Friday, 2016-04-08

*** stingaci has joined #openstack-keystone00:01
*** david-lyle has quit IRC00:02
*** david-lyle has joined #openstack-keystone00:03
*** mylu has quit IRC00:04
*** ayoung has joined #openstack-keystone00:08
*** ChanServ sets mode: +v ayoung00:08
*** mylu has joined #openstack-keystone00:09
openstackgerritOpenStack Proposal Bot proposed openstack/keystone: Updated from global requirements  https://review.openstack.org/30062600:10
*** shaleh has quit IRC00:13
*** mylu has quit IRC00:14
*** dflorea has quit IRC00:15
*** mylu has joined #openstack-keystone00:25
openstackgerritOpenStack Proposal Bot proposed openstack/keystone: Updated from global requirements  https://review.openstack.org/30062600:27
openstackgerritOpenStack Proposal Bot proposed openstack/keystoneauth: Updated from global requirements  https://review.openstack.org/30313000:27
openstackgerritOpenStack Proposal Bot proposed openstack/keystonemiddleware: Updated from global requirements  https://review.openstack.org/30313100:27
*** spandhe has quit IRC00:28
*** dflorea has joined #openstack-keystone00:29
*** furface has quit IRC00:30
*** furface has joined #openstack-keystone00:30
openstackgerritOpenStack Proposal Bot proposed openstack/pycadf: Updated from global requirements  https://review.openstack.org/30315000:32
openstackgerritOpenStack Proposal Bot proposed openstack/python-keystoneclient: Updated from global requirements  https://review.openstack.org/30315600:32
*** mylu has quit IRC00:46
*** dflorea has quit IRC00:49
*** diazjf1 has joined #openstack-keystone00:51
*** diazjf1 has quit IRC00:51
*** mylu has joined #openstack-keystone00:51
*** fpatwa has joined #openstack-keystone00:52
*** diazjf has quit IRC00:54
fpatwaI am trying to check if we can restrict the catalog sent in a token for non-admin users to NOT contain the admin and/or internal endpoints00:54
fpatwaAnyone know if that is possible to do ?00:54
*** spandhe has joined #openstack-keystone01:01
*** fpatwa has quit IRC01:01
*** sdake has joined #openstack-keystone01:02
*** fpatwa has joined #openstack-keystone01:02
*** stingaci has quit IRC01:05
*** fpatwa has quit IRC01:07
*** spandhe has quit IRC01:08
*** browne has quit IRC01:09
*** fpatwa has joined #openstack-keystone01:09
*** ksnihyr has joined #openstack-keystone01:14
*** sdake has quit IRC01:14
*** ksnihyr has quit IRC01:18
*** sdake has joined #openstack-keystone01:19
jamielennoxmorgan: our very own TC - they grow up so fast!01:19
jamielennoxmorgan: contrats mate01:19
jamielennoxcongrats01:19
*** EinstCrazy has joined #openstack-keystone01:21
*** sdake_ has joined #openstack-keystone01:22
*** mylu has quit IRC01:23
*** sdake has quit IRC01:25
*** fpatwa has quit IRC01:26
*** fpatwa has joined #openstack-keystone01:27
*** fpatwa has quit IRC01:31
morgan^_^01:36
*** mylu has joined #openstack-keystone01:37
*** edmondsw has quit IRC01:38
*** dave-mccowan has quit IRC01:45
*** browne has joined #openstack-keystone01:59
*** mylu has quit IRC02:04
*** mylu has joined #openstack-keystone02:10
*** stingaci has joined #openstack-keystone02:15
*** ksnihyr has joined #openstack-keystone02:15
*** cburgess has quit IRC02:16
*** cburgess has joined #openstack-keystone02:19
cloudnullcongrats morgan02:19
*** openstackstatus has quit IRC02:19
*** aginwala has quit IRC02:19
*** ksnihyr has quit IRC02:19
*** woodster_ has quit IRC02:27
*** aginwala has joined #openstack-keystone02:29
*** richm has quit IRC02:33
*** richm1 has joined #openstack-keystone02:33
*** sdake_ is now known as sdake02:33
*** richm1 has quit IRC02:38
*** aginwala has quit IRC02:39
*** aginwala has joined #openstack-keystone02:52
*** EinstCrazy has quit IRC02:53
*** EinstCrazy has joined #openstack-keystone02:55
*** EinstCrazy has quit IRC02:57
*** aginwala has quit IRC02:59
*** EinstCrazy has joined #openstack-keystone02:59
*** agrebennikov has quit IRC03:15
*** ksnihyr has joined #openstack-keystone03:15
*** knikolla has joined #openstack-keystone03:15
*** mylu has quit IRC03:20
*** ksnihyr has quit IRC03:20
*** fpatwa has joined #openstack-keystone03:24
*** fpatwa has quit IRC03:28
*** mylu has joined #openstack-keystone03:40
*** anush_ has joined #openstack-keystone03:41
*** fpatwa has joined #openstack-keystone03:45
*** jasonsb has joined #openstack-keystone03:47
*** jasonsb has quit IRC03:49
*** knikolla has quit IRC03:50
*** jasonsb has joined #openstack-keystone03:51
*** browne has quit IRC03:53
*** links has joined #openstack-keystone03:53
*** knikolla has joined #openstack-keystone03:54
*** knikolla has quit IRC03:56
stevemarmorgan: congrats -- third, yay04:04
morganhehe04:05
morganstevemar: thnx04:05
morganstevemar: also... do you ever sleep?04:06
stevemarmorgan: it's .. oh it's midnight04:06
morganstevemar: i know... cause my flight left the ground at 2300 EASTERN04:06
stevemarmorgan: late flight04:07
morganmassive delays04:07
morgani walked into the airport 3 minutes before boarding the plane04:07
morganTSA PRE *AND* zero people in the line04:08
openstackgerritRodrigo Duarte proposed openstack/keystone: Add identity providers integration tests  https://review.openstack.org/30229904:10
*** ksnihyr has joined #openstack-keystone04:16
*** sdake_ has joined #openstack-keystone04:19
morganstevemar: also -- lookin forward to the summit more and more.04:19
*** sdake has quit IRC04:19
jamielennoxstevemar, morgan: it's a bit of a mind bending review, but when you get a chance: https://review.openstack.org/#/c/255686/04:20
patchbotjamielennox: patch 255686 - keystone - Make AuthContext depend on auth_token middleware04:20
*** ksnihyr has quit IRC04:21
stevemarmorgan: i'm totally stoked for the summit04:22
navidpstevemar, can you become coauthor on this patch ?04:22
navidpstevemar, https://review.openstack.org/#/c/276350/04:22
patchbotnavidp: patch 276350 - python-openstackclient - Moving authentication from keystoneclient to keyst...04:22
stevemarnavidp: i haven't contributed enough to warrant co-authorship IMO04:23
stevemari just changed one line04:23
navidpstevemar, ok04:25
stevemarnavidp: i also rebased, but meh04:25
navidpstevemar, about osc generic password, what is the path change it or make it work the old way then change it in the future04:26
navidpstevemar, my concern was first make it work then fix issues like over wirting a plugin as jamie mentioned04:27
stevemarnavidp: we could go either way on that one i guess, we could put up a follow-on patch that fixes it04:31
morganjamielennox: oh. neat04:32
morganjamielennox: unfortunately, i don't think i can review much on a plane.. hard to git clone the repo04:32
morgan(the gerrit stuff is the easy part)04:32
morgani need to re-setup gertty04:32
*** dflorea has joined #openstack-keystone04:32
jamielennoxmorgan: no worries, it's just been there for a while untouched and i want to give it a shove now we're open again04:33
morganyeah we should move that one through04:33
*** mylu has quit IRC04:33
morganjamielennox: i think wwe have a few things to really move through soon04:34
navidpstevemar, ok then04:34
*** mylu has joined #openstack-keystone04:37
morganstevemar: dude, co-author all the patches! (/sarcasm)04:38
morganstevemar:  New Rule: All patches must list "Steve Martinelli" as co-author :P04:38
stevemarmorgan: i have too many commits as it is04:38
*** aginwala has joined #openstack-keystone04:39
morgandude we should figure out what is going on here: https://review.openstack.org/#/c/300626/04:50
patchbotmorgan: patch 300626 - keystone - Updated from global requirements04:50
*** dflorea has quit IRC04:51
*** dflorea has joined #openstack-keystone04:53
*** rcernin has joined #openstack-keystone04:55
*** fpatwa has quit IRC04:55
*** fpatwa has joined #openstack-keystone04:56
*** alex_xu has quit IRC04:56
*** dflorea has quit IRC04:59
jamielennoxmorgan: that's weird, it didn't pick up the oslo.db[fixtures] extras04:59
*** fpatwa has quit IRC05:01
jamielennoxoh - well that: https://review.openstack.org/#/c/300626/11/test-requirements.txt05:01
patchbotjamielennox: patch 300626 - keystone - Updated from global requirements05:01
*** alex_xu has joined #openstack-keystone05:01
*** dflorea has joined #openstack-keystone05:03
*** sdake has joined #openstack-keystone05:03
*** GB21 has joined #openstack-keystone05:04
morganlol05:05
*** mylu has quit IRC05:06
*** sdake_ has quit IRC05:06
*** links has quit IRC05:07
stevemarjamielennox: yeah, i think the proposal bot needs an update to handle optional dependencies05:08
*** mylu has joined #openstack-keystone05:10
*** anush_ has quit IRC05:11
*** sdake_ has joined #openstack-keystone05:13
*** sdake has quit IRC05:17
*** ksnihyr has joined #openstack-keystone05:17
*** ksnihyr has quit IRC05:22
*** prometheanfire has joined #openstack-keystone05:23
prometheanfirekeystoneclient 2.3.1 doesn't support v3?05:23
prometheanfireguess all functionality is in openstackclient then?05:24
*** links has joined #openstack-keystone05:24
stevemarprometheanfire: oh yeah - the CLI bundled in keystoneclient has been deprecated for a year and a half now05:24
stevemarand it totally going to be removed in 3.0.005:24
stevemarprometheanfire: keystoneclient still has all the python bindings05:25
* prometheanfire shugs05:25
prometheanfireguess I'll finally fully switch then05:25
prometheanfirewas half and half before05:25
prometheanfirethanks, parting05:26
*** prometheanfire has left #openstack-keystone05:26
*** stingaci has quit IRC05:29
*** furface has quit IRC05:29
*** markvoelker has quit IRC05:30
*** mylu has quit IRC05:32
*** stingaci has joined #openstack-keystone05:34
stevemarthat was the most nonchalant response to "we're removing keystone CLI" i've ever received05:35
stevemarwhy can't they all be like that05:35
*** mkrcmari__ has joined #openstack-keystone05:37
*** mvk_ has quit IRC05:41
*** aginwala has quit IRC05:53
*** sdake has joined #openstack-keystone05:55
*** dflorea has quit IRC05:56
*** sdake_ has quit IRC05:57
*** links has quit IRC06:03
*** sdake_ has joined #openstack-keystone06:09
*** sdake has quit IRC06:11
*** ksnihyr has joined #openstack-keystone06:18
*** links has joined #openstack-keystone06:19
*** henrynash has quit IRC06:21
*** ksnihyr has quit IRC06:24
openstackgerritOpenStack Proposal Bot proposed openstack/keystone: Imported Translations from Zanata  https://review.openstack.org/30324006:30
*** markvoelker has joined #openstack-keystone06:30
*** markvoelker has quit IRC06:36
*** furface has joined #openstack-keystone06:38
jamielennoxit's friday, i'll take on one small-ish requirements problem, turns out to be a real PITA...06:45
*** furface has quit IRC06:49
*** furface has joined #openstack-keystone06:52
*** fpatwa has joined #openstack-keystone06:56
*** e0ne has joined #openstack-keystone07:01
*** fpatwa has quit IRC07:01
*** jaosorior has joined #openstack-keystone07:04
*** jed56 has joined #openstack-keystone07:06
*** ksnihyr has joined #openstack-keystone07:12
*** stingaci has quit IRC07:19
*** stingaci has joined #openstack-keystone07:24
*** alex_xu_ has joined #openstack-keystone07:32
*** alex_xu has quit IRC07:34
*** GB21 has quit IRC07:35
*** real56 has joined #openstack-keystone07:41
*** pnavarro_ has joined #openstack-keystone07:42
jamielennoxstevemar, morgan: filed fix against requirements: https://review.openstack.org/30327107:43
*** jaosorior has quit IRC07:47
*** jaosorior has joined #openstack-keystone07:47
*** stingaci has quit IRC07:48
*** pcaruana has joined #openstack-keystone07:49
*** stingaci has joined #openstack-keystone07:49
*** jamielennox is now known as jamielennox|away07:53
*** fhubik has joined #openstack-keystone07:53
*** stingaci has quit IRC07:54
*** henrynash has joined #openstack-keystone07:54
*** ChanServ sets mode: +v henrynash07:54
*** GB21 has joined #openstack-keystone08:02
*** GB21 has quit IRC08:24
*** jistr has joined #openstack-keystone08:28
*** markvoelker has joined #openstack-keystone08:31
*** e0ne has quit IRC08:32
*** openstackgerrit has quit IRC08:33
*** openstackgerrit has joined #openstack-keystone08:33
*** markvoelker has quit IRC08:36
*** e0ne has joined #openstack-keystone08:40
openstackgerritMerged openstack/keystone: Imported Translations from Zanata  https://review.openstack.org/30324008:57
*** fpatwa has joined #openstack-keystone08:57
*** fpatwa has quit IRC09:02
*** harlowja has quit IRC09:06
*** mhickey has joined #openstack-keystone09:14
*** sdake_ has quit IRC09:17
*** real56 has quit IRC09:17
*** stingaci has joined #openstack-keystone09:19
*** stingaci has quit IRC09:24
*** e0ne has quit IRC09:25
*** mhickey has quit IRC09:34
*** GB21 has joined #openstack-keystone09:35
*** subscope has joined #openstack-keystone09:38
*** subscope has quit IRC09:44
*** subscope has joined #openstack-keystone09:45
*** GB21 has quit IRC09:47
*** GB21 has joined #openstack-keystone09:47
*** mhickey has joined #openstack-keystone09:52
*** subscope has quit IRC10:10
*** subscope has joined #openstack-keystone10:15
*** e0ne has joined #openstack-keystone10:20
*** subscope has quit IRC10:20
*** fhubik has quit IRC10:21
*** rk4n has joined #openstack-keystone10:24
*** rk4n has quit IRC10:25
*** EinstCrazy has quit IRC10:30
*** stingaci has joined #openstack-keystone10:32
*** markvoelker has joined #openstack-keystone10:32
*** stingaci has quit IRC10:36
*** fpatwa has joined #openstack-keystone10:38
*** markvoelker has quit IRC10:39
*** rk4n has joined #openstack-keystone10:39
*** rk4n has quit IRC10:41
*** aswadr_ has joined #openstack-keystone10:47
*** GB21 has quit IRC10:58
*** GB21 has joined #openstack-keystone10:58
*** fpatwa has quit IRC10:59
*** fpatwa has joined #openstack-keystone11:00
*** fpatwa has quit IRC11:05
*** GB21 has quit IRC11:05
*** GB21 has joined #openstack-keystone11:06
*** pece has joined #openstack-keystone11:07
*** fpatwa has joined #openstack-keystone11:07
peceWSGIScriptAlias / /usr/bin/keystone-wsgi-public   is it possible to pass different keystone.conf file here? Or maybe different root /etc/keystone directory?11:08
*** rk4n has joined #openstack-keystone11:09
*** fpatwa has quit IRC11:14
*** fpatwa has joined #openstack-keystone11:15
*** rk4n has quit IRC11:16
*** fpatwa has quit IRC11:19
*** henrynash has quit IRC11:23
*** GB21 has quit IRC11:37
*** GB21 has joined #openstack-keystone11:38
*** subscope has joined #openstack-keystone11:38
openstackgerritOpenStack Proposal Bot proposed openstack/keystone: Updated from global requirements  https://review.openstack.org/30062611:43
*** GB21 has quit IRC11:43
*** GB21 has joined #openstack-keystone11:44
*** subscope has quit IRC11:50
*** ayoung has quit IRC11:54
*** e0ne has quit IRC11:59
*** GB21 has quit IRC12:02
*** GB21 has joined #openstack-keystone12:02
*** fpatwa has joined #openstack-keystone12:03
*** subscope has joined #openstack-keystone12:03
*** fpatwa has quit IRC12:07
*** markvoelker has joined #openstack-keystone12:21
*** raildo-afk is now known as raildo12:26
*** GB21 has quit IRC12:27
*** trown|outtypewww is now known as trown12:29
*** gordc has joined #openstack-keystone12:31
odyssey4meis anyone else getting a failure to resolve review.openstack.org ?12:31
odyssey4meoops, wrong channel12:31
*** henrynash has joined #openstack-keystone12:32
*** ChanServ sets mode: +v henrynash12:32
*** edmondsw has joined #openstack-keystone12:33
*** stingaci has joined #openstack-keystone12:33
*** pauloewerton has joined #openstack-keystone12:34
*** stingaci has quit IRC12:38
openstackgerritSamuel de Medeiros Queiroz proposed openstack/keystone: Add API Change Tutorial  https://review.openstack.org/30278912:41
samueldmqmorning keystoners12:44
samueldmqdstanek: you around ?12:44
samueldmqdstanek: re: https://review.openstack.org/#/c/212957/9/keystone/tests/unit/policy/backends/test_base.py12:45
patchbotsamueldmq: patch 212957 - keystone - Create unit tests for the policy drivers12:45
dstaneksamueldmq: here12:45
samueldmqdstanek: are you okay with changing that abstractmethod _get_driver() to a @property thing ?12:45
dstaneksamueldmq: looking...12:47
dstaneksamueldmq: sure, but we probably don't need that to be a private property12:49
dstaneksamueldmq: make it an abstract property12:49
samueldmqdstanek: @abstractproperty ?12:51
samueldmq@abc.abstractproperty12:52
samueldmqdstanek: thanks12:52
dstaneksamueldmq: yessir12:56
samueldmqdstanek: and how do I override in the subclasses ?12:57
dstaneksamueldmq: you just create a property12:57
samueldmqdstanek: hmm, I thoght that I simply use @property it would be a property to the instance, not to the class12:57
*** e0ne has joined #openstack-keystone12:58
dstanek@property acts on the class, but has an effect on the instance - it creates a descriptor on the class12:59
*** EinstCrazy has joined #openstack-keystone12:59
dstanekjust like any function than is defined as a method12:59
openstackgerritSamuel de Medeiros Queiroz proposed openstack/keystone: Create unit tests for the policy drivers  https://review.openstack.org/21295713:00
samueldmqdstanek: ^13:00
samueldmqdstanek: hmm, interesting, it makes sense, because it is a regular class function anyways13:00
samueldmqdstanek: that always get executed when called (even without () ) and returns the value, that's why it can be called like an attribute13:02
samueldmq(that's my assumption on how it works ) :)13:02
dstaneksamueldmq: look at the example here that doesn't use the decorator syntax: https://docs.python.org/2/library/functions.html#property13:03
dstaneksamueldmq: you can see that self is passed into the get/set/del methods just like a method13:03
*** real56 has joined #openstack-keystone13:05
raildostevemar: ping, the cross project session about v3 migration :) https://www.openstack.org/summit/austin-2016/summit-schedule/events/948113:05
*** shoutm has joined #openstack-keystone13:15
*** EinstCrazy has quit IRC13:17
*** henrynash has quit IRC13:22
*** EinstCrazy has joined #openstack-keystone13:22
*** rodrigods has quit IRC13:22
*** rodrigods has joined #openstack-keystone13:22
*** anush_ has joined #openstack-keystone13:24
*** richm has joined #openstack-keystone13:26
*** rderose has joined #openstack-keystone13:27
*** jsavak has joined #openstack-keystone13:29
*** subscope has quit IRC13:30
*** EinstCrazy has quit IRC13:31
*** stingaci has joined #openstack-keystone13:34
*** stingaci has quit IRC13:38
lbragstado/13:40
*** dims is now known as dimsum__13:41
samueldmqlbragstad: o/13:41
*** dansmith is now known as superdan13:42
*** mylu has joined #openstack-keystone13:43
dolphm\o13:45
dolphmPSA: morgan was elected to the TC!13:46
dolphmmorgan: congratulations!13:46
morgan:)13:49
*** EinstCrazy has joined #openstack-keystone13:50
dstanekmorgan: congrats13:51
*** links has quit IRC13:51
morganwhee13:52
stevemarraildo: nice13:55
*** real56 has quit IRC13:55
*** EinstCrazy has quit IRC13:55
samueldmqmorgan: congrats13:56
*** subscope has joined #openstack-keystone13:56
*** ayoung has joined #openstack-keystone13:57
*** ChanServ sets mode: +v ayoung13:57
raildomorgan: congrats sir :)13:58
*** knikolla has joined #openstack-keystone14:03
*** darren_wang has joined #openstack-keystone14:04
*** ayoung has quit IRC14:04
*** mylu has quit IRC14:05
darren_wangHi, dear keystone team, is there any possibility that during deleting contents of a domain, the machine suddenly powers off and the contents could not be clean deleted?14:06
lbragstadmorgan congrats!14:11
*** mylu has joined #openstack-keystone14:11
morgandarren_wang: i could see some issues with that.14:12
morgandarren_wang: but not sure where that impact ends up being without digging a bit more. likely it could just be issued (the delete) again14:12
morgansince we are fairly good at being stateless in our queries, and we wrap everything in transactions so worst case is data is left over - but a fully failed/db break is unlikely un;ess you also crash the db server AND the system hand;es that/corrupts the filesystem14:14
*** dave-mccowan has joined #openstack-keystone14:16
*** jaosorior has quit IRC14:20
darren_wangmorgan: Thanks. I'm not sure about the possible impacts either. It just came to me after I saw the codes that contents are deleted after receiving notifications, and I think alike when that extreme situation does happen, some garbage data may be left.14:20
morgandarren_wang: in general, keystone is very hard to break like that14:21
morganmost cases just reissuing the delete should be sufficient14:21
darren_wangI see14:21
morganbut that is also hard to test ;)14:22
darren_wangsomething abnormal14:22
morganto be fair, i've never heard of a case since i started working on keystone where that was the root of a serious problem14:22
morganno one has reported that type of issue in the wild/production that ive seen14:23
darren_wanghah, then it's just a theoretic assumption14:24
morganconceptually, it is possible data is left in the db, but it is highly unlikely to break keystone operations14:24
morganbecause of the way we deelete. either we handle the associated data directly before removing the domain record, or we rely on the db for FKs14:25
morgani expect if it did happen, someone just did "delete domain" again and it worked.14:25
morgan:)14:25
*** anush_ has quit IRC14:26
morgandarren_wang: def. feel free to look more closely and if you see something weird let us know/help us fix it (write code), etc. happy to make keystone better14:26
darren_wangOK, thank you morgan.14:26
morganand also welcome to our little corner of IRC :)14:27
*** mylu has quit IRC14:27
*** spzala has joined #openstack-keystone14:27
darren_wangno problem~ :)14:29
*** nisha has joined #openstack-keystone14:31
*** darren_wang has quit IRC14:32
*** spzala has quit IRC14:33
*** EinstCrazy has joined #openstack-keystone14:34
*** subscope has quit IRC14:34
*** daemontool has quit IRC14:35
*** EinstCrazy has quit IRC14:37
*** trown is now known as trown|relocating14:37
*** nisha has quit IRC14:41
*** sdake has joined #openstack-keystone14:45
*** slberger has joined #openstack-keystone14:47
*** sdake_ has joined #openstack-keystone14:49
openstackgerritThomas Bechtold proposed openstack/keystoneauth: Fix doc build if git is absent  https://review.openstack.org/30346614:50
*** sdake has quit IRC14:51
*** agrebennikov has joined #openstack-keystone14:52
openstackgerritRodrigo Duarte proposed openstack/keystone-specs: Add note about service provider fields  https://review.openstack.org/30347114:53
*** EinstCrazy has joined #openstack-keystone14:53
*** david_cu has joined #openstack-keystone14:54
*** david_cu has quit IRC14:54
*** agrebennikov has quit IRC14:59
*** spzala has joined #openstack-keystone14:59
*** nisha has joined #openstack-keystone15:01
samueldmqdstanek: need your python expertise in https://review.openstack.org/#/c/212006/18/keystone/tests/unit/endpoint_policy/backends/test_sql.py15:01
patchbotsamueldmq: patch 212006 - keystone - Create unit tests for endpoint policy drivers15:01
samueldmqdstanek: why does the commented code does not work ?15:01
*** mylu has joined #openstack-keystone15:02
dstaneksamueldmq: looking...15:02
samueldmqdstanek: I think it's related to how python calls setUp when we hve multiple inheritance15:02
*** EinstCrazy has quit IRC15:02
dstaneksamueldmq: what is the error that you are getting?15:03
samueldmqrole table doesn't exist15:03
samueldmqlet me paste it15:03
morgansamueldmq: it is multi-inheritence ick i'm sure :(15:04
dstaneksamueldmq: probably the order of the parents15:04
samueldmqdstanek: http://paste.openstack.org/show/493485/15:04
*** subscope has joined #openstack-keystone15:04
samueldmqmorgan: I think so too :(15:04
samueldmqdstanek: yes, and if I change the order of parents, the other setUp is not called15:05
dstaneksamueldmq: let me pull it down and take a look15:05
samueldmqbut if I do the proxy class, both setUp get called15:06
samueldmqit's very weird15:06
morganthere is a reason the inverted order doesn't work15:06
morganit's MRO in python15:06
morgansamueldmq: basically, you end up calling super in the baseTest class, which goes to it's parent, now it doesn't chace back up to the other inherited class15:07
morganwith the object ahead of the base class, the super from the mix-in calls the base class setup with super. If the mixin had a parent with setup, it would skip the basetest class setup15:08
morganpython MRO is a bit wonky with multiple inheritence15:08
morgani mean... it acts the same every time, but it's not intuative. but neither is multiple inheritence15:09
dstaneksamueldmq: this is why i was removing a lot of the inheritence from our tests :-(15:10
morgandstanek: ++15:10
samueldmq:(15:14
*** real56 has joined #openstack-keystone15:15
samueldmqdstanek: morgan: it's still a bit confusing to me :(15:15
*** ayoung has joined #openstack-keystone15:16
*** ChanServ sets mode: +v ayoung15:16
*** agrebennikov has joined #openstack-keystone15:16
samueldmqmorgan: actually looks like none of the orders work15:20
samueldmqmorgan: when I don't define the "proxy" class (BaseSqlDriverTestCase=15:20
morgansamueldmq: the proxy class might also have super inheritence issues15:20
morganagain MRO is odd. i'd need to dig a bit further15:20
samueldmqmorgan: it does work with the proxy class15:20
morganhave a few things to look at later today15:20
samueldmqthe uncommented code works, the commented one doesn't :(15:21
samueldmqhttps://review.openstack.org/#/c/212006/18/keystone/tests/unit/endpoint_policy/backends/test_sql.py15:21
patchbotsamueldmq: patch 212006 - keystone - Create unit tests for endpoint policy drivers15:21
samueldmqmorgan: sure15:21
samueldmqmorgan: I 100% agree with you, something very odd hapenning there, and it's very likely to be MRO15:21
samueldmqjust wanted to understand15:21
*** nisha__ has joined #openstack-keystone15:21
dstaneksamueldmq: it appears to be an order of operations thing to me. because of the mro the broken one tries to do db work before the db setup work is completed15:22
morgandstanek: yah15:22
dstaneksamueldmq: usually this requires rearranging the setUps15:22
samueldmqdstanek: but the db is only used in tests15:23
samueldmqwhich is always after all setUp calls15:23
samueldmqat least should be15:23
dstaneksamueldmq: you don't setup the db until after the rest of the setup15:23
dstaneksamueldmq: but your test_base tries to to api calls15:24
dstanekso if you don't have a DB until after that how would it work?15:24
dstanekbrb15:24
samueldmqdstanek: ah15:24
samueldmqdstanek: yes, you're correct man15:24
samueldmqmorgan: this guy found it ^15:24
*** lhcheng has quit IRC15:25
*** aginwala has joined #openstack-keystone15:25
*** nisha has quit IRC15:25
*** aginwala has quit IRC15:29
*** aginwala has joined #openstack-keystone15:30
openstackgerritRodrigo Duarte proposed openstack/keystone: Add service providers integration tests  https://review.openstack.org/30350215:31
*** trown|relocating is now known as trown15:31
*** spzala has quit IRC15:32
*** mylu has quit IRC15:33
openstackgerritSamuel de Medeiros Queiroz proposed openstack/keystone: Create unit tests for endpoint policy drivers  https://review.openstack.org/21200615:33
samueldmqmorgan: dstanek: working fine now, thanks for your help15:34
*** mylu has joined #openstack-keystone15:34
*** nisha__ is now known as nisha15:34
samueldmqbknudson: ^ all comments addressed sir, thanks for reviewing, it's much better now15:34
openstackgerritRodrigo Duarte proposed openstack/keystone-specs: Add note about service provider fields  https://review.openstack.org/30347115:34
*** stingaci has joined #openstack-keystone15:35
*** mylu has quit IRC15:38
*** mylu has joined #openstack-keystone15:38
*** shoutm has quit IRC15:38
*** stingaci has quit IRC15:39
nishahey all :)15:39
openstackgerritSamuel de Medeiros Queiroz proposed openstack/keystone: Create unit tests for the policy drivers  https://review.openstack.org/21295715:40
dstaneksamueldmq: my pleasure15:41
openstackgerritSamuel de Medeiros Queiroz proposed openstack/keystone: Create unit tests for endpoint policy drivers  https://review.openstack.org/21200615:41
*** mhickey has quit IRC15:42
*** flaper87 has quit IRC15:42
samueldmqnisha: hi15:42
*** sdake has joined #openstack-keystone15:43
*** sdake_ has quit IRC15:44
*** flaper87 has joined #openstack-keystone15:44
*** nisha has quit IRC15:45
*** nisha has joined #openstack-keystone15:46
ayoungsamueldmq, did you take a look at the hack job I did on your keystone.common.controller cleanup?15:51
ayoungsamueldmq, https://review.openstack.org/#/c/279263/15:52
patchbotayoung: patch 279263 - keystone - enforcement logic refactored15:52
*** aginwala has quit IRC15:53
*** aginwala has joined #openstack-keystone15:54
*** spzala has joined #openstack-keystone15:54
*** aginwala has quit IRC15:56
*** aginwala_ has joined #openstack-keystone15:56
*** EinstCrazy has joined #openstack-keystone15:56
*** mhickey has joined #openstack-keystone15:57
*** EinstCrazy has quit IRC15:58
*** spzala has quit IRC15:59
*** jsavak has quit IRC16:00
*** e0ne has quit IRC16:00
openstackgerritayoung proposed openstack/keystone: [WIP]Make fernet default token provider  https://review.openstack.org/25865016:03
*** dflorea has joined #openstack-keystone16:05
*** jsavak has joined #openstack-keystone16:06
*** dflorea has quit IRC16:06
ayoungraildo, OK  Got it down to 3 fail trust tests and....keystone.tests.unit.test_v3_assignment.AssignmentTestCase.test_token_revoked_once_group_role_grant_revoked16:06
ayoungthat one is...nasty16:07
ayoungbut it should resolve to a "revoke by userid"  for all users in the group....16:07
*** dflorea has joined #openstack-keystone16:12
*** daemontool has joined #openstack-keystone16:13
*** dflorea has quit IRC16:15
*** browne has joined #openstack-keystone16:17
*** ksnihyr has quit IRC16:19
*** aginwala_ has quit IRC16:19
*** aginwala has joined #openstack-keystone16:19
*** dflorea has joined #openstack-keystone16:20
raildoayoung: ok, I'll working on it today16:21
*** aginwala has quit IRC16:24
ayoungraildo, tackle th trusts ones.  I've got a handle on the other failure16:24
raildoayoung: so I'll take a look in the previous patch16:24
ayoungraildo, we were not calling the revoke API for non persisted tokens16:24
*** jsavak has quit IRC16:24
ayoungwhich is strange.  It was a deliberate choice, but does not make sense16:25
ayounganyway, I've almost got that test fixed.  Update patch as soon as I doo16:25
ayoungdoo bee doo bee doo16:25
*** mvk_ has joined #openstack-keystone16:25
raildoayoung: haha, ok sir16:25
*** mylu has quit IRC16:26
*** mylu has joined #openstack-keystone16:26
*** timcline has joined #openstack-keystone16:27
*** mkrcmari__ has quit IRC16:29
*** aginwala has joined #openstack-keystone16:31
*** jsavak has joined #openstack-keystone16:32
*** roxanaghe has joined #openstack-keystone16:32
*** slberger1 has joined #openstack-keystone16:32
*** slberger has quit IRC16:32
*** spzala has joined #openstack-keystone16:33
*** nisha has quit IRC16:34
*** subscope has quit IRC16:35
*** mylu has quit IRC16:40
*** jsavak has quit IRC16:42
*** jsavak has joined #openstack-keystone16:43
*** anush_ has joined #openstack-keystone16:43
*** mylu has joined #openstack-keystone16:44
ayoungraildo, can I get a second et of eyes on this?  I want to make sure I am reading it right.  Once I fix the token provider validation, this call fails, but with a 401, not a 404:  http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_assignment.py#n34516:44
ayoungraildo, and, it fails prior to hitting the controller code16:45
ayoungwhich means that the token used to validate the token is bad, but it does not look like a token is passed in excpet for the headers={'x-subject-token': token},16:45
ayoungshouldn't that call require an Auth token in addition to the subject token?16:45
*** harlowja has joined #openstack-keystone16:46
*** Ephur has joined #openstack-keystone16:46
raildoayoung: this error 401 will occur for every request after a delete grant...16:46
*** pcaruana has quit IRC16:46
ayounghow did that test ever work?  Should we be able to call validate token without a valid token?16:46
raildoayoung: i had the same error in other tests16:46
ayoungraildo, is that a known issue?16:46
*** anush_ has quit IRC16:46
raildoayoung: that why i tried to skip the revocation event16:47
ayoungraildo, ooh, no.  That is a sign of something more significant wrong16:47
ayoungraildo, so, in that call to self.head('/auth/tokens',    is it authentication somehow?16:48
ayoungah16:48
*** david_cu has joined #openstack-keystone16:48
ayoungself.head is a unit test call, so the unit test holds on to an admin token?16:49
raildoyes, since every request will get a new token16:49
ayoungraildo, AH...so that test uses the same user for validating the token that it using for the system under test... Let me try rewritting the test with a different user16:51
ayoungwe need to pslit admin off the test user...maybe more than just there16:51
*** edmondsw has quit IRC16:52
raildoayoung: yes... I'm trying to find the exactly method16:52
ayoungraildo, its OK, I have it.16:52
ayoungraildo, fooled myself thinkg the post call was standard requres, not unit test16:52
*** rcernin has quit IRC16:55
*** e0ne has joined #openstack-keystone16:56
*** ksnihyr has joined #openstack-keystone16:57
*** mhickey has quit IRC16:57
*** e0ne has quit IRC16:57
*** rderose has quit IRC17:00
*** aginwala has quit IRC17:00
htrutaguys, when someone instantiates a service client, such as novaclient with a ksa session, what do we expect to have as user_agent?17:00
*** aginwala has joined #openstack-keystone17:00
*** aginwala has quit IRC17:01
*** anush_ has joined #openstack-keystone17:01
*** aginwala has joined #openstack-keystone17:02
*** dflorea has quit IRC17:03
*** can8dnSix has joined #openstack-keystone17:03
*** mylu has quit IRC17:06
*** shaleh has joined #openstack-keystone17:07
*** roxanaghe has quit IRC17:07
shalehare Domain scoped tokens still valid?17:07
shalehno project specified, just Domain17:08
*** roxanaghe has joined #openstack-keystone17:09
morganuhm.17:15
morgani mean, i don't think we got rid of those17:15
*** e0ne has joined #openstack-keystone17:15
shalehmorgan: I was testing out the new oslopolicy-checker and the tool only worked for project scoped tokens. It seemed like an omission, but maybe I was missing something.17:16
*** aginwala has quit IRC17:16
morganayoung: ^cc?17:17
shalehmorgan: I was experimenting with domain admins and my policy was failing inside keystone.17:17
morgandomain scope has never been well tested17:17
morganand def needs some custom policy.json stuff17:17
shalehmorgan: which is why I started asking questions before I started debugging and hacking code :-)17:18
morganyeah17:18
morganit was not the intention to make domain-scope go away as far as I know17:18
* morgan looks at flights for Austin.17:19
shalehfunny thing is, when i modded the oslopolicy-checker to take domain too my token passed policy in the tool17:19
morganLOL17:19
morganthat kinda... wow17:19
*** roxanaghe has quit IRC17:19
shalehwhich leads me to believe I hit a weird code path in Keystone17:19
morganyou might have17:19
*** aginwala has joined #openstack-keystone17:20
*** roxanaghe has joined #openstack-keystone17:23
*** spzala has quit IRC17:24
*** jistr has quit IRC17:25
*** jsavak has quit IRC17:25
*** jsavak has joined #openstack-keystone17:26
*** jasonsb has quit IRC17:29
*** timcline has quit IRC17:36
*** timcline has joined #openstack-keystone17:37
*** fawadkhaliq has joined #openstack-keystone17:39
*** timcline has quit IRC17:41
knikollamorgan, congrats on TC election.17:41
morganknikolla: thnx17:41
fawadkhaliqhi folks, quick question..is there a way to associate arbitrary tags to tenant/project in Keystone? Either some existing mechanism or a way to enhance to introduce this mechanism. The use case is to introduce tenant/project based affinity using tags.17:42
morganfawadkhaliq: waht would tenant/project based affinity do?17:42
morganbefore i answer, a concrete "what are you solving" helps :)17:42
*** roxanaghe has quit IRC17:43
morganand where would these tags be consumed?17:43
fawadkhaliqmorgan: in a very large scale deployment, would like to schedule all the tenant resources (workloads, networks etc) into a particular zone/cell/region etc.17:43
morganah.17:43
fawadkhaliqmorgan: tags would be consumed by the likes of Nova schedular17:43
*** sdake_ has joined #openstack-keystone17:44
morganso, you *can* add arbitrary data to just about any object in keystone17:44
morganit gets serailized into the object within the DB17:44
morgannow. nothing outside of keystone really knows how to consume this/use it17:44
morganalso, it's not a great idea to do.17:44
*** sdake has quit IRC17:45
fawadkhaliqmorgan: I see, can you please point me to how that works?17:45
morgani am not sure the best place to put such a tag thing, because nova/cinder/scheduler at best receives just the tenant/project id (uuid) and not the whole object17:45
*** jsavak has quit IRC17:45
fawadkhaliqmorgan: would like feedback on why its not a good idea, maybe I am missing something.17:45
morganbecause it's unbounded data and not validated.17:45
morganit's not good from a standpoint of the implementation in keystone.17:45
morganand that it's very isolated to keystone17:46
*** jsavak has joined #openstack-keystone17:46
fawadkhaliqmorgan: makes sense and agree.17:46
shalehfawadkhaliq: in other words, the admin adds the data but the user discovers the data is wrong.17:46
morgani think this type of tagging should almost go as close to the consumer (like quotas do) as possible17:46
morganbut - i could see why it's useful to put in keystone17:46
morgansince it's centralized17:47
*** roxanaghe has joined #openstack-keystone17:47
fawadkhaliqgot it, can you please point me to how to utilize these tags today? I would like play around with it a bit.17:47
shalehmorgan: perhaps some Congress rules to keep items together?17:48
*** mylu has joined #openstack-keystone17:51
fawadkhaliqmorgan: ^17:52
*** stingaci has joined #openstack-keystone17:52
*** e0ne has quit IRC17:52
*** mylu has quit IRC17:55
shalehfawadkhaliq: there is a 'extra' blob stored with the project when stored in SQL. You would add data there.17:56
fawadkhaliqshaleh: thanks, I see it. Quick check..is this available in both v2 and v3? :)17:56
*** jsavak has quit IRC17:57
shalehwhat is this v2 thing you speak of? :-)17:57
fawadkhaliqshaleh: lol..some people still use them, sorry ;-)17:57
*** jsavak has joined #openstack-keystone17:57
shalehfawadkhaliq: I _think_ so. But you need to test it.17:58
shalehwhen you REST call the create project, just include an extra: { data....} and see what happens.17:58
shalehErrors will lead you to the places to hack :-)17:58
fawadkhaliqshaleh: got it. thank you so much yourself and morgan. it was helpful. :)17:59
*** sdake has joined #openstack-keystone17:59
shalehfawadkhaliq: if you are feeling adventurous a middleware could intercept the project creation and perform your business logic for validation17:59
*** stingaci has quit IRC18:00
*** stingaci has joined #openstack-keystone18:00
*** tellesnobrega is now known as tellesnobrega_af18:00
*** tellesnobrega_af is now known as tellesnobrega18:00
fawadkhaliqshaleh:  yes, just the start. this is definitely going to evolve. its certainly better than what our colleagues at cern did by using conf files. but definite room for improvement.18:01
shalehfawadkhaliq: piece of advice, prefix your key names in extras with something specific to the cloud/site/whatever. This way no one else steps on you by accident.18:01
*** sdake_ has quit IRC18:01
*** jsavak has quit IRC18:02
fawadkhaliqshaleh: thanks. lots of risks here, true.18:02
*** jsavak has joined #openstack-keystone18:03
*** fawadkhaliq has quit IRC18:03
*** fawadkhaliq has joined #openstack-keystone18:03
amakarovstevemar, o/18:04
amakarovstevemar, what should be done to target unified delegations to newton?18:05
*** jsavak has quit IRC18:05
*** jsavak has joined #openstack-keystone18:05
*** spzala has joined #openstack-keystone18:05
*** roxanaghe has quit IRC18:06
*** stingaci has quit IRC18:13
*** roxanaghe has joined #openstack-keystone18:14
*** timcline has joined #openstack-keystone18:14
*** ayoung has quit IRC18:14
morganamakarov: if no work was landed in Mitaka, just retarget the spec (review)18:22
morganamakarov: move it to newton18:22
*** e0ne has joined #openstack-keystone18:22
morganif things were done in mitaka... there might be a little more to do18:22
*** roxanaghe has quit IRC18:23
*** stingaci has joined #openstack-keystone18:23
*** aginwala has quit IRC18:24
*** e0ne has quit IRC18:25
*** e0ne has joined #openstack-keystone18:26
amakarovmorgan, thanks, can I target a blueprint too then?18:26
*** e0ne has quit IRC18:26
morganblueprint uhm sure - i don't know who can set blueprint targets. but def. do the spec review to retarget to newton18:27
*** aginwala has joined #openstack-keystone18:28
amakarovmorgan, got it. btw, can you please give me a feedback about closure table: https://review.openstack.org/#/c/285521/18:28
patchbotamakarov: patch 285521 - keystone - Closure table for HMT18:28
morganamakarov: it's been on my list i need to followup with DinaBelova's changes today too18:29
amakarovmorgan, we agreed to try it on midcycle18:29
morganfor profiling18:29
*** stingaci has quit IRC18:29
amakarovmorgan, profiling is more important18:29
shalehamakarov: is it less cute than last time?18:29
morganamakarov: i think they're about the same importance fwiw18:29
morgan:)18:29
morganamakarov: because we're early in the cycle.18:30
morganboth should land.18:30
morganthe closure table > profiling late in the cycle18:30
morganif that makes sense18:30
morganbecause if we are going to use HMT, we need it.18:30
amakarovmorgan, btw, are we going to actually use HMT at long last? :)18:31
morganno idea18:31
morganbut closure table is a requirement to make it fully featured.18:32
morganand usable imo18:32
amakarovshaleh, it depends on the target of the comparison )18:32
shalehamakarov: haneef showed my the profilerproxy class yesterday. It was trying too hard :-)18:33
shalehshowed me18:33
*** jsavak has quit IRC18:33
*** jsavak has joined #openstack-keystone18:34
amakarovmorgan, it has a weak point though: multiple inserts will be slow as index gets... well... reindexed )18:34
morganprojects change infrequently compared to reads18:34
*** slberger1 has quit IRC18:35
amakarovshaleh, that thing gives us very interesting results already (in house, of course) and it is definitely a thing to have in upstream18:36
shalehamakarov: agreed on value, disagreed on coding style.18:36
shalehunless there is some reason it needs to use dynamic type generation gymnastics18:37
amakarovshaleh, is it that terrible?18:37
shalehamakarov: coworkers yesterday asked me why it was so complex. Zero comments. Assumption without checking of "%s" formatting.18:38
*** slberger has joined #openstack-keystone18:38
shalehI'd like to understand why it needs to go through the get_profiler_proxy() function.18:38
amakarovshaleh, oh, that... forgot to ask Dina today about it - a shame (18:38
shalehIt looked like the class could just provide __init__18:38
shalehwhen you employ magic in code you need comments explaining why18:39
morganshaleh: possibly. i think osprofiler needs to be rethought as a consumer of hook points in the various projects vs. directly including osprofiler referenced code in the projects18:39
morganshaleh: but that is a convo i'm going to start with DinaBelova at the summit18:39
shalehmorgan: reasonable18:39
morganso we can make it x-project18:39
amakarovshaleh, I cam imagine they want to use it as a template, but even named it "prefix" ))18:40
morganbut until then, we should seriously consider (which what changes are needed) to land the current code.18:40
amakarovSo yes - %s is redundant18:40
*** roxanaghe has joined #openstack-keystone18:40
shalehmorgan: if the cute factor goes away it looks reasonable now18:40
morganshaleh: yeah the technical perspective looks good. hence why we should cover what is needed to land it18:41
morganshaleh: and then we work to make profiling better in *all* of openstack18:41
shalehmorgan: agreed18:41
morganwith a good pattern that allows tools that aren't "osprofiler only" to hook in18:41
shalehmorgan: you wear the TC hat well18:42
morganLOL18:42
*** sdake has quit IRC18:42
shalehmorgan: It would be nice to have something more like the US and less like the UN18:42
shalehamakarov: is there sufficient detail on the review for DinaBelova to update it quickly?18:43
amakarovshaleh, I think so. Otherwise I'll fix it myself and then you'll have to merge it before she noticed :D18:46
*** stingaci has joined #openstack-keystone18:46
shalehamakarov: :-)18:47
*** aginwala has quit IRC18:49
*** aginwala has joined #openstack-keystone18:50
amakarovshaleh, jamielennox|away left a comment about moving proxy to oclo.cache18:50
shalehamakarov: following morgan's thinking18:51
*** sdake has joined #openstack-keystone18:52
*** vgridnev has joined #openstack-keystone18:57
*** aginwala has quit IRC18:57
*** roxanaghe has quit IRC19:02
*** jsavak has quit IRC19:02
*** jsavak has joined #openstack-keystone19:03
*** jed56 has quit IRC19:03
*** ayoung has joined #openstack-keystone19:09
*** ChanServ sets mode: +v ayoung19:09
openstackgerritRodrigo Duarte proposed openstack/keystone: Add identity providers integration tests  https://review.openstack.org/30229919:12
openstackgerritRodrigo Duarte proposed openstack/keystone: Add service providers integration tests  https://review.openstack.org/30350219:12
*** roxanaghe has joined #openstack-keystone19:12
*** david_cu has left #openstack-keystone19:12
rodrigodsbknudson, ^ think it is much better now :)19:13
*** dave-mccowan has quit IRC19:15
*** roxanaghe has quit IRC19:23
*** sheel has quit IRC19:27
*** daemontool has quit IRC19:33
*** browne has quit IRC19:34
*** daemontool has joined #openstack-keystone19:37
*** mylu has joined #openstack-keystone19:39
*** anush_ has quit IRC19:40
*** mylu has quit IRC19:48
*** anush_ has joined #openstack-keystone19:48
*** mylu has joined #openstack-keystone19:48
*** mylu has quit IRC19:49
*** cdcasey has joined #openstack-keystone19:49
*** real56 has quit IRC19:50
*** slberger has quit IRC19:54
stevemarctracey: sent you a *hip* email - heyooo19:55
*** bigjools has quit IRC19:56
*** SpamapS has quit IRC19:58
*** zeus has quit IRC19:58
*** jdennis has quit IRC19:59
*** jdennis has joined #openstack-keystone19:59
*** aswadr_ has quit IRC20:00
*** daemontool has quit IRC20:02
*** zeus has joined #openstack-keystone20:03
*** zeus is now known as Guest8001920:03
*** roxanaghe has joined #openstack-keystone20:04
*** bigjools has joined #openstack-keystone20:06
*** bigjools has quit IRC20:06
*** bigjools has joined #openstack-keystone20:06
shalehstevemar: I was playing with oslopolicy-checker, nice addition. I ran into a potential edge case.20:09
shalehstevemar: what if my token is domain scoped instead of project scoped? I noticed there was no support for that currently.20:10
*** dave-mccowan has joined #openstack-keystone20:10
*** stingaci has quit IRC20:10
*** fawadkhaliq has quit IRC20:11
*** fawadkhaliq has joined #openstack-keystone20:11
*** stingaci has joined #openstack-keystone20:12
stevemarshaleh: domain scoped are slowly going away, but yeah, probably a hole in functionality20:12
shalehstevemar: why is that? if I have a domain admin user, why scope to a project too?20:13
*** SpamapS has joined #openstack-keystone20:15
*** browne has joined #openstack-keystone20:16
*** dan_nguyen has joined #openstack-keystone20:17
*** slberger has joined #openstack-keystone20:18
*** arunkant has quit IRC20:20
*** sdake_ has joined #openstack-keystone20:26
*** tellesnobrega is now known as tellesnobrega_af20:26
*** daemontool has joined #openstack-keystone20:26
*** sdake has quit IRC20:27
*** woodster_ has joined #openstack-keystone20:33
*** vgridnev has quit IRC20:36
*** vgridnev has joined #openstack-keystone20:40
*** sdake_ has quit IRC20:42
*** vgridnev has quit IRC20:42
*** edmondsw has joined #openstack-keystone20:43
*** stingaci has quit IRC20:47
samueldmqcrinkle: the _placeholder things on the migrations are for backport20:53
*** roxanaghe has quit IRC20:53
crinklesamueldmq: ah cool, thanks20:54
samueldmqcrinkle: for example, something that affects current version (Newton) and should be placed in Mitaka too20:54
samueldmqcrinkle: then it replaces a _placeholder :)20:54
crinklegotcha, that makes sense20:55
*** roxanaghe has joined #openstack-keystone20:58
*** stingaci has joined #openstack-keystone20:58
*** can8dnSix has quit IRC21:04
*** daemontool has quit IRC21:05
*** pece has quit IRC21:05
*** trown is now known as trown|outtypewww21:05
*** jsavak has quit IRC21:06
*** raildo is now known as raildo-afk21:07
*** clenimar has quit IRC21:07
*** pauloewerton has quit IRC21:11
*** gordc has quit IRC21:12
Anticimexis this example up to date? http://docs.openstack.org/developer/python-keystoneclient/using-api-v3.html#authenticating-using-sessions21:13
*** pnavarro_ has quit IRC21:13
*** timcline has quit IRC21:16
morganAnticimex: unlikely, as keystoneclient.session is deprecated in favour of keystoneauth21:17
morganAnticimex: let me see if i can find the right link21:17
morganAnticimex: this is keystoneauth http://docs.openstack.org/developer/keystoneauth/using-sessions.html#sessions-for-users21:18
morganAnticimex: and for "developers" http://docs.openstack.org/developer/keystoneauth/using-sessions.html#sessions-for-client-developers21:18
shalehAnticimex: good news is not much difference21:18
morganshaleh: ++21:18
*** mylu has joined #openstack-keystone21:21
*** agrebennikov has quit IRC21:24
*** mylu has quit IRC21:25
Anticimexmorgan: ah, thanks for that pointer.  i was doubting a bit, mostly due to my vim pylint thingy not understanding my virtualenv i guess :)21:26
*** anush_ has quit IRC21:26
*** mylu has joined #openstack-keystone21:26
*** vgridnev has joined #openstack-keystone21:27
Anticimexshaleh: looks very similar yes21:28
*** mylu has quit IRC21:28
*** vgridnev has quit IRC21:32
*** roxanaghe has quit IRC21:38
*** roxanaghe has joined #openstack-keystone21:39
*** roxanaghe has quit IRC21:40
*** roxanaghe has joined #openstack-keystone21:40
*** roxanaghe has quit IRC21:40
Anticimexthe old example had region_name, as keystoneclient.v3.client.Client has.  is that now replaced with endpoint_filter as per http://docs.openstack.org/developer/keystoneauth/using-sessions.html#service-discovery ?21:42
*** wwriverrat has joined #openstack-keystone21:42
*** agrebennikov has joined #openstack-keystone21:43
Anticimexi'll figure it out21:44
*** slberger has quit IRC21:44
*** ksnihyr has quit IRC21:45
*** wwriverrat has left #openstack-keystone21:45
dtroyer_zzanyone who wants to talk ksa still around?21:46
dtroyer_zzI'm wondering if the fixtures are fair game for use in OSC's tests:  keystoneauth1/fixture/v2.py and v3.py21:47
*** cdcasey has quit IRC21:47
*** browne has quit IRC21:50
*** slberger has joined #openstack-keystone21:53
*** Ephur has quit IRC21:58
morgandtroyer_zz: hey21:59
morganso whats up?21:59
morganoh in keystoneauth1/fixture22:00
morganyes that should be consumable outside of keystoneauth22:00
morganit's designed explicitly for that22:00
dtroyer_zzyeah… I've made jamielennox|away happy in the past using things I shouldn't have ;)22:00
dtroyer_zzcoolio22:00
dtroyer_zzI need some changes, but I can work around that for now22:00
morganlet me know what changes are needed22:00
*** stingaci has quit IRC22:01
morganwe've done a *really* good job in keystoneauth marking things private that should be private22:01
morganwe started everything private then moved the symbols public when we spun it out of keystoneclient22:01
dtroyer_zzI'll pull that together later, almost dinner time… I'll point them out in the review I'm going to put up in a few minutes if I make flake8 happy22:01
morganok22:02
morgansounds good :)22:02
dtroyer_zzit's not privacy so much as _service makes some assumptions I wish it didn't, helpful things that don't mimic the real world22:02
*** slberger has left #openstack-keystone22:02
dtroyer_zz_Service22:02
morganright, see _Service (directly) isn't fair game :P22:02
morganbut Token is22:02
morganetc.22:03
dtroyer_zzI'm going through Token, but the assumption is in _Service22:03
morganright22:03
morganand happy to fix if we can :)22:03
dtroyer_zzI want to take a crack at it and get my one commit to ksa ;)22:04
dtroyer_zzoh, and congrats on TC22:04
morganthanks :P22:04
morgani honeslyt didn't expect to win22:04
morgandtroyer_zz: ^_^22:04
morgannot upset about it, but surprised :)22:04
*** sheel has joined #openstack-keystone22:04
morganit'll be definitely interesting22:05
dtroyer_zzI haven't looked at the votes, there are sometimes interesting things in there, who else got high votes when you did or didn't, etc.22:05
morganeh, not sure.22:05
dtroyer_zzso do you have a new gig yet?22:05
morgannot yet22:05
morganworking on some details for one22:05
dtroyer_zzWell then you've repeated my achievement last april, getting elected without a job!22:06
morgandtroyer_zz: ^5!22:06
morganI was required to put a company name on my badge for the summit22:06
morganit might say "-"22:06
morganat the moment22:06
Anticimexhalp.  https://gist.github.com/Millnert/ad1eadd53824a5e405acd451ef5338cd  - doesn't work, getting 401 unauthorised e.g. on the keystone.users.list()22:07
dtroyer_zzI had already done all that, mine said Nebula anyway, but I started at Intel a couple weeks before, so I just slapped a sticker on it22:08
*** stingaci has joined #openstack-keystone22:08
morgandtroyer_zz: doing it right, for sure!22:08
Anticimexlooking for other sample usage of keystoneauth /keystonev3?22:08
Anticimexs/?//22:08
morganAnticimex: are you an admin there or have admin rights22:09
Anticimexyea22:09
morganuser list is admin action iirc.22:09
morganok.22:09
Anticimexadmin perms22:09
Anticimexit's just an example22:09
morgandoes everything fail or just user.list?22:09
morganeveryhitng in keystoneclient that is22:09
*** kragniz has quit IRC22:09
Anticimexah, no, first error is that attempt22:10
*** kragniz has joined #openstack-keystone22:10
Anticimexbut i get same error when i don't run that22:10
morganright22:10
Anticimexeg instead let the code continue and attempt to get endpoints22:10
morganso you get the same error even w/o user.list22:10
morgan?22:10
morganok22:10
morganso, check to see if your credentials are correct22:10
morganand your scope is correct.22:10
Anticimexack, i'll doublecheck that.22:11
shalehAnticimex: side comment, dict.get() returns None if the item is not found. So you can collapse the front matter in your function22:11
morganand make sure you have the right CACert if the endpoint is ssl22:11
Anticimexi guess openstackclient is wise.  my rc file is a bit of a mess, trying to work with eg "glance" cli and openstack etc22:12
morganbecause requests is ... weird.22:12
Anticimexhaven't figured out what's the correct envs to use for all cases22:12
Anticimexyeah, cacert works though, otherwise i'd get ssl error22:12
morgancorect "envs for all cases: is hard.22:12
Anticimexright, it's not actually what i want22:12
Anticimexi'm using multiple now22:12
Anticimexone for global admin22:12
Anticimexanother for v3 domain admin22:12
shalehos-client-config and clouds.yaml FTW22:13
shalehenv vars are so 30 years ago22:13
morganshaleh: ++22:13
Anticimexinteresting22:13
Anticimexone thing at a time :]22:13
shalehAnticimex: heh, are you new to OpenStack? It is guzzle the ocean and try not to drown. :-)22:13
*** lhcheng has joined #openstack-keystone22:14
*** ChanServ sets mode: +v lhcheng22:14
Anticimexshaleh: thx for feedback on the .get()  some minutes earlier i was using env['foo'] which gives keyerror22:14
dtroyer_zzmorgan: https://review.openstack.org/#/c/276350/22/openstackclient/tests/identity/v2_0/fakes.py has the stuffs I was asking about…22:15
patchbotdtroyer_zz: patch 276350 - python-openstackclient - Moving authentication from keystoneclient to keyst...22:15
shalehAnticimex: dict.get() returns None. dict.get(key, base_value) returns base_value if key is not in dict22:15
Anticimexshaleh: well, first time writing clients yeah. but we have three deployments up and running within a week.  we just haven't.. eh, used them so much yet :)22:15
shalehAnticimex: both have their uses22:15
Anticimexack22:15
morgandtroyer_zz: ahh22:15
*** lhcheng has quit IRC22:15
*** markvoelker has quit IRC22:16
*** agrebennikov has quit IRC22:17
*** lhcheng has joined #openstack-keystone22:18
*** ChanServ sets mode: +v lhcheng22:18
*** fawadkhaliq has quit IRC22:25
*** fawadkhaliq has joined #openstack-keystone22:25
*** agrebennikov has joined #openstack-keystone22:26
*** kragniz has quit IRC22:27
*** lhcheng has quit IRC22:28
*** kragniz has joined #openstack-keystone22:28
*** agrebennikov has quit IRC22:29
*** Guest80019 is now known as zeus22:34
*** zeus has joined #openstack-keystone22:34
*** arunkant has joined #openstack-keystone22:40
*** dan_nguyen has quit IRC22:40
*** knikolla has quit IRC22:43
Anticimexmorgan: got it working, by stroke of genius (openstack --debug $foo-op) and check what params are sent22:44
Anticimex;)22:44
*** dan_nguyen has joined #openstack-keystone22:44
Anticimexnow though, all examples i find for "getting the glance endpoint" are quite old.  so once i have the working client instance, should i go services -> endpoints or endpoints -> blackbox to figure out which endpoint i need for glanceclient?22:44
*** fpatwa has joined #openstack-keystone22:45
Anticimexi'm fumbling around at http://docs.openstack.org/developer/python-keystoneclient/api/keystoneclient.v3.html#22:45
*** stingaci has quit IRC22:45
*** kragniz has quit IRC22:49
*** kragniz has joined #openstack-keystone22:49
*** stingaci has joined #openstack-keystone22:55
*** BigWillie has joined #openstack-keystone22:56
*** shaleh has quit IRC22:56
*** BigWillie has quit IRC23:00
*** browne has joined #openstack-keystone23:05
*** richm has quit IRC23:10
*** markvoelker has joined #openstack-keystone23:16
*** markvoelker has quit IRC23:21
*** wolsen has quit IRC23:24
*** wolsen has joined #openstack-keystone23:25
*** dan_nguyen has quit IRC23:26
*** dan_nguyen has joined #openstack-keystone23:30
*** cburgess has quit IRC23:30
*** cburgess has joined #openstack-keystone23:32
*** lhcheng has joined #openstack-keystone23:40
*** ChanServ sets mode: +v lhcheng23:40
*** browne has quit IRC23:45
*** browne has joined #openstack-keystone23:45
openstackgerritRodrigo Duarte proposed openstack/keystone: Add service providers integration tests  https://review.openstack.org/30350223:47
*** fawadkhaliq has quit IRC23:48
*** fawadkhaliq has joined #openstack-keystone23:48
*** markvoelker has joined #openstack-keystone23:52

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