*** shrasool has quit IRC | 00:16 | |
*** shrasool has joined #openstack-keystone | 00:16 | |
*** erus has quit IRC | 00:38 | |
*** erus has joined #openstack-keystone | 00:40 | |
*** erus has quit IRC | 01:05 | |
*** erus has joined #openstack-keystone | 01:08 | |
*** gyee has quit IRC | 01:09 | |
*** shrasool has quit IRC | 01:17 | |
*** dave-mccowan has quit IRC | 02:00 | |
*** Dinesh_Bhor has joined #openstack-keystone | 02:12 | |
*** imacdonn has quit IRC | 02:53 | |
*** imacdonn has joined #openstack-keystone | 02:53 | |
*** erus has quit IRC | 03:16 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system reader role for users https://review.openstack.org/605485 | 03:17 |
---|---|---|
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system member role user test coverage https://review.openstack.org/623317 | 03:17 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system admin role in users API https://review.openstack.org/623318 | 03:17 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement domain reader functionality for user API https://review.openstack.org/623319 | 03:17 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement domain member functionality for user API https://review.openstack.org/623320 | 03:17 |
*** erus has joined #openstack-keystone | 03:19 | |
*** raildo has joined #openstack-keystone | 03:30 | |
*** erus has quit IRC | 03:47 | |
*** erus has joined #openstack-keystone | 03:48 | |
*** erus has quit IRC | 03:54 | |
*** erus has joined #openstack-keystone | 04:07 | |
*** erus has quit IRC | 04:13 | |
*** erus has joined #openstack-keystone | 04:22 | |
*** lbragstad has quit IRC | 04:23 | |
*** erus has quit IRC | 04:28 | |
*** erus has joined #openstack-keystone | 04:36 | |
*** erus has quit IRC | 04:43 | |
*** erus has joined #openstack-keystone | 04:44 | |
*** erus has quit IRC | 04:46 | |
*** erus has joined #openstack-keystone | 04:47 | |
*** erus has quit IRC | 04:54 | |
*** erus has joined #openstack-keystone | 05:03 | |
*** erus has quit IRC | 05:10 | |
*** erus has joined #openstack-keystone | 05:14 | |
*** erus has quit IRC | 05:21 | |
*** erus has joined #openstack-keystone | 05:30 | |
*** erus has quit IRC | 05:36 | |
*** erus has joined #openstack-keystone | 05:44 | |
*** erus has quit IRC | 05:50 | |
*** erus has joined #openstack-keystone | 06:02 | |
*** Dinesh_Bhor has quit IRC | 06:04 | |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Ensure change is addressed for unified limit table https://review.openstack.org/621497 | 06:37 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Add domain_id column for limit https://review.openstack.org/620202 | 06:37 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Add domain level limit support - Manager https://review.openstack.org/621468 | 06:37 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Add domain level limit support - API https://review.openstack.org/622773 | 06:37 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: [WIP] Add domain level support for strict-two-level-model https://review.openstack.org/623153 | 06:37 |
*** Dinesh_Bhor has joined #openstack-keystone | 06:53 | |
*** rcernin has quit IRC | 07:01 | |
*** dklyle has quit IRC | 07:09 | |
*** dklyle has joined #openstack-keystone | 07:10 | |
*** trident has quit IRC | 07:23 | |
*** trident has joined #openstack-keystone | 07:25 | |
*** dims has quit IRC | 07:44 | |
*** dims has joined #openstack-keystone | 07:47 | |
*** awalende has joined #openstack-keystone | 08:11 | |
*** Dinesh_Bhor has quit IRC | 08:14 | |
*** Dinesh_Bhor has joined #openstack-keystone | 08:19 | |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Add domain level limit support - API https://review.openstack.org/622773 | 08:28 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: [WIP] Add domain level support for strict-two-level-model https://review.openstack.org/623153 | 08:28 |
*** awalende has quit IRC | 08:30 | |
*** Dinesh_Bhor has quit IRC | 09:11 | |
*** Dinesh_Bhor has joined #openstack-keystone | 09:12 | |
*** Dinesh_Bhor has quit IRC | 09:40 | |
*** trident has quit IRC | 10:18 | |
*** trident has joined #openstack-keystone | 10:21 | |
*** sapd1_ has joined #openstack-keystone | 10:50 | |
sapd1_ | Hi anyone, How to use openstack command line with keystone-to-keystone config? Using user in keystone identity provider to access keystone service provider. | 10:51 |
*** trident has quit IRC | 11:04 | |
cmurphy | sapd1_: example here https://docs.openstack.org/keystone/latest/advanced-topics/federation/federated_identity.html#testing-it-all-out | 11:05 |
*** trident has joined #openstack-keystone | 11:06 | |
openstackgerrit | Merged openstack/ldappool master: Change openstack-dev to openstack-discuss https://review.openstack.org/622571 | 11:10 |
*** sapd1_ has quit IRC | 11:17 | |
*** sapd1_ has joined #openstack-keystone | 11:20 | |
*** shrasool has joined #openstack-keystone | 11:26 | |
*** sapd1_ has quit IRC | 11:35 | |
*** sapd1_ has joined #openstack-keystone | 11:51 | |
*** sapd1_ has quit IRC | 11:56 | |
*** shrasool has quit IRC | 12:03 | |
*** shrasool has joined #openstack-keystone | 12:07 | |
openstackgerrit | Merged openstack/python-keystoneclient master: Change openstack-dev to openstack-discuss https://review.openstack.org/621926 | 12:09 |
openstackgerrit | Merged openstack/oslo.limit master: Update mailinglist from dev to discuss https://review.openstack.org/621795 | 12:15 |
*** shrasool has quit IRC | 12:16 | |
openstackgerrit | Merged openstack/keystonemiddleware master: Change openstack-dev to openstack-discuss https://review.openstack.org/622764 | 12:31 |
*** trident has quit IRC | 12:39 | |
*** trident has joined #openstack-keystone | 12:41 | |
*** shrasool has joined #openstack-keystone | 12:42 | |
*** shrasool has quit IRC | 12:45 | |
*** sapd1_ has joined #openstack-keystone | 13:02 | |
*** sapd1_ has quit IRC | 13:07 | |
*** cenekzach has joined #openstack-keystone | 13:18 | |
*** mhen has joined #openstack-keystone | 13:25 | |
mhen | Keystone API v3 provides a "/policies" endpoint. I struggle to find information on what this is actually used for. How does this relate to the "policy.json" file used in Keystone and other services? | 13:32 |
cenekzach | Hello, question about keystone-ldap integration. Keystone uses 2 methods [1] from python ldap module for LDAP queries 'search_s' and 'search_ext'. While the former is 'synchronous' (as the ldap module calls it) the latter is not (it has synchronous variant called 'search_ext_s' [2]). It there a reason why the 'ext' variant is not synchchronous? The ldap module implements reconnect in case of connection failure only for the sy | 13:35 |
cenekzach | nchronous methods. Our ldap connections get dropped (RST from the server side) and keystone reconnects after several minutes, all requests fail till then. | 13:35 |
cenekzach | [1] keystone/keystone/identity/backends/ldap/common.py | 13:35 |
cenekzach | [2] https://github.com/python-ldap/python-ldap/blob/master/Lib/ldap/ldapobject.py | 13:35 |
*** jdennis has quit IRC | 13:36 | |
*** edmondsw has quit IRC | 13:42 | |
cmurphy | mhen: it's a partially baked API that we've deprecated, we don't really refer to it anywhere because we don't encourage anyone to use it | 13:43 |
cmurphy | cenekzach: I'm not totally sure but I think these comments give a hint that it was intentional http://git.openstack.org/cgit/openstack/keystone/tree/keystone/identity/backends/ldap/common.py#n772 http://git.openstack.org/cgit/openstack/keystone/tree/keystone/identity/backends/ldap/common.py#n798 | 13:45 |
openstackgerrit | Merged openstack/keystone master: Update role policies for system reader https://review.openstack.org/622524 | 13:45 |
mhen | cmurphy, thanks for your response! So essentially the API is not actually used anywhere in OpenStack? | 13:46 |
cmurphy | mhen: correct | 13:46 |
*** alexchadin has joined #openstack-keystone | 13:47 | |
mhen | ok, thanks | 13:51 |
*** dave-mccowan has joined #openstack-keystone | 14:05 | |
*** lbragstad has joined #openstack-keystone | 14:06 | |
*** ChanServ sets mode: +o lbragstad | 14:06 | |
*** dave-mccowan has quit IRC | 14:10 | |
*** edmondsw has joined #openstack-keystone | 14:12 | |
openstackgerrit | Colleen Murphy proposed openstack/keystone master: Consolidate Keystone docs: admin/identity-external-authentication.rst https://review.openstack.org/547087 | 14:13 |
*** lbragstad has quit IRC | 14:25 | |
*** lbragstad has joined #openstack-keystone | 14:29 | |
*** ChanServ sets mode: +o lbragstad | 14:29 | |
*** raildo has joined #openstack-keystone | 14:32 | |
lbragstad | o/ | 14:33 |
openstackgerrit | Colleen Murphy proposed openstack/keystone master: Consolidate catalog management guide https://review.openstack.org/623501 | 14:33 |
cmurphy | lbragstad: o/ | 14:33 |
*** jdennis has joined #openstack-keystone | 14:33 | |
cmurphy | lbragstad: easy review plz https://review.openstack.org/622977 | 14:33 |
* lbragstad loves easy reviews on friday morning | 14:34 | |
lbragstad | i don't think we really need to wait for another core there? | 14:35 |
cmurphy | i wouldn't | 14:35 |
*** Dinesh_Bhor has joined #openstack-keystone | 14:42 | |
*** mvkr has quit IRC | 14:42 | |
*** takamatsu has joined #openstack-keystone | 14:43 | |
*** alexchadin has quit IRC | 14:43 | |
*** jdennis has quit IRC | 14:45 | |
*** shrasool has joined #openstack-keystone | 14:54 | |
*** Dinesh_Bhor has quit IRC | 14:56 | |
openstackgerrit | Merged openstack/ldappool master: Fix releasenotes build https://review.openstack.org/622977 | 14:59 |
*** jdennis has joined #openstack-keystone | 15:03 | |
*** trident has quit IRC | 15:25 | |
*** trident has joined #openstack-keystone | 15:27 | |
openstackgerrit | Merged openstack/ldappool master: Add py36 tox environment https://review.openstack.org/615847 | 15:31 |
*** shrasool has quit IRC | 15:37 | |
*** shrasool has joined #openstack-keystone | 15:49 | |
*** shrasool has quit IRC | 15:52 | |
lbragstad | cmurphy i like the federation introduction doc | 15:57 |
cmurphy | \o/ | 15:58 |
*** shrasool has joined #openstack-keystone | 15:59 | |
*** shrasool has quit IRC | 16:05 | |
*** erus has quit IRC | 16:11 | |
lbragstad | so - it looks like the old policy.v3cloudsample.json file allowed project and domain admins to list all roles in the deployment | 16:12 |
lbragstad | with how we're doing things moving forward | 16:13 |
lbragstad | i'm not sure i see the value in allowing project admins to call GET /v3/roles ? | 16:13 |
lbragstad | it might make sense for a domain admin, if they're looking to add role assignments to users and projects within their domain | 16:14 |
lbragstad | but we also have domain-specific roles... | 16:14 |
lbragstad | wondering if people have an opinion here | 16:14 |
*** erus has joined #openstack-keystone | 16:17 | |
*** gyee has joined #openstack-keystone | 16:25 | |
lbragstad | i guess a deployment could have global roles that are specific to other users | 16:26 |
*** mvkr has joined #openstack-keystone | 16:27 | |
lbragstad | domain and project users would be able to see those and that might not be right? | 16:27 |
lbragstad | i guess a safer alternative would be to block all access to global roles for domain and project users | 16:27 |
lbragstad | and domain admins can create domain specific roles, which they do have access to | 16:27 |
knikolla | o/ | 16:31 |
knikolla | lbragstad: cmurphy: thanks for reviewing the renewable application credentials spec | 16:32 |
knikolla | i'll wait for kmalloc as well and then incorporate the feedback. | 16:32 |
lbragstad | ack | 16:33 |
lbragstad | idk - the second path feels more inline with the actual defaults we have today and safer in general | 16:34 |
lbragstad | the policy.v3cloudsample.json was the official defaults anyway | 16:34 |
lbragstad | wasn't* | 16:34 |
lbragstad | i'll write things up for the second approach and document why in the commit messages... then we can just iterate in review | 16:35 |
*** raildo has quit IRC | 16:36 | |
*** shrasool has joined #openstack-keystone | 16:44 | |
kmalloc | Let me get coffee | 16:46 |
lbragstad | grab me some, too kthx | 16:47 |
kmalloc | lbragstad: FYI, vacation for me will be starting Dec 18, and I'll be back jan 14 | 16:47 |
lbragstad | sounds good | 16:47 |
kmalloc | I might check in some, but the goal is to be mostly offline from work. | 16:48 |
lbragstad | as it should be | 16:48 |
kmalloc | Yup | 16:48 |
*** shrasool has quit IRC | 16:49 | |
*** erus has quit IRC | 17:00 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add role tests for system member role https://review.openstack.org/622525 | 17:02 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update role policies for system admin https://review.openstack.org/622526 | 17:02 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with roles https://review.openstack.org/622527 | 17:02 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with roles https://review.openstack.org/622528 | 17:02 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove role policies from policy.v3cloudsample.json https://review.openstack.org/622529 | 17:02 |
*** erus has joined #openstack-keystone | 17:04 | |
*** shrasool has joined #openstack-keystone | 17:10 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update endpoint policies for system admin https://review.openstack.org/619331 | 17:13 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with endpoints https://review.openstack.org/619332 | 17:13 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with endpoints https://review.openstack.org/619281 | 17:13 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove endpoint policies from policy.v3cloudsample.json https://review.openstack.org/619333 | 17:13 |
*** dims has quit IRC | 17:21 | |
cenekzach | cmurphy thanks, I will dig deeper | 17:36 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update registered limit policies for system admin https://review.openstack.org/621016 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with registered limits https://review.openstack.org/621017 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with registered limits https://review.openstack.org/621018 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove registered limit policies from policy.v3cloudsample.json https://review.openstack.org/621019 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add limit protection tests https://review.openstack.org/621020 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add limit tests for system member role https://review.openstack.org/621021 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update limit policies for system admin https://review.openstack.org/621022 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with limits https://review.openstack.org/621023 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with limits https://review.openstack.org/621024 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove limit policies from policy.v3cloudsample.json https://review.openstack.org/621025 | 17:55 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add region protection tests for system readers https://review.openstack.org/619085 | 18:16 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add region tests for system member role https://review.openstack.org/619086 | 18:16 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update region policies to use system admin https://review.openstack.org/619241 | 18:16 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with regions https://review.openstack.org/619242 | 18:16 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with regions https://review.openstack.org/619243 | 18:16 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove region policies from policy.v3cloudsample.json https://review.openstack.org/619244 | 18:16 |
* lbragstad goes to make coffee | 18:41 | |
*** shrasool has quit IRC | 18:46 | |
*** dims has joined #openstack-keystone | 18:50 | |
kmalloc | knikolla: commented on the app-cred spec | 19:14 |
kmalloc | lbragstad: onto my 4th cup now. :P | 19:15 |
lbragstad | nice | 19:16 |
lbragstad | sounds to me like you're "review ready" | 19:16 |
lbragstad | at least that's what i'm hearing ;) | 19:16 |
kmalloc | "review ready" is such a loaded term | 19:17 |
lbragstad | ready to review all the things * | 19:17 |
kmalloc | on the plus side, i got my office almost all cleaned up and my desk is almost all setup like it was supposed to be | 19:17 |
*** shrasool has joined #openstack-keystone | 19:20 | |
*** shrasool has quit IRC | 19:22 | |
kmalloc | lbragstad: i already reviewed the app creds thing | 19:24 |
knikolla | kmalloc: thanks! responded. | 19:28 |
knikolla | i like the renew on login. | 19:28 |
knikolla | kmalloc: with regards to the conversation we had during the keystone meeting, do you want to the idp associated with an app cred to be immutable? | 19:29 |
kmalloc | yes. | 19:35 |
kmalloc | please | 19:35 |
knikolla | will do. | 19:35 |
knikolla | you also think that we should not force a ttl on app creds created through a federated login that only have concrete roles? | 19:36 |
kmalloc | no | 19:37 |
kmalloc | explicitly not force a ttl in that case | 19:37 |
kmalloc | sorry if i wasn't clear | 19:38 |
kmalloc | a user may set a TTL in that case. | 19:38 |
kmalloc | a user is not forced into a TTL because the roles exist explicitly within keystone | 19:38 |
kmalloc | vs. conveyed by the IDP Auth | 19:38 |
knikolla | what about the case of a disabled user still having access? | 19:39 |
kmalloc | if a user has concrete roles in keystone | 19:40 |
kmalloc | chances are they can login locally to keystone anyway | 19:40 |
kmalloc | this is the side-band "go disable the user in keystone" problem | 19:40 |
kmalloc | oath was already looking into | 19:40 |
kmalloc | and we should make that an explicit case. | 19:40 |
kmalloc | maybe not? | 19:41 |
kmalloc | i tend to err to the side of only roles that require refreshing are forced into the IDP ttl | 19:41 |
kmalloc | as long as the behavior is clearly outlined and documented i think we're ok | 19:41 |
knikolla | not if they only rely on the idp for authN rather than authZ | 19:42 |
knikolla | they may have concrete roles, but no account locally | 19:42 |
knikolla | (the MOC case) | 19:42 |
knikolla | also I think autoprovisioning actually creates the concrete role assignments according to this snippet https://github.com/openstack/keystone/blob/b25a655793db0859f9c3e77a013fa26346ec8435/keystone/auth/plugins/mapped.py#L173-L176 | 19:44 |
kmalloc | right. | 19:44 |
kmalloc | i'm ok with that going either way | 19:45 |
kmalloc | as long as the behavior is clearly outlined | 19:45 |
kmalloc | and documented | 19:45 |
knikolla | i could provide a config option to toggle it on or off | 19:45 |
kmalloc | ugh. no. | 19:45 |
kmalloc | we can add a toggle later if needed | 19:45 |
knikolla | as forcing ttl on app creds created through federation changes the behavior of current app creds | 19:46 |
knikolla | current behavior of creating app creds* | 19:46 |
kmalloc | do we allow app creds with federated logins? | 19:46 |
knikolla | for concrete roles yes | 19:46 |
knikolla | i think | 19:46 |
kmalloc | hm. | 19:47 |
kmalloc | lets confirm | 19:47 |
kmalloc | if we do, then the behavior has to remain the same =/ | 19:47 |
knikolla | a quick look a the code makes me think we do | 19:49 |
knikolla | as we only check that the user has the roles in the project | 19:49 |
knikolla | https://github.com/openstack/keystone/blob/b25a655793db0859f9c3e77a013fa26346ec8435/keystone/application_credential/core.py#L132 | 19:49 |
kmalloc | bah | 19:51 |
lbragstad | is it just me or is this test asserting the exact opposite of what we want? https://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_protection.py#n961 | 19:53 |
kmalloc | lbragstad: that a non-domain admin shouldn't be able to do things? | 20:05 |
kmalloc | it's asserting a non-domain-admin can't do management and then a domain admin can | 20:05 |
lbragstad | is it? https://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_protection.py#n970 | 20:06 |
lbragstad | https://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_protection.py#n979 | 20:06 |
nsmeds | I know this isn't exactly correct channel - but you guys appear to be most active with the policy.json work | 20:09 |
nsmeds | I've managed to get some good overrides working for keystone and cinder. But neutron completely ignored the policy.json changes. | 20:09 |
nsmeds | anyone aware of issue? | 20:09 |
nsmeds | (posted in their channel but tis a Friday) | 20:09 |
kmalloc | lbragstad: _test_user_management | 20:10 |
kmalloc | lbragstad: not "test_user_management" | 20:10 |
kmalloc | lbragstad: https://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_protection.py#n797 | 20:10 |
kmalloc | so we pass in the user data. it does appear to be testing the right thing. | 20:11 |
kmalloc | or at least we're getting a forbidden | 20:11 |
kmalloc | https://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/unit/test_v3_protection.py#n802 | 20:11 |
lbragstad | oh ... wtf | 20:11 |
kmalloc | we use self.auth | 20:11 |
* lbragstad shakes head | 20:12 | |
kmalloc | we don't re-auth in _Test_user_management | 20:12 |
kmalloc | we auth with "just a user" and set that to self.auth | 20:12 |
kmalloc | it's not straightforward. | 20:12 |
lbragstad | no at all | 20:12 |
lbragstad | not* | 20:12 |
kmalloc | but it does work and is asserting the correct behavior | 20:12 |
lbragstad | bah - oik | 20:12 |
kmalloc | also git.openstack.org is way harder to read than github | 20:12 |
kmalloc | it takes me ~10x as long because of lack of syntax highlighing | 20:13 |
*** shrasool has joined #openstack-keystone | 20:22 | |
*** cenekzach has left #openstack-keystone | 20:26 | |
*** shrasool has quit IRC | 21:13 | |
*** shrasool has joined #openstack-keystone | 21:15 | |
*** shrasool_ has joined #openstack-keystone | 21:23 | |
*** shrasool has quit IRC | 21:23 | |
*** shrasool_ is now known as shrasool | 21:23 | |
*** jdennis has quit IRC | 21:27 | |
*** shrasool has quit IRC | 21:30 | |
*** jdennis has joined #openstack-keystone | 21:44 | |
lbragstad | kmalloc the domain-id/domain_id normalization stuff made it into the flask refactor? | 21:47 |
lbragstad | i'm noticing something odd with query strings and the user API | 21:48 |
lbragstad | but i could just be uneducated | 21:48 |
lbragstad | for example -this is what i have locally | 21:49 |
lbragstad | https://pasted.tech/pastes/980db6c1c3111b1358f4149c136ab76f190beec5 | 21:49 |
lbragstad | and this is the failure i'm seeing https://pasted.tech/pastes/a976b0a0350601826aa90f43e20330d99b0f2c46.raw | 21:49 |
*** imus has quit IRC | 21:52 | |
*** jdennis has quit IRC | 21:53 | |
*** shrasool has joined #openstack-keystone | 22:09 | |
kmalloc | lbragstad: uhm. | 22:14 |
kmalloc | the normalization stuff might have made it in | 22:15 |
kmalloc | i mean... i don't remember specifically | 22:15 |
lbragstad | i see bits of it | 22:15 |
kmalloc | we have SOME normalization bits | 22:15 |
kmalloc | i don't think i changed anything behavior wise with body-key normalization | 22:16 |
lbragstad | we have normalization on the body and on query parameters | 22:17 |
*** erus has quit IRC | 22:17 | |
*** erus has joined #openstack-keystone | 22:18 | |
kmalloc | i don't see explicitly where that is happening atm. | 22:18 |
kmalloc | but doesn't mean it didn't land in the flask refactor | 22:18 |
lbragstad | sure - it was a lot of code | 22:21 |
kmalloc | oh i see | 22:21 |
lbragstad | i'm trying to maintain the behavior we have in these tests | 22:21 |
kmalloc | any and all calls for "._normalize_dict" does it | 22:21 |
kmalloc | because normalize_dict -> normalize_arg | 22:21 |
kmalloc | which does `-` -> `_` | 22:21 |
*** trident has quit IRC | 22:22 | |
kmalloc | https://github.com/openstack/keystone/blob/6dd1c7dae82b55d9d7126da7f4356eac2cc494c1/keystone/server/flask/common.py#L902-L908 | 22:22 |
kmalloc | there used to be other normalization mechanisms | 22:22 |
lbragstad | https://github.com/openstack/keystone/blob/master/keystone/tests/unit/test_v3_protection.py#L981-L1056 | 22:22 |
*** dmellado has quit IRC | 22:22 | |
kmalloc | they were all collapsed | 22:22 |
*** trident has joined #openstack-keystone | 22:22 | |
kmalloc | query args, that one i am not seeingf | 22:22 |
kmalloc | yet. | 22:22 |
kmalloc | but it might be a flask-ism | 22:22 |
lbragstad | i think we get that with | 22:23 |
lbragstad | flask.requests.args.get('domain_id') | 22:23 |
kmalloc | yeah. | 22:23 |
kmalloc | so we might be more normalized now... | 22:24 |
lbragstad | which might work with ?domain-id=$domain-id and ?domain_id=$domain-id | 22:24 |
kmalloc | though that is odd | 22:24 |
kmalloc | ?domain-id afaict is ... never supposed to work? | 22:24 |
kmalloc | also... extras still suck | 22:24 |
kmalloc | a lot | 22:24 |
lbragstad | well - yeah.. actually that's the part that tripping me up i think | 22:25 |
lbragstad | because we actually use it in policy enforcement if domain-id is a qs | 22:25 |
kmalloc | well thats broken | 22:25 |
lbragstad | https://github.com/openstack/keystone/blob/master/keystone/tests/unit/test_v3_protection.py#L1018-L1025 | 22:25 |
lbragstad | ^ right? | 22:25 |
kmalloc | so my view is pretty straightforward | 22:25 |
kmalloc | if the *code* says domain_id is the way we enforce | 22:26 |
kmalloc | policy is wrong | 22:26 |
kmalloc | and we fix it. policy is config | 22:26 |
kmalloc | not API contract | 22:26 |
lbragstad | policy.v3cloudsample.json you mean | 22:26 |
lbragstad | yeah | 22:26 |
kmalloc | yeah. | 22:26 |
lbragstad | domain_id is a proper filter, i think | 22:26 |
kmalloc | doesn't matter what policy file. | 22:26 |
kmalloc | domain_id is the filter | 22:26 |
kmalloc | i'd -2 code to add domain-id as a filter | 22:26 |
kmalloc | to fix policy.json-isms that were wrong | 22:26 |
lbragstad | domain-id should be ignored to be consistent with the rest of the keystone | 22:26 |
kmalloc | yes. | 22:27 |
lbragstad | ok | 22:27 |
lbragstad | so if that's the case... | 22:27 |
kmalloc | in v4 we can consider not ignoreing query params | 22:27 |
lbragstad | https://github.com/openstack/keystone/blob/master/keystone/tests/unit/test_v3_protection.py#L1018-L1025 shoudl assert empty lists? | 22:27 |
kmalloc | and i am still seirous a v4 should happen | 22:27 |
lbragstad | and not 403s | 22:27 |
kmalloc | hmmmmm. | 22:27 |
kmalloc | yes. | 22:27 |
kmalloc | fix the policy file to be not stupid, assert on that behavior | 22:27 |
lbragstad | right? because if i'm a domain admin ask for users in another domain that i'm not an admin of, an empty list should be what the filter returns | 22:28 |
kmalloc | yep | 22:29 |
kmalloc | 100% | 22:29 |
lbragstad | ok - i'll have to rework these tests | 22:29 |
kmalloc | sorry | 22:29 |
lbragstad | no worriesy | 22:29 |
lbragstad | this stuff looks old anyway | 22:29 |
lbragstad | other stuff probably evolved around it without us knowing | 22:30 |
kmalloc | yeh | 22:30 |
kmalloc | it's annoying | 22:30 |
lbragstad | well - we have some better organization now, so hopefully it'll happen less in the future | 22:31 |
kmalloc | this is part of why the endless pluggability is so bad. | 22:31 |
kmalloc | weird things happen and grow around it. | 22:31 |
*** dmellado has joined #openstack-keystone | 23:11 | |
*** shrasool has quit IRC | 23:29 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Bump oslo.policy and oslo.context versions https://review.openstack.org/623248 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system reader role for users https://review.openstack.org/605485 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system member role user test coverage https://review.openstack.org/623317 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement system admin role in users API https://review.openstack.org/623318 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement domain reader functionality for user API https://review.openstack.org/623319 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement domain member functionality for user API https://review.openstack.org/623320 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Implement domain admin functionality for user API https://review.openstack.org/623321 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add explicit testing for project users and the user API https://review.openstack.org/623322 | 23:42 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove user policies from policy.v3cloudsample.json https://review.openstack.org/623323 | 23:42 |
lbragstad | kmalloc should be all fixed now ^ | 23:42 |
lbragstad | most of the changes we talked about are in the last patch in that series, where i remove the tests from test_v3_protection and add them to keystone.tests.unit.protection.v3.test_users | 23:42 |
lbragstad | ported your original comments, too | 23:43 |
lbragstad | where applicable | 23:43 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!