13:59:14 #startmeeting glance 13:59:14 Meeting started Thu Aug 3 13:59:14 2023 UTC and is due to finish in 60 minutes. The chair is pdeore. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:59:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:59:14 The meeting name has been set to 'glance' 13:59:14 #topic roll call 13:59:14 #link https://etherpad.openstack.org/p/glance-team-meeting-agenda 13:59:22 o/ 13:59:25 o/ 13:59:37 o/ 14:00:42 let's wait for abhishekk & croelandt 14:00:45 o/ 14:01:45 ok let's start 14:01:50 #topic Updates 14:02:01 Oct 2023 vPTG dates are finalized and will be during October 23-27, 2023 14:02:13 so you can register yourself for the virtual PTG 14:02:15 #link http://ptg2023.openinfra.dev/ 14:02:31 I've registered our team for the same.. 14:02:32 oops 14:03:02 it is during my PTO :/ 14:03:15 ohhhu 14:04:14 it's not possible to prepone or postpone your pto? :P 14:04:27 I will add my topics for PTG and let you discuss those :P 14:04:50 no no :P 14:05:13 it depends what topics you add btw :D 14:05:18 no other option 14:05:28 yup :) 14:05:36 * abhishekk will be watching recording afterwards ;) 14:05:44 moving ahead 14:06:00 PTL and TC Election dates are also finalized, during Aug 16 to Aug 30 nominations will be opened 14:06:02 #link https://governance.openstack.org/election/ 14:06:29 moving to next 14:06:32 #topic release/periodic jobs updates 14:06:35 cool, congratulations in advance 14:06:46 :/ :P 14:07:01 M3 is 4 weeks from now 14:07:26 Periodic jobs are all green except failure on fips job with error - "FileNotFoundError: [Errno 2] No such file or directory: 'multipathd'" 14:07:40 need to check more about this ^ 14:07:50 I've seen it elsewhere, 14:08:05 so I suspect it's a job configuration issue and we should check with cinder peopel 14:08:26 +1 I think need to involve whoami-rajat here 14:08:26 on the milestone stuff, 14:08:59 I'm a bit concerned about the location api stuff in terms of not having any cinder or nova patches that show it working for those consumers in various job configurations 14:09:12 do those exist (for cinder I assume) already to show that it works the way we want and expect? 14:09:30 I haven't seen client/sdk patches either 14:10:16 I don't think there are patches from cinder or nova side to test glance work 14:10:35 I can convince whoami-rajat to put a quick PoC for cinder side 14:10:43 ack, I just found the sdk patch, I hadn't seen that before, but it looks like it has -1 for months so I assume not very active 14:11:15 yeah it's not updated as per the revised spec , I will update it asap 14:11:15 abhishekk: I think that'd be good to make sure we don't merge an api that doesn't work the way we want/need 14:11:48 +1, I can assist him if he is busy 14:12:03 nova uses glanceclient afaik, so there will probably be a larger effort needed to transition nova to sdk and then to this 14:12:06 today definitely he is out due to stomachache 14:12:52 on the other hand it also depends on service role for which keystone patch is not merged yet 14:13:01 ack 14:13:02 but we can create dependency chain I guess 14:13:18 so I guess I'm wondering what the likelihood is that all those things will get into place by m3 14:13:18 it has 2 +2s but not approved yet I think 14:13:52 https://review.opendev.org/c/openstack/keystone/+/863420 14:13:59 yup 14:14:18 very less 14:15:15 I'm working on getting that merged 14:15:30 what? keystone? 14:15:35 yeah 14:15:56 that will be good for us if it gets in sooner 14:16:08 d34dh0r53, thanks for the update 14:16:20 np :) 14:17:04 * abhishekk wondering but why deadhorse :D 14:17:15 let's at least try to get the glance & sdk side work in m3 .. 14:17:37 well, I want to see at least one of cinder or nova using it before I'm +2 I think 14:17:41 glance side needs glanceclient work as well 14:17:42 ideally both 14:17:56 if glanceclient is coming that will make the nova PoC easier 14:18:10 yeah 14:18:37 agree with dansmith we need to confirm it is working as we expected with consumers 14:18:46 ++ 14:18:55 I will try to speed up this work 14:19:34 I think to write tempest side tests also needs client and keystone change 14:20:02 that we can postpone if we get PoC working with nova and/or cinder 14:20:29 yeah 14:20:29 obviously need a tempest test before +2 :) 14:20:37 before *I* +2 anyway :) 14:20:43 :D 14:20:48 :( 14:20:48 I rest my case 14:21:46 move to next ? 14:22:00 I think we discussed the next part here 14:22:21 yup 14:22:23 yeah, almost 14:22:28 but yeah, lets officially move to next one 14:22:45 #topic Important Reviews 14:23:12 so, we have sqlalchemy 2.0 migration related patches up for review which is also a high priority for m3, so kindly please have a look 14:23:16 #link https://review.opendev.org/q/topic:sqlalchemy-20+project:openstack/glance 14:23:32 why is it a high priority for m3? 14:23:50 I think its a community goal? 14:24:03 the tc goal for this isn't even on the radar, FYI, because we're not even sure when sa 2.0 will be in a distro 14:24:29 obviously it's good to get done early, but just saying, I don't think there's any actual priority here 14:24:46 ack, sounds good then 14:25:27 so we can miss m3? 14:25:29 ohh that great then, i was assumed so since it's a community goal and other projects have already done it 14:26:04 let's not wait for the last minute anyway, but nice to know it's not taht urgent 14:26:08 we should do it for sure, but I'm not aware of any reason we need to prioritize getting it into m3 14:26:13 definitely need to review those things with care 14:26:20 ack 14:26:23 +1 14:26:32 https://review.opendev.org/c/openstack/governance/+/887083 14:26:34 fwiw ^ 14:26:50 I think priority should be given to CI failures 14:26:54 YES' 14:27:12 I've spent the last month doing nothing but improving gate issues 14:27:32 and progress has been made on general issues, but project-specific things need help 14:27:53 as whoami-rajat noted, there have been some glance image state conflict things (rarely) which is something that people here could help with 14:28:00 and plenty of cinder issues 14:28:19 we've been nearly wedged for a couple weeks until recently with several things that landed 14:28:34 you've probably noticed a lot fewer actual job timeouts lately, for example 14:28:46 if we let that stuff slip, it will just go back to terrible 14:28:51 agree 14:29:08 the condition will get worse during last couple of weeks as well 14:29:34 I will try to have a look at a failure pointed by whoami-rajat 14:29:38 yup 14:30:26 and sorry about bare rechecks 14:30:36 * dansmith glares at abhishekk 14:30:37 :) 14:31:04 I can understand your motive behind it 14:33:58 * abhishekk am i disconnected? 14:34:00 let's move ahead, 14:34:36 my laptop got hanged :/ 14:34:38 #topic Open Discussions 14:35:44 i think we have already disucssed the points dansmith mentioned in the etherpad 14:35:55 yup 14:36:01 anyone has anything else ? 14:36:18 * croelandt has nothing 14:36:57 * abhishekk don't have anything as well 14:37:12 cool, let's conclude for today ! 14:37:25 Thanks everyone for joining !! 14:37:53 #endmeeting