in index.rst  https://review.openstack.org/52312122:30
openstackgerritMerged openstack/governance master: Update URL for TC tracker on the wiki  https://review.openstack.org/52338122:30
openstackgerritMerged openstack/governance master: Add log-classify to infra project  https://review.openstack.org/52216622:30
openstackgerritMerged openstack/governance master: Update policy artifacts for telemetry/panko  https://review.openstack.org/52345222:30
openstackgerritMerged openstack/governance master: Add ansible-role-k8s-glance  https://review.openstack.org/52413022:30
*** slaweq has quit IRC22:33
*** baoli has quit IRC22:33
*** baoli has joined #openstack-dev22:34
*** baoli has quit IRC22:35
*** vladikr has quit IRC22:36
*** slaweq has joined #openstack-dev22:36
openstackgerritMerged openstack/governance master: Mark the completion of tempest plugin split goal for cloudkitty team  https://review.openstack.org/52458222:36
openstackgerritMerged openstack/governance master: Update tempest plugin split goal for Cyborg team  https://review.openstack.org/52458422:36
openstackgerritMerged openstack/governance master: Update tempest plugin split goal for Loci team  https://review.openstack.org/52458622:36
openstackgerritMerged openstack/governance master: Update tempest plugin split goal for Masakari team  https://review.openstack.org/52458722:36
openstackgerritMerged openstack/governance master: Update tempest plugin split goal for OpenStack-Helm team  https://review.openstack.org/52458922:36
openstackgerritMerged openstack/governance master: Mark the completion of Tempest plugin split goal for Tripleo team  https://review.openstack.org/52459522:36
openstackgerritMerged openstack/governance master: stop linking to documentation from governance  https://review.openstack.org/52319522:38
*** achampion has joined #openstack-dev22:39
*** diablo_rojo has quit IRC22:43
*** bobh has quit IRC22:49
*** yamamoto has joined #openstack-dev22:50
*** figleaf is now known as edleafe22:53
*** yamamoto has quit IRC22:54
*** kumarmn has quit IRC23:01
*** slaweq has quit IRC23:04
*** bobh has joined #openstack-dev23:05
*** bobh has quit IRC23:09
*** esberglu has quit IRC23:10
*** baoli has joined #openstack-dev23:11
*** signed8b_ has joined #openstack-dev23:12
*** kbringard has quit IRC23:13
*** signed8bit has quit IRC23:15
*** penick has quit IRC23:15
*** baoli has quit IRC23:16
*** signed8b_ has quit IRC23:16
*** bobh has joined #openstack-dev23:17
*** bobh has quit IRC23:22
*** kumarmn has joined #openstack-dev23:23
*** kumarmn has quit IRC23:28
*** purplerbot has quit IRC23:29
*** bobh has joined #openstack-dev23:32
*** bobh has quit IRC23:37
*** yamahata has joined #openstack-dev23:37
*** vnogin has quit IRC23:37
*** vnogin has joined #openstack-dev23:38
*** mdorman has quit IRC23:38
*** felipemonteiro has joined #openstack-dev23:39
*** xarses has quit IRC23:44
*** kumarmn has joined #openstack-dev23:48
*** markvoelker has quit IRC23:49
*** markvoelker has joined #openstack-dev23:50
*** yamamoto has joined #openstack-dev23:51
*** slaweq has joined #openstack-dev23:52
*** markvoelker has quit IRC23:54
*** yamamoto has quit IRC23:55
*** slaweq has quit IRC23:57
*** vnogin has quit IRC23:57

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