18:00:48 <dolphm> #startmeeting keystone 18:00:49 <openstack> Meeting started Tue Jan 14 18:00:48 2014 UTC and is due to finish in 60 minutes. The chair is dolphm. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:50 <gyee_> \o 18:00:51 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:53 <openstack> The meeting name has been set to 'keystone' 18:00:55 <fabiog> hi 18:01:09 <dolphm> #topic Hackathon 18:01:30 <dolphm> the agenda is basically empty, and lots of people are en route to san antonio today for the hackathon tomorrow 18:01:32 <dolphm> so... 18:01:38 <dolphm> #topic Open discussion 18:01:40 <dolphm> :) 18:01:58 <dolphm> i'm mostly looking at blueprint-targeted patchsets today 18:02:19 <nkinder> I'd like to discuss the keystone OSSN I'm working on. 18:02:45 <dolphm> nkinder: the one we discussed on friday? 18:02:58 <nkinder> dolphm: yep. Just want to know if there's anything to change. 18:03:20 <bknudson> nkinder: is it posted somewhere? 18:03:20 <nkinder> dolphm: I still don't mention anything about potential fixes in Havana 18:03:37 <nkinder> bknudson: let me grab the launchpad link... 18:04:02 <fabiog> dolphm, I was wondering if the extensions notification is ready for merging ... 18:04:05 <nkinder> https://bugs.launchpad.net/ossn/+bug/1254619/comments/24 18:04:09 <uvirtbot> Launchpad bug 1254619 in ossn "external.Default authentication plugin only considers leftmost part of the REMOTE_USER split by "@"" [Undecided,In progress] 18:04:23 <fabiog> #link https://review.openstack.org/#/c/57811/ 18:05:37 <dolphm> fabiog: i started reviewing it this morning again, and ended up spending an hour rewriting that EXTENSIONS_HOWTO doc a bit 18:05:52 <dolphm> fabiog: so i still need to get back on track there 18:05:57 <bknudson> nkinder: seems like with the recommended action working around the problem that no fix to havana is required. 18:06:40 <fabiog> dolphm, ok. Thanks! 18:07:08 <dolphm> fabiog: worst case, i'm sure we'll spend some time on it during the hackathon... but my goal is to review it this afternoon 18:07:24 <nkinder> bknudson: the only restriction that havana introduced is that you can't have external users with "@" characters in their name. 18:07:38 <dolphm> nkinder: so, we landed a fix to master... bknudson, i'd be okay backporting that to havana as-is... thoughts? 18:08:20 <bknudson> nkinder: the changes to havana also made it so that we don't ship a plugin that works with both v2 and v3 auth. 18:08:36 <bknudson> dolphm: I don't think the changes are backwards-compatible? 18:09:17 <bknudson> I haven't looked at the review closely... it's on my list of things to do. 18:09:21 <dolphm> bknudson: well, it would provide a backwards compatible solution, but you'd have to opt-in to use the "broken" class after upgrading 18:09:38 <dolphm> bknudson: otherwise, we'd be hot swapping one implementation for another, and yeah... that might break people 18:11:21 <nkinder> dolphm: so are you OK with the OSSN as is, or are changes you'd like to see? 18:13:20 <dolphm> nkinder: i like it as is 18:14:26 <nkinder> dolphm: ok, great. Would you please approve it in the launchpad bug? We're trying to always get PTL signoff for OSSNs now. 18:14:43 <dolphm> nkinder: sure 18:14:47 <nkinder> dolphm: thanks! 18:15:36 <dolphm> nkinder: done 18:25:18 <lbragstad> dolphm: quick question on trusts if we are still in open discussion? 18:25:30 <dolphm> lbragstad: absolutely 18:25:44 <lbragstad> Should the manager class here be implemented to call driver? 18:25:44 <lbragstad> https://github.com/openstack/keystone/blob/master/keystone/trust/core.py#L36 18:25:47 <lbragstad> like... 18:25:55 <lbragstad> https://github.com/openstack/keystone/blob/master/keystone/catalog/core.py#L72 18:26:36 <lbragstad> so that, if we wanted to use notifications on trusts, we could by wrapping the Manager methods with the notifications wrapper? 18:28:30 <dolphm> lbragstad: it does call the driver -- everything is proxied through automatically 18:28:44 <dolphm> lbragstad: if you want to wrap specific calls, you can override them and do so 18:29:17 <jamielennox> hi all, sorry i'm late 18:29:20 <dolphm> jamielennox: o/ 18:29:30 <lbragstad> dolphm: ok, so that would only be methods for trust CUD that would need to be in the trust.Manager 18:29:51 <dolphm> jamielennox: there's only a few of us on, so it's just an hour of open discussion 18:30:15 <jamielennox> yea, i thought it would be quiet today 18:30:17 <dolphm> lbragstad: if that's what you want to emit notifications for, then yes 18:30:23 <lbragstad> ok, sounds good 18:30:44 <dolphm> lbragstad: create_trust, delete_trust? (there's no update) 18:30:50 <lbragstad> right 18:30:53 <dolphm> ++ 18:31:05 <lbragstad> cool, thanks! 18:31:59 <jamielennox> not that it's not nice to chat but is there anything for me? otherwise i might go back to bed 18:34:39 <jamielennox> dolphm: can you have a loot at your -1 on https://review.openstack.org/#/c/62546/ 18:34:55 <jamielennox> you wanted a rebase - but then the -1 got carried over 18:35:18 <dolphm> jamielennox: hrm, i asked for a rebase because i thoguht there was a merge conflict to be resolved 18:35:24 <dolphm> maybe i did a bad checkout 18:35:51 <dolphm> jamielennox: approved 18:36:06 <jamielennox> it's a little old now, but it shouldn't conflict with anything 18:38:13 <jamielennox> dolphm: i also rebased and opend up the the auth_plugins patch: https://review.openstack.org/#/c/60751/ 18:38:49 <jamielennox> if you get a chance try to wrap your head around it because it's going to be difficult to get the compatibility right and figure out how to test the thing 18:40:20 <jamielennox> alright - enjoy the hackathon - wish i could be there 18:40:27 <dolphm> jamielennox: +2'd the underlying patch 18:40:35 <dolphm> jamielennox: /salute go to bed! 18:42:25 <dolphm> bknudson: i posted a diff in a review comment on https://review.openstack.org/#/c/54647/ 18:42:32 <dolphm> bknudson: i think that completes the bp? 18:42:57 <bknudson> dolphm: the problem is it'll never pass tempest 18:43:11 <bknudson> because tempest has tests that verify the old behavior 18:43:11 <dolphm> bknudson: eek, what does tempest assert? 18:43:16 <dolphm> ah. 18:43:34 <bknudson> dolphm: I posted a change to tempest but was -1d https://review.openstack.org/#/c/56106/ 18:43:38 <bknudson> and then I didn't get back to it. 18:43:59 <bknudson> dolphm: maybe it can be accepted (or a version of it) given the blueprint. 18:44:24 <bknudson> I'm easily distracted 18:44:43 <dolphm> lol 18:45:42 <dolphm> bknudson: it sounds like they just want to see the bp on the keystone side? 18:46:18 <dolphm> which would just be no-check-id 18:53:01 <marekd> dolphm: https://review.openstack.org/#/c/60244 - some comments here, mainly about naming conventions. These are minor things, i don't mind changing the names, but if you leave some trace i would fix it tomorrow. 18:53:48 <dolphm> marekd: leave a trace? 18:54:27 <dolphm> marekd: go ahead and clean up the names, and include the changes from this patch (which i'll abandon) https://review.openstack.org/#/c/66638/ 18:55:43 <marekd> dolphm: ok, i will do so. 18:55:44 <marekd> thanks. 18:58:12 <dolphm> just a couple minutes left, so i'm going to #endmeeting before i forget :) 18:58:15 <dolphm> #endmeeting