18:00:08 <lbragstad> #startmeeting keystone 18:00:09 <openstack> Meeting started Tue Aug 1 18:00:08 2017 UTC and is due to finish in 60 minutes. The chair is lbragstad. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:10 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:12 <openstack> The meeting name has been set to 'keystone' 18:00:16 <lbragstad> ping ayoung, breton, cmurphy, dstanek, edmondsw, gagehugo, henrynash, hrybacki, knikolla, lamt, lbragstad, lwanderley, notmorgan, rderose, rodrigods, samueldmq, spilla, aselius, dpar 18:00:20 <cmurphy> ohai 18:00:21 <edmondsw> o/ 18:00:22 <knikolla> o/ 18:00:22 <rodrigods> hey 18:00:22 <gagehugo> o/ 18:00:25 <lbragstad> #link https://etherpad.openstack.org/p/keystone-weekly-meeting 18:00:29 <lbragstad> agenda ^ 18:00:30 <spilla> o/ 18:00:30 * cmurphy actually here today 18:00:41 <lamt> o/ 18:00:49 <lbragstad> wow - y'all *almost* broke down the door 18:01:10 <rmascena> o/ 18:01:43 <lbragstad> let's give it a minute or two for others to show up 18:01:45 <knikolla> the light agenda maybe :P 18:01:57 <lbragstad> knikolla: possibly! 18:02:21 <lbragstad> i won't lie - i don't mind the meeting being done in 30 minutes or less 18:03:36 <lbragstad> #topic announcements: release status 18:03:39 <rarora> o/ 18:03:52 <lbragstad> last week we released pike-3 on schedule 18:04:10 <lbragstad> #link https://review.openstack.org/#/c/488411/ 18:04:19 <lbragstad> we also released the final version of python-keystoneclient for pike 18:04:24 <lbragstad> #link https://review.openstack.org/#/c/488413/ 18:04:40 <lbragstad> the only this is that pike-3 went out the door without project tags 18:05:17 <lbragstad> so - that's going to have to be one of the first things on our plate for queens 18:05:39 <lbragstad> thanks to rarora gagehugo lamt spilla and others for putting in all the work there 18:06:12 <lbragstad> that should be a pretty easy win to start queens off with 18:06:20 <lbragstad> #topic announcements: specs 18:06:24 <gagehugo> yeah it's pretty close 18:06:47 <lbragstad> but - we also need to clean up the specs repository to accurately reflect the things we didn't make into pike 18:06:52 <lbragstad> #link https://review.openstack.org/#/c/487534/ 18:06:57 <lbragstad> #link https://review.openstack.org/#/c/487533/ 18:07:36 <lbragstad> i've proposed bumping application creds and federated attributes to backlog until we're ready to start parsing what we want to go through for queens 18:07:49 <lbragstad> project tags will also needs a patch to bump it 18:07:52 <lbragstad> to backlog 18:07:53 <gagehugo> I can edit https://review.openstack.org/#/c/484529/ for that 18:08:08 <gagehugo> move it to backlog with those edits 18:08:18 <lbragstad> gagehugo: cool - thanks 18:08:36 <edmondsw> might want to move it to backlog separately and then get agreement on the edits 18:08:45 <lbragstad> #action gagehugo to propose https://review.openstack.org/#/c/484529/ to backlog until queens opens for development 18:09:54 <lbragstad> we have a lot of work that spilled over from pike this release - i think a big part of the PTG is going to be keeping discussions fresh to that we can carry things over gracefully 18:10:19 <knikolla> yeah 18:10:23 <gagehugo> edmondsw will do 18:10:31 <lbragstad> i'll get into that a bit more later 18:10:40 <lbragstad> now that we are in feature freeze 18:10:43 <lbragstad> #topic bugs 18:10:57 <lbragstad> we should be ultra focused on bugs, bug review, and triage 18:11:04 <lbragstad> #link https://goo.gl/F7Pc8x 18:11:14 <lbragstad> ^ that's a list of all bugs that are targetting pike-rc 18:11:25 <lbragstad> so - ideally, we need to have those fixed before we can get pike out the door 18:11:45 <lbragstad> if there is a bug in that list that you don't feel needs to be included in pike, let me know 18:12:21 <lbragstad> on the contrary, if there *is* a release blocking bug that pops up - let me know 18:12:41 <lbragstad> i'll be attending to that list daily for the rest of the release - reviewing and picking up whatever changes i can 18:13:14 <lbragstad> any questions on the current bug list? 18:14:18 <lbragstad> the good thing is that most of those bugs are in progress, so reviews should really help trim that list down 18:14:21 <edmondsw> microversions made the rc bug cut? 18:14:38 <lbragstad> edmondsw: this one? 18:14:41 <lbragstad> #link https://bugs.launchpad.net/keystone/+bug/1689644 18:14:41 <openstack> Launchpad bug 1689644 in OpenStack Identity (keystone) "Keystone does not report microversion headers" [Medium,In progress] - Assigned to Rohan Arora (ra271w) 18:14:56 <lbragstad> edmondsw: or microversion support in general? 18:15:01 <edmondsw> yeah... how quickly things can change. It was nearly unanimous at the PTG that we wouldn't do microversions :) 18:15:16 <knikolla> well… that's not really doing microversions :P 18:15:29 <lbragstad> knikolla: right - it's just communicating it to the caller 18:15:41 <lbragstad> real support for microversions still needs to be done 18:15:43 <edmondsw> how can you report microversions if you don't have microversions? 18:16:00 <lbragstad> edmondsw: the current approach returns 3.0 and 3.8 18:16:11 <lbragstad> as the min and max api versions 18:16:21 <lbragstad> for v3 18:16:24 <lbragstad> and 2.0 for v2.0 18:16:32 <lbragstad> but the approach is still in review 18:17:13 <edmondsw> sounds fun 18:18:20 <lbragstad> moving on 18:18:25 <lbragstad> #topic ptg planning 18:18:43 <lbragstad> i have an etherpad up to capture ideas we want to discuss at the PTG 18:18:45 <lbragstad> #link https://etherpad.openstack.org/p/keystone-queens-ptg 18:19:15 <lbragstad> and we're getting to the point where I need to start grouping things in bucket and organizing meeting times with other groups if we want to discuss cross-project things 18:19:29 <lbragstad> i assume the PTG is going to break down much like it did in Atlanta 18:19:42 <lbragstad> where the first couple days are going to be dedicated to cross-project initiatives 18:19:56 <lbragstad> and the last three are going to be project-focused 18:20:18 <gagehugo> will there be a policy meeting for cross-project? 18:20:30 <lbragstad> gagehugo: i fully expect to have one - yes 18:20:34 <gagehugo> cool 18:20:45 <lbragstad> it's one of the first topics i added for the cross-project section in the planning etherpad 18:20:54 <cmurphy> there's a schedule https://docs.google.com/spreadsheets/d/1xmOdT6uZ5XqViActr5sBOaz_mEgjKSCY7NEWcAEcT-A/edit#gid=397241312 18:21:00 <lbragstad> along with unified limits, application creds, etc... 18:21:15 <lbragstad> cmurphy: oh - good call 18:21:38 <cmurphy> and yeah looks like cross project stuff will be more monday/tuesday 18:21:55 <lbragstad> cool 18:22:16 <lbragstad> also - i sent out a reminder asking who was going to be attending 18:22:23 <gagehugo> there was short 1 hr meetings too at ATL, for some cross-project stuff 18:22:30 <lbragstad> the foundation would like to have an accurate estimate so that they can line up rooms properly 18:22:31 <gagehugo> idk how that works with the schedule 18:23:20 <lbragstad> if you're planning on attending the PTG please add your name to https://etherpad.openstack.org/p/keystone-queens-ptg if you haven't already done so 18:23:28 <lbragstad> gagehugo: yeah - i wasn't there the first two days in ATL 18:23:38 <edmondsw> lbragstad I will be there, but will probably be sitting in the nova room more than keystone 18:24:11 <lbragstad> edmondsw: ack 18:24:18 <lbragstad> edmondsw: i think we have a few people in that boat 18:24:20 <lbragstad> which is fine 18:24:34 <knikolla> as long as you join us for beer after, we're good 18:24:48 <lbragstad> this is what i wanted to spend the majority of the time on today 18:25:16 <edmondsw> :) 18:25:34 <lbragstad> is there anything on the current planning doc that people don't agree with? 18:25:39 <lbragstad> is there anything you *do* agree with? 18:25:56 <lbragstad> I'd like to get an idea of where people's interests lie and what they are hoping to work on and accomplish in pike 18:26:13 <lbragstad> (as a way to keep expectations in line and sane) 18:26:55 <knikolla> hmm.. is dstanek going to be there for the native saml topic? 18:27:01 <lbragstad> Personally for me - i'm going to be dedicating time to global role assignments (along with championing community goals for policy) 18:27:02 <knikolla> he was championing it 18:27:18 <lbragstad> knikolla: he was - but I doubt he will make it to the PTG 18:27:23 <lbragstad> knikolla: i haven't heard anything yet 18:27:30 <cmurphy> app creds definitely a high priority for pike 18:27:37 <lbragstad> i completely agree 18:27:38 <knikolla> cmurphy: ++ 18:28:04 <knikolla> i also want to see federated attributes. i'll probably pick it up if intel folks don't find the time to work on it. 18:28:06 <lbragstad> then again - unified limits has some pretty awesome payoffs and advantages that would be nice to have 18:28:51 <lbragstad> we also have to take a peak at oslo.policy to see if we can enhance it so that we can emit deprecation warnings when policy changes 18:29:05 * lbragstad is already feeling overwhelmed 18:29:12 <cmurphy> @.@ 18:29:36 <knikolla> that's why we have 3 days at the ptg 18:30:23 <knikolla> lbragstad: how is read-only role a project specific thing? 18:30:49 <edmondsw> lbragstad looking at the room schedule, which of those rooms do you plan to have policy discussions in? 18:31:17 <gagehugo> yeah we should grab a room for that 18:31:41 <lbragstad> knikolla: good catch - moved 18:31:47 <lbragstad> edmondsw: the biggest one we can find? 18:31:53 <edmondsw> lol 18:32:00 <edmondsw> row 18? 18:32:02 <knikolla> lol 18:32:02 <lbragstad> edmondsw: the discussion we had in ATL was packed 18:32:11 <knikolla> i think that'll be detrimental 18:32:12 <lbragstad> or 18:32:36 <lbragstad> we coax people off-site? 18:33:03 <edmondsw> the API WG room was packed in ATL, and covered more than API WG stuff... maybe talk to them about carving out time there? 18:33:04 <lbragstad> but row 18 looks like it will work 18:33:23 <gagehugo> yeah the API-WG room in ATL was interesting 18:33:23 <lbragstad> we also don't need *every* project developer there 18:33:40 <lbragstad> i also assume there are going to be project developers who don't care about policy 18:33:51 <knikolla> contact your elected policy representative today :P 18:33:54 <lbragstad> but I'd figure at least 3 - 5 developers from each interested project? 18:33:58 <edmondsw> I'd kinda like it to be a dedicated room so people know that's going to be the topic and come accordingly 18:34:21 <knikolla> yeah, it being the nova room last time wasn't good. 18:34:24 <edmondsw> lbragstad that's some serious optimism! would be great 18:34:51 <edmondsw> having policy changes as a goal for queens should help attendance 18:34:58 <knikolla> honestly, i think teams should have a policy liason going forward 18:34:59 <edmondsw> and we can talk about more than policy-in-code 18:35:00 <lbragstad> yeah 18:35:10 <edmondsw> knikolla ++ 18:35:27 <lbragstad> right - we really need to talk about breaking the scope check and putting it into code 18:35:48 * edmondsw nods vigorously 18:35:49 <lbragstad> it'd be nice to make some progress on a standard set of roles we should provide by default 18:36:38 <edmondsw> lbragstad that's a short list in my opinion... admin, member, observer 18:37:21 <lbragstad> yeah - but we can couple that with fixing admin-ness and isolating scope to in-code only and we're in a much better place 18:37:24 <edmondsw> maybe add a few internal roles, e.g. service, that we wouldn't expect end-users to ever have 18:37:36 <edmondsw> absolutely 18:38:06 <lbragstad> i'm really looking forward to doing what's needed to have those discussions in Denver 18:38:31 <lbragstad> but i'm also thinking it's going to take up most of my bandwidth for queens 18:38:58 <knikolla> seems fair. given that it's the single most important thing for now. 18:39:15 <lbragstad> yeah - in addition to application creds 18:39:24 <lbragstad> which i think have massive usability benefits 18:39:35 <knikolla> yeah, but requires less coordination with other teams 18:39:55 <lbragstad> yeah - we need to come to consensus on how to use them 18:40:32 <knikolla> besides the lifetime of an app cred. i thought that pretty much there was consensus? 18:40:52 <lbragstad> knikolla: yeah- i think the lifecycle is really the only thing left to discuss 18:41:05 <lbragstad> (there is some additional details about roping it into policy in the future, too) 18:41:12 <lbragstad> s/is/are/ 18:42:14 <lbragstad> does anyone have things they'd like to get on the map for the PTG? 18:42:36 * lbragstad feels like he's been doing a lot of talking 18:42:59 <knikolla> i think the list is pretty comprehensive. 18:43:23 <lbragstad> ++ some topics are going to be pretty dense 18:43:35 <knikolla> i'd like to get another shot at revising 'role check in middleware' if folks want 18:43:53 <lbragstad> knikolla: hrybacki was talking to me about that a few weeks ago 18:44:00 <lbragstad> knikolla: you should be sure to check in with him 18:44:23 <knikolla> sounds good. 18:44:42 <lbragstad> #topic open discussion 18:44:44 <lbragstad> floor is open 18:46:48 <lbragstad> looks like we can get some time back 18:46:55 <lbragstad> office hours starts in about 14 minutes 18:46:58 <lbragstad> thanks for coming! 18:47:00 <lbragstad> #endmeeting