17:00:13 <gibi> #startmeeting nova notification
17:00:18 <openstack> Meeting started Tue Jan 17 17:00:13 2017 UTC and is due to finish in 60 minutes.  The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:22 <gibi> o/
17:00:22 <openstack> The meeting name has been set to 'nova_notification'
17:01:05 <aunnam> o/
17:01:56 <sneti_> o/
17:01:57 <gibi> let's wait a minute or two
17:05:07 <gibi> OK
17:05:08 <gibi> let's start
17:05:25 <gibi> good news, the flavor notification bp has been completed
17:05:32 <gibi> congrats
17:05:44 <gibi> there is one follow up patch to make the test nicer
17:05:54 <gibi> https://review.openstack.org/#/c/419086/
17:06:01 <gibi> but that is not a must have for Ocata
17:06:13 <gibi> also there is a related bugfix on the lazy loaded fields
17:06:28 <gibi> https://review.openstack.org/#/c/415857/
17:06:39 <gibi> unfortunately this quite big and complex change
17:06:48 <gibi> and hard to split into smaller pieces
17:07:01 <gibi> however this would be really nice to merge in Ocata
17:07:12 <gibi> to have tha flavor.project in the flavor notifications
17:07:34 <gibi> so please try to review the patch https://review.openstack.org/#/c/415857/
17:08:03 <gibi> any question, comment on the flavor notification bp?
17:09:13 <gibi> OK lets move forward
17:10:00 <gibi> the first patch of the searchlight related notifiaction bp is also merged
17:10:32 <gibi> but there is couple of patches still needs review
17:10:45 <gibi> also the instance.tags patch is affected by the lazy-load bug
17:11:28 <gibi> any questions, comments on the bp/additional-notification-fields-for-searchlight ?
17:12:00 <sneti_> gibi, we thought of working on lazy load for tags after instance.tags gets merged
17:12:12 <gibi> sneti_: that is fine I think
17:12:31 <sneti_> gibi, ya..so its not affected by lazy load
17:12:44 <gibi> sneti_: until bugfix https://review.openstack.org/#/c/415857/ is merged the lazy-loaded tags field will not appear in the payload
17:13:01 <gibi> sneti_: so it would be good to merge at least the bugfix before the instance.tag patch
17:13:32 <gibi> sneti_: but I'm fine with not making tags loaded by default for now
17:13:40 <aunnam> gibi, when you find time can you review this patch https://review.openstack.org/#/c/419730/ on keypairs
17:14:07 <gibi> aunnam: sure, I will check it tomorrow
17:14:16 <aunnam> so for searchlight bp we have patches up for part1
17:14:29 <aunnam> gibi, thanks
17:14:54 <gibi> aunnam: yes, I think part1 is now up for review and part2 most probably will slip to Pike
17:15:39 <gibi> any other thing related to the searchlight bp?
17:16:30 <gibi> OK let's move forward
17:16:32 <sneti_> gibi,  https://review.openstack.org/#/c/415857 will not have any effect on tags patch right
17:17:31 <gibi> sneti_: Let me check
17:18:23 <gibi> sneti_: right, it is not effected by the lazy-load bug (I was wrong about that before)
17:18:39 <gibi> sneti_: because you are not using the schema to copy the tags from the instance
17:18:46 <sneti_> gibi, right
17:18:54 <gibi> sneti_: but you have this from_instance method that will trigger the lazy-load
17:19:01 <gibi> sneti_: cool! :
17:19:02 <gibi> :)
17:19:26 <sneti_> gibi, right.. we can avoid only with changing tags to default
17:19:47 <gibi> sneti_: yes, but that is OK to be done later
17:20:02 <gibi> sneti_: hence my +1 on the instance.tags patch :)
17:20:15 <sneti_> gibi, ok :)
17:20:21 <gibi> good
17:21:56 <gibi> moving forward to bp/json-schema-for-versioned-notifications
17:22:27 <gibi> unfortunately that needs to be moved to Pike due to the outstanding oslo dependencies
17:22:51 <gibi> anyhow I think there was good progress on the oslo side so now in Pike we will have change to finish the feature in nova as well
17:23:27 <gibi> and last but not least the notification transformation bp
17:24:04 <gibi> I think it has lower prio in the cores' heads during this last rush before the freeze
17:24:15 <gibi> we haven't merged patches there for weeks
17:24:28 <gibi> but we have a lot that is ready for core review
17:24:33 <gibi> I will try to ask the cores again
17:24:50 <gibi> but it is clear that lot of the work will spill over to Pike
17:25:47 <gibi> I proposed couple of cleanup and refactor patch in the transformation series
17:25:59 <gibi> https://review.openstack.org/#/c/418488/
17:26:05 <gibi> https://review.openstack.org/#/c/418489
17:26:24 <gibi> these would help with the future transformations
17:26:38 <gibi> so any feedback / review on these patches are very welcome :)
17:26:49 <gibi> Anything else to discuss?
17:27:18 <aunnam> gibi,saw this reported today https://bugs.launchpad.net/nova/+bug/1657087
17:27:18 <openstack> Launchpad bug 1657087 in OpenStack Compute (nova) "Functional versioned notification test test_create_delete_server_with_instance_update randomly fails" [Medium,Confirmed]
17:27:32 <aunnam> any idea on this bug?
17:28:11 <gibi> It seems Matt already proposed a fix for it
17:28:14 <gibi> https://review.openstack.org/#/c/421221/
17:28:20 <gibi> haven't looked it yet
17:28:38 <gibi> ahh it is not fix it is some extra log to see what happens
17:28:55 <aunnam> ya just the log
17:29:16 <aunnam> somehow some notifications are missing
17:29:48 <gibi> I think we shall try to reproduce the failure with the log included to know more
17:30:00 <gibi> at the moment I cannot guess
17:30:35 <aunnam> will try to reproduce and see what's happening there
17:31:08 <gibi> aunnam: thanks, if you can reproduce then please post the logs to the bug report so any of us can look at it
17:32:16 <aunnam> gibi, sure. only problem is it randomly fails, so hard to reproduce i guess
17:33:06 <gibi> aunnam: yes, so we have to try a lot :)
17:33:35 <gibi> aunnam: or we could merge matt's patch and let the gate try a lot
17:33:47 <gibi> aunnam: I will ask matt about it
17:34:02 <aunnam> gibi: ok
17:35:44 <gibi> anything else for today?
17:36:00 <aunnam> nothing from my end
17:36:24 <sneti_> no
17:36:29 <gibi> OK
17:36:38 <gibi> thanks for the fruitefull discussion
17:36:50 <gibi> see you around!
17:36:57 <gibi> #endmeeting