ayoung | jamielennox, depends on what stage of the development cycle we clone. | 00:23 |
---|---|---|
ayoung | jamielennox, good to hear from you. Did you read the context? Horizon is doing something unsavory with the user_id and the client | 00:24 |
ayoung | rafaelweingartne, for K2K? | 00:24 |
jamielennox | i didn't really look, something about context? | 00:24 |
rafaelweingartne | k2k? | 00:30 |
rafaelweingartne | well, I would like to integrate OpenStack in a federation | 00:30 |
rafaelweingartne | we already have IdPs, which are implemented with Keycloak | 00:30 |
rafaelweingartne | Therefore, OpenStack would be an SP | 00:30 |
rafaelweingartne | I need to know the endpoints that OpenStack uses to configure in Keycloak (as a client), and it is always easier to get that data from a metadata file that describer the provider (IdP or SP) | 00:31 |
openstackgerrit | Vishakha Agarwal proposed openstack/keystone master: Add openstack_groups to assertion https://review.openstack.org/588211 | 01:00 |
*** vishakha has joined #openstack-keystone | 01:05 | |
cmurphy | rafaelweingartne: for keystone as a saml identity provider then the metadata endpoint is /v3/OS-FEDERATION/saml2/metadata, for keystone as a service provider it depends on the auth module you're using, for shibboleth it's /Shibboleth.sso/Metadata i think, for mellon it's something different | 01:36 |
rafaelweingartne | Thanks | 01:39 |
vishakha | cmurphy: Hey! | 01:48 |
cmurphy | hi vishakha | 01:52 |
vishakha | cmurphy: Again a question from federation | 01:56 |
vishakha | cmurphy: I was trying with samltest.id. But after logging with my saml credentials in horizon, I am getting unauthorized user | 01:57 |
vishakha | cmurphy: I have set up my keystone apache | 01:57 |
*** rafaelweingartne has quit IRC | 01:57 | |
vishakha | cmurphy: Have setup my shibholeth2.xml, but only error I am getting is of authorization | 01:58 |
cmurphy | vishakha: which part is giving you the unauthorized message, is it the samltest.id site, or the horizon dashboard, or is it from keystone? you should be able to tell based on the url in the browser bar | 02:00 |
*** Dinesh_Bhor has joined #openstack-keystone | 02:02 | |
cmurphy | vishakha: i suspect it's coming from keystone, in which case you can set insecure_debug=true and it should tell you more | 02:02 |
vishakha | cmurphy: I have set insecure_debug = True, But still the same error . No more information | 02:05 |
cmurphy | vishakha: can you tell where it's coming from? is it keystone, shibboleth/apache, horizon, or is it on the samltest.id site itself? | 02:10 |
vishakha | cmurphy: getting these logs from horizon_error | 02:23 |
vishakha | https://www.irccloud.com/pastebin/p5480Vmw/ | 02:23 |
vishakha | cmurphy: No logs in keystone | 02:23 |
cmurphy | vishakha: those tracebacks in horizon are normal actually | 02:24 |
cmurphy | it's weird :( | 02:24 |
vishakha | cmurphy: when logging , no logs are there in keystone. | 02:25 |
cmurphy | vishakha: when you see the unauthorized message in the browser, does the url in the browser bar have a /identity path in it? or /dashboard? or is it coming from the samltest.id site itself? | 02:25 |
vishakha | cmurphy: http://127.0.0.1/identity/v3/auth/OS-FEDERATION/websso/saml2?origin=http://localhost:5440/dashboard/auth/websso/ | 02:25 |
cmurphy | vishakha: okay, so it's coming from keystone | 02:26 |
cmurphy | keystone or shib | 02:26 |
vishakha | https://www.irccloud.com/pastebin/C5T4K5FN/ | 02:26 |
cmurphy | vishakha: is there any error in /var/log/shibboleth/shibd.log or shibd_warn.log? | 02:26 |
cmurphy | or in the main apache error log | 02:27 |
cmurphy | or sometimes those logs end up in the horizon logs so keep looking for other errors in the horizon logs | 02:27 |
vishakha | cmurphy: Yes I looked in all the log files,No error like log | 02:28 |
vishakha | cmurphy: let me share with you | 02:28 |
openstackgerrit | wangxiyuan proposed openstack/keystone master: Update contributor doc https://review.openstack.org/617513 | 02:29 |
vishakha | cmurphy: In shibholeth, no logs for today | 02:31 |
vishakha | cmurphy: I dont understand why there are so less logs | 02:33 |
cmurphy | :/ | 02:37 |
vishakha | cmurphy, wxy-xiyuan : Also updated patch for adding openstack_groups in SAML. https://review.openstack.org/#/c/588211/. Pl review | 02:39 |
cmurphy | vishakha: thanks for that, will take a look soon | 02:39 |
vishakha | cmurphy: :) | 02:41 |
*** Dinesh_Bhor has quit IRC | 02:44 | |
*** Dinesh_Bhor has joined #openstack-keystone | 02:52 | |
*** jmlowe has quit IRC | 02:59 | |
*** jmlowe has joined #openstack-keystone | 03:00 | |
*** itlinux has joined #openstack-keystone | 04:33 | |
openstackgerrit | Merged openstack/keystone master: Document user options https://review.openstack.org/618823 | 05:35 |
openstackgerrit | Merged openstack/keystone master: Add missing ws seperator between words https://review.openstack.org/618689 | 05:35 |
openstackgerrit | Merged openstack/keystone master: changed port in tools/sample_data.sh https://review.openstack.org/618196 | 06:01 |
*** annp has joined #openstack-keystone | 06:17 | |
*** artem_vasilyev has joined #openstack-keystone | 06:49 | |
openstackgerrit | Shuayb Popoola proposed openstack/keystone master: Fix --bootstrap-password's default password error https://review.openstack.org/619101 | 07:01 |
*** rcernin has quit IRC | 07:26 | |
*** sapd1__ has quit IRC | 08:54 | |
*** sapd1 has joined #openstack-keystone | 08:55 | |
*** jaosorior has quit IRC | 09:26 | |
*** pcaruana has joined #openstack-keystone | 09:48 | |
*** jaosorior has joined #openstack-keystone | 10:03 | |
*** breton has quit IRC | 10:08 | |
*** Emine has joined #openstack-keystone | 10:10 | |
*** Emine has quit IRC | 10:31 | |
*** shrasool has joined #openstack-keystone | 10:34 | |
openstackgerrit | Shuayb Popoola proposed openstack/keystone master: Fix --bootstrap-password's default password error https://review.openstack.org/619101 | 10:44 |
*** Emine has joined #openstack-keystone | 10:54 | |
*** mvkr has quit IRC | 11:03 | |
*** Dinesh_Bhor has quit IRC | 11:37 | |
*** xek has quit IRC | 11:39 | |
*** mvkr has joined #openstack-keystone | 11:43 | |
*** mugsie has joined #openstack-keystone | 11:49 | |
*** raildo has joined #openstack-keystone | 11:59 | |
*** erus has quit IRC | 12:01 | |
*** xek has joined #openstack-keystone | 12:04 | |
*** pvradu has joined #openstack-keystone | 12:31 | |
*** pvradu has quit IRC | 12:32 | |
*** pvradu has joined #openstack-keystone | 12:33 | |
lbragstad | kmalloc yeah - i was more or less curious if we should have bugs open for that | 12:36 |
lbragstad | but i think that answers my question and that we should | 12:36 |
lbragstad | originally - i was just opening bugs for keystone subsystems that didn't take into account system-scope, but there are parts that do but don't use the new default roles | 12:37 |
mbuil | vishakha: did you manage to switch to the SP in horizon successfully? | 12:38 |
lbragstad | vishakha i responded to your comments here - let me know if you have further questions https://review.openstack.org/#/c/605851/4 | 12:39 |
*** amoralej is now known as amoralej|lunch | 13:13 | |
*** artem_vasilyev has quit IRC | 13:27 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add region protection tests for system readers https://review.openstack.org/619085 | 13:43 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update region policies to include system member https://review.openstack.org/619086 | 13:43 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update region policies to use system admin https://review.openstack.org/619241 | 13:43 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with regions https://review.openstack.org/619242 | 13:43 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with regions https://review.openstack.org/619243 | 13:43 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove region policies from policy.v3cloudsample.json https://review.openstack.org/619244 | 13:43 |
*** erus has joined #openstack-keystone | 13:49 | |
*** BlackDex has quit IRC | 14:02 | |
*** pvradu_ has joined #openstack-keystone | 14:03 | |
*** BlackDex has joined #openstack-keystone | 14:04 | |
*** pvradu has quit IRC | 14:07 | |
*** pvradu_ has quit IRC | 14:07 | |
*** pvradu has joined #openstack-keystone | 14:08 | |
*** amoralej|lunch is now known as amoralej | 14:13 | |
*** jaosorior has quit IRC | 14:25 | |
openstackgerrit | John Dennis proposed openstack/oslo.policy master: Fully log RBAC enforcement data https://review.openstack.org/619260 | 14:31 |
openstackgerrit | Jens Harbott (frickler) proposed openstack/keystone master: DNM: Test jobs running on bionic instead of xenial https://review.openstack.org/611563 | 14:41 |
lbragstad | jdennis nice - thanks the for the patch | 14:44 |
lbragstad | for the* | 14:44 |
lbragstad | i'm clearly under-caffeinated | 14:45 |
jdennis | lbragstad: you're welcome | 14:53 |
*** shrasool has quit IRC | 15:13 | |
*** Emine has quit IRC | 15:17 | |
*** Emine has joined #openstack-keystone | 15:18 | |
*** jaosorior has joined #openstack-keystone | 15:24 | |
kmalloc | jdennis: woot! | 15:25 |
kmalloc | lbragstad: yep | 15:25 |
lbragstad | did gerrit just die? | 15:32 |
lbragstad | oh - nevermind | 15:32 |
lbragstad | it's back | 15:32 |
kmalloc | lbragstad: hehe | 15:37 |
*** erus has quit IRC | 15:38 | |
kmalloc | jdennis: the patch looks good. the mask_password / mask_dict_password should solve dhellman's concerns | 15:47 |
kmalloc | on the "logged data" bit the best we can. | 15:48 |
*** devx has quit IRC | 15:48 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update service policies for system reader https://review.openstack.org/619277 | 15:49 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update service policies for system member https://review.openstack.org/619278 | 15:49 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update service policies for system admin https://review.openstack.org/619279 | 15:49 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with services https://review.openstack.org/619280 | 15:49 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with services https://review.openstack.org/619281 | 15:49 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove service policies from policy.v3cloudsample.json https://review.openstack.org/619282 | 15:49 |
jdennis | kmalloc: I see I added comments essentially echoing your thoughts just after you posted | 15:50 |
kmalloc | :) | 15:52 |
kmalloc | jdennis: whelp, i think we're on the same page. that's easy :) | 15:53 |
lbragstad | if anyone is looking to do some reviews | 16:20 |
lbragstad | https://review.openstack.org/#/c/605539/ is the start of several linear series of reviews | 16:20 |
lbragstad | to incorporate default roles into APIs, implement system scope, and start removing the policy.v3cloudsample.json file | 16:20 |
lbragstad | https://review.openstack.org/#/c/603319/ also fixes a bug | 16:22 |
kmalloc | lbragstad: <whine>but i just did reviews last week</whine> -- i'll add them to my list :) | 16:26 |
lbragstad | cool - thanks | 16:26 |
kmalloc | lbragstad: probably will have some comments/score for them before lunch | 16:26 |
lbragstad | i should have the entire catalog api squared away by EOD | 16:26 |
lbragstad | incorporating default roles, system-scope, and removing obsolete policies from policy.v3cloudsample.json | 16:27 |
kmalloc | cool. | 16:43 |
kmalloc | and like i said i should be able to generate an architecture diagram for next week so that should help for forward looking planning | 16:44 |
*** shrasool has joined #openstack-keystone | 16:48 | |
*** pcaruana has quit IRC | 16:50 | |
*** xek_ has joined #openstack-keystone | 16:54 | |
*** xek has quit IRC | 16:57 | |
*** Emine has quit IRC | 17:05 | |
*** pvradu has quit IRC | 17:08 | |
*** erus has joined #openstack-keystone | 17:13 | |
*** xek_ is now known as xek | 17:18 | |
tobias-urdin | i've been talking about ip restricting users before, is this a super bad idea? | 17:19 |
tobias-urdin | http://paste.openstack.org/show/735912/ | 17:19 |
*** prashkre has joined #openstack-keystone | 17:21 | |
prashkre | lbragstad: Hi. Have you got a chance to look at https://bugs.launchpad.net/keystone/+bug/1800077? | 17:22 |
openstack | Launchpad bug 1800077 in OpenStack Identity (keystone) "LDAP Referrals were returned and ignored" [Undecided,New] | 17:22 |
prashkre | lbragstad: If not, could you please take a look? | 17:22 |
*** prashkre has quit IRC | 17:30 | |
*** shrasool has quit IRC | 17:30 | |
*** prashkre has joined #openstack-keystone | 17:31 | |
*** jaosorior has quit IRC | 17:41 | |
lbragstad | prashkre just wrapping something up now, and then i can take a look | 17:50 |
lbragstad | tobias-urdin that use case came up again last week | 17:50 |
lbragstad | specifically if it would be possible to restrict users access to projects based on IP addresses | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with endpoints https://review.openstack.org/619281 | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update endpoint policies for system reader https://review.openstack.org/619329 | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update endpoint policies for system member https://review.openstack.org/619330 | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update endpoint policies for system admin https://review.openstack.org/619331 | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with endpoints https://review.openstack.org/619332 | 17:51 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove endpoint policies from policy.v3cloudsample.json https://review.openstack.org/619333 | 17:51 |
prashkre | lbragstad: sure. will be waiting for you update. | 17:53 |
lbragstad | prashkre in that bug report | 17:54 |
lbragstad | are you just expecting an update to the log message? | 17:55 |
lbragstad | it looks like the code is only expecting to deal with a single referral and not multiple | 17:55 |
lbragstad | prashkre is ^ the direction you're looking for | 18:01 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Disambiguate between multiple referrals and config https://review.openstack.org/619336 | 18:01 |
lbragstad | er... ^ | 18:01 |
prashkre | lbragstad: I think your changes should be fine. | 18:07 |
prashkre | looks good to me. | 18:07 |
kmalloc | tobias-urdin: i thinkw e want to do this in the user-options functionality | 18:07 |
kmalloc | based upon waht we talked abouty in berlin | 18:07 |
lbragstad | or do we want to implement it in project options? | 18:11 |
lbragstad | i guess it boils down to setting it on the project or the user... | 18:11 |
lbragstad | gagehugo did https://review.openstack.org/#/c/616286/ make it to stable/rocky? | 18:12 |
ayoung | $ openstack endpoint list | 18:12 |
ayoung | You are not authorized to perform the requested action: identity:list_endpoints. (HTTP 403) (Request-ID: req-ba6d4ab2-aad5-40ec-bb30-730e9eb6332d) | 18:12 |
ayoung | Gah | 18:12 |
ayoung | "We'll give you all this info if you know where to look in the token data, but if you ask for it explicitly, we say 'NO' and slam the door." | 18:13 |
ayoung | We are not very nice | 18:13 |
lbragstad | gagehugo nevermind... i don't know how timestamps work | 18:16 |
ayoung | lbragstad, the word of the day is Disambiguate. I did not even know that word existed. Thank you. | 18:21 |
*** prashkre has quit IRC | 18:31 | |
tobias-urdin | lbragstad: kmalloc what i'm after is protecting all service and admin users which is a insane security risk since they have the admin role, i guess that one will have to do for now | 18:32 |
cmurphy | tobias-urdin: lbragstad kmalloc i opened a placeholder bug for this https://bugs.launchpad.net/keystone/+bug/1804042 | 18:32 |
openstack | Launchpad bug 1804042 in OpenStack Identity (keystone) "RFE: Add ability to restrict auth by forwarded IP" [Undecided,New] | 18:32 |
cmurphy | ayoung: it's openstack catalog show | 18:33 |
tobias-urdin | cmurphy: thanks | 18:35 |
cmurphy | tobias-urdin: if you want to write the spec there's still time ;) | 18:37 |
lbragstad | ayoung :) | 18:48 |
*** shrasool has joined #openstack-keystone | 18:49 | |
*** amoralej is now known as amoralej|off | 18:56 | |
*** xek has quit IRC | 19:00 | |
kmalloc | cmurphy: NICE | 19:05 |
kmalloc | lbragstad: i'd like to do it as project options and user options | 19:05 |
kmalloc | both | 19:05 |
kmalloc | tobias-urdin: ^ | 19:05 |
*** xek has joined #openstack-keystone | 19:08 | |
lbragstad | ack | 19:12 |
ayoung | cmurphy, um, yeah, but quick, without looking, what do you put in there to show swift? | 19:32 |
ayoung | the tools may be there, but they don't reflect how we've talked about this for years. endpoint list could do the same thing, or service list, and get the data from the token | 19:33 |
ayoung | and if someone wants to do it using curl, as the API is supposed to be the first class citizen, it does not work | 19:34 |
ayoung | and, by the way, the command is: openstack catalog show object-store | 19:34 |
ayoung | which I was only able to figure out by parsing the token output | 19:35 |
ayoung | I think that for catalog API calls, we should allow the member role to make those calls, but to implicitly add in the catalog filters for the token if they are not an admin. I'm not sure how to write that up. | 19:36 |
*** dklyle has quit IRC | 19:38 | |
*** mvkr has quit IRC | 19:40 | |
*** nsmeds has joined #openstack-keystone | 19:41 | |
*** raildo has quit IRC | 19:43 | |
*** raildo has joined #openstack-keystone | 19:43 | |
*** dklyle has joined #openstack-keystone | 19:46 | |
*** shrasool has quit IRC | 20:06 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Disambiguate between multiple referrals and config https://review.openstack.org/619336 | 20:17 |
jdennis | kmalloc: ping, https://github.com/openstack/oslo.policy/blob/f79650325f459e12bbac8f6967dcfabece1de7a4/oslo_policy/policy.py#L824 | 20:31 |
jdennis | kmalloc: but it's really oslo_context.context._DeprecatedPolicyValues | 20:32 |
kmalloc | Yeah | 20:32 |
jdennis | kmalloc: should the isintance test for both classes? | 20:32 |
kmalloc | Hmm | 20:32 |
kmalloc | Dunno | 20:33 |
kmalloc | lbragstad: ^ | 20:33 |
kmalloc | I am not sure about that choice atm | 20:33 |
kmalloc | I have to think on it. Lance might have more context to base an answer on immediately | 20:33 |
jdennis | kmalloc: yeah, it's opaque to me atm as well | 20:33 |
kmalloc | Exactly | 20:33 |
*** raildo has quit IRC | 20:35 | |
*** mvkr has joined #openstack-keystone | 20:35 | |
jdennis | lbragstad: '_DeprecatedPolicyValues' object has no attribute 'to_policy_values' | 20:39 |
lbragstad | _DeprecatedPolicyValues is kinda weird | 20:43 |
lbragstad | i had to grok it for a while before i thought i understood it | 20:43 |
jdennis | lbragstad: I think I have a solution, give me a moment to test it | 20:44 |
lbragstad | jdennis well - to answer your question, we recently made a change to oslo.policy so that we overload creds to be two different types | 20:45 |
lbragstad | it can either be a Context object or a dictionary | 20:45 |
lbragstad | as much as i'm not a fan of overloading stuff, we did it so that it was easier for services to just pass their context objects into the enforcer as opposed to constructing a dictionary from scratch | 20:46 |
lbragstad | (example - https://review.openstack.org/#/c/605539/ ) | 20:47 |
nsmeds | Hey guys, question of the curious. I'm looking to create a custom role, and any of the users in this role can create new domains and then have full admin powers within the domains they create (create/edit/delete projects, neutron resources, nova resources, cinder resources). | 20:48 |
jdennis | lbragstad: heres the problem, strutils.mask_dict_password() only will accept a dict object, we expect creds to be a dict but sometimes it's a MutableMapping (apparently) so my fix is in the test for MutableMapping to convert it to a dict, | 20:48 |
nsmeds | In the middle of reading anything related I can get my hands on - but if anyone has suggestions to point me in the right direction they'd be much appreciated. | 20:48 |
jdennis | lbragstad: but the real fix is that strutils.mask_dict_password() should accept a MutableMapping because a dict is a MutableMapping | 20:49 |
lbragstad | jdennis correct https://github.com/openstack/oslo.policy/blob/f79650325f459e12bbac8f6967dcfabece1de7a4/oslo_policy/policy.py#L825 should return a dict though | 20:49 |
lbragstad | oh - wait... | 20:49 |
lbragstad | ah - yeah.. i see what you mean | 20:50 |
jdennis | lbragstad: I changed the sense of the test here: https://github.com/openstack/oslo.policy/blob/f79650325f459e12bbac8f6967dcfabece1de7a4/oslo_policy/policy.py#L832 | 20:50 |
lbragstad | we can operate on it like a dict because its an instance of MutableMapping | 20:50 |
jdennis | and did the conversion to dict inside that test | 20:51 |
lbragstad | sure - that makes sense after reading the comment i apparently wrote, too | 20:52 |
lbragstad | that seems like it will be safe, i don't think anything is expecting to use `creds` after enforcement | 20:53 |
lbragstad | so converting it to another type is probably fine? | 20:53 |
lbragstad | (i know oslo.policy doesn't had it back to the service after enforcement) | 20:53 |
lbragstad | hand it* | 20:53 |
jdennis | lbragstad: yeah, I think the early code is responsible for assuring by the time you begin to use creds it's a dict | 20:54 |
lbragstad | nsmeds we're currently in the middle of that work and we are tracking status in bugs | 20:54 |
lbragstad | jdennis ++ yeah i agree.. the work to overload creds with something more standardized came later | 20:55 |
lbragstad | nsmeds for example - https://bugs.launchpad.net/keystone/+bug/1794376 | 20:55 |
openstack | Launchpad bug 1794376 in OpenStack Identity (keystone) "Domains API should account for system-scope and default roles" [High,In progress] - Assigned to Lance Bragstad (lbragstad) | 20:55 |
lbragstad | nsmeds we're tracking the rest of the work with https://bugs.launchpad.net/keystone/+bugs?field.tag=policy | 20:58 |
nsmeds | lbragstad: appreciate the reply. I'll go read through the open issues now to try and better understand current state of things | 21:00 |
lbragstad | nsmeds sounds good - ping if you have more questions | 21:00 |
*** shrasool has joined #openstack-keystone | 21:04 | |
lbragstad | nsmeds some of the patches to do what you're looking for are actually up for review here - https://review.openstack.org/#/c/605485/10 | 21:17 |
openstackgerrit | Merged openstack/keystone master: Region update extra support https://review.openstack.org/517726 | 21:20 |
*** erus has quit IRC | 21:20 | |
tobias-urdin | kmalloc: that would be preferable to have on both | 21:26 |
*** xek has quit IRC | 21:40 | |
lbragstad | our federated identity provider api is a bit weird | 21:56 |
lbragstad | we implement it with a PUT | 21:56 |
lbragstad | and return a 201 | 21:56 |
nsmeds | lbragstad: so you're working on creating separate roles: admin, member, reader - scoped either at system or domain level | 22:01 |
lbragstad | nsmeds yep | 22:01 |
nsmeds | but this means (as far as I can tell) that to make new domains, you'd need to be system admin | 22:02 |
*** shrasool has quit IRC | 22:02 | |
lbragstad | by default yes | 22:02 |
nsmeds | ah fair - you're improving the defaults | 22:03 |
lbragstad | nsmeds that's not to say you couldn't implement a new role | 22:03 |
nsmeds | my goal of having 1 role which allows new domains created, then admin control within domains that user created | 22:03 |
nsmeds | yep, I'll be making something custom | 22:03 |
nsmeds | gotcha | 22:03 |
*** shrasool has joined #openstack-keystone | 22:04 | |
lbragstad | nsmeds mind explaining the use case a bit more? | 22:04 |
lbragstad | i'm just curious is all | 22:04 |
nsmeds | sure. There's a team in company developing a product on openstack, they basically need "admin" permissions but we want to avoid giving them access to modify existing domains/resources in those domains | 22:08 |
nsmeds | and part of their product will require being able to create new domains and have full control within those domains | 22:08 |
nsmeds | (word of warning: joined only a few months ago and still getting familiar with Openstaack, so this is my basic understanding) | 22:08 |
*** rcernin has joined #openstack-keystone | 22:09 | |
lbragstad | aha | 22:10 |
lbragstad | intersting | 22:10 |
lbragstad | so the whole requirement for creating a new domain is kinda like signing up new users who interface with the product? | 22:10 |
nsmeds | yep, new user signs up and gets a new openstack domain | 22:11 |
nsmeds | we'll need to fine-tune the rbac stuff more as this develops, but for now giving the team a "admin-but-not-admin" role which we can add their users to would be super helpful | 22:12 |
nsmeds | so that's what I'm working on now | 22:12 |
lbragstad | gothca | 22:14 |
lbragstad | will customers be able to create new domains outside of the one they get at sign up? | 22:15 |
nsmeds | tbh, not sure | 22:16 |
*** shrasool has quit IRC | 22:17 | |
lbragstad | cool | 22:17 |
lbragstad | well - one thing we're also working that might help is application credentials | 22:17 |
lbragstad | for example | 22:18 |
lbragstad | you could give an instance of your product an application credential with the system admin role, but tighting it down to only be able to call the POST /v3/domains API if you wanted | 22:18 |
lbragstad | so you end up minimizing damage done if the product is compromised (not saying it would), just coding defensively | 22:19 |
lbragstad | that's something we're trying to get done for Stein http://specs.openstack.org/openstack/keystone-specs/specs/keystone/stein/capabilities-app-creds.html | 22:19 |
nsmeds | ok gotcha =) saved lots of links you've shared, much appreciated. | 22:24 |
nsmeds | somewhat related: what a user creates a domain, they gain "owner" role for anything inside that domain yes? | 22:25 |
nsmeds | i'd like to get some short-term solution working for them right now, which I understand means I'm editing the policy file - and picturing this https://pastebin.com/QtdzYBFS | 22:28 |
nsmeds | does that make sense? and I edit the policies for any actions they require | 22:29 |
*** Dinesh_Bhor has joined #openstack-keystone | 22:30 | |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Pass context objects to policy enforcement https://review.openstack.org/605539 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update idp policies for system reader https://review.openstack.org/619371 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update idp policies for system member https://review.openstack.org/619372 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Update idp policies for system admin https://review.openstack.org/619373 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for domain users interacting with idps https://review.openstack.org/619374 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Add tests for project users interacting with idps https://review.openstack.org/619375 | 22:30 |
openstackgerrit | Lance Bragstad proposed openstack/keystone master: Remove idp policies from policy.v3cloudsample.json https://review.openstack.org/619376 | 22:30 |
lbragstad | nsmeds yeah - that's kinda the girst | 22:31 |
lbragstad | gist* | 22:31 |
lbragstad | you'd have to create some special role | 22:31 |
lbragstad | and then explicitly override any policies (in policy files) that the user would expect to use | 22:31 |
nsmeds | yep - understood. thanks a bunch =) | 22:34 |
lbragstad | no problem | 22:34 |
*** imacdonn has quit IRC | 22:42 | |
*** imacdonn has joined #openstack-keystone | 22:42 | |
*** prashkre has joined #openstack-keystone | 22:56 | |
*** Dinesh_Bhor has quit IRC | 23:20 | |
openstackgerrit | John Dennis proposed openstack/oslo.policy master: Fully log RBAC enforcement data https://review.openstack.org/619260 | 23:22 |
*** rm_work has quit IRC | 23:33 | |
*** rm_work has joined #openstack-keystone | 23:33 | |
*** lbragstad is now known as lbragstad_turkey | 23:46 | |
jdennis | gobble gobble | 23:51 |
*** lbragstad_turkey has quit IRC | 23:56 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!