14:01:54 #startmeeting glance 14:01:55 Meeting started Thu Dec 17 14:01:54 2020 UTC and is due to finish in 60 minutes. The chair is abhishekk. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:56 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:58 The meeting name has been set to 'glance' 14:02:05 #topic roll call 14:02:08 o/ 14:02:11 #link https://etherpad.openstack.org/p/glance-team-meeting-agenda 14:02:13 o/ 14:02:31 o/ 14:02:32 nothing much on agenda, same as last week 14:02:38 lets start 14:02:55 #topic release/periodic jobs update 14:03:03 Milestone 1 yet to tag, gate is unblocked 14:03:13 https://review.opendev.org/763920 - Should we wait for this? 14:03:19 looking 14:03:22 This is cyril's patch 14:03:47 its in good shape 14:04:15 i think we should get it into M-1 release 14:04:19 we still need to figure out what to do with the default vaue so we can detect if it's set or not 14:04:47 Yes, I added comment about that 14:05:15 so IMHO M-1 will be earliest next week then, which brings the question are we tagging M-1 or should we just skip it like last cycle and go directly for M-2? Do we actually have a reason to tag it? 14:05:41 we did tag M1 last cycle 14:05:53 but skipped it in ussuri 14:06:00 ah, time flies 14:06:26 IMO as M2 is just 4 weeks away I think we should club both and release directly at M2 14:06:45 that works for me 14:06:50 cool 14:07:12 sounds good to me too 14:07:15 So we are skipping M1 14:07:46 Periodic job was green till this morning, seeing 1 failure today 14:07:54 holidays are coming so M1&M2 would both be tiny changes anyays compared 14:08:11 right 14:08:25 Reason is lock is not busted in time on copy image task 14:09:00 Will go through logs in detail after the meeting to see why lock is not released 14:09:22 that's it from this topic 14:09:24 hmm-m, weird 14:09:35 yeah 14:09:52 Hopefully will find the cause in logs 14:09:57 ++ 14:10:01 Moving ahead 14:10:09 #topic Milestone 2 priorities 14:10:21 Distrbuted Image Import via request proxying - https://review.opendev.org/c/openstack/glance-specs/+/763574 14:10:44 needs new revision, so that we can start working on PoC 14:10:52 Secure RBAC - https://review.opendev.org/q/project:openstack/glance+topic:secure-rbac 14:11:24 I have started manual testing with secure rbac changes, recording results in https://etherpad.opendev.org/p/glance-rbac-testing 14:12:24 We need changes in python-glanceclient as well to work with system persona's 14:12:50 I am coordinating with lance for the same 14:13:29 any questions/suggestions? 14:14:11 I just realized that the lc job is broken in client too, will have patch up tonight for it and check the store 14:14:34 should be pretty straight forward now when I know how to tackle it 14:14:40 jokke, ++ 14:14:57 moving ahead, we can discuss this later if required 14:15:10 #topic Important spces for reviews 14:15:43 I have pushed new revision for [Spec lite] Discovery API for task - https://review.opendev.org/c/openstack/glance-specs/+/763740 14:15:57 Just forgot to change the title of commit message :( 14:16:27 rosmaita, jokke kindly go through it as per your convenient time 14:16:39 ack 14:16:56 We also need to sync with luzi for image encryption status 14:17:08 will do 14:17:21 jokke, thank you 14:17:25 i believe luzi is away until 2021 14:17:36 ahh 14:17:50 Any update about barbican changes? 14:17:50 at least the popup team meetings have been cancelled through the remainder of the year 14:18:02 yes 14:18:09 last i heard, they were still being worked on and had not merged yet 14:18:22 ^^ 14:18:36 I haven't heard anything beyond that either 14:18:39 rosmaita, jokke ack thank you 14:18:56 ok, that brings us to our next topic 14:19:00 i'll check the meeting logs to be sure, i have not been reliable about attending 14:19:07 #topic Meetings during christmas and new year week? 14:19:23 cancel 24 December for sure! 14:19:39 and 31 December 14:19:46 I will not be around on 31st 14:19:56 yeah, sack 'em 14:20:08 that would make next meeting 7 January, that sounds ok 14:20:18 ok, I will send mail about the same 14:20:40 that's it for today 14:20:42 I'm officially off from Tue onwards, now reality might be different based on the fact that my XMas plans got cancelled and I'll be stuck at home 14:20:43 i just dropped a -1 on cyril's patch 14:20:55 see if my comment makes sense 14:21:07 jokke: sorry to hear about your plans going awry 14:21:14 "Christmas" and "plans" are not a thing these days 14:21:23 agree 14:21:47 rosmaita: too much of the family in the risk group so just not gonna risk the travels through 2 airports :( 14:22:01 i applaud your doing the sensible thing 14:22:09 +1 14:22:15 at least you have a good coffee grinder at home! 14:22:31 he has many things at home 14:22:47 And I just ordered a kilo more of beans to have stocks to carry me over new year :P 14:22:47 really nice sauna, for one! 14:23:54 moving to Open discussion 14:23:55 #topic Open discussion 14:23:59 yeah, the downside is that I'll be here on my own so I might go cabin crazy and do some work over the period, not promising anything 'though 14:25:11 I will be around as well 14:25:19 rosmaita, comment makes sense 14:25:31 ok, cool 14:25:40 Only thing part from the coming holidays I had was the lc jobs for client and store I hopefully have up for review by tomorrow morn for you abhishekk. I'll do a pass on the specs too 14:25:59 ack 14:26:18 they are likely broken for stable branches so might be that I'll tackle them tomorrow 14:26:21 here's the log from the last encryption pop-up team meeting: http://eavesdrop.openstack.org/meetings/image_encryption/2020/image_encryption.2020-12-07-13.00.log.html 14:26:35 rosmaita, thank you 14:26:49 no progress on barbican secret consumer api 14:27:10 yeah, i think all our stable branches are broken in cinder 14:27:13 for all deliverables 14:27:22 master python-cinderclient is still hosed 14:27:43 what was the outcome of the suggestion on the ML to drop the l-c jobs? 14:28:06 nothing concrete 14:28:50 I think the discussion pretty much died. I for one expressed my disagreement with it. While it's inconvenient now, it really revealed some neglect on our dependency management which is good to get reminder of 14:28:51 bummer 14:29:38 jokke: i would agree, except things were so freaking out of date that i think we need a better mechanism for maintaining requirements 14:29:49 or maybe a better policy for how to do updates 14:30:01 better policy * 14:30:55 rosmaita: yeah, like I mentioned at some point we used to have more eyes on it and centralized distribution, but it was thought that this way is better. At least wit the pip change we get bummed about it more frequently now 14:31:13 yeah, at least this didn't happen at RC-time 14:31:27 yarp 14:31:31 figures it would happen in 2020, though 14:31:37 this year absolutely sucks 14:31:54 just remembered and want to tell you guys in advance, My sisters wedding falls during milestone 2 week and I am going on leave between 15 - 20 January 14:31:58 I think that would have been just simple "Lets cap pip for now and figure it out after release" 14:32:19 but you're right for once we got it on the time when there is no massive fires to put out elsewhere 14:32:20 abhishekk: congratulations to your sister! hope you are able to have a good celebration 14:32:28 so either of you need to take care of release 14:32:38 abhishekk: np 14:32:46 I'll look after it 14:33:02 remind me before you get to celebrations 'though ;P 14:33:28 :D I will definitely enjoy when I know you guys are here 14:34:03 rosmaita, thank you 14:34:29 Anything else or should we wrap up early 14:34:48 i will take the time to look over your discovery spec 14:35:05 nothing else from me 14:35:07 ++ 14:35:16 lets wrap up 14:35:26 thank you all and have a nice week ahead 14:35:31 thanks all and Happy Holidays! 14:35:42 Next meeting will be on 7 January 2021 14:35:43 happy holidays, see you in 2021 14:35:54 happy holidays 14:36:00 #endmeeting