14:01:14 #startmeeting glance 14:01:14 Meeting started Thu May 15 14:01:14 2025 UTC and is due to finish in 60 minutes. The chair is croelandt. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:14 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:14 The meeting name has been set to 'glance' 14:01:23 #topic roll call 14:01:25 o/ 14:01:27 #link https://etherpad.openstack.org/p/glance-team-meeting-agenda 14:01:39 It's probably juste going to be the two of us :) 14:02:17 well, the *3* of us  \o/ 14:02:22 #topic Release/periodic job updates 14:02:35 I only see a failure for glance-multistore-cinder-import-fips 14:02:54 that is related to dependency between taskflow and one other library 14:03:09 ok I see the Python version for Nova is not the right one 14:03:19 so it's out of our hands? 14:03:37 I think so, we can wait and keep watch on it 14:03:50 yes 14:03:57 #topic Important stable patches 14:04:05 #link https://review.opendev.org/c/openstack/glance/+/948810 14:04:07 #link https://review.opendev.org/c/openstack/glance/+/948812 14:04:22 So these are backports of Dan's fix for a bug that was holding the gate in master 14:04:28 the bug is also present in stable/2024.* 14:04:47 I think we should backport the fix so we can easily merge backports in stable/2024.* (that is, without hitting "recheck" 12 times) 14:04:51 any thoughts? 14:04:55 sounds good 14:05:01 I will ninja approve it one by one 14:05:06 ok, thanks a lot! 14:05:19 np! 14:05:20 #topic glance-gaps patches 14:05:26 o/ 14:05:38 I will take over the SDK/CLI patches next week since mrjoshi is going to leave us :-( 14:05:53 mrjoshi: anything we need to know regarding the work that is left to be done? Anything important you want to highlight? 14:06:56 croelandt: all the best ;) 14:07:22 Thank you so croelandt , I have addressed all the comments and the patches are up, they are failing on zuul jobs, intermittent failure. 14:07:46 are those related to glance or your patches? 14:09:09 No it's not related to my patch 14:09:27 it's been occuring in other sdk patches aswell 14:09:37 ok so that's fine :) 14:10:28 cool 14:10:29 Well thanks a lot for all your work on the SDK/CLI, we made huge progress thanks to you 14:10:46 ++, thank you for all your efforts mrjoshi!! 14:11:07 #topic Issues with secure hash 14:11:16 #link https://review.opendev.org/c/openstack/glance/+/949672 14:11:22 #link https://review.opendev.org/c/openstack/glance/+/949668 14:11:30 so I'm a bit out of the loop here 14:11:39 First of all, I think the doc patch should be merged with the code patch 14:11:48 so e don't risk having inaccurate docs 14:12:09 and then, I think we want to decide whether it's ok to rely on the "old" behaviour like some Tempest do 14:12:12 am I right? 14:12:13 First of all, is it a good idea to flip the config value? rosmaita wdyt? 14:13:11 This feature is still in early phase, and I am sure no one is still using it (talking about hash calculation in new location API) 14:14:08 :/ 14:14:13 We have enabled it by default, so that all the images/snapshot etc will have hash/checksum calculated 14:14:27 Maybe Tempest could use the right config value instead of us changing the default, right? 14:15:17 that is why I have disabled it in the job, we don't need to calculate the hashes in tests IMO 14:16:07 But, we need to disable it for all the jobs which are using ceph as backend 14:16:40 Right now I have disabled it only for glance_store job, but we need to do it for nova-ceph-multistore job as well 14:16:55 so we'd drop Rajat's patches? 14:17:12 We definitely need some input whether it is really good to flip the config value 14:18:02 may be dansmith, rosmaita can comment here about the same 14:18:05 yeah I think it would solve our testing issues, but it might be worse for real users 14:18:35 ok so I think we need to get Dan, Brian, Rajat and the two of us in a 15 minute IRC meeting at some point 14:18:43 and figure this out 14:18:46 ALSO you are the boss now, so whatever you say!!! 14:18:58 I haven't looked at this so I need time to digest 14:19:01 If I had it my way, we'd remove all features 14:19:13 bang!! 14:19:21 I don't think it's urgent, so maybe we can try and find a time slot next week maybe? 14:19:28 dansmith: ack 14:19:34 unless this gets resolved in Gerrit comments by then :) 14:20:58 ok moving on, then 14:21:08 #topic One easy patch per core dev 14:21:16 #link https://review.opendev.org/q/topic:%22bug/2110185%22 14:21:35 ok so I guess we'd want to have all these patches merged in the same release, so the behaviour is consistent across all drivers, right? 14:22:32 yes, we need it for our main feature, pass image size in advance to glance 14:23:01 ok so I'll start reviewing those but I'll only push +W once they all seem good 14:23:09 ideally we would have them all in M2 14:23:37 sounds good, we also need a release for glance store after that, so it will be good if we get it early as well 14:23:44 o/ 14:24:22 Luzi: hi 14:24:47 yes 14:24:53 croelandt: I have approved your patch related to unused method 14:24:54 by the way we have a release for glacne_store M1 14:25:05 is there anything else M1-related I need to do as PTL? 14:25:20 cool! I think that is enough 14:25:38 nice 14:25:46 #topic Open Discussion 14:25:53 Anything else to discuss? 14:25:58 hi 14:26:17 we do have some time to work on the image encryption again starting next week 14:26:28 great! 14:27:14 we are here if you guys need anything, just ping us! 14:27:29 yes don't hesitate to ping on #openstack-glance 14:27:46 were there any other points except for the copy image and glance_direct still missing in the patch? 14:28:23 i think markus answered a few open review questions - it would be good if someone could get over the patch again 14:28:39 I think those are important, I will again have a look tomorrow and comment if something else is missing! 14:28:47 thank you :) 14:28:51 np!! 14:29:22 ok, I think we can wrap up 14:31:10 thank you!! 14:31:15 #endmeeting