13:59:40 <pdeore> #startmeeting glance
13:59:40 <opendevmeet> Meeting started Thu Aug 31 13:59:40 2023 UTC and is due to finish in 60 minutes.  The chair is pdeore. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:59:40 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:59:40 <opendevmeet> The meeting name has been set to 'glance'
13:59:40 <pdeore> #topic roll call
13:59:40 <pdeore> #link https://etherpad.openstack.org/p/glance-team-meeting-agenda
13:59:50 <pdeore> o/
14:00:52 <abhishekk> o/
14:00:56 <dansmith> o/
14:01:03 <mrjoshi> o/
14:01:12 <pdeore> let's start
14:01:23 <pdeore> #topic Release/periodic jobs updates
14:02:03 <pdeore> We are skipping to tag m3 and will directly tag rc1, since there are some patches of stephen which still needs to merged
14:02:37 <pdeore> what do you think on this ?
14:02:54 <abhishekk> sounds good
14:03:03 <pdeore> cool
14:03:14 <pdeore> glance_store 4.6.1 released
14:03:29 <pdeore> and glanceclient we are not releasing since there is nothing merged after last release
14:03:40 <abhishekk> ack
14:03:56 <pdeore> Periodic jobs all green except fips jobs failure, need to look into the exact error
14:04:23 <pdeore> moving to next
14:04:31 <pdeore> #topic Raise FFE for new Locations APIs ?
14:04:52 <pdeore> I want to know team's opinion on raising this FFE for this
14:05:18 <dansmith> tbh, I'm not sure it's a good idea
14:05:22 <abhishekk> you need two votes for FFEs
14:05:36 <pdeore> I think things are pretty much in good shape and if we get some extra time we can verify all the corner cases and if it gets in by rc then we can backport it to stable/bobcat
14:05:44 <dansmith> I know I've not been helping with the last flurry of reviews, because I'm swamped with other stuff
14:06:12 <pdeore> abhishekk, yeah that's why thought to confirm it here :)
14:06:14 <dansmith> so I apologize for that, but I'm trying to follow the review comments and I'm just a bit worried about rushing that in with the kind of stuff that has been changing
14:06:33 <dansmith> and I've not had a chance to dig deeply into the nova part to convince myself that everything is fine
14:06:43 <abhishekk> Even though I am reviewing it carefully, I definitely need second eye on it
14:06:54 <dansmith> it also seems like the nova patches are still not passing reliably or at all? (/me checks)
14:07:07 <pdeore> well, on nova patch nova-ceph-multistore job is passing now
14:07:20 <dansmith> okay, what did you change to fix it?
14:07:41 <abhishekk> I have pointed the error on glance patch
14:07:45 <abhishekk> why it was failing
14:08:16 <dansmith> the nova patch still has this same error: https://81b3bd201171c89feee7-a605dbf95134d478b50bec2dfa092555.ssl.cf2.rackcdn.com/891036/4/check/nova-ceph-multistore/14600f0/testr_results.html
14:08:23 <pdeore> well, earlier i was trying trying set the image back to queued from active in one of the revert call
14:08:26 <dansmith> that's on the latest recheck
14:09:05 <abhishekk> comment on PS 20
14:09:41 <pdeore> yeah on the WIP patch it's failing, and i think that's too bcz of the same reason, the revert of SetImageTOActive state,
14:09:48 <pdeore> I will fix that
14:10:07 <dansmith> but the WIP patch is using just new glanceclient with old glance
14:11:03 <pdeore> this is what i noticed in the logs , I'm still not sure about this dependencies :(
14:11:44 <abhishekk> I can see location import executed in above logs
14:11:49 <dansmith> I don't understand the concern over the dependencies.. this is the only order in which it can go
14:12:25 <abhishekk> but if still this is failing I think we can not rush over it
14:13:45 <pdeore> client patch is dependant on glance changes, so I think the new location part will execute, that what i understand
14:14:21 <abhishekk> And next time when you fix it, try running this tempest test in local environment
14:14:31 <pdeore> abhishekk, ok if everyone suggests so then let's move it to next cycle
14:14:52 <dansmith> pdeore: oh, that's a problem indeed
14:14:56 <pdeore> I 've already run it this time, and it didn't failed
14:15:16 <dansmith> pdeore: we need to test nova with new glanceclient and old glance separately from new glanceclient and new glance
14:15:17 <abhishekk> this should not happen I think
14:15:18 <dansmith> so I'm even more concerned
14:15:34 <dansmith> can you drop the glance dep on the glanceclient patch and do a DNM on top for testing in the similar way?
14:15:50 <pdeore> dansmith, sure I will do that
14:15:55 <dansmith> the thing that concerns me though is the DNM nova patch was failing consistently, WIP had passed, then you made a change and now it's flipped
14:16:12 <dansmith> so there's something we're not understanding about which version of which components are being tested
14:16:14 <pdeore> yeah
14:16:34 <dansmith> and this is what I mean about me not having had a chance to deeply inspect the stack
14:16:52 <dansmith> I worry that it's passing locally because versions are all lined up, but that it'll fail if they're not
14:17:41 <pdeore> yeah i got it, need to check everything thoroughly
14:19:02 <pdeore> let's move to next topic then
14:19:04 <abhishekk> This will also give us chance to refactor the tests which are using lots of same code
14:19:39 <pdeore> yeah and some test scenarios which you have highlighted
14:19:51 <abhishekk> right
14:20:47 <pdeore> shall we move ahead ?
14:20:53 <abhishekk> yes
14:21:01 <pdeore> #topic Review priority
14:21:10 <pdeore> #link https://review.opendev.org/q/topic:sqlalchemy-20+project:openstack/glance - SQLAlchemy 2.0 migration patches
14:21:31 <abhishekk> I have reviewed most of the patches, but some patches needs another vote
14:21:35 <pdeore> as I mentioned at the start, there are still some patches to get in, which has 1 +2 from abhishekk so need one more +2, so kindly please have a look
14:21:46 <pdeore> yeah
14:22:04 <abhishekk> Also need to keep eye on failures and add rechecks with reason
14:22:31 <pdeore> yes
14:22:44 <pdeore> I think croelandt is gonna be back in next week , right ?
14:23:08 <abhishekk> I don't know the exact date
14:23:16 <abhishekk> but next week he will be ther
14:23:27 <abhishekk> may be on 4th
14:23:35 <pdeore> hmm, ok
14:24:01 <pdeore> so that's it from me for today
14:24:12 <pdeore> moving to open discussions
14:24:15 <pdeore> #topic Open Discussions
14:24:43 <pdeore> anyone has anything else to discuss ?
14:25:07 <abhishekk> I have nothing, I am creating a document to list common CI failures, which we can discuss during PTG
14:25:27 <dansmith> common glance-specific ones?
14:25:31 <abhishekk> yeah
14:25:50 <abhishekk> Will share it with team once have enough cases
14:25:50 <dansmith> cool
14:26:09 <pdeore> abhishekk, yeah sure, I will add it in PTG etherpad once created
14:26:15 <abhishekk> thanks
14:27:04 <pdeore> cool, then let's conclude for the day !
14:27:15 <pdeore> Thanks everyone for joining !
14:27:17 <abhishekk> thanks
14:27:35 <pdeore> #endmeeting