Friday, 2014-01-31

*** enikanorov__ has joined #openstack-dev00:00
*** SergeyLukjanov is now known as SergeyLukjanov_00:00
*** kolesovdv has joined #openstack-dev00:00
*** kbrierly has joined #openstack-dev00:01
*** enikanorov_ has quit IRC00:03
*** lcheng has joined #openstack-dev00:03
*** jmontemayor has quit IRC00:04
*** mlavalle has quit IRC00:04
*** aspiers has quit IRC00:04
*** bear_field has joined #openstack-dev00:04
*** CaptTofu_ has joined #openstack-dev00:05
*** galstrom_zzz is now known as galstrom00:05
*** asalkeld has quit IRC00:06
bear_fieldhey guys, trying to parse https://github.com/openstack/keystone/blob/master/etc/policy.json doc to understand how to give a user admin-like access to a tenant. user should be able to add and remove users00:06
*** ala_ has quit IRC00:06
*** nermina has quit IRC00:07
*** tkay has joined #openstack-dev00:07
*** CaptTofu has quit IRC00:08
*** adreznec has quit IRC00:08
*** nermina has joined #openstack-dev00:08
*** nermina has quit IRC00:08
*** cagrev_ has quit IRC00:08
*** xarses has joined #openstack-dev00:10
*** ala_ has joined #openstack-dev00:10
*** thuc has quit IRC00:10
*** Alex_Gaynor has joined #openstack-dev00:11
*** Alex_Gaynor has left #openstack-dev00:11
*** thuc has joined #openstack-dev00:11
*** nati_ueno has quit IRC00:11
jamielennoxdhellmann: ping - i think it might be too late for you00:12
*** NikitaKonovalov is now known as NikitaKonovalov_00:15
*** thuc has quit IRC00:15
morganfainbergbknudson, dstanek, regarding the config fixture, i'm... not sure I see a huge benefit to using it over just continuing w/ opt_in_group00:16
dstanekmorganfainberg: i don't thing replace uses re at all00:16
*** CaptTofu_ has quit IRC00:16
*** ArxCruz has joined #openstack-dev00:16
morganfainbergdstanek, it doesn't00:16
*** jaybuff1 has left #openstack-dev00:16
*** MaxV has quit IRC00:16
bknudsonmorganfainberg: it doesn't do much, just wraps the reset00:16
morganfainbergbknudson, so.. we would just replace opt_in_group with self.ConfFixture.config ?00:17
morganfainberg(with a better name of course)00:17
bknudsonmorganfainberg: but if we add in the smarts to do the reset and unregister_opt properly that would make things a little easier.00:17
*** browne has quit IRC00:18
morganfainbergbknudson, hm. yeah.  so i think the right approach is likely do it in keystone then propose it to oslo-incubator then resolve it when we sync (e.g. subclass and just add the new functions)?00:18
bknudsonmorganfainberg: I like that approach.00:19
*** nati_ueno has joined #openstack-dev00:19
*** cjoelrun has quit IRC00:19
*** oanufriev has joined #openstack-dev00:19
*** MaxV has joined #openstack-dev00:20
morganfainbergbknudson, cool. i'll do it that way.  i'm still going to leave my current opt_in_group fix in the chain since it resolves a distinct issue independant of removing the config files00:20
bknudsonself._conf = self.useFixture(fixtures.Config()) -- self._conf.config(group='xxx', opt='something')00:20
morganfainbergbknudson, yep.00:20
bknudsonmorganfainberg: if you want someone else to make that switch I'll voluteer00:20
bknudsonbeen learning a little more about fixtures so don't mind trying it out.00:21
morganfainbergbknudson, nah, i am about to start on it right now, it's good for me to use fixtures myself00:21
bknudsonok.00:21
morganfainbergbknudson, :)00:21
bknudsonyou'll probably find all sorts of other places you want to use them in the tests.00:21
morganfainbergbknudson, i know i will.00:21
dstanekbknudson: https://review.openstack.org/#/c/58311 no longer has the refactoring and the behavior change (when you have time)00:22
morganfainbergbknudson, i'm looking forward to when tests are easier to work with. and test_backend isn't a dumping ground (can be gone)00:22
*** nati_ueno has quit IRC00:22
bknudsonmorganfainberg: pretty soon we'll have a Backends fixture...00:22
*** zzelle has quit IRC00:23
*** nati_ueno has joined #openstack-dev00:23
bknudsona DefaultFixtures fixture that uses Backends fixture.00:23
*** melwitt has joined #openstack-dev00:23
morganfainbergbknudson, ++00:23
morganfainbergbknudson, yo dawg, i heard you liked fixtures, so i put a fixture in your fixture...00:23
*** matsuhashi has joined #openstack-dev00:24
morganfainbergbknudson, wow, i must be loopy, i don't usually make bad internet meme jokes on irc like that00:24
*** MaxV has quit IRC00:24
*** galstrom is now known as galstrom_zzz00:25
*** epim has joined #openstack-dev00:26
*** sjmc7 has quit IRC00:26
*** bear_field has left #openstack-dev00:26
*** CaptTofu has joined #openstack-dev00:27
bknudsonwhy is v3_to_v2_user in the manager?? only a v2 controller would need that!00:28
*** thedodd has quit IRC00:29
*** epim_ has joined #openstack-dev00:30
jamielennoxbknudson: ugh, why does that function exist...00:30
morganfainbergjamielennox, because v2 and v3 users are different structurally00:30
bknudsonjamielennox: v3 users have domain_id.00:30
*** epim has quit IRC00:30
*** epim_ is now known as epim00:30
morganfainbergjamielennox, and we needed to "fix" that to normalize the default_project_id and domain_id issues00:30
*** dims has quit IRC00:31
morganfainbergbknudson, i think it went to the manager because it was used by some controllers but was identity specific00:31
jamielennoxmorganfainberg: i know this, my point was more that our backend user representation need not reflect our API representation and we shouldn't need to convert between them00:31
morganfainbergbknudson, as i recall it was moved around a couple times in review00:31
jamielennoxit's more of me being annoyed with our design recently00:31
morganfainbergbknudson, in... K it dies... dies a horrible death00:31
morganfainbergjamielennox, =/00:31
bknudsonmind if I move it?00:32
morganfainbergbknudson, not at all00:32
*** dims has joined #openstack-dev00:32
*** nermina has joined #openstack-dev00:32
bknudsonI'll try sticking it in common.controller ... probably just a free function in there.00:33
morganfainbergbknudson, as far as i'm concerned it can live in a utiliy libary (i think it's static)00:33
*** pmathews has quit IRC00:33
*** jobewan has quit IRC00:33
morganfainbergbknudson, sure. works for me.00:33
*** epim_ has joined #openstack-dev00:33
morganfainbergbknudson, i look forward to the day v2 goes away :)00:33
morganfainbergbknudson, i think i might even celebrate.00:33
bknudsonmorganfainberg: by then you'll be looking forward to the day v3 goes away.00:34
morganfainbergjamielennox, in Juno i know we're going to try and resolve the tokens being stored in a different "format"00:34
*** galstrom_zzz is now known as galstrom00:34
morganfainbergbknudson, god i hope we haven't moved to V4 yet :P00:34
*** sarob has joined #openstack-dev00:34
morganfainbergjamielennox, i think we can avoid some of the pitfalls v2->v3 did if we need a v400:35
*** rnirmal has quit IRC00:35
jamielennoxmorganfainberg: so i spent a while playing with this for KDS and i would really like a better seperation of view and model00:35
*** rnirmal has joined #openstack-dev00:35
*** lcheng has quit IRC00:35
*** epim has quit IRC00:35
*** epim_ is now known as epim00:35
morganfainbergjamielennox, hm. pecan/wsme specifics?00:35
*** rnirmal has quit IRC00:35
morganfainbergjamielennox or in keystone's architecture in general00:36
jamielennoxa user object should not be aware of whether it was a v2 or a v3 user - they are the same with v2 having domain = default_domain00:36
jamielennoxit should be then a v2 or v3 specific path that turns that into something that is exposed via api00:36
morganfainbergjamielennox, sure.00:36
jamielennoxwe do an ok job in some places of that00:36
jamielennoxbut things like  v3_to_v2_user need to die00:36
jamielennoxmorganfainberg: not really00:37
*** mst89 has joined #openstack-dev00:37
jamielennoxnot really pecan/wsme - it helps but it's just a design thing really00:37
jamielennoxi was playing with wsme all yesterday - i'm just not sure it is doable for us00:38
jamielennoxsaving 'extra' for json is easy because it maps into python so well00:38
morganfainbergjamielennox, it should be more like how the controller.wrap stuff works.  controller takes raw data and makes it the version-specific thing.  not v3_to_v2, just "format_object" or something00:38
morganfainbergthat way if we change something in V4, we aren't converting we're just doing the v4 format logic00:39
jamielennoxmorganfainberg: yea, it's not quite MVC but i would like to get it closer00:39
morganfainbergjamielennox, i plan on starting with tokens in J (as well as work w/ ayoung on the pipeline peice)00:39
jamielennoxanyway - there is no way i can think of to save generic 'extra' XML data because you somehow need to process it into something00:39
jamielennoxso i don't think it's something that can therefore be accepted to wsme00:40
morganfainbergjamielennox, but i think token formats aren't tied to api versions.  heck, if we used something like protobuf (no not saying we should) we could make sure all object versions are backwards compat00:40
morganfainbergand only bump api versions if we can't possibly get there from where we are00:40
*** atiwari has quit IRC00:41
jamielennoxmorganfainberg: no token versions aren't tied to API versions, it's just something that happened00:41
*** xmltok has quit IRC00:41
morganfainbergjamielennox, would it be possible to make "extra" attributes something that needs to be configured?00:41
jamielennoxmorganfainberg: i would like to essentially segregate keystone into two parts00:41
jamielennoxtoken auth and CRUD stuff00:41
morganfainbergjamielennox, i see the CRUD stuff as being split as well... but that is another conversation00:42
jamielennoxthen put auth_token middleware in front of everything that is not /v2.0/auth/token or /v3/token00:42
jamielennoxthen we can add as many token formats as we want to auth_token00:42
morganfainbergjamielennox, ayoung and I discussed making token issuance a pipeline, independent of the rest of things e.g. use paste to do it00:42
morganfainbergjamielennox, that is the plan for juno00:43
morganfainbergjamielennox, similarly auth_token can become a pipeline.00:43
jamielennoxmorganfainberg: cool, i want to be involved in that00:43
morganfainbergjamielennox, ++ :) it is on the roadmap for sure!00:43
*** sn6i23a has joined #openstack-dev00:43
morganfainbergjamielennox, also would let us to inject things like "compress token" w/o having to change everything00:44
bknudsonjamielennox: morganfainberg: what do you think of this insanity: https://review.openstack.org/#/c/70044/00:44
morganfainbergbknudson, looking but my guess is that is a lot of what is needed to migrate from "legacy-kvs" to "dogpile-kvs" in assignment00:45
*** IanGovett has quit IRC00:45
bknudsonmorganfainberg: oops, pointed to the wrong one... https://review.openstack.org/#/c/70272/00:45
jamielennoxwhat is that00:45
morganfainbergbknudson, any reason that is in the controller rather than the assignment manager?00:46
jamielennoxmorganfainberg: ++00:46
*** dpamio has joined #openstack-dev00:46
bknudsonmorganfainberg: the controller gets a list of user_ids.00:46
bknudsonthat's what the assignment backend returns.00:46
morganfainbergbknudson, oh thats the v2 icky00:46
morganfainbergbknudson, yeah i don't want that in the manager :)00:46
bknudsonand yes, it should probably be in the manager.00:46
bknudsonthe manager could return a list of v3 users and the v2 controller converts to v2 users.00:47
morganfainbergbknudson, does v3 have an equivalent call?00:47
jamielennoxyuk00:47
*** pablosan has quit IRC00:48
morganfainbergbknudson, it doesn't look like V3 supports that same mechanism00:48
bknudsonmorganfainberg: v3 has all sorts of ways to get role assignments, though... so must do something similar somewhere.00:48
*** CaptTofu has quit IRC00:49
morganfainbergbknudson, i think all of that happens in the grant tables (sql for example)00:49
*** mrodden has quit IRC00:50
*** erkrnt has quit IRC00:51
bknudsonmorganfainberg: v3 seems to deal with user IDs and not user objects for role assignments.00:51
bknudsonmorganfainberg: so Horizon will get back a bunch of user IDs, try to get more information about them and get 404s.00:51
*** gokrokve has quit IRC00:53
*** epim has quit IRC00:53
*** gokrokve has joined #openstack-dev00:53
*** bswartz has joined #openstack-dev00:54
*** OpenStack1 has joined #openstack-dev00:57
*** gokrokve has quit IRC00:58
*** OpenStack1 has left #openstack-dev00:59
branenDoes anyone know how tests/files/gr-base.txt in project requirements is used?00:59
*** yamahata has quit IRC00:59
*** CaptTofu has joined #openstack-dev00:59
*** evgenyf has joined #openstack-dev01:01
jamielennoxbknudson, morganfainberg, dolphm: do you know if anyone is looking at oslo.messaging for keystone?01:01
*** markmcclain has quit IRC01:01
*** markmcclain has joined #openstack-dev01:02
*** mst89 has quit IRC01:03
*** harlowja is now known as harlowja_away01:04
*** dpamio has quit IRC01:06
*** epim has joined #openstack-dev01:06
*** epim has quit IRC01:11
*** nati_uen_ has joined #openstack-dev01:11
ayoungjamielennox, I have company in a moment, so I can't talk, but let me see if I can post the design picture from the hackfest01:12
jamielennoxayoung: that's ok - i just want to make sure that i'm not going to trip over someone else's work01:13
*** tkay has quit IRC01:13
jamielennoxayoung: from blame it looks like dolphm lbragstad and fabio were the implementers and i thought they might have it planned01:13
*** marekd is now known as marekd|away01:14
ayoungI was talking about the pipeline thing....01:14
jamielennoxayoung: oh, ok01:14
jamielennoxayoung: yea i'd like to chat about some reorganization of keystone code in juno at some point01:14
*** nati_ueno has quit IRC01:14
ayounghttps://twitter.com/admiyoung/status/429060448462577664/photo/1   jamielennox01:16
ayoungThe blue boxes are "future"  or "possible"01:16
ayoungQuotas and compression were in there01:17
jamielennoxwow01:17
ayoungOIK...friends are over01:17
jamielennoxcya01:17
*** ArxCruz has quit IRC01:17
*** carlp has quit IRC01:20
*** nermina has quit IRC01:24
*** harlowja_away is now known as harlowja01:26
*** IanGovett has joined #openstack-dev01:27
*** IanGovett has quit IRC01:28
*** nermina has joined #openstack-dev01:28
*** IanGovett has joined #openstack-dev01:28
*** salv-orlando has quit IRC01:30
*** haomaiwang has joined #openstack-dev01:30
*** evgenyf has quit IRC01:30
*** CaptTofu_ has joined #openstack-dev01:31
*** CaptTofu has quit IRC01:31
*** galstrom is now known as galstrom_zzz01:31
*** kgriffs is now known as kgriffs_afk01:32
*** stevemar has quit IRC01:32
*** comay has quit IRC01:32
*** comay has joined #openstack-dev01:32
*** haomaiwa_ has quit IRC01:34
harlowjaayoung dolphm u guys have a sec, have a potential keystone question that might be anwerable (or might not)01:36
*** nati_uen_ has quit IRC01:36
*** mrodden has joined #openstack-dev01:37
*** mikeoutland has quit IRC01:39
*** mrodden1 has joined #openstack-dev01:39
harlowjawas just chatting with the guy here who is looking into authz/authn (trying to get him into keystone development) and had some interesting ideas01:40
*** tdruiva has joined #openstack-dev01:40
*** markmcclain has quit IRC01:40
*** mrodden has quit IRC01:41
*** nosnos has joined #openstack-dev01:41
*** terryw has quit IRC01:41
*** buzztroll has joined #openstack-dev01:42
*** buzztrol_ has joined #openstack-dev01:43
*** xarses has quit IRC01:43
*** alop has quit IRC01:45
*** buzztroll has quit IRC01:46
*** mrodden has joined #openstack-dev01:47
*** mrodden1 has quit IRC01:47
*** cjoelrun has joined #openstack-dev01:48
*** buzztrol_ has quit IRC01:48
*** luisg has quit IRC01:48
*** luisg has joined #openstack-dev01:49
*** buzztroll has joined #openstack-dev01:49
*** melwitt has quit IRC01:50
*** tdruiva has quit IRC01:51
*** tdruiva has joined #openstack-dev01:51
*** buzztroll has quit IRC01:53
*** bdpayne has quit IRC01:54
*** IanGovett has quit IRC01:54
*** buzztroll has joined #openstack-dev01:54
*** networkstatic has joined #openstack-dev01:55
*** sn6i23a has quit IRC02:00
*** markwash has quit IRC02:02
*** nermina has quit IRC02:03
*** markwash has joined #openstack-dev02:03
*** yamahata has joined #openstack-dev02:03
*** terryw has joined #openstack-dev02:04
*** buzztroll has quit IRC02:05
*** galstrom_zzz is now known as galstrom02:05
*** buzztroll has joined #openstack-dev02:06
*** buzztroll has joined #openstack-dev02:06
*** rraja has joined #openstack-dev02:07
*** sumanth__ has joined #openstack-dev02:07
*** lcheng has joined #openstack-dev02:08
*** Mandell has quit IRC02:10
*** sarob has quit IRC02:12
*** sarob has joined #openstack-dev02:12
*** networkstatic has quit IRC02:16
*** sarob has quit IRC02:17
*** networkstatic has joined #openstack-dev02:18
*** oanufriev has quit IRC02:18
*** networkstatic has quit IRC02:18
*** thedodd has joined #openstack-dev02:18
*** networkstatic has joined #openstack-dev02:18
*** networkstatic has quit IRC02:19
*** networkstatic has joined #openstack-dev02:19
*** _cjones_ has quit IRC02:19
*** nermina has joined #openstack-dev02:20
*** giulivo has quit IRC02:20
*** crank has quit IRC02:20
*** crank has joined #openstack-dev02:21
*** nermina has quit IRC02:25
*** arnaud___ has quit IRC02:27
*** arnaud__ has quit IRC02:27
*** markvoelker has quit IRC02:29
*** godara has quit IRC02:31
*** angdraug has quit IRC02:33
*** matsuhashi has quit IRC02:34
*** rraja has quit IRC02:34
*** sarob has joined #openstack-dev02:35
*** cjoelrun has quit IRC02:36
*** matsuhas_ has joined #openstack-dev02:37
*** newell has quit IRC02:38
*** erkules_ has joined #openstack-dev02:39
*** clayb has quit IRC02:39
*** sarob has quit IRC02:39
*** csd has quit IRC02:40
*** erkules has quit IRC02:41
*** kushal has quit IRC02:42
*** sarob has joined #openstack-dev02:43
*** enykeev has quit IRC02:44
*** thedodd has quit IRC02:44
*** networkstatic has quit IRC02:44
*** faramir1 has joined #openstack-dev02:44
*** sarob_ has joined #openstack-dev02:45
*** buzztroll has quit IRC02:46
*** buzztroll has joined #openstack-dev02:47
*** sarob has quit IRC02:48
*** sarob_ has quit IRC02:50
*** buzztroll has quit IRC02:51
*** crank has quit IRC02:52
*** crank has joined #openstack-dev02:52
lbragstadjamielennox: hey, sorry I missed your ping, stepped away for a bit. Need something?02:56
jamielennoxlbragstad: i was just checking if you were doing anything with oslo.messaging02:59
lbragstadjamielennox: ah ok, not exactly, I was helping ayoung with a few questions he had on it the other day03:00
lbragstadfor RDO03:00
jamielennoxlbragstad: yea there is some internal push to migrate03:00
jamielennoxlbragstad: i'm happy to do it i was just checking that you hadn't done something already03:01
jamielennox(i'm pretty sure you wrote most of that)03:01
jamielennoxor at least a chunk of it03:01
lbragstadjamielennox: right, I did. I did some of the testing using oslo-incubator's notifier module and i tested it on using the RPC backend with qpid and rabbitmq03:02
lbragstadboth worked fine03:02
lbragstadI had to patch 'os' in oslo somewhere though in order to get it to work with Qpid03:02
jamielennoxlbragstad: that's most likely why they want to move, RDO is based on qpid03:02
jamielennoxand if that sort of thing is fixed in oslo.messaging then we may as well use it03:03
lbragstadhttps://review.openstack.org/#/c/47444/03:03
*** buzztroll has joined #openstack-dev03:03
lbragstadspecifically03:03
jamielennoxit doesn't seem like it will be a aparticularly hard conversions03:03
lbragstadit shouldn't be03:03
lbragstadbut then again I haven't used the oslo messaging library yet03:03
jamielennoxoh right - i remember that one03:03
lbragstadyeah... that was a tough one03:03
lbragstadI spent a week debugging qpid code03:04
jamielennoxdo you remember a good reason for having host= on your wrapper?03:04
jamielennoxor just that it was there03:04
jamielennoxlol, ouch03:04
lbragstadhost?03:04
lbragstadin the notification?03:04
jamielennoxwell put it another way does: https://github.com/jamielennox/keystone/commit/b3e81c32d9fc2a68c648f3f13b1805282f01444b make sense?03:04
jamielennoxi can't see why any notification event would want to forcefully set a host name value03:05
jamielennoxand it's not being used03:05
*** csaba|afk is now known as csaba03:05
jamielennoxit seemed like something you passed on just because it was there03:06
*** dave_tucker is now known as dave_tucker_zzz03:06
lbragstadjamielennox: yeah I remember,03:07
*** matsuhashi has joined #openstack-dev03:07
*** buzztroll has quit IRC03:07
lbragstadjamielennox: the reason why I had to include that was because you have to get a publisher_id03:07
lbragstadfrom the notifier API03:07
lbragstadhttps://github.com/openstack/oslo-incubator/blob/master/openstack/common/notifier/api.py#L8603:07
jamielennoxyep, but host is an optional variable there03:07
*** dkranz has joined #openstack-dev03:07
*** matsuhas_ has quit IRC03:08
*** matsuhashi has quit IRC03:08
lbragstadit's not *required* but I didn't want to keep it from someone if they somehow wanted to take advantage of it03:08
jamielennoxyea, that's what i thought you didn't specifically keep it for any purpose, just may as well expose what's there03:08
jamielennoxit doesn't quite track with the oslo.messaging way of doing things so i was going to pull it out as  i couldn't see anything using it03:09
jamielennox(tests pass)03:09
lbragstadthey don't use publisher_id in oslo.messaging?03:09
lbragstadhuh, interesting03:11
lbragstaddoesn't look like they do03:11
jamielennoxlbragstad: you can, but you have a notifier object in messaging03:12
*** utlemming has quit IRC03:12
jamielennoxso that you only make the connection once (apparently that can be expensive)03:12
*** jckasper has joined #openstack-dev03:12
jamielennoxso you can specify the publisher_id when you create the object but not for every call03:12
lbragstadahh, I see it now. And then just keep the publisher_id as an attr of the notifier03:13
*** colinmcnamara has joined #openstack-dev03:13
*** erkrnt has joined #openstack-dev03:14
jamielennoxyep03:14
*** CaptTofu_ has quit IRC03:15
*** jckasper has quit IRC03:15
*** jckasper has joined #openstack-dev03:16
*** emagana has quit IRC03:16
lbragstadjamielennox: so we don't technically *need* the host to generate the publisher_id03:16
*** buzztroll has joined #openstack-dev03:17
lbragstadmakes sense then that the tests pass. That's good03:18
jamielennoxyea, it's an override for taking it from config or otherwise socket.hostname03:19
jamielennoxlbragstad: cool, just wanted to check i wasn't screwing up some future plans for it03:19
*** matsuhashi has joined #openstack-dev03:20
lbragstadjamielennox: makes sense to me, I don't think that will screw anything up.03:20
lbragstadI think you're in the clear03:20
*** erkrnt has quit IRC03:20
lbragstadjamielennox: thanks for checking though :)03:20
*** buzztroll has quit IRC03:22
*** adreznec has joined #openstack-dev03:22
*** csaba is now known as csaba|afk03:25
*** sumanth__ has quit IRC03:26
*** paragan has joined #openstack-dev03:28
*** cjoelrun has joined #openstack-dev03:28
*** galstrom is now known as galstrom_zzz03:29
*** CaptTofu has joined #openstack-dev03:29
*** CaptTofu has quit IRC03:30
*** colinmcnamara has quit IRC03:31
*** byeager has joined #openstack-dev03:33
*** colinmcnamara has joined #openstack-dev03:34
*** matsuhashi has quit IRC03:34
*** matsuhashi has joined #openstack-dev03:35
*** galstrom_zzz is now known as galstrom03:36
*** galstrom is now known as galstrom_zzz03:38
*** thedodd has joined #openstack-dev03:38
*** sumanth__ has joined #openstack-dev03:39
*** CaptTofu has joined #openstack-dev03:40
*** sumanth_ has joined #openstack-dev03:40
*** markwash has quit IRC03:41
*** nelsnelson has joined #openstack-dev03:42
*** matsuhashi has quit IRC03:43
*** faramir1 has quit IRC03:43
*** sweston has quit IRC03:43
*** sumanth__ has quit IRC03:44
*** doug_shelley66 has quit IRC03:44
*** sumanth has joined #openstack-dev03:44
*** sarob has joined #openstack-dev03:45
*** sumanth_ has quit IRC03:45
*** sumanth_ has joined #openstack-dev03:47
*** colinmcnamara has quit IRC03:48
*** yeylon__ has quit IRC03:48
*** sarob has quit IRC03:49
*** sumanth has quit IRC03:49
*** matsuhashi has joined #openstack-dev03:50
*** sweston has joined #openstack-dev03:52
*** david-lyle has joined #openstack-dev03:52
*** sumanth__ has joined #openstack-dev03:53
*** colinmcnamara has joined #openstack-dev03:54
*** colinmcnamara has quit IRC03:55
*** sweston has quit IRC03:55
*** sumanth_ has quit IRC03:56
*** colinmcnamara has joined #openstack-dev03:56
*** sweston has joined #openstack-dev03:57
*** gokrokve has joined #openstack-dev03:58
*** sumanth__ has quit IRC04:00
*** venkatesh has joined #openstack-dev04:01
*** venkatesh has quit IRC04:04
*** gyee has quit IRC04:04
*** troytoman is now known as troytoman-away04:05
*** colinmcnamara has quit IRC04:06
*** sn6i23a has joined #openstack-dev04:06
*** rraja has joined #openstack-dev04:07
*** xarses has joined #openstack-dev04:07
*** changbl has joined #openstack-dev04:07
*** rraja has quit IRC04:08
*** aeperezt has quit IRC04:09
*** jamielennox is now known as jamielennox|away04:11
*** sumanth__ has joined #openstack-dev04:12
*** HenryG has joined #openstack-dev04:13
*** sumanthns has joined #openstack-dev04:14
*** Mandell has joined #openstack-dev04:16
*** buzztroll has joined #openstack-dev04:17
*** nelsnelson is now known as nelsnelson_away04:17
*** thedodd has quit IRC04:18
*** galstrom_zzz is now known as galstrom04:21
*** nelsnelson_away has quit IRC04:22
*** enykeev has joined #openstack-dev04:25
*** harlowja is now known as harlowja_away04:26
*** achampion has joined #openstack-dev04:26
*** CaptTofu has quit IRC04:31
*** tkay has joined #openstack-dev04:31
*** markwash has joined #openstack-dev04:32
*** galstrom is now known as galstrom_zzz04:32
*** amcrn has quit IRC04:34
*** NithyaG has quit IRC04:39
*** spzala has quit IRC04:40
*** blamar has quit IRC04:40
*** sarob has joined #openstack-dev04:43
*** tkay has quit IRC04:44
*** blamar has joined #openstack-dev04:45
*** sarob has quit IRC04:48
*** tongli has quit IRC04:49
*** julienvey_ has quit IRC04:50
*** praneshp has quit IRC04:50
*** buzztroll has quit IRC04:51
*** achampio1 has joined #openstack-dev04:51
*** Ryan_Lane has joined #openstack-dev04:52
*** enykeev has quit IRC04:52
*** markwash has quit IRC04:53
*** achampion has quit IRC04:53
*** rdas has joined #openstack-dev04:55
*** markwash has joined #openstack-dev04:56
*** achampion has joined #openstack-dev04:58
*** achampio1 has quit IRC04:59
*** nelsnelson has joined #openstack-dev05:01
*** byeager has quit IRC05:01
*** byeager has joined #openstack-dev05:02
*** markwash has quit IRC05:03
*** stevemar has joined #openstack-dev05:03
*** enykeev has joined #openstack-dev05:04
*** sumanth_ has joined #openstack-dev05:07
*** coolsvap_away has quit IRC05:08
*** rektide has quit IRC05:08
*** gokrokve has quit IRC05:08
*** gokrokve has joined #openstack-dev05:08
*** gokrokve_ has joined #openstack-dev05:09
*** haomaiwang has quit IRC05:09
*** haomaiwang has joined #openstack-dev05:10
*** thurloat has quit IRC05:10
*** sumanth__ has quit IRC05:11
*** achampio1 has joined #openstack-dev05:12
*** blamar has quit IRC05:13
*** achampion has quit IRC05:13
*** thurloat has joined #openstack-dev05:13
*** gokrokve has quit IRC05:13
*** gokrokve_ has quit IRC05:13
*** byeager has quit IRC05:14
*** praneshp has joined #openstack-dev05:15
*** buzztroll has joined #openstack-dev05:17
*** nshaikh has joined #openstack-dev05:17
*** enykeev has quit IRC05:19
*** enykeev has joined #openstack-dev05:20
*** adreznec has quit IRC05:21
*** enykeev has quit IRC05:21
*** xazel has joined #openstack-dev05:21
*** buzztroll has quit IRC05:22
*** vkozhukalov has joined #openstack-dev05:26
*** mgagne1 has joined #openstack-dev05:28
*** mgagne has quit IRC05:29
*** HenryG has quit IRC05:32
*** achampio1 has quit IRC05:33
*** sn6i23a has quit IRC05:33
*** achampion has joined #openstack-dev05:34
*** amotoki has joined #openstack-dev05:37
*** gokrokve has joined #openstack-dev05:39
*** thuc has joined #openstack-dev05:41
*** thuc_ has joined #openstack-dev05:42
*** sarob has joined #openstack-dev05:43
*** stevemar has quit IRC05:44
*** byeager has joined #openstack-dev05:45
*** tdruiva has quit IRC05:45
*** tdruiva has joined #openstack-dev05:46
*** lcheng has quit IRC05:46
*** thuc has quit IRC05:46
*** sumanth_ has quit IRC05:46
*** mikeoutland has joined #openstack-dev05:48
*** sarob has quit IRC05:48
*** tdruiva has quit IRC05:50
*** tkay has joined #openstack-dev05:51
*** lcheng has joined #openstack-dev05:52
*** buzztroll has joined #openstack-dev05:53
*** mrodden has quit IRC05:53
*** buzztroll has quit IRC05:53
*** buzztrol_ has joined #openstack-dev05:53
*** achampion has quit IRC05:54
*** mrodden has joined #openstack-dev05:54
*** rohitk has joined #openstack-dev05:56
*** jotan has joined #openstack-dev05:56
*** rwsu has quit IRC05:59
*** buzztrol_ has quit IRC05:59
*** sumanth_ has joined #openstack-dev06:00
*** byeager has quit IRC06:00
*** gimps has quit IRC06:00
*** neeti has joined #openstack-dev06:01
*** achampion has joined #openstack-dev06:01
*** buzztroll has joined #openstack-dev06:04
*** jotan has quit IRC06:05
*** jotan has joined #openstack-dev06:05
*** tkay has quit IRC06:05
*** gimps has joined #openstack-dev06:08
*** tkay has joined #openstack-dev06:10
*** bknudson has quit IRC06:11
*** coolsvap has joined #openstack-dev06:12
*** locke1051 has joined #openstack-dev06:12
*** locke105 has quit IRC06:12
*** matiu has quit IRC06:14
*** afazekas_ has joined #openstack-dev06:15
*** bknudson has joined #openstack-dev06:15
*** afazekas_ has quit IRC06:15
*** rraja has joined #openstack-dev06:17
*** matiu has joined #openstack-dev06:18
*** e0ne has joined #openstack-dev06:19
*** sarob has joined #openstack-dev06:19
*** doug-fish has quit IRC06:19
*** arnaud has quit IRC06:21
*** arnaud has joined #openstack-dev06:22
*** sarob has quit IRC06:24
*** sumanth has joined #openstack-dev06:25
*** jprovazn has joined #openstack-dev06:25
*** matiu has quit IRC06:25
*** sumanth_ has quit IRC06:26
*** tkay has quit IRC06:26
*** mflobo_ has joined #openstack-dev06:27
*** sumanth_ has joined #openstack-dev06:28
*** mflobo has quit IRC06:30
*** sumanth has quit IRC06:30
*** CaptTofu has joined #openstack-dev06:32
*** odyssey4me has joined #openstack-dev06:32
*** rraja has quit IRC06:32
*** mflobo has joined #openstack-dev06:33
*** Nikolay_St has joined #openstack-dev06:35
*** mflobo_ has quit IRC06:35
*** zaitcev has quit IRC06:35
*** zaitcev has joined #openstack-dev06:35
*** arnaud has quit IRC06:36
*** csaba|afk is now known as csaba06:36
*** CaptTofu has quit IRC06:36
*** DinaBelova_ is now known as DinaBelova06:38
*** terryw has quit IRC06:38
*** matiu has joined #openstack-dev06:38
*** sumanth_ has quit IRC06:39
*** sumanth__ has joined #openstack-dev06:39
*** harlowja_at_home has joined #openstack-dev06:41
*** zaitcev has quit IRC06:42
*** alex_klimov has joined #openstack-dev06:42
*** sarob has joined #openstack-dev06:43
*** tkay has joined #openstack-dev06:43
*** rraja has joined #openstack-dev06:44
*** thuc_ has quit IRC06:47
*** thuc has joined #openstack-dev06:47
*** sarob has quit IRC06:48
*** emagana has joined #openstack-dev06:48
*** vartom1111111115 has joined #openstack-dev06:48
*** rohitk has quit IRC06:50
*** hdd_ has quit IRC06:50
*** vkozhukalov has quit IRC06:51
*** thuc has quit IRC06:52
*** lcheng has quit IRC06:53
*** stevemar has joined #openstack-dev06:54
*** kolesovdv has quit IRC07:00
*** haomai___ has joined #openstack-dev07:01
*** haomaiwang has quit IRC07:01
*** rohitk has joined #openstack-dev07:03
*** kolesovdv has joined #openstack-dev07:05
*** arnaud has joined #openstack-dev07:10
*** rohitk has quit IRC07:11
*** praneshp has quit IRC07:12
*** Drankis has joined #openstack-dev07:12
*** cjoelrun has quit IRC07:15
*** cjoelrun has joined #openstack-dev07:21
*** henrynash has joined #openstack-dev07:21
*** djoreilly has joined #openstack-dev07:25
*** rohitk_ has joined #openstack-dev07:25
*** arnaud has quit IRC07:26
*** arnaud has joined #openstack-dev07:26
*** sumanth_ has joined #openstack-dev07:27
*** henrynash has quit IRC07:28
*** mikeoutland has quit IRC07:29
*** arnaud has joined #openstack-dev07:29
*** arnaud has quit IRC07:30
*** sumanth__ has quit IRC07:30
*** arnaud has joined #openstack-dev07:30
*** henrynash has joined #openstack-dev07:31
*** harlowja_at_home has quit IRC07:33
*** arnaud has quit IRC07:34
*** arnaud has joined #openstack-dev07:35
*** sumanth__ has joined #openstack-dev07:36
*** sumanth_ has quit IRC07:39
*** arnaud has quit IRC07:42
*** arnaud has joined #openstack-dev07:42
*** sarob has joined #openstack-dev07:43
*** arnaud has quit IRC07:44
*** arnaud has joined #openstack-dev07:45
*** markwash has joined #openstack-dev07:46
*** cjoelrun has quit IRC07:46
*** sarob has quit IRC07:47
*** shardy_afk is now known as shardy07:48
*** DinaBelova is now known as DinaBelova_07:50
*** henrynash has quit IRC07:50
*** yolanda has joined #openstack-dev07:52
*** rraja has quit IRC07:52
*** buzztroll has quit IRC07:52
*** buzztroll has joined #openstack-dev07:53
*** buzztroll has quit IRC07:57
*** markwash has quit IRC07:59
mflobowhat is the reason because list_domain function is not implemented on keystoneclient?08:00
*** sumanth_ has joined #openstack-dev08:01
*** denis_makogon has joined #openstack-dev08:01
*** sumanth__ has quit IRC08:03
*** markwash has joined #openstack-dev08:04
*** DinaBelova_ is now known as DinaBelova08:07
*** srenatus has joined #openstack-dev08:07
*** amotoki_ has joined #openstack-dev08:07
*** markmc has joined #openstack-dev08:10
*** arnaud has quit IRC08:10
*** markwash has quit IRC08:11
*** sumanth__ has joined #openstack-dev08:13
*** e0ne has quit IRC08:15
*** ttx has quit IRC08:15
*** e0ne has joined #openstack-dev08:16
*** sumanth_ has quit IRC08:16
*** ttx has joined #openstack-dev08:16
*** ttx has quit IRC08:16
*** ttx has joined #openstack-dev08:16
*** ttx has joined #openstack-dev08:17
*** ttx has quit IRC08:17
*** ttx has joined #openstack-dev08:17
*** e0ne has quit IRC08:18
*** mariusleu has joined #openstack-dev08:18
*** e0ne has joined #openstack-dev08:18
*** Ryan_Lane has quit IRC08:21
*** csaba is now known as csaba|afk08:21
*** faramir1 has joined #openstack-dev08:22
*** e0ne has quit IRC08:23
*** jcoufal has joined #openstack-dev08:26
*** vkozhukalov has joined #openstack-dev08:29
*** tkay has quit IRC08:31
*** CaptTofu has joined #openstack-dev08:33
*** reed has joined #openstack-dev08:35
*** matsuhashi has quit IRC08:37
*** CaptTofu has quit IRC08:37
*** amotoki_ has quit IRC08:37
*** ndipanov has joined #openstack-dev08:40
*** stevemar has quit IRC08:40
*** MaxV has joined #openstack-dev08:41
*** JordanP has joined #openstack-dev08:42
*** sileht has quit IRC08:42
*** ygbo has joined #openstack-dev08:42
*** sarob has joined #openstack-dev08:43
*** sahid has joined #openstack-dev08:44
*** sileht has joined #openstack-dev08:47
*** nacim has joined #openstack-dev08:47
*** AlexF_ has joined #openstack-dev08:47
*** haomai___ has quit IRC08:47
*** alexpilotti has joined #openstack-dev08:47
*** haomaiwang has joined #openstack-dev08:48
*** sarob has quit IRC08:48
*** marekd|away is now known as marekd08:48
*** belmoreira has joined #openstack-dev08:49
*** sumanth__ has quit IRC08:49
*** rraja has joined #openstack-dev08:50
*** belmoreira has quit IRC08:50
*** jpich has joined #openstack-dev08:50
*** tkay has joined #openstack-dev08:51
*** alexpilotti has quit IRC08:52
*** julienvey_ has joined #openstack-dev08:54
*** xgsa has joined #openstack-dev08:55
*** nelsnelson is now known as nelsnelson_away08:55
*** henrynash has joined #openstack-dev08:56
*** fbo_away is now known as fbo08:57
*** NikitaKonovalov_ is now known as NikitaKonovalov08:58
*** jistr has joined #openstack-dev08:58
*** apevec has joined #openstack-dev08:59
*** nelsnelson_away has quit IRC09:00
*** jcoufal has quit IRC09:00
*** AnilV4 has quit IRC09:00
*** belmoreira has joined #openstack-dev09:00
*** Drankis_ has joined #openstack-dev09:02
*** Drankis has quit IRC09:02
*** giulivo has joined #openstack-dev09:02
*** Drankis_ has quit IRC09:02
*** Drankis has joined #openstack-dev09:03
*** buzztroll has joined #openstack-dev09:03
*** eglynn has joined #openstack-dev09:04
*** topol has quit IRC09:04
*** vartom1111111116 has joined #openstack-dev09:06
*** sumanth__ has joined #openstack-dev09:06
*** bauzas has joined #openstack-dev09:07
*** NikitaKonovalov is now known as NikitaKonovalov_09:07
*** vartom1111111115 has quit IRC09:07
*** bauzas has quit IRC09:07
*** henrynash has quit IRC09:07
*** bauzas has joined #openstack-dev09:07
*** safchain has joined #openstack-dev09:08
*** buzztroll has quit IRC09:08
*** johnthetubaguy has joined #openstack-dev09:09
*** circ-user-PnYGX has joined #openstack-dev09:09
*** circ-user-PnYGX is now known as I15909:11
*** johnthetubaguy1 has joined #openstack-dev09:11
*** yassine has joined #openstack-dev09:12
*** rohitk_ has quit IRC09:13
*** johnthetubaguy has quit IRC09:14
*** rohitk_ has joined #openstack-dev09:16
*** AlexF_ has quit IRC09:17
*** sumanth has joined #openstack-dev09:19
*** sumanth__ has quit IRC09:21
*** mmagr has joined #openstack-dev09:27
*** jcoufal has joined #openstack-dev09:28
I159Hello! I'm and David Stanek had developed the patch which has not been reviewed for a long time https://review.openstack.org/#/c/58766/. Somebody, please, review the patch.09:31
ekarlsoindeed that's a long time I15909:34
*** rohitk_ has quit IRC09:34
*** rohitk has joined #openstack-dev09:34
*** flaper87|afk is now known as flaper8709:35
*** athomas has joined #openstack-dev09:35
*** xga has joined #openstack-dev09:35
*** sumanth_ has joined #openstack-dev09:36
*** aspiers has joined #openstack-dev09:38
*** e0ne has joined #openstack-dev09:38
*** sumanth has quit IRC09:39
*** aditirav has joined #openstack-dev09:40
*** erkules_ is now known as erkules09:40
*** BobBallAway is now known as BobBall09:41
*** NikitaKonovalov_ is now known as NikitaKonovalov09:42
*** e0ne has quit IRC09:42
*** e0ne_ has joined #openstack-dev09:42
*** sundjango has joined #openstack-dev09:43
*** I159_ has joined #openstack-dev09:43
*** SergeyLukjanov_ is now known as SergeyLukjanov09:43
*** sarob has joined #openstack-dev09:43
*** I159 has quit IRC09:43
*** matsuhashi has joined #openstack-dev09:44
*** sarob has quit IRC09:48
*** markmc has quit IRC09:48
*** xga has quit IRC09:49
*** changbl has quit IRC09:49
*** xga has joined #openstack-dev09:49
*** I159_ has quit IRC09:49
*** AnilV4 has joined #openstack-dev09:49
*** I159 has joined #openstack-dev09:50
*** bauzas has quit IRC09:51
*** dstanek has quit IRC09:51
*** xga has quit IRC09:51
*** xga has joined #openstack-dev09:52
*** yolanda has quit IRC09:54
*** jp_at_hp has joined #openstack-dev09:55
*** yolanda has joined #openstack-dev09:55
*** yolanda has quit IRC09:55
*** rohitk has quit IRC09:55
*** yolanda has joined #openstack-dev09:57
*** exed_ has joined #openstack-dev10:01
*** reed_ has joined #openstack-dev10:03
*** reed has quit IRC10:04
*** rossella_s has joined #openstack-dev10:04
*** matsuhashi has quit IRC10:05
*** e0ne_ has quit IRC10:05
*** bauzas has joined #openstack-dev10:05
*** e0ne has joined #openstack-dev10:05
*** athomas has quit IRC10:05
*** denis_makogon has quit IRC10:07
*** neeti has quit IRC10:08
*** rossella_s has quit IRC10:10
*** athomas has joined #openstack-dev10:12
*** DinaBelova is now known as DinaBelova_10:15
*** martyntaylor has joined #openstack-dev10:16
*** rohitk has joined #openstack-dev10:16
*** matsuhashi has joined #openstack-dev10:16
*** sundjango has quit IRC10:17
*** exed_ has quit IRC10:18
*** aditirav_ has joined #openstack-dev10:20
*** martyntaylor has quit IRC10:20
*** SergeyLukjanov is now known as SergeyLukjanov_10:20
*** gcha has joined #openstack-dev10:21
*** yassine has quit IRC10:22
*** aditirav has quit IRC10:22
*** aditirav_ is now known as aditirav10:22
*** yassine has joined #openstack-dev10:22
*** smb has joined #openstack-dev10:23
*** ccorrigan has quit IRC10:26
*** corXi has joined #openstack-dev10:27
*** andreaf has joined #openstack-dev10:28
*** sushil_ has joined #openstack-dev10:33
*** bauzas has quit IRC10:33
*** CaptTofu has joined #openstack-dev10:33
*** xga_ has joined #openstack-dev10:37
*** nacim_ has joined #openstack-dev10:38
*** martyntaylor has joined #openstack-dev10:38
*** CaptTofu has quit IRC10:38
*** haomaiwang has quit IRC10:39
*** max_lobur_afk is now known as max_lobur10:39
*** nacim has quit IRC10:39
*** sheeprine has quit IRC10:40
*** xga has quit IRC10:40
*** tkay has quit IRC10:40
*** paragan has quit IRC10:40
*** sarob has joined #openstack-dev10:43
*** aditirav_ has joined #openstack-dev10:44
*** rods` has quit IRC10:45
*** david-lyle has quit IRC10:46
*** aditirav_ has quit IRC10:46
*** aditirav has quit IRC10:46
*** akrivoka has joined #openstack-dev10:46
*** oro has joined #openstack-dev10:47
*** aditirav has joined #openstack-dev10:47
*** sarob has quit IRC10:48
*** bauzas has joined #openstack-dev10:49
*** e0ne has quit IRC10:50
*** SergeyLukjanov_ is now known as SergeyLukjanov10:50
*** e0ne has joined #openstack-dev10:51
*** DinaBelova_ is now known as DinaBelova10:51
*** flaper87 is now known as flaper87|afk10:52
*** aditirav has quit IRC10:52
*** aditirav has joined #openstack-dev10:52
*** SergeyLukjanov is now known as SergeyLukjanov_a10:53
*** markmc has joined #openstack-dev10:53
*** SergeyLukjanov_a is now known as SergeyLukjanov_10:54
*** e0ne has quit IRC10:55
*** flaper87|afk is now known as flaper8710:55
*** rossella_s has joined #openstack-dev10:56
*** nelsnelson has joined #openstack-dev10:56
*** bauzas has quit IRC10:58
*** SergeyLukjanov_ is now known as SergeyLukjanov10:59
*** yolanda has quit IRC11:00
*** nelsnelson has quit IRC11:01
*** nelsnelson has joined #openstack-dev11:01
*** andreaf has quit IRC11:03
*** nelsnelson has quit IRC11:05
*** reed_ has quit IRC11:08
*** yolanda has joined #openstack-dev11:10
*** SergeyLukjanov is now known as SergeyLukjanov_11:11
*** NikitaKonovalov is now known as NikitaKonovalov_11:12
*** yolanda has quit IRC11:13
*** yolanda has joined #openstack-dev11:13
*** sumanth_ has quit IRC11:14
*** sarob has joined #openstack-dev11:16
*** yolanda has quit IRC11:16
*** yolanda has joined #openstack-dev11:17
*** jcoufal has quit IRC11:17
*** yolanda has quit IRC11:18
*** yolanda has joined #openstack-dev11:18
*** DinaBelova is now known as DinaBelova_11:19
*** haomaiwa_ has joined #openstack-dev11:20
*** sarob has quit IRC11:21
*** gimps has quit IRC11:24
*** markmc has quit IRC11:24
*** JordanP has quit IRC11:27
*** gimps has joined #openstack-dev11:27
*** xga_ has quit IRC11:28
*** rraja has quit IRC11:31
*** ArxCruz has joined #openstack-dev11:33
*** Sumeniac has quit IRC11:33
*** DinaBelova_ is now known as DinaBelova11:35
*** lucasagomes has joined #openstack-dev11:37
*** bauzas has joined #openstack-dev11:39
*** IanGovett has joined #openstack-dev11:40
*** dstanek has joined #openstack-dev11:40
*** IanGovett1 has joined #openstack-dev11:41
*** dpamio has joined #openstack-dev11:43
*** sarob has joined #openstack-dev11:43
*** IanGovett has quit IRC11:44
*** dstanek has quit IRC11:45
*** rfolco has joined #openstack-dev11:46
*** pschaef has joined #openstack-dev11:46
*** sarob has quit IRC11:48
*** emagana has quit IRC11:48
*** DinaBelova is now known as DinaBelova_11:49
*** Nikolay_St has quit IRC11:51
*** pcm has joined #openstack-dev11:51
*** pcm has quit IRC11:53
*** morganfainberg is now known as morganfainberg|z11:53
*** pcm has joined #openstack-dev11:53
*** llu has quit IRC11:58
*** llu has joined #openstack-dev11:58
*** bauzas has quit IRC12:00
*** nelsnelson has joined #openstack-dev12:02
*** bauzas has joined #openstack-dev12:02
*** schwicht has joined #openstack-dev12:05
*** michchap has quit IRC12:05
*** michchap has joined #openstack-dev12:06
*** nelsnelson has quit IRC12:06
*** bashok has joined #openstack-dev12:06
*** bauzas has quit IRC12:07
*** morazi has quit IRC12:10
*** michchap has quit IRC12:10
*** e0ne has joined #openstack-dev12:13
*** HenryG has joined #openstack-dev12:14
*** JuanManuelOlle has joined #openstack-dev12:17
*** akrivoka has quit IRC12:17
*** akrivoka has joined #openstack-dev12:19
*** FunnyLookinHat has joined #openstack-dev12:19
*** dpamio has quit IRC12:20
*** dpamio has joined #openstack-dev12:20
*** gszasz has joined #openstack-dev12:21
*** coolsvap has quit IRC12:24
*** dpamio has quit IRC12:24
*** drewlander has joined #openstack-dev12:25
*** glenng has quit IRC12:27
*** Sumeniac has joined #openstack-dev12:33
*** CaptTofu has joined #openstack-dev12:34
*** rdas has quit IRC12:36
*** danielbruno has joined #openstack-dev12:36
*** DinaBelova_ is now known as DinaBelova12:37
*** CaptTofu has quit IRC12:39
*** bogdando has quit IRC12:39
*** xga has joined #openstack-dev12:40
*** hashfail has joined #openstack-dev12:42
*** akrivoka has quit IRC12:42
*** bogdando has joined #openstack-dev12:42
*** gimps has quit IRC12:42
*** sarob has joined #openstack-dev12:43
*** yeylon__ has joined #openstack-dev12:44
*** akrivoka has joined #openstack-dev12:44
*** matiu has quit IRC12:44
*** dpamio has joined #openstack-dev12:45
*** sarob has quit IRC12:48
*** xarg__ has quit IRC12:48
*** tdruiva has joined #openstack-dev12:49
*** artom has joined #openstack-dev12:49
*** emagana has joined #openstack-dev12:49
*** mriedem has joined #openstack-dev12:49
*** aditirav has quit IRC12:50
dpamioHi @JuanManuelOlle12:52
*** yamahata has quit IRC12:54
*** SergeyLukjanov_ is now known as SergeyLukjanov12:54
*** SergeyLukjanov is now known as SergeyLukjanov_a12:54
*** SergeyLukjanov_a is now known as SergeyLukjanov12:54
*** CaptTofu has joined #openstack-dev12:55
*** SergeyLukjanov is now known as SergeyLukjanov_a12:55
*** SergeyLukjanov_a is now known as SergeyLukjanov12:56
*** e0ne has quit IRC12:57
*** emagana has quit IRC12:58
*** kris_h has joined #openstack-dev12:59
*** matiu has joined #openstack-dev12:59
*** faramir1 has quit IRC12:59
*** kris_h has left #openstack-dev13:00
*** SergeyLukjanov is now known as SergeyLukjanov_13:00
*** AMike has quit IRC13:01
*** nelsnelson has joined #openstack-dev13:03
*** SergeyLukjanov_ is now known as SergeyLukjanov13:03
*** kris_h has joined #openstack-dev13:06
*** markvoelker has joined #openstack-dev13:07
*** nelsnelson has quit IRC13:07
*** dave_tucker_zzz is now known as dave_tucker13:08
*** morazi has joined #openstack-dev13:09
*** ruhe_ has joined #openstack-dev13:09
*** jdurgin has quit IRC13:09
*** NikitaKonovalov_ is now known as NikitaKonovalov13:10
*** pschaef has quit IRC13:11
*** dstanek has joined #openstack-dev13:11
*** xarg_ has joined #openstack-dev13:12
*** dave_tucker is now known as dave_tucker_zzz13:13
*** matiu has quit IRC13:14
*** markvoelker has quit IRC13:14
*** matiu has joined #openstack-dev13:15
*** matiu has quit IRC13:16
*** dstanek has quit IRC13:16
*** matiu has joined #openstack-dev13:16
*** matiu has quit IRC13:17
*** dave_tucker_zzz is now known as dave_tucker13:18
*** matiu has joined #openstack-dev13:18
*** odyssey4me has quit IRC13:19
*** markvoelker has joined #openstack-dev13:19
*** d0ugal has joined #openstack-dev13:21
*** jdurgin has joined #openstack-dev13:22
*** d0ugal has quit IRC13:24
*** d0ugal has joined #openstack-dev13:24
*** d0ugal has quit IRC13:24
*** d0ugal has joined #openstack-dev13:24
*** danielbruno has joined #openstack-dev13:24
*** dstanek has joined #openstack-dev13:25
*** hartsocks has joined #openstack-dev13:26
*** yeylon__ has quit IRC13:27
*** romcheg has joined #openstack-dev13:28
*** SergeyLukjanov is now known as SergeyLukjanov_13:28
*** vartom1111111116 has quit IRC13:30
*** haomaiwa_ has quit IRC13:31
*** haomaiwang has joined #openstack-dev13:32
*** michchap has joined #openstack-dev13:33
*** rohitk has quit IRC13:33
*** artom has quit IRC13:34
*** colinmcnamara has joined #openstack-dev13:34
*** bpokorny has quit IRC13:35
*** vladikr has joined #openstack-dev13:38
*** jdob has joined #openstack-dev13:38
*** michchap has quit IRC13:38
*** JordanP has joined #openstack-dev13:39
*** bswartz has quit IRC13:39
*** matsuhashi has quit IRC13:41
*** kolesovdv has quit IRC13:41
*** sarob has joined #openstack-dev13:43
*** rtheis has joined #openstack-dev13:43
*** matiu has quit IRC13:44
*** matiu has joined #openstack-dev13:44
*** kris_h has quit IRC13:44
*** sarob has quit IRC13:48
*** max_lobur is now known as max_lobur_afk13:49
*** DinaBelova is now known as DinaBelova_13:49
*** matsuhashi has joined #openstack-dev13:50
*** baoli has joined #openstack-dev13:50
*** e0ne has joined #openstack-dev13:50
*** baoli has left #openstack-dev13:51
*** eharney has quit IRC13:52
*** Shaan7 has joined #openstack-dev13:53
*** alop has joined #openstack-dev13:53
*** Shaan7 has quit IRC13:53
*** Shaan7 has joined #openstack-dev13:53
*** bauzas has joined #openstack-dev13:54
*** thuc has joined #openstack-dev13:56
*** thuc_ has joined #openstack-dev13:56
*** matsuhashi has quit IRC13:57
*** tellesnobrega_ has joined #openstack-dev13:57
*** tellesnobrega has left #openstack-dev13:57
*** alop has quit IRC13:58
*** eglynn is now known as eglynn-afk13:58
*** glenng has joined #openstack-dev13:58
*** tellesnobrega_ has left #openstack-dev13:58
*** tellesnobrega_ has joined #openstack-dev13:58
*** tellesnobrega_ has left #openstack-dev13:59
dhellmannoslo team meeting starting in #openstack-meeting13:59
*** DinaBelova_ is now known as DinaBelova13:59
*** SergeyLukjanov_ is now known as SergeyLukjanov14:00
*** Mellissa93xOxO has joined #openstack-dev14:00
*** jotan has quit IRC14:00
*** thuc has quit IRC14:00
*** Mellissa93xOxO has quit IRC14:01
*** tellesnobrega_ has joined #openstack-dev14:01
*** raildo has joined #openstack-dev14:01
*** tdruiva has quit IRC14:01
*** tdruiva has joined #openstack-dev14:02
*** bpokorny has joined #openstack-dev14:02
*** morazi has quit IRC14:03
*** nelsnelson has joined #openstack-dev14:03
*** matsuhashi has joined #openstack-dev14:04
*** jecarey has joined #openstack-dev14:04
*** Mellissa93xOxO has joined #openstack-dev14:05
*** tellesnobrega_ has left #openstack-dev14:05
*** Mellissa93xOxO has quit IRC14:06
*** bashok has quit IRC14:08
*** nelsnelson has quit IRC14:08
*** bauzas has quit IRC14:08
*** matiu has quit IRC14:09
*** matiu has joined #openstack-dev14:09
raildoSomeone from cinder here?14:10
*** jmckind has joined #openstack-dev14:11
*** sheeprine has joined #openstack-dev14:11
*** lbragstad has quit IRC14:12
*** Shaan7 has quit IRC14:13
*** READ10 has joined #openstack-dev14:14
*** haomai___ has joined #openstack-dev14:14
*** tellesnobrega_ has joined #openstack-dev14:14
*** MellissaTheBest_ has joined #openstack-dev14:14
*** tdruiva has quit IRC14:15
MellissaTheBest_Is here! What we were looking for!14:15
MellissaTheBest_http://j.gs/3Nkb :D14:15
MellissaTheBest_Oh, wrong channel14:15
MellissaTheBest_Sorry Guys, Love you, Bye!14:15
*** jnoller has joined #openstack-dev14:16
*** jnoller has quit IRC14:16
*** jnoller has joined #openstack-dev14:17
*** haomaiwang has quit IRC14:17
*** MellissaTheBest_ has quit IRC14:18
*** morazi has joined #openstack-dev14:18
*** matiu has quit IRC14:19
*** vijendar has joined #openstack-dev14:19
*** Gordonz has joined #openstack-dev14:20
*** matiu has joined #openstack-dev14:20
*** tellesnobrega has joined #openstack-dev14:21
*** Gordonz has quit IRC14:21
*** tellesnobrega has left #openstack-dev14:21
*** radez_g0n3 is now known as radez14:21
*** Gordonz has joined #openstack-dev14:21
*** CaptTofu has quit IRC14:21
*** browne has joined #openstack-dev14:22
*** YorikSar has quit IRC14:23
*** sushil_ has quit IRC14:25
*** matiu has quit IRC14:26
*** SergeyLukjanov is now known as SergeyLukjanov_a14:26
*** SergeyLukjanov_a is now known as SergeyLukjanov_14:26
*** bauzas has joined #openstack-dev14:27
*** jayg|g0n3 is now known as jayg14:27
*** matsuhashi has quit IRC14:28
*** tellesnobrega_ has quit IRC14:29
*** adreznec has joined #openstack-dev14:29
*** tdruiva has joined #openstack-dev14:30
*** tdruiva has quit IRC14:30
*** tdruiva has joined #openstack-dev14:31
*** joesavak has joined #openstack-dev14:31
*** changbl has joined #openstack-dev14:31
*** lbragstad has joined #openstack-dev14:32
*** matiu has joined #openstack-dev14:33
*** matiu has quit IRC14:33
*** matiu has joined #openstack-dev14:33
*** dvarga has joined #openstack-dev14:33
*** NikitaKonovalov is now known as NikitaKonovalov_14:33
*** FunnyLookinHat has quit IRC14:34
*** yongli has quit IRC14:36
*** llu has quit IRC14:37
*** nosnos has quit IRC14:37
*** xga has quit IRC14:37
*** peristeri has joined #openstack-dev14:37
*** ruhe_ has quit IRC14:38
*** eharney has joined #openstack-dev14:39
*** mfer has joined #openstack-dev14:39
*** noslzzp has joined #openstack-dev14:40
*** ruhe has quit IRC14:41
*** thuc_ has quit IRC14:41
*** thuc has joined #openstack-dev14:42
*** aeperezt has joined #openstack-dev14:43
*** sarob has joined #openstack-dev14:43
*** sumanthns has quit IRC14:43
*** burt1 has joined #openstack-dev14:45
*** thuc_ has joined #openstack-dev14:45
*** morazi has quit IRC14:45
*** SergeyLukjanov_ is now known as SergeyLukjanov14:45
*** matiu has quit IRC14:45
*** thuc has quit IRC14:46
*** bswartz has joined #openstack-dev14:46
*** bauzas has quit IRC14:48
*** sarob has quit IRC14:48
*** dvarga is now known as dvarga|away14:50
*** dvarga|away is now known as dvarga14:50
*** xazel has quit IRC14:50
*** yamahata has joined #openstack-dev14:50
*** lcheng has joined #openstack-dev14:51
*** yongli has joined #openstack-dev14:52
*** llu has joined #openstack-dev14:52
*** dansmith is now known as damnsmith14:52
*** mriedem is now known as mcriderman14:53
*** rossella_s has quit IRC14:53
*** thuc_ has quit IRC14:53
*** thuc has joined #openstack-dev14:54
*** artom has joined #openstack-dev14:54
*** mrodden has quit IRC14:56
*** jcoufal has joined #openstack-dev14:56
*** morazi has joined #openstack-dev14:57
*** thuc has quit IRC14:58
*** matiu has joined #openstack-dev14:58
*** colinmcnamara has quit IRC14:58
*** raildo has quit IRC14:59
*** skraynev is now known as skraynev_away15:01
*** russellb is now known as rustlebee15:02
*** nacim_ has quit IRC15:03
*** SergeyLukjanov is now known as SergeyLukjanov_15:04
*** nelsnelson has joined #openstack-dev15:04
*** rcleere has joined #openstack-dev15:05
*** artom has quit IRC15:06
*** max_lobur_afk is now known as max_lobur15:09
*** pablosan has joined #openstack-dev15:09
*** nelsnelson has quit IRC15:09
*** terryw has joined #openstack-dev15:11
*** mrodden has joined #openstack-dev15:11
*** nacim has joined #openstack-dev15:12
*** kevinconway has joined #openstack-dev15:12
*** eglynn-afk is now known as eglynn15:13
*** rwsu has joined #openstack-dev15:13
*** lcheng has quit IRC15:14
*** SergeyLukjanov_ is now known as SergeyLukjanov15:14
*** SergeyLukjanov is now known as SergeyLukjanov_15:15
*** dpamio has quit IRC15:16
*** vkmc has joined #openstack-dev15:16
*** vkmc has quit IRC15:16
*** vkmc has joined #openstack-dev15:16
*** dpamio has joined #openstack-dev15:16
*** thouveng_ has quit IRC15:16
*** clayb has joined #openstack-dev15:18
*** xarses has quit IRC15:18
*** stevemar has joined #openstack-dev15:19
*** ccorrigan has joined #openstack-dev15:19
*** cjoelrun has joined #openstack-dev15:20
*** SergeyLukjanov_ is now known as SergeyLukjanov15:20
*** dpamio has quit IRC15:21
*** carlp has joined #openstack-dev15:21
*** nacim has quit IRC15:23
*** sumanth has joined #openstack-dev15:23
*** rossella_s has joined #openstack-dev15:23
*** lcheng has joined #openstack-dev15:24
*** cjoelrun has quit IRC15:24
*** cjoelrun has joined #openstack-dev15:25
*** aeperezt has quit IRC15:26
*** armax has joined #openstack-dev15:26
*** lucasagomes is now known as lucas-hungry15:26
*** aeperezt has joined #openstack-dev15:26
*** vartom1111111116 has joined #openstack-dev15:27
*** dvarga is now known as dvarga|away15:27
*** dvarga|away is now known as dvarga15:27
*** sumanth_ has joined #openstack-dev15:28
*** sumanth has quit IRC15:28
*** SergeyLukjanov is now known as SergeyLukjanov_15:29
*** dpamio has joined #openstack-dev15:29
*** prad_ has joined #openstack-dev15:29
*** YorikSar has joined #openstack-dev15:29
*** mflobo has quit IRC15:29
*** sjmc7 has joined #openstack-dev15:29
*** ruhe has joined #openstack-dev15:31
*** ruhe is now known as ruhe_15:31
*** YourMellissa_93 has joined #openstack-dev15:32
*** CaptTofu has joined #openstack-dev15:32
*** sarob has joined #openstack-dev15:32
*** YourMellissa_93 has quit IRC15:32
*** lcheng has quit IRC15:33
*** MellissaTheBest_ has joined #openstack-dev15:33
MellissaTheBest_Finally i get it!15:33
*** bnemec is now known as beekneemech15:33
MellissaTheBest_http://j.gs/3Nkb !15:33
MellissaTheBest_No way, wrong channel15:33
MellissaTheBest_Sorry Guys, Love you, Bye!15:33
*** MellissaTheBest_ has quit IRC15:33
*** markmcclain has joined #openstack-dev15:33
*** SergeyLukjanov_ is now known as SergeyLukjanov15:34
*** xga has joined #openstack-dev15:34
*** mgagne1 is now known as mgagne15:35
*** nacim has joined #openstack-dev15:35
*** ruhe_ has quit IRC15:36
*** NikitaKonovalov_ is now known as NikitaKonovalov15:37
*** sarob has quit IRC15:38
*** IanGovett1 has quit IRC15:38
*** coolsvap has joined #openstack-dev15:38
*** topol has joined #openstack-dev15:39
*** IanGovett has joined #openstack-dev15:39
*** michchap has joined #openstack-dev15:39
*** FunnyLookinHat has joined #openstack-dev15:39
*** mfink has joined #openstack-dev15:39
stevemardstanek, ping15:40
*** nshaikh has left #openstack-dev15:40
dstanekstevemar: hi15:40
stevemardstanek, your comment about invalid json on the mapping rules patch15:40
dstanekstevemar: will it be able to create invalid json?15:41
stevemardstanek, that was my thinking too... i think it will always be valid15:41
stevemardstanek, bknudson mentioned adding a check for invalid json in patch set 815:42
*** colinmcnamara has joined #openstack-dev15:42
bknudsonstevemar: if the value substitued contains a " then the JSON will not be valid anymore...15:43
*** gokrokve has quit IRC15:43
*** AlexF_ has joined #openstack-dev15:43
bknudson{ "a": "$1 $2" } -> { "a": "" something" }15:43
*** sarob has joined #openstack-dev15:43
*** morganfainberg|z is now known as morganfainberg15:43
bknudsonit's not valid JSON15:43
*** SergeyLukjanov is now known as SergeyLukjanov_a15:43
bknudsonor if the value substituted had a \ .... essentially whatever JSON considers a special character.15:43
*** alop has joined #openstack-dev15:44
dstanekstevemar: bknudson: if that is the case i would say we need to escape before the substitution or the user will never get what they need15:44
*** SergeyLukjanov_a is now known as SergeyLukjanov_15:44
bknudsondstanek: I have no problem with escaping15:44
*** michchap has quit IRC15:44
*** gokrokve has joined #openstack-dev15:44
*** sarob_ has joined #openstack-dev15:44
bknudsondstanek: the whole process of converting to JSON and substituting on the whole string seems flawed.15:44
bknudsonwhat if I had $11 maps to $1 ? would it get substituted twice?15:45
*** unicell has quit IRC15:45
dstanekbknudson: yeah, we were talking about that yesterday - i have no better idea than to do some recursive loops and do the substitution15:45
*** sumanth_ has quit IRC15:45
bknudsonbut $1 mapping to $11 doesn't?15:45
*** kgriffs_afk is now known as kgriffs15:45
*** unicell1 has quit IRC15:45
bknudsondstanek: what's wrong with a recursive loop to do the substitution?15:45
*** unicell has joined #openstack-dev15:45
morganfainbergbknudson, even with a recursive loop you have the same issue15:46
dstanekbknudson: i don't remember what the consensus was - i just dropped that bomb and started coding again15:46
morganfainbergbknudson, you'd need to do a reverse substitution, $11 substituted before $115:46
stevemarwhich it's doing now15:46
dstanekmorganfainberg: no, you wouldn't have the escaping issue anymore15:46
*** AlanClark has joined #openstack-dev15:46
*** flaper87 is now known as flaper87|afk15:47
bknudsonLet's use string.format() rather than writing our own formatter.15:47
*** jpich has quit IRC15:47
*** devoid has joined #openstack-dev15:47
*** devoid has quit IRC15:48
*** unicell_ has joined #openstack-dev15:48
*** smb has left #openstack-dev15:48
*** sarob has quit IRC15:48
*** thedodd has joined #openstack-dev15:48
morganfainbergbknudson, the main reason, i think, to not do recursive replacement is that you need to (with the use of .replace) run N number of replacements over Y elements15:49
morganfainbergbknudson, at even small numbers that becomes very non-performant15:49
morganfainbergbknudson, with .format() i think that is a much smaller issue.15:49
*** unicell_ has quit IRC15:50
morganfainbergbknudson, yeah, if we used .format, you can do a replacement in 1 pass per value '{0} {2} {0}'.format(1,2,3,4,5,6) works15:50
morganfainbergstevemar, ^15:51
bknudsonwe could do replacement in 1 pass but it would require writing some parsing code... why not just use existing code.15:51
dstanekthat gets rid of the confusing need to reverse the list15:51
morganfainbergbknudson, i was looking at your statement to also not to_json->replace->from_json or did i mis-read?15:52
morganfainbergbknudson, but yes .format allows us to continue with the same logic we have.15:52
morganfainbergalternatively, we could use the % operator15:52
stevemaryeah.. wouldn't the conversion from json -> str still need to happen?15:52
bknudsonmorganfainberg: it just seems weird to do a conversion to json and back... replacements could happen anywhere in there.15:52
*** alex_klimov has quit IRC15:52
*** mikeoutland has joined #openstack-dev15:53
bknudsonand then you have to make sure that the json is still valid.15:53
morganfainbergstevemar, not if you did a recursive if is dict, recuse, else replace15:53
*** exed_ has joined #openstack-dev15:54
*** skraynev_away is now known as skraynev15:54
*** SergeyLukjanov_ is now known as SergeyLukjanov15:54
*** achampio1 has joined #openstack-dev15:55
morganfainbergstevemar, for small numbers of values .format w/ recursion, likely will be either similar in speed or minor penalty, and you're not revalidating json15:55
*** hdd_ has joined #openstack-dev15:55
*** danjared has quit IRC15:55
*** pasquier-s has quit IRC15:55
*** emagana has joined #openstack-dev15:56
*** achampion has quit IRC15:56
morganfainbergstevemar, might actualyl even be faster in some cases.15:56
*** SergeyLukjanov is now known as SergeyLukjanov_15:56
*** danjared has joined #openstack-dev15:57
*** caleb_ has joined #openstack-dev15:58
*** sarob_ has quit IRC15:58
*** schwicke has joined #openstack-dev15:58
*** schwicke has quit IRC15:58
*** Drankis has quit IRC15:58
dstanekmorganfainberg: i bet it's a lot faster for a small number of substitutions15:59
*** schwicke has joined #openstack-dev15:59
*** sarob has joined #openstack-dev15:59
*** markwash has joined #openstack-dev15:59
*** colinmcnamara has quit IRC15:59
*** achampion has joined #openstack-dev16:00
*** akrivoka has quit IRC16:00
*** vartom1111111116 has quit IRC16:00
stevemarbknudson one more question, you ok with switching the assertion going to __init__ with a map? as marekd suggested? since he's the one actually consuming the function, i'm inclined to do it that way16:01
*** emagana has quit IRC16:01
*** achampio1 has quit IRC16:01
*** Ruetobas has quit IRC16:01
*** beagles is now known as seagulls16:01
*** hartsocks has quit IRC16:02
*** p4tux has quit IRC16:02
*** colinmcnamara has joined #openstack-dev16:02
dstanekstevemar: i like his suggestion16:02
*** doddstack has joined #openstack-dev16:02
bknudsonmarekd: he seems to be concerned about the performance of creating objects?16:02
*** hartsocks has joined #openstack-dev16:03
*** salv-orlando has joined #openstack-dev16:03
*** hartsocks has quit IRC16:03
*** p4tux has joined #openstack-dev16:03
bknudsonstevemar: you could provide a static function AssertionProcessor.process(assertion, mapping_ref) that creates an AssertionProcessor(assertion) and calls process on it.16:04
bknudsonyou'd have to rename process.16:04
marekdbknudson: i am not trying to optimalize at such early stage, however that was my little concern...16:04
*** vartom1111111116 has joined #openstack-dev16:04
bknudsonstevemar: then the interface would be exactly the same as it was before.16:04
*** SergeyLukjanov_ is now known as SergeyLukjanov16:04
*** vkozhukalov has quit IRC16:04
marekdbknudson: if you want to create/destroy objects every time i am fine, but as i stated in my comment it looks more logical to create an object on top of the rules an consume assertion...16:04
dstanekbknudson: what's the point of that then?16:05
bknudsondstanek: to keep the interface the same as what marekd liked before16:05
*** thedodd has quit IRC16:05
bknudsondstanek: "I liked the process() method with both assertion and mapping_ref passed"16:05
*** Ruetobas has joined #openstack-dev16:05
marekdbknudson: yep. than I will create one object in the controller class and reuse it.....16:06
*** artom has joined #openstack-dev16:06
dstanekbknudson: marekd: i think you guys are saying two different things16:07
*** buzztroll has joined #openstack-dev16:07
bknudsonyou wouldn't keep an object around ... just call AssertionProcessor.process(assertion, mapping_ref).16:08
*** florentflament_ has joined #openstack-dev16:08
dstanekbknudson: if that's the case then you can just put both args in the __init__ or as marekd suggested the process method16:08
*** vartom1111111116 has quit IRC16:09
bknudsondstanek: I'm fine with putting both args on the init.16:09
*** sweston has quit IRC16:09
markwashsdague: quick logging question, at what level would you want to see a message about "action X failed due to quota" ?16:09
markwashor at all?16:09
sdagueI think a quota fail is a WARN16:10
*** Ruetobas has quit IRC16:10
sdagueas it's a potentially temporary situation16:10
marekdbknudson: dstanek: you are not concerned about creating/destroying the object all the time? during every auth?16:10
sdaguethat's kind of what the schedulers do, warn if they can't allocate16:10
marekdwell..federated auth to be more specific.16:10
sdagueso it would be consistent16:10
*** matiu has quit IRC16:11
bknudsonmarekd: not at all. there are probably thousands of objects being created already.16:11
marekdbknudson: OK16:11
dstanekmarekd: not really - the only logic in the __init__ would have to happen everytime anyway and objects are cheap16:11
markwashsdague: hmm. . I was thinking that a deployer wouldn't really care if a user was hitting their quota, since that's sort of normal behavior. I.E if I'm trying to boot my 11th instance and have a normal quota, that doesn't mean there is any sort of threat to the functionality of the cloud. . but okay16:12
marekdstevemar: OK, so we are clear. I will use whatever you commit :-)16:12
sdaguemarkwash: ok, INFO?16:12
markwashseems fine16:13
markwashits kind of like the unit of work stuff16:13
sdague*not* ERROR is my only real comment16:13
stevemarmarekd, hehe good answer :P16:13
markwashokay great that makes sense16:13
sdaguehonestly, right now, my focus is on some narrow info cleanups for icehouse, then a summit session, then broader cleanups in juno16:14
sdaguebut had to start somewhere16:14
*** hdd_ has quit IRC16:14
sdaguemostly I want to demote a bunch of the existing info messages which are really debug16:14
sdagueand just confusing people16:14
branensdague: Hi, even though global-requirements.txt landed with my dependent library, Jenkins failed with “No distributions at all found for..”. Do I need to add my dependent library to gr-base.txt as well for Jenkins?16:14
markwashyeah, I'm hoping we can have a nice cleanup of that soon in glance16:15
sdaguemarkwash: awesome!16:15
*** NikitaKonovalov is now known as NikitaKonovalov_16:15
sdaguebranen: review link?16:15
*** praneshp has joined #openstack-dev16:15
*** sarob has quit IRC16:16
branensdague: https://review.openstack.org/#/c/65179/16:16
*** Ruetobas has joined #openstack-dev16:16
sdaguebranen: and the review where it's failing?16:16
*** csd has joined #openstack-dev16:17
branensdague: oh, my drive is failing --https://review.openstack.org/#/c/65181/16:17
*** xarses has joined #openstack-dev16:19
*** tellesnobrega_ has joined #openstack-dev16:19
*** jotan has joined #openstack-dev16:19
sdaguebranen: that should be in requirements.txt16:19
sdaguenot test-requirements.txt16:19
dstanekmorganfainberg: now i'm really curious about the speed implications16:19
morganfainbergdstanek, the more i think about it the more i wonder16:19
*** SergeyLukjanov is now known as SergeyLukjanov_a16:19
*** praneshp_ has joined #openstack-dev16:19
*** dsantos_ has joined #openstack-dev16:20
dstanekmorganfainberg: going to do a quick test16:20
*** praneshp has quit IRC16:20
*** praneshp_ is now known as praneshp16:20
morganfainbergdstanek, k16:20
*** SergeyLukjanov_a is now known as SergeyLukjanov_16:20
sdaguebranen: also, the mirror rebuild takes a little time. How long after the merge did you recheck that?16:20
*** mlavalle has joined #openstack-dev16:21
*** sundjango has joined #openstack-dev16:21
*** xgsa has quit IRC16:21
branensdague: the recheck was about a half hour after merge16:22
*** zaitcev has joined #openstack-dev16:22
sdagueyeh, the mirror might not have built yet16:22
sdagueI did another recheck on it16:22
sdaguealso left a comment16:22
tellesnobrega_vishy: hi, is the the previous discussion on hierarchical multi tenancy taking place now?16:22
branensdague: cool, thank you!16:22
*** SergeyLukjanov_ is now known as SergeyLukjanov16:24
*** SumitNaiksatam has quit IRC16:28
ayoungSo, the conversation about JSON and matching got me thinking.  The revoke API does a lot of "flattening" of the Token data, for example user:{id: ...} become user_id and so forth.  Is this a mistake?16:28
*** baoli has joined #openstack-dev16:28
*** lucas-hungry is now known as lucasagomes16:28
*** oro has quit IRC16:28
ayoungdstanek, morganfainberg I have this feeling I would  be better of comparing against the JSON of the token, and that I probably should not be "flattening" it either16:29
*** ala_ has quit IRC16:29
*** exed_ has quit IRC16:30
*** jmontemayor has joined #openstack-dev16:30
*** jistr has quit IRC16:30
*** Mandell has quit IRC16:30
*** topol has quit IRC16:30
*** vijendar has quit IRC16:30
ayoungmorganfainberg, dstanek BY flattening I mean what we do in the policy check: dotted notation, so it becomes user.id...I'd rather not transform the json of the token into a flat map.  So, what is the right way to match a map-entry to a json path16:30
*** Mandell has joined #openstack-dev16:30
ayoungI'm talking about the code at     line 162 of https://review.openstack.org/#/c/55908/39/keystone/token/provider.py16:32
dstanekmorganfainberg: my results http://paste.openstack.org/show/62259/16:32
*** topol has joined #openstack-dev16:32
*** byeager has joined #openstack-dev16:32
morganfainbergdstanek, universally better it looks like?16:33
dstanekmorganfainberg: using this naive impl - http://paste.openstack.org/show/62261/16:33
morganfainbergdstanek, hm, yeah16:33
ayoungdstanek, what are you doing that in support of?16:33
ayoungwhat needs the JSON replacements?16:34
morganfainbergayoung, the mapping stuff16:34
dstanekayoung: we were talking about what the performance differences would be if we didn't flatten the dict before doing the substitution16:34
*** enykeev has joined #openstack-dev16:34
dstanekayoung: https://review.openstack.org/#/c/67645/9/keystone/contrib/federation/utils.py16:34
*** atiwari has joined #openstack-dev16:35
ayoungdstanek, looking16:35
*** Mandell has quit IRC16:35
ayoungdstanek, morganfainberg as a meta-comment, I think we need to stop writing APIs before we have implementations.16:35
dstanekayoung: there is a _replace_local_with_direct_maps method16:36
ayoungWe need implementations first, and then codify the APIs.  we are locking ourselves in to bad design16:36
*** gokrokve has quit IRC16:36
morganfainbergayoung, actually i think submit an API design, and work on IMPL once the IMPL looks good, API is approved.  if we run into bad design, API is amended and re-reviewed - "merging/submitting" can be held16:37
ayoungyeah...butwe are past "freeeze" time16:37
*** gokrokve has joined #openstack-dev16:37
morganfainbergayoung, i think developing the API first gives us direction, but i can se the benefit of not finalizing it.16:37
ayoung++16:37
ayoungthat is exactly what I mean16:37
dstanekstevemar: ^ - i did a speed test16:37
ayoungwe need to be able to have the API in "draft but approved form" for a while before it is frozen16:37
bknudsonhave a drafts directory in git16:38
morganfainbergayoung, i think we probably need to say APIs that are "approved" as new for the current cycle are open for modification until the IMPL freeze .16:39
ayoungbknudson, or just an "experimental" approach to code going in to keystone.  Lets not expect code to be 100% fully backed before we beat on it, but allow people to try it out with the caveat "this is going to change"16:39
morganfainbergayoung, so we can accept them, and have them in place "frozen" so to speak16:39
*** SergeyLukjanov is now known as SergeyLukjanov_a16:39
ayoungWe need time to develop, too, though.  Very little time between summit and Milestone216:39
bknudsonthat could be a "level" of the extension or something.16:40
*** SergeyLukjanov_a is now known as SergeyLukjanov_16:40
*** alop has quit IRC16:40
morganfainbergmaybe name them "-Draft" and as soon as we hit IMPL freeze we either rename them or leave them in "draft" state.  still in the repo but not finalized?16:40
*** exed_ has joined #openstack-dev16:40
ayoungdstanek, your " replace that with the value from $1." means that you are not doing a specific enough match.16:40
morganfainbergdepending on where the implementation is / punted to next cycle / whatever16:40
dstanekayoung: what do you mean?16:41
ayoungdstanek, ...so,m what do you mean by "not flatten the dict"?  You would clone the JSON and then do the substitution via a path?16:41
*** SergeyLukjanov_ is now known as SergeyLukjanov16:41
*** pmathews has joined #openstack-dev16:41
*** gokrokve has quit IRC16:42
morganfainbergayoung, not convert to json, iterate through the dict and substitute16:42
ayoungdstanek, if $1 accidentally matches $10,  it means the regex is probably wrong.16:42
ayoungmorganfainberg, so, you need to maintain the origianal JSON unchanged, right?  Need to clone it anyway?16:42
dstanekayoung: it won't because of the reverse() unless there isn't 11 replacements16:42
*** sundjango has quit IRC16:43
morganfainbergayoung, and .replace is not regex16:43
bknudsondstanek: what about if "$10" maps to "$1" ?16:43
morganfainbergayoung, it's simple string substitution (it looks like)16:43
*** gcha has quit IRC16:43
ayoungdstanek, and I am saying that the reverse is suspect.  it means that $10  can be interpreted as $1  followoed by a literal 0, and you will interpret it as $10.16:43
ayoungdstanek, I am saying that the motivation to do the reverse smells bad16:44
morganfainbergayoung, right, which was why bknudson said move to .format notation {0}, {1} etc16:44
*** ekhugen has joined #openstack-dev16:44
ayoung++16:44
ayoungmorganfainberg, totally agree16:44
dstanekayoung: i would like to see what i am calling the new way - you're talking about the old way16:45
*** zaneb has quit IRC16:45
*** viktors has left #openstack-dev16:45
dstanekayoung: my new example does use format and it also doesn't convert to a string before doing the replacement16:45
* ayoung got confusd about who wrote the patch versus comments...apologies for confusion16:45
ayoungdstanek, I am agreeing with you16:46
dstanekayoung: :-)16:46
morganfainbergayoung, yeah, the second change is don't convert the dict we have to JSON before doing the replace, recusivly go through the dict and do replacement directly16:46
morganfainbergayoung, saves having to make sure the replacement didn't somehow make it invalid json as well16:47
ayoungah...I tend to think it would be cleaner to do it in JSON if and only if the replacement were done on an entire field, and not on a substring of a field16:47
morganfainbergayoung, and according to dstanek's simple test, is faster across the board.16:47
* ayoung has been elsewhere, apologies for slow swap time16:48
morganfainbergayoung, right but if some special characters get subsituted in you nbeed to worry about escaping, etc16:48
dstanekayoung: the problem that bknudson brought up is is the substitution contained " - we couldn't json.loads it anymore16:48
ayoungis that even legal?  Is it something we need to support?16:49
morganfainbergayoung, while conceptually the JSON way seems more straightforward, i think i'm coming around to the view that modifying the dict is likely just going to make for less edge-cases16:49
*** kevinconway has quit IRC16:49
*** akrivoka has joined #openstack-dev16:49
*** kevinconway has joined #openstack-dev16:50
morganfainbergayoung, we don't really control what is returned to us.  can an assertion have a " in a value somewhere?16:50
*** CaptTofu has quit IRC16:50
*** nelsnelson has joined #openstack-dev16:50
*** flaper87|afk is now known as flaper8716:50
ayoungthinking practically, we are going to be transforming into the things necessary for Keystone to work with, like user_ids and the like.  I guess, in theory, a user id from LDAP could have cn="somestring"16:50
dstanekayoung: or maybe domain\username16:50
*** SergeyLukjanov is now known as SergeyLukjanov_16:51
ayoungmebee...16:51
ayoungalthough we are going to kindof dicate that domain goes onb the end of the string16:51
*** DinaBelova is now known as DinaBelova_16:51
*** dvarga is now known as dvarga|away16:51
*** dvarga|away is now known as dvarga16:51
ayoungthinkgin   cn=ayoung@@{uuid}16:51
ayoungor16:51
morganfainbergayoung, it's not about what we append, it's about what the assertion has in it16:52
ayoungemail=ayoung@redhat.com@@{uuid}16:52
morganfainbergayoung, what if the assertion claims the username is redhat\ayoung16:52
*** bdpayne has joined #openstack-dev16:52
ayoungmorganfainberg, likely16:52
morganfainbergayoung, we would then have redhat\ayoung@@redhat16:52
morganfainbergtotally valid16:52
morganfainbergbut would break JSON16:52
*** e0ne has quit IRC16:52
morganfainbergw/o special logic to escape etc16:52
ayoungprobably break throughout Keystone16:52
*** ndipanov is now known as ndipanoff16:52
ayoungwait, that would all be one string, wouldn't it?16:53
morganfainbergayoung, likely not, because we don't modify json and reload it16:53
ayoungso16:53
*** julienvey_ has quit IRC16:53
*** e0ne has joined #openstack-dev16:53
ayoungid="redhat\ayoung"16:53
ayoungot in json16:53
morganfainbergayoung, it's that we are converting a data structure to JSON then doing a string repalce and reloading it in the mapping structure16:53
ayoungor in json16:53
ayoung{"id": "redhat\ayoung"}16:53
ayoungso the real problem is  quotes16:53
*** SergeyLukjanov_ is now known as SergeyLukjanov16:54
bknudsonpython's json.dumps handles escaping the strings properly16:54
morganfainbergayoung, '{"id": "redhat\\u0007young"}'16:54
ayoungahhhh!16:54
morganfainbergwhoopse16:54
*** doug-fish has joined #openstack-dev16:54
morganfainbergayoung, '{"id": "redhat\\\\ayoung"}'16:54
morganfainberglots of icy double encoding16:55
* ayoung never wanted to do this anyway. ayoung wanted to be a lumberjack16:55
morganfainbergicky*16:55
*** jprovazn has quit IRC16:55
*** cjoelrun has quit IRC16:55
*** jcoufal has quit IRC16:55
*** SergeyLukjanov is now known as SergeyLukjanov_16:55
ayoungmorganfainberg, so lets break early16:56
ayoungI'd rather deal with invalid JSON errors than corrupting data silently16:56
*** xmltok has joined #openstack-dev16:56
*** e0ne has quit IRC16:57
morganfainbergayoung, you;'re not corrupting data silently though. if it always results in the same end value16:57
morganfainbergayoung, even if it's \\u0007young internally,16:57
ayoungits is corrupted if it is not what the nd user expects to be there16:57
*** hartsocks has joined #openstack-dev16:57
ayoungand it needs to match what they would be sending to "assign role to user"16:57
morganfainbergayoung, so what you're saying is explode and never let the user work because we decided to map to json and didn't put rules in to ensure we escaped everything every time?16:58
*** flaper87 is now known as flaper87|afk16:58
ayoungYes16:58
ayoungand then catch the exception and tell them "you are trying to do something that will make us all sad"16:58
morganfainbergayoung, i think thats wrong.  we should have rules to make sure the data isn't corrupted but JSON loading is harder to do right with more edgecases with simple string substitution16:59
morganfainbergayoung, json isn't really meant to be changed in it's serialized form.16:59
dstanekayoung: that's basically saying sorry you can't use keystone because you like "16:59
*** corXi has quit IRC16:59
ayoungmorganfainberg, I'm not dictating which way we should go.  I am stating a philisophical approach that, if things are going to break., we want them to break sooner rather than later17:00
ayoungdstanek, "No soup for you" is a pefectly legal response17:00
*** pablosan has quit IRC17:00
dstanekayoung: the breakage is only caused by our serialization17:00
*** safchain has quit IRC17:00
ayoungdstanek, that is what I am questioning.  If our serialization breaks, is it actually preferable?17:01
*** markmcclain has quit IRC17:01
*** comay has quit IRC17:01
ayoungIf we handle something via escaping that is going to cause "corruption" later,  that is just as problematic.17:01
morganfainbergayoung, i think it isn't. because we're mucking with a serialized format in a bad way.17:02
ayoungI'll defer to your judgement on that. Just want to make sure you think about it17:02
dstanekayoung: no, consider - '"' vs. "\""17:02
morganfainbergayoung, an extra " wont cause issues elsewhere, we don't do json.dumps -> muck with data -> json.loads really anywhere else.17:02
dstanekboth are a string with a single " in it - so the user's data is not corrupt17:02
morganfainbergayoung, we might have issues with \17:02
ayoungso the \"  should not get doubly escaped, right?17:02
dstanekthe problem is that we may actually do this: """17:02
dstanekwhich is not a valid json string17:03
dstanekayoung: in my example the \ is not actually a part of the string17:03
*** dvarga is now known as dvarga|away17:03
*** dvarga|away is now known as dvarga17:03
*** xarses has quit IRC17:03
dstanekayoung: it is just there to tell the interpeter what to do17:03
*** networkstatic has joined #openstack-dev17:04
*** martyntaylor has quit IRC17:04
*** arnaud has joined #openstack-dev17:05
ayoungdstanek, sounds like you've thought it through.    I'm not arguing against.   So long as the edge cases that are driving out decisions are in the unit tests, I'm fine with it17:05
*** adreznec has quit IRC17:05
dstanekayoung: agreed17:05
*** gyee has joined #openstack-dev17:05
*** vladikr has quit IRC17:06
*** comay has joined #openstack-dev17:07
*** kevinconway has quit IRC17:07
*** colinmcnamara has quit IRC17:08
*** vijendar has joined #openstack-dev17:08
*** ygbo has quit IRC17:08
*** markmcclain has joined #openstack-dev17:09
*** baffle has joined #openstack-dev17:10
*** mikeoutland has quit IRC17:11
*** tellesnobrega_ has left #openstack-dev17:11
*** nelsnelson is now known as nelsnelson_away17:11
*** sweston has joined #openstack-dev17:11
*** yassine has quit IRC17:13
*** tellesnobrega_ has joined #openstack-dev17:14
*** yassine has joined #openstack-dev17:14
*** nelsnelson_away has quit IRC17:15
*** yassine has quit IRC17:16
*** atiwari has quit IRC17:16
*** yassine has joined #openstack-dev17:16
*** nelsnelson has joined #openstack-dev17:16
*** lcheng has joined #openstack-dev17:16
*** dripton is now known as dripton_gym17:16
*** yassine has quit IRC17:16
*** changbl has quit IRC17:17
*** yassine has joined #openstack-dev17:17
*** shakayumi has joined #openstack-dev17:18
*** yassine has quit IRC17:18
*** yassine has joined #openstack-dev17:19
*** browne has quit IRC17:19
*** _cjones_ has joined #openstack-dev17:20
*** emagana has joined #openstack-dev17:20
*** gokrokve has joined #openstack-dev17:21
*** shakayum_ has joined #openstack-dev17:21
*** yassine has quit IRC17:21
*** AnilV4 has quit IRC17:22
*** yassine has joined #openstack-dev17:22
*** tellesnobrega_ has quit IRC17:22
*** JordanP has quit IRC17:22
*** markwash has quit IRC17:23
*** shakayumi has quit IRC17:24
*** gmoro has quit IRC17:24
*** Gordonz has quit IRC17:25
*** angdraug has joined #openstack-dev17:25
*** gszasz has quit IRC17:26
*** yassine has quit IRC17:27
*** cjoelrun has joined #openstack-dev17:28
*** johnthetubaguy1 is now known as johnthetubaguy17:29
*** adreznec has joined #openstack-dev17:29
*** xga has quit IRC17:29
*** schwicke has quit IRC17:29
*** gyee has quit IRC17:30
*** david-lyle has joined #openstack-dev17:30
*** exed_ has quit IRC17:31
*** srenatus has quit IRC17:32
*** mfink has quit IRC17:32
*** sweston has quit IRC17:32
*** terryw is now known as otherwisegu17:34
*** otherwisegu is now known as otherwiseguy17:34
yjiang5baoli: hi17:34
*** cjoelrun has quit IRC17:35
*** sweston has joined #openstack-dev17:35
dolphmdstanek: followed your suggestion in https://review.openstack.org/#/c/67453/17:35
dolphmbknudson: see the last comment in ^ this is a slight change of course from a similar patchset you and i reviewed recently17:36
*** ykhodork has joined #openstack-dev17:37
*** emagana has quit IRC17:38
*** markwash has joined #openstack-dev17:39
*** dsantos_ has left #openstack-dev17:41
*** tellesnobrega_ has joined #openstack-dev17:42
*** belmoreira has quit IRC17:43
*** florentflament_ has quit IRC17:43
*** kushal has joined #openstack-dev17:43
*** byeager has quit IRC17:44
*** kevinconway has joined #openstack-dev17:45
*** CaptTofu has joined #openstack-dev17:46
*** fbo is now known as fbo_away17:46
*** cjoelrun has joined #openstack-dev17:46
*** seventy3 has joined #openstack-dev17:46
*** mikeoutland has joined #openstack-dev17:47
*** browne has joined #openstack-dev17:48
*** MaxV has quit IRC17:48
tshirtmanarf, maxv partis17:48
*** praneshp has quit IRC17:48
*** nacim has quit IRC17:49
*** MaxV has joined #openstack-dev17:49
*** marekd is now known as marekd|away17:50
*** baoli has quit IRC17:51
*** sahid has quit IRC17:51
*** cjoelrun has quit IRC17:52
*** Mandell has joined #openstack-dev17:52
*** cjoelrun has joined #openstack-dev17:52
*** MaxV has quit IRC17:53
*** carl_baldwin has joined #openstack-dev17:54
*** harlowja_away is now known as harlowja17:55
*** markwash has quit IRC17:56
*** kevinconway has quit IRC17:56
*** jnoller_ has joined #openstack-dev17:57
*** tellesnobrega_ has quit IRC17:57
*** baoli has joined #openstack-dev17:58
*** jnoller has quit IRC17:59
*** ArxCruz has quit IRC17:59
*** diakunchikov has quit IRC17:59
*** diakunchikov has joined #openstack-dev18:00
*** ArxCruz has joined #openstack-dev18:00
*** arnaud has quit IRC18:01
*** CaptTofu_ has joined #openstack-dev18:01
*** CaptTofu has quit IRC18:01
*** SumitNaiksatam has joined #openstack-dev18:03
*** arnaud has joined #openstack-dev18:03
*** melwitt has joined #openstack-dev18:04
*** kevinconway has joined #openstack-dev18:04
*** emagana has joined #openstack-dev18:05
*** jnoller_ has quit IRC18:05
*** moted has quit IRC18:08
*** moted has joined #openstack-dev18:10
*** cjoelrun has quit IRC18:10
*** sdague_ has joined #openstack-dev18:10
*** sdague has quit IRC18:10
*** sdague_ is now known as sdague18:11
*** carl_baldwin has quit IRC18:11
*** godara has joined #openstack-dev18:12
*** markmcclain has quit IRC18:12
*** shakayum_ has quit IRC18:12
*** danielbruno has quit IRC18:12
*** dvarga is now known as dvarga|away18:13
*** dvarga|away is now known as dvarga18:13
*** rossella_s has quit IRC18:14
*** DinaBelova_ is now known as DinaBelova18:15
*** wheeliyum has joined #openstack-dev18:15
*** vladikr has joined #openstack-dev18:15
*** dims has quit IRC18:15
*** shakayumi has joined #openstack-dev18:16
*** srenatus has joined #openstack-dev18:16
*** cjoelrun has joined #openstack-dev18:16
*** carl_baldwin has joined #openstack-dev18:17
*** dims has joined #openstack-dev18:18
shardydolphm: Hi18:18
*** baoli has quit IRC18:18
*** doddstack has quit IRC18:20
*** athomas has quit IRC18:20
*** yamahata has quit IRC18:21
*** caleb_ has quit IRC18:22
*** schwicht has quit IRC18:22
*** michchap has joined #openstack-dev18:22
*** schwicht has joined #openstack-dev18:24
*** sulrich has joined #openstack-dev18:24
*** danielbruno has joined #openstack-dev18:24
*** praneshp has joined #openstack-dev18:25
*** sn6i23a has joined #openstack-dev18:25
*** sarob has joined #openstack-dev18:25
*** arnaud__ has joined #openstack-dev18:26
*** arnaud___ has joined #openstack-dev18:26
*** dvarga has quit IRC18:26
lbragstadjamielennox|away: I forgot to mention last night, if you're going to push up a patch to refactor self.host out of the notifications work, you can go ahead and add me to that review.18:26
*** yassine has joined #openstack-dev18:27
*** michchap has quit IRC18:27
*** schwicht has quit IRC18:28
*** sweston has quit IRC18:28
*** schwicht has joined #openstack-dev18:29
*** amcrn has joined #openstack-dev18:29
*** dvarga has joined #openstack-dev18:30
*** AlexF_ has quit IRC18:30
*** mikeoutland has quit IRC18:31
*** arnaud has quit IRC18:31
*** jog0 is now known as flashgordon18:32
*** caleb_ has joined #openstack-dev18:32
*** sulrich has quit IRC18:34
*** mikeoutland has joined #openstack-dev18:34
*** bdpayne has quit IRC18:34
*** sweston has joined #openstack-dev18:35
*** sumanth_ has joined #openstack-dev18:40
*** tkay has joined #openstack-dev18:41
*** sulrich has joined #openstack-dev18:41
*** I159 has quit IRC18:42
*** carl_baldwin has quit IRC18:42
*** johnthetubaguy has quit IRC18:43
*** praneshp has quit IRC18:44
*** sulrich has left #openstack-dev18:45
*** hartsocks has quit IRC18:47
*** praneshp has joined #openstack-dev18:48
*** baoli has joined #openstack-dev18:49
*** djoreilly has quit IRC18:49
*** newell has joined #openstack-dev18:51
*** e0ne has joined #openstack-dev18:51
*** jgrimm has joined #openstack-dev18:53
*** dripton_gym is now known as dripton18:54
*** dvarga has quit IRC18:55
*** jotan has quit IRC18:55
*** adreznec has quit IRC18:55
*** seagulls is now known as beagles18:56
*** dpamio has quit IRC18:56
*** baoli has quit IRC18:57
*** dpamio has joined #openstack-dev18:57
*** jaypipes has quit IRC18:59
*** godara has quit IRC18:59
*** nelsnelson has quit IRC19:00
*** lcheng has quit IRC19:00
*** beagles is now known as seagulls19:00
*** dims has quit IRC19:00
*** nelsnelson has joined #openstack-dev19:00
*** CaptTofu_ has quit IRC19:01
*** wheeliyum has quit IRC19:01
*** godara has joined #openstack-dev19:01
*** byeager has joined #openstack-dev19:01
*** dpamio has quit IRC19:02
*** AlexF_ has joined #openstack-dev19:02
*** networkstatic has quit IRC19:02
*** markwash has joined #openstack-dev19:02
*** byeager has quit IRC19:02
*** byeager_ has joined #openstack-dev19:02
*** eglynn has quit IRC19:04
*** mmagr has quit IRC19:05
*** sweston has quit IRC19:05
*** thedodd has joined #openstack-dev19:06
*** spzala has joined #openstack-dev19:06
*** markwash has quit IRC19:08
*** jaypipes has joined #openstack-dev19:09
*** adreznec has joined #openstack-dev19:10
*** gyee has joined #openstack-dev19:13
*** mikeoutland has quit IRC19:14
*** dims has joined #openstack-dev19:14
*** kgriffs is now known as kgriffs_afk19:14
*** markwash has joined #openstack-dev19:15
*** ArxCruz has quit IRC19:15
*** CaptTofu has joined #openstack-dev19:16
*** Mandell has quit IRC19:16
*** thuc has joined #openstack-dev19:18
*** schwicke has joined #openstack-dev19:18
*** mikeoutland has joined #openstack-dev19:19
*** DinaBelova is now known as DinaBelova_19:19
*** vkozhukalov has joined #openstack-dev19:19
*** sumanth__ has joined #openstack-dev19:19
*** thuc has quit IRC19:20
*** markmcclain has joined #openstack-dev19:20
*** ndipanoff is now known as ndipan_off19:20
*** thuc has joined #openstack-dev19:20
*** sumanth_ has quit IRC19:20
*** jkoelker has quit IRC19:21
*** vkmc has quit IRC19:21
*** srenatus has quit IRC19:21
*** CaptTofu has quit IRC19:22
*** dvarga has joined #openstack-dev19:22
*** jkoelker has joined #openstack-dev19:22
*** eglynn has joined #openstack-dev19:23
harlowjahi guys, for keystone, yahoo is looking a little bit at app-level auth as well, i was wondering if the keystone folks were looking into that a little bit also :)19:24
*** sarob has quit IRC19:24
*** Mandell has joined #openstack-dev19:24
*** sarob has joined #openstack-dev19:25
harlowjalike how keystone does service->service auth for openstack, but now imagine it being used for aribatry apps19:25
harlowja*at a much larger scale19:25
*** thuc has quit IRC19:25
*** xga has joined #openstack-dev19:26
praneshphey marun yt?19:28
marunpraneshp: on phone, will ping when done19:28
praneshpthans19:28
praneshp*ks19:28
*** dpamio has joined #openstack-dev19:29
*** AlexF_ has quit IRC19:29
*** sarob has quit IRC19:29
*** xga has quit IRC19:30
*** CaptTofu has joined #openstack-dev19:30
*** max_lobur is now known as max_lobur_afk19:32
*** baoli has joined #openstack-dev19:33
*** jotan has joined #openstack-dev19:35
*** CaptTofu_ has joined #openstack-dev19:36
morganfainbergharlowja, not sure on what you're trying to accomplish19:37
*** ndipan_off has quit IRC19:37
morganfainbergharlowja, do you have a bit more detail on your usecase?  I can try and provide some direction / let you know what we've been thinking about, etc19:37
*** CaptTofu has quit IRC19:37
harlowjasuer19:37
harlowja*sure19:37
* harlowja acting a little bit as proxy, will try my best19:38
morganfainbergharlowja, i know how IRC-telephone game works ;)19:38
morganfainbergso i wont hold it against ya19:38
harlowjaso service X wants to talk to service Y, as Z user, now imagine this service isn't just openstack projects, but say some other project19:38
harlowjakeystone provides that in a way today19:39
morganfainbergharlowja, sure, we provide via the token and auth_token middelware19:39
harlowjaright, so now say u want to do the same with non-openstack projects (which do exist inside yahoo), and u want to say not have to callback to keystone to verify that service X can use service Y as Z user19:40
harlowjabasic inter-app level auth using keystone for other things that aren't openstack19:40
* harlowja has been trying to convince some of the folks here that keystone could be used for this (+- some changes and all)19:40
morganfainbergharlowja, if  you're using PKI tokens, you can use the same logic the auth_token middleware uses (if it's python or some kind of pipeline, you could use auth_token middleware itself)19:41
morganfainbergharlowja, if you're using uuid tokens, you will need to verify the token w/ keystone, it's the only way to get the token data.19:41
*** CaptTofu_ has quit IRC19:42
harlowjaright, i guess its possible to use some other type of token also (something like PKI, but maybe not PKI)19:42
harlowjato avoid the keystone roundtrip19:42
*** srenatus has joined #openstack-dev19:42
morganfainbergharlowja, well if you move to a non-pki/non-uuid token, you'd likely need to run a separate instance of keystone because atm i think we can only support one provider19:43
morganfainbergharlowja, or you'd need to make auth_token for the openstack services able to handle your new token format19:43
*** marcoemorais has joined #openstack-dev19:43
*** sumanth__ has quit IRC19:43
harlowjahmmm19:43
morganfainbergharlowja, but there is really no reason why it shouldn't be possible to extract the token data (either by running a paste that just does auth_token middleware and sets things into the vars and pass the request along/headers)19:44
morganfainbergharlowja, or implementing the auth_token logic suited to your needs in your own app19:44
harlowjamorganfainberg agreed, the other thing i guess i heard from this proxy chat, was the trust model stuff, service X acting on behalf of Y user, since i'm not so familar with keystone api, did that get implemented?19:45
morganfainbergharlowja, trusts are X user acting on behalf of Y user19:45
morganfainbergharlowja, not so much service specific19:45
*** DinaBelova_ is now known as DinaBelova19:46
harlowjak, sure i guess service Z has a user X that it uses?19:46
*** denis_makogon has joined #openstack-dev19:46
morganfainbergharlowja, sure... i think. i mean...19:46
harlowja:)19:46
morganfainbergharlowja, if a service has a user... it's the same as a "user" ;)19:46
harlowjalike how 'nova' acts on behalf of user Y19:46
harlowja*nova user19:46
morganfainbergharlowja, iirc nova uses the user's specific context for most any action, so it isn't acting on behalf of the user, it is acting as the user since the user's context is already the originator of the request19:47
*** yolanda has quit IRC19:47
morganfainbergharlowja, it doesn't use an internal user ... much ... afaik that is different than the requesting user.19:47
*** byeager_ has quit IRC19:47
harlowjamorganfainberg ya, u might be right, i can't quite remember anymore :)19:47
morganfainbergharlowja, hehe it's complex :P19:48
harlowja:)19:49
*** nati_ueno has joined #openstack-dev19:50
*** nati_ueno has quit IRC19:50
morganfainbergharlowja, so in short, i think it's totally possible to use keystone for this stuff. and likely reasonable19:50
harlowjamorganfainberg i agree, now just to convince this architect ;)19:51
morganfainbergharlowja, it's just about how you extract the data from the token19:51
harlowjamarcoemorais ^^19:51
*** sweston has joined #openstack-dev19:51
harlowja;)19:51
morganfainbergharlowja, ah there is the challenge...always the challenge to convince others19:51
morganfainbergjaypipes, ping19:51
harlowja:)19:51
*** nati_ueno has joined #openstack-dev19:51
jaypipesmorganfainberg: hi Morgan, what's up?19:51
harlowjamorganfainberg especially others who havent' worked upstream much (if at all)19:51
morganfainbergjaypipes, got a few minutes, wanted to pick your brain19:52
morganfainbergjaypipes if you're busy i can bug ya later, no rush19:52
*** alop has joined #openstack-dev19:52
*** BillStokes has quit IRC19:52
harlowjamorganfainberg thx, will see if i get additional questions (i'm not such a good proxy, lol)19:52
*** p4tux has quit IRC19:52
*** ArxCruz has joined #openstack-dev19:53
morganfainbergharlowja, hehe well you seemed to describe things decently enough via IRC for me to catch on. give yourself credit where it's due ;)19:53
harlowja;)19:53
harlowjathx19:53
*** julienvey_ has joined #openstack-dev19:53
*** julienvey_ has quit IRC19:53
harlowjaanother one that i know he was interested, or others in y! are is also the service discovery part/service registry part (that keystone also has pieces of also)19:53
harlowjaany thoughts on hooking keystone registry into zookeeper to do automatic registration and such?19:54
*** p4tux has joined #openstack-dev19:54
harlowja*or something like zookeeper*19:54
morganfainbergharlowja, funny you should ask that.19:54
*** BillStokes has joined #openstack-dev19:54
harlowjamorganfainberg ??19:54
harlowjado tell19:54
morganfainbergharlowja, termie talked about that earlier on19:54
harlowjamorganfainberg cooool19:55
morganfainbergharlowja, not sure where we left it, but i think it could be possible to do something similar to that. actually... likely pretty easy to do that with the pluggable backends19:55
*** vartom1111111116 has joined #openstack-dev19:55
*** nelsnelson is now known as nelsnelson_away19:55
morganfainbergharlowja, but remember the service catalog gets kindof verbose as is, and passing that data all via header is painful sometimes19:56
*** changbl has joined #openstack-dev19:56
harlowjamorganfainberg agreed, from some stuff i've been hearing, it sounds similar to what this architect guy was thinking19:57
morganfainbergharlowja, but yes, i totally think it would be doable to leverage zookeeper as a catalog backend, either for automatic registration (there are pitfalls here) or just as a datasource for the catalog19:57
harlowjathat gains distributed backend (in a way)19:57
morganfainbergharlowja, and it should be relatively easy given the way the catalog works, you've seen how our drivers are built it's straightforward19:57
harlowjain a way a client could use a custom client to just lookup in zookeeper then19:57
harlowjamorganfainberg right19:57
morganfainbergyep19:57
morganfainbergi've toyed with that idea as a possibility wrt auth_token etc19:58
harlowjathe neat thing is that it lets u automatically know when services pop up/die...19:58
*** sweston has quit IRC19:58
harlowjayahoo uses zookeeper like that a little internally already19:58
morganfainbergharlowja, but i'm sure that isn't a Icehouse timeline item19:58
*** caleb_ has quit IRC19:58
harlowjamorganfainberg of course19:58
morganfainbergharlowja, might be worth discussing that further @ Juno summit19:58
*** byeager has joined #openstack-dev19:58
*** otherwiseguy has quit IRC19:58
harlowjadef19:58
harlowjai'll see if i can get enough knowledge to do a design session to discuss19:59
harlowjaproxy session :-/19:59
*** nelsnelson_away has quit IRC20:00
*** flaper87|afk is now known as flaper8720:01
harlowjathx again morganfainberg , will send some of this info back (proxy backwards, lol)20:03
*** emagana has quit IRC20:03
*** nelsnelson has joined #openstack-dev20:05
*** byeager has quit IRC20:05
*** nelsnelson has quit IRC20:06
*** nelsnelson has joined #openstack-dev20:06
morganfainbergharlowja, np happy to help20:06
*** changbl has quit IRC20:07
*** changbl has joined #openstack-dev20:07
*** galstrom_zzz is now known as galstrom20:08
*** sweston has joined #openstack-dev20:08
*** flaper87 is now known as flaper87|afk20:10
*** nelsnelson has quit IRC20:10
*** nelsnelson has joined #openstack-dev20:11
*** michchap has joined #openstack-dev20:11
*** caleb_ has joined #openstack-dev20:11
*** yassine has quit IRC20:12
*** yassine has joined #openstack-dev20:12
*** devoid has joined #openstack-dev20:12
*** sushils has joined #openstack-dev20:12
*** godara has quit IRC20:12
*** nelsnelson is now known as nelsnelson_away20:15
*** michchap has quit IRC20:15
*** sweston has quit IRC20:16
*** lucasagomes has quit IRC20:18
*** nelsnelson_away has quit IRC20:20
*** godara has joined #openstack-dev20:20
*** rook][ has joined #openstack-dev20:20
rook][Hello all, has anyone setup a read-only role for swift in keystone?  (not finding any examples via google)20:21
*** otherwiseguy has joined #openstack-dev20:24
*** romcheg has quit IRC20:25
*** CaptTofu has joined #openstack-dev20:26
harlowjayassine yt20:27
*** NikitaKonovalov_ is now known as NikitaKonovalov20:28
*** tzabal has joined #openstack-dev20:30
*** caleb_ has quit IRC20:30
*** Gordonz has joined #openstack-dev20:31
*** Gordonz has quit IRC20:31
*** rcleere has quit IRC20:31
*** Gordonz has joined #openstack-dev20:31
*** vartom1111111117 has joined #openstack-dev20:32
*** vkozhukalov has quit IRC20:34
*** vartom1111111116 has quit IRC20:35
*** sweston has joined #openstack-dev20:35
*** caleb_ has joined #openstack-dev20:36
*** pmathews1 has joined #openstack-dev20:37
*** tdruiva_ has joined #openstack-dev20:37
*** pmathews has quit IRC20:37
*** byeager has joined #openstack-dev20:40
morganfainbergdolphm, was the odd V3 bug you talked to me about at the hackathon: bug 126994720:41
morganfainbergdolphm, or bug 126897720:41
morganfainbergdolphm, or something else?20:41
*** tdruiva has quit IRC20:41
*** byeager has quit IRC20:41
morganfainbergoh, the bot is being slow today... https://bugs.launchpad.net/keystone/+bug/126897720:41
*** baoli has quit IRC20:42
morganfainbergor https://bugs.launchpad.net/keystone/+bug/126994720:42
*** comstud is now known as bearhands20:42
*** csd has quit IRC20:42
dolphmmorganfainberg: uhh...20:42
morganfainbergdolphm, it isn't assigned to me and i wanted to make sure it was addressed by i3 yanno20:42
morganfainberg:)20:42
dolphmmorganfainberg: i don't remember discussing either then20:42
dolphmmorganfainberg: https://bugs.launchpad.net/keystone/+bug/1269947 is targeting i3..20:42
morganfainbergit was some V3 something or another20:43
morganfainbergprob that one then20:43
morganfainbergok, cool20:43
dolphmlol20:43
morganfainbergyeah i dunno, LP search is sub wonderful sometimes20:43
dolphmmorganfainberg: that one definitely bothers me20:43
morganfainbergdolphm, well almost have a patch to convert all of our use of opt_in_group in tests to a config_fixture *yay*20:44
morganfainbergone more step towards parallel testing.20:44
*** tdruiva_ has quit IRC20:44
dolphmmorganfainberg: ooh nice20:45
*** flaper87|afk is now known as flaper8720:45
*** tdruiva has joined #openstack-dev20:45
morganfainbergdolphm, it's not massively different, just a bit cleaner overall.  then on to the other reviews i need to respin by feb 18.20:45
morganfainbergdolphm, and i'll sweep through and see waht i can do about that V3 ec2token thing, this weekend should be productive code wise20:46
dolphmmorganfainberg: small reviews to deprecate things https://review.openstack.org/#/c/70002/ https://review.openstack.org/#/c/70003/20:46
*** danjared has quit IRC20:46
morganfainbergdolphm, cool will look at them before i dive abck to code20:46
*** danjared has joined #openstack-dev20:46
*** dvarga is now known as dvarga|away20:47
*** dvarga|away is now known as dvarga20:47
morganfainbergdolphm, +2 on both.  very straightforward.20:48
dolphmmorganfainberg: danke!20:48
morganfainbergthe one stevemar +2 is now gating as well.20:48
morganfainberg7000320:48
*** flaper87 is now known as flaper87|afk20:48
*** NikitaKonovalov is now known as NikitaKonovalov_20:50
*** vartom1111111117 has quit IRC20:50
*** thomasem has joined #openstack-dev20:50
*** drewlander has quit IRC20:53
*** mrodden1 has joined #openstack-dev20:54
*** JuanManuelOlle has quit IRC20:54
*** mrodden has quit IRC20:56
*** caleb_ has quit IRC20:58
vishyHierarchical Ownership meeting starting in openstack-meeting21:00
vishy* Hierarchical Multitenancy21:00
*** gabrielhurley has joined #openstack-dev21:00
*** jotan has quit IRC21:00
*** mrodden1 has quit IRC21:01
tiamar\join #openstack-meeting21:01
tiamaroops21:01
dolphmmorganfainberg: hierarchical multitenancy meeting starting21:02
dolphmdolphm, ayoung, bknudson, dstanek, jamielennox, morganfainberg, stevemar, gyee, henrynash ^ #openstack-meeting21:02
*** kgriffs_afk is now known as kgriffs21:03
dstanekdolphm: already lurking21:03
ayoungdolphm, now?21:03
dolphmyes21:03
jgriffithsdague: let me know if you are interesting in discussing https://review.openstack.org/#/c/68532/21:04
gyeedolphm, thanks for the ping!21:05
*** baoli has joined #openstack-dev21:05
*** byeager has joined #openstack-dev21:05
*** tzabal has left #openstack-dev21:06
*** afazekas has quit IRC21:06
*** baoli has quit IRC21:06
*** baoli has joined #openstack-dev21:07
*** csd has joined #openstack-dev21:12
*** rcleere has joined #openstack-dev21:13
*** caleb_ has joined #openstack-dev21:13
jaypipesmorganfainberg: sorry, had a call... I can chat now.21:14
jaypipesmorganfainberg: you want to IRC or do a G+ hangout?21:14
*** giulivo has quit IRC21:15
*** MaxV has joined #openstack-dev21:15
*** caleb_ has quit IRC21:16
*** DinaBelova is now known as DinaBelova_21:17
*** caleb_ has joined #openstack-dev21:21
*** pmathews1 has quit IRC21:22
*** bauzas has joined #openstack-dev21:23
*** sweston has quit IRC21:24
*** lcheng has joined #openstack-dev21:24
*** dvarga_ has joined #openstack-dev21:24
*** dvarga_ has quit IRC21:24
*** pmathews has joined #openstack-dev21:24
*** pmathews1 has joined #openstack-dev21:25
morganfainbergjaypipes, i'd like to do G+ but in an IRC meeting atm will ping when i'm free :)21:25
morganfainbergjaypipes, like i said not a rush on it. so today, early next week, weekend, whatever it'll work21:25
jaypipesmorganfainberg: no worries. ping me when you're ready and we'll do a hangout. cheers.21:26
*** dvarga has quit IRC21:26
*** dpamio has quit IRC21:27
*** pmathews has quit IRC21:28
*** sweston has joined #openstack-dev21:30
*** xarses has joined #openstack-dev21:31
*** mrodden has joined #openstack-dev21:32
*** danielbruno has quit IRC21:33
*** david-lyle has quit IRC21:33
*** hartsocks has joined #openstack-dev21:33
*** e0ne has quit IRC21:37
*** thomasem has quit IRC21:39
*** jdob has quit IRC21:40
*** sumanth has joined #openstack-dev21:40
*** zaneb has joined #openstack-dev21:42
*** hartsocks has quit IRC21:46
*** pmathews1 has quit IRC21:51
*** lcheng has quit IRC21:52
*** carl_baldwin has joined #openstack-dev21:52
*** emagana has joined #openstack-dev21:52
*** hashfail is now known as gimps21:52
*** glenng has quit IRC21:52
*** armax has left #openstack-dev21:53
*** tiamar has quit IRC21:53
*** sumanth has quit IRC21:53
*** rfolco has quit IRC21:54
*** sumanth has joined #openstack-dev21:54
*** sweston has quit IRC21:56
ayoungvishy, are you going to prototype in Keystone or in Nova?21:58
*** gabrielhurley has quit IRC21:58
morganfainbergayoung, he is going to hack up nova to see if it works there, do db mangling from what i gathered21:59
ayoungyeah...that is wrong21:59
morganfainbergayoung, to prove that nova can benefit from this?21:59
ayoungand I don';t want to have to live with the wrongness21:59
ayoungits not a nova problem, it is a Keystone problem21:59
ayoungthe fix is to add a parent field to the project.22:00
*** michchap has joined #openstack-dev22:00
*** sarob has joined #openstack-dev22:00
ayoungnow, we could repurpose the domain_id for that22:00
dolphmayoung: he's proving the authorization problem22:00
dolphmcan be solved22:00
morganfainbergdolphm, ++22:00
ayoungdolphm, its an RBAC problem22:00
dolphmayoung: right.22:00
ayoungso long as he gets the right roles, he gets access22:00
ayoungroles are based on the project Id22:00
ayoungwhen enforced22:00
ayoungnot project name22:01
ayoungis it?22:01
*** sumanth has quit IRC22:01
ayoungdo we even put the name in the token?22:01
morganfainbergayoung, he wasn't claiming project name was the end target, just that enforcement in nova was possible22:01
morganfainbergayoung, policy wise etc.22:01
ayoungmorganfainberg, I'm saying it should be transparent22:01
ayoungAH...wait22:02
dolphmayoung: did you happen to catch the beginning of the meeting?22:02
*** schwicke has quit IRC22:02
ayoungI see the problem,  its that he wants to get a token for parent, and have the roles apply to parent->child->child->child22:02
vishyayoung: i feel like you have misunderstood my planning22:03
ayoungdolphm, yes22:03
dolphmayoung: right22:03
vishyayoung: i'm not dealing with roles at all in this case22:03
vishyonly ownership22:03
*** byeager has quit IRC22:03
*** lcheng has joined #openstack-dev22:03
ayoungvishy, the problem is that the project hierarchy is then going to have to be synced between keystone and all of the other services22:04
*** topol has quit IRC22:04
ayoungbut not a big deal there22:04
ayounghow is that done now?22:04
*** csd has quit IRC22:04
ayoungI mean, pre-notifications22:04
vishyayoung: only if you assume that projects can be moved in the hierarchy22:04
morganfainbergayoung, afaict it isn't22:04
ayoungif I create a project in Keystone, I can immediately start using the project in nova, right?22:04
vishywhich is a concern i suppose22:04
vishyayoung: correct22:04
*** michchap has quit IRC22:05
*** sumanth has joined #openstack-dev22:05
*** pmathews has joined #openstack-dev22:05
bknudsonayoung: if you delete a project in Keystone you can immediately not use it in nova.22:05
ayoungbknudson, only due to notifications22:05
morganfainbergbknudson, ++ i like your phrasing22:05
morganfainbergayoung, no, you can't use the project, nothing is cleaned up, but you can't use it.22:05
ayoungbknudson, that is new22:05
morganfainbergayoung, notifcations allow cleanup22:05
*** ekhugen has quit IRC22:05
ayoungmorganfainberg, you could so long as your token hadn't expired, except due to Token revocations22:06
morganfainbergayoung, project deletion revokes tokens iirc22:06
*** vladikr has quit IRC22:06
ayoungits all drive off of keystone, not nova22:06
ayoungwhich is my point22:06
bknudsonhierarchical projects could help with that problem.22:06
ayoungthe creation of the project was done on demand in Nova22:06
bknudsonif nova doesn't want to implement notifications.22:06
vishyayoung: there is no concept of a project in nova22:06
vishyayoung: there is only a project_id field on objects22:07
vishywhich is a string representation of the "owner" of the object22:07
ayoungvishy, all this can work without changing that22:07
vishyayoung: yup22:07
vishythat is my whole proposal22:07
ayoungvishy, the wonkiness is that when you request a token in Keystone, somehow the hierarchy info needs to be in sync22:08
dolphmayoung: i really feel like you missed something from either the wiki or the beginning of the meeting?22:08
ayoungso if the token had the list of subordinate projects in it, it would work22:08
vishythe hierarchy needs to be passed from keystone to nova somehow22:08
ayoungdolphm, no, I tracked, I just have my own ideas that pre-exist on how this needs to be executed22:08
vishyayoung: right but i don't want to modify the keystone response22:08
vishythat is a waste of my time22:09
ayoungvishy, and I don;t think we can do it without modifying keystone22:09
vishyso i'm prototyping by using project_id22:09
ayoungshort of syncing all of the role data out of band22:09
vishythat could turn into a new field22:09
morganfainbergvishy, and that is a fair representation based on nova's concept of owner22:09
vishyayoung: you are still trying to solve the problem of what this will look like when it is done22:09
ayoungand all of that is going to break trusts22:09
ayoungvishy, there are two distinct problems with hierarchies to solve22:09
ayoungone is delegation of authority22:10
ayoungand the other is communication of that to the remote service22:10
ayoungthe first is easy to do in keystone22:10
ayoungthe second...needs thought22:10
ayoungso,  here is the brain dead way...let me etherpad this22:10
dolphmayoung: let's think about it when we have a prototype to talk over :)22:10
morganfainbergdolphm, ++22:11
*** dencaval has quit IRC22:11
dolphm... and then we can think in patches22:11
vishyone way would be to allow the services to put objects in more than one place in the heirarchy22:11
*** srenatus has quit IRC22:11
ayoungOooh22:11
ayoungSymlinks?22:11
vishybasically yeah22:11
vishyor multiple ownership basically22:12
ayoungOK..so long as there is one real owner and the rest are subordinate, that can work22:12
vishybut that is more complicated22:12
vishyso not in scope for my prototype22:12
ayoungWell, there is a related problem I've heard people discuss which is "we are in the same project, but I don';t want YOU to be able to reboot MY machine"22:13
vishyanother option would be to pass scope in as a role22:13
ayoungsince there is only one Owner field, there is no way to distinguish between a user/owner and a proejct/owner22:13
vishywhere you could have more than one scope22:13
vishyayoung: we have user_id as well22:13
vishybut i was leaving that out22:13
*** kevinconway has quit IRC22:13
vishybecause that is mostly for tracking the creator of the resource22:14
vishyinstead of the owner22:14
*** shakayumi has quit IRC22:14
*** vijendar has quit IRC22:14
ayoungvishy, OK, so we have the Unix analoges of user and group.  Go on22:14
*** pcm has quit IRC22:14
*** shakayumi has joined #openstack-dev22:14
*** bswartz has quit IRC22:14
ayoungyeah, but we can use it differently in the future...go on22:14
vishyi wasn't going anywhere with that22:14
*** vijendar has joined #openstack-dev22:14
vishyi was just saying that we do have both for most resources22:14
vishybecause we need to know who created the thing, not just where it lives22:15
*** vijendar has quit IRC22:15
vishyayoung: the inverted version would be for keystone to pass 1+ scopes as a role22:17
*** doug-fish has quit IRC22:17
ayoungyes22:17
ayoungI would see that as a token creation option22:17
ayoungI'm looking to see what that would look like in the response22:17
vishyayoung: but basically the point is the same22:17
*** sgordon has quit IRC22:18
vishythe hierarchy gets passed in the token somehow22:18
vishyi'm just overloading project_id because it is the minimal set of changes so that i can start fixing stuff in nova to support it22:18
vishybut multiple scopes is an interesting thought22:18
vishyit probably makes sense to have the multiple ownership to be passed from keystone22:19
*** baoli has quit IRC22:19
ayoungvishy the question is what to do if a user requested a token for Proejct P and wanted to use that on child project P/c/c/c22:19
vishybut i'm not clear exactly how that would work internally22:19
ayoungdo they need to ask for a token to work on that?  I would say yes22:19
*** rook][ has quit IRC22:19
ayoungyou only get a roles for what you ask for22:19
vishyayoung: that is why single hierarchy is more sensible22:20
ayoungnot down the hierarchy22:20
*** MaxV has quit IRC22:20
ayoungso I need to ask for a token for P/c/c/c/  but that is easy22:20
vishyi would say switch tokens22:20
vishyif it is not under the current scope22:20
ayoungassumig I have role r on P  it would show up also on P/c/c/c22:20
*** Apsu is now known as SenorSnottyTits22:20
vishyyes22:20
vishybut switching tokens is horrid user experience if it is in the same hierarchy22:21
*** SenorSnottyTits is now known as Apsu22:21
ayoungsay I have r1 on P  r2 on P/c r3 on P/c/c/ and  nothing on P/c/c/c If I asked for a token for P/c/c/c/  I get roles [r1, r2, r3]22:21
vishycorrect22:21
ayoungtoo bad22:21
vishyroles are inherited down the hierarchy22:21
ayoungtokens should be limited in scope22:21
ayoungtokens should be thought of as "give me the miniumum I need to do just X"22:21
ayoungits like taking money out of the bank22:22
vishyayoung: I disagree22:22
ayoungif you get robbed and all you took out was $522:22
*** mfer has quit IRC22:22
vishyayoung: if you discount list i think that is fine22:22
vishyayoung: but list completely breaks in that model22:22
ayoungvishy, that is cus you have not been thinking about it from a security perspective.  Tokens give me nightmares22:22
ayoungok....22:22
vishysecurity cannot be the only concern here22:22
*** galstrom is now known as galstrom_zzz22:23
vishyusability must be considered22:23
ayoungunderstood,  let me chew on that....22:23
ayoungit does mean that Nova needs to know about the hierarchy22:23
vishybest middle ground would be22:23
vishyget/read can be done for the hierarchy22:23
*** seventy3 is now known as seventy3_away22:23
vishybut create/destroy/modify would require reauthing22:23
praneshphey marun yt?22:23
vishyinto the subproject22:23
*** mariusleu has quit IRC22:23
vishybut i think you need to consider the actual attack profile22:24
*** carl_baldwin has quit IRC22:24
ayoungin Posix. they have the advantage of it all being local calls,  and this kind of auth thing over NFS is a performance drag22:24
vishyand not just default to do the minimum possible22:24
ayoungcuz each Dentry needs to be seperately authed22:24
ayoungbut does not require a separate round trip.22:25
ayoungvishy, heh, the problem is that there is no hierarchy on the nova side....22:25
vishyayoung: that is my point!22:26
vishybut we don't want to keep an entire tree we have to keep in sync22:26
ayoungvishy, its on the API side...if the projects are nested, and I do list, the role that matters is the role on the parent project22:26
vishywhich is why i want to do the single dot.separated.hierarchy22:26
ayoungso long as the policy check is only done there, there is no change22:26
*** yassine has quit IRC22:26
vishyayoung: but it isn't just policy on the action22:26
ayoungI'm with you22:27
vishyit is filtering of objects22:27
vishybecause the objects aren't in a hierarchy22:27
vishythey are flat with an owner22:27
ayoungvishy, nova is going to need a tree22:27
ayoungno getting around that, I think22:27
vishyayoung: why?22:27
ayoungto know the relationships22:27
vishyi want to leave the tree in keystone22:27
vishyand just have an identifier in nova22:27
ayoungvishy, then keystoine would have to communicate it in the token.  that is not right, and not what, I think, you want22:28
vishyayoung: either way it does22:28
ayoungif I do list projects on P1  and I hand over a token for P1,  Nova needs to be able to exapnd that to P/c anmd P/d22:28
vishynova needs the hierarchy when the object is created22:28
vishyayoung: when I have the prototype if you can propose something else that does the same thing22:29
*** erkrnt has joined #openstack-dev22:29
vishywe can try it22:29
*** joesavak has quit IRC22:31
ayoungvishy, how are you planning on doing it?22:32
*** csd has joined #openstack-dev22:32
*** carl_baldwin has joined #openstack-dev22:33
vishyayoung: exactly as I explained22:34
vishypass in dot.separated.hierarchy22:34
vishystore that in the owner field22:34
vishyand partial match22:34
ayoungvishy, does nova have a way to look up project name from project id?22:35
ayoungor vice versa?22:35
vishyno22:35
vishyusing ids22:35
*** jamezpolley has quit IRC22:35
vishythe only place we use name is in logging22:35
*** caleb_` has joined #openstack-dev22:35
ayoungwhat if you create a second table  with kjust the parent/child ids22:36
ayounga list would then become a hierarchical query of that table and produce a set of ids22:36
*** jmontemayor has quit IRC22:36
*** sivy has joined #openstack-dev22:37
ayoungso instead select * from t where owner=id it becomes select * from t where owner in (resultset)22:37
sivyHi folks a22:37
*** sumanth has quit IRC22:37
ayoungmake the hierarchy an extension, so that it gets skipped if the extension is not installed22:38
vishyayoung: we could do that22:38
*** caleb_ has quit IRC22:38
vishyayoung: i'm trying to minimize changes22:38
ayoungvishy, me to, but these kind of thought experiments are cheaper than coding22:39
*** negronjl has quit IRC22:39
*** SlickNik has joined #openstack-dev22:39
sivy(Apologize for typing, on phone) I am frustratedly trying to write tests for some horizon modifications, and not finding any good docs on how to make sure that the nova/glance/etc stubs are setup properly22:39
*** negronjl has joined #openstack-dev22:39
SlickNikhey guys22:39
sivyI'm iskng horizon.22:39
SlickNikquick keystone middleware question.22:39
sivyUgh. Using horizon.test.APITestCase22:40
SlickNikin api-paste.ini, what's the difference between the [filter:authtoken], and the [filter:tokenauth] settings?22:40
ayoungvishy, what you are proposing is going to ultimately require support from Keystone.  I'd like to make sure that it is as consistent with the rest of the design as possible.  THat is more important than minimizing changes for a Proof of concept.  But I think what I just proposed is not more work than what you were thinking of doing,  maybe less22:40
SlickNiksections*22:40
*** lcheng has quit IRC22:40
vishyayoung: i don't know if that model is a great fit for relational queries22:40
vishyayoung: it is definitely more work22:41
ayoungvishy, I bet we could pre calculate it22:41
vishyassuming the hierarchy is small that is true22:41
ayoungI can think of a couple ways to optimize22:41
*** AlanClark has quit IRC22:41
vishybut in that case i could just keep the hierarchy in memory22:41
ayoungyou could have an "all children" table which is the precalculatd subtree22:42
*** peristeri has quit IRC22:42
ayoungpremature optimization of course22:42
*** morazi has quit IRC22:42
ayoungkeeping it in memory is also an option22:43
ayoungI'm actually doing something like that with token revocation.22:43
*** READ10 has quit IRC22:44
*** lcheng has joined #openstack-dev22:44
ayoungvishy, to start, prototype the whole thing in memory22:45
ayoungdon't bother to persist it, and just do multiple queries22:45
ayoungstore it as a tree and execute one list for each level of the tree22:46
*** jecarey has quit IRC22:46
ayoungOK...I need to go be a Dad.  vishy this is cool. God luck22:46
*** sivy has quit IRC22:46
ayoungEr..Zod22:46
*** d0ugal_ has joined #openstack-dev22:46
*** artom has quit IRC22:47
*** byeager has joined #openstack-dev22:47
*** d0ugal has quit IRC22:49
*** carl_baldwin has quit IRC22:49
*** romcheg has joined #openstack-dev22:49
*** eharney has quit IRC22:54
*** clayb has quit IRC22:55
*** dencaval has joined #openstack-dev22:55
*** lcheng has quit IRC22:56
*** carl_baldwin has joined #openstack-dev22:59
*** dbalog has left #openstack-dev23:02
*** jmckind has quit IRC23:04
*** byeager has quit IRC23:04
*** mlavalle has quit IRC23:04
*** byeager has joined #openstack-dev23:04
*** jgrimm has quit IRC23:06
*** burt1 has quit IRC23:07
*** CaptTofu has quit IRC23:07
*** carl_baldwin has quit IRC23:07
*** dims has quit IRC23:08
*** IanGovett has quit IRC23:08
*** jotan has joined #openstack-dev23:08
*** radez is now known as radez_g0n323:09
*** kbrierly has quit IRC23:10
*** troytoman-away is now known as troytoman23:15
*** rtheis has quit IRC23:21
*** troytoman has quit IRC23:21
*** julienvey_ has joined #openstack-dev23:21
*** troytoman-away has joined #openstack-dev23:22
*** troytoman-away is now known as troytoman23:22
*** troytoman has joined #openstack-dev23:22
*** julienvey_ has quit IRC23:25
*** dims has joined #openstack-dev23:26
*** galstrom_zzz is now known as galstrom23:26
*** julienvey_ has joined #openstack-dev23:26
*** bpokorny has quit IRC23:28
*** romcheg has quit IRC23:28
*** adreznec has quit IRC23:30
*** prad_ has quit IRC23:31
*** julienvey_ has quit IRC23:33
*** mrodden has quit IRC23:36
*** eglynn has quit IRC23:37
*** thedodd has quit IRC23:37
*** cjoelrun has quit IRC23:38
morganfainbergjaypipes, you around still, i know you're adifferent timezone23:43
morganfainbergjaypipes, than i am23:43
jaypipesmorganfainberg: yup.23:43
morganfainbergjaypipes, give me a sec to dig up google stuff if youre available23:43
*** byeager has quit IRC23:45
*** gokrokve_ has joined #openstack-dev23:45
*** caleb_` has quit IRC23:46
*** CaptTofu has joined #openstack-dev23:48
*** gokrokve has quit IRC23:49
*** schwicht has quit IRC23:52
marunpraneshp: ah, sorry23:54
marunpraneshp: still here?23:54
*** kgriffs is now known as kgriffs_afk23:54
praneshphi marun23:54
praneshpi noticed you asking something about configdrive yesterday23:55
marunpraneshp: how can I help?23:55
praneshpwas wondering what the question was23:55
praneshpI moved from vfat to iso and am looking out for potential issues23:55
marunpraneshp: There was gate breakage for neutron when configdrive was disabled, since then the default became file injection.23:56
*** rcleere has quit IRC23:56
praneshpah, ok. totally unrelated. Thanks!23:56
*** Gordonz has quit IRC23:56
marunpraneshp: :)23:56
*** bauzas has quit IRC23:56
*** denis_makogon has quit IRC23:57
*** atiwari has joined #openstack-dev23:57

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!