15:00:46 <sjmc7> #startmeeting openstack search
15:00:47 <openstack> Meeting started Thu Apr 20 15:00:46 2017 UTC and is due to finish in 60 minutes.  The chair is sjmc7. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:48 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:51 <openstack> The meeting name has been set to 'openstack_search'
15:01:08 <sjmc7> morning folks
15:01:15 <Kevin_Zheng> ：）
15:02:11 <sjmc7> or evening
15:03:04 <sjmc7> i’ll give it a couple of minutes, see if anyone else drifts in
15:04:31 <TravT> o/
15:05:56 <sjmc7> ok. probably a quick one this week as travis and i have been really busy with other stuff
15:06:11 <sjmc7> #topic nova cells
15:06:55 <sjmc7> i’ve updated the versioned notifications patch https://review.openstack.org/#/c/453352/ so that the tests pass (and i’ve added the bare minimum of unit tests for the versioned code)
15:07:26 <Kevin_Zheng> cool, I have one question
15:07:45 <sjmc7> shoot
15:07:46 <Kevin_Zheng> Are we able to do part-update?
15:07:59 <sjmc7> i haven’t done so in the patch but yes, we can do
15:08:11 <Kevin_Zheng> Seems notification payload will be different
15:08:14 <sjmc7> however… something we’re missing is the updated_at timestamp which makes that harder
15:08:15 <Kevin_Zheng> Cool
15:08:19 <sjmc7> different?
15:08:35 <Kevin_Zheng> I haven't read all the post yet
15:08:52 <sjmc7> i’m missing something..
15:09:14 <Kevin_Zheng> But seems in nova side payload might be different
15:10:01 <sjmc7> from what?
15:10:05 <Kevin_Zheng> For e.g. Payload in instance action may not include tag field
15:10:55 <Kevin_Zheng> I'm not sure yet, I will read the whole post soon
15:11:25 <sjmc7> hmmm. ok, that would be a bit unfortunate but not too big a deal
15:11:34 <sjmc7> i’ll file a bug for missing updated_at
15:11:41 <Kevin_Zheng> Yeah
15:13:57 <sjmc7> the only other big bit of work for the cells stuff is the soft-delete action
15:14:11 <sjmc7> i don’t know if nova ever sends a hard delete notification
15:14:21 <sjmc7> i.e. whether instances sit around forever
15:15:04 <Kevin_Zheng> by soft delete
15:15:30 <Kevin_Zheng> you mean the soft-delete action
15:15:33 <sjmc7> i.e. setting a flag rather than actually deleting the instance
15:15:44 <sjmc7> err.. maybe. how many deletes are there? :)
15:15:54 <Kevin_Zheng> Or deleted but still in do
15:16:00 <sjmc7> yeah, that one
15:16:20 <Kevin_Zheng> This one doesn't have flag
15:16:42 <Kevin_Zheng> Soft-delete action has it's own notification
15:16:54 <sjmc7> ok. yeah, i meant the actual delete but leaving htings in the DB
15:17:53 <Kevin_Zheng> The record can be deleted by a cmd
15:18:02 <Kevin_Zheng> Direct db
15:18:21 <Kevin_Zheng> Action so no info sent
15:18:24 <sjmc7> right, but normally it’d sit around in the database forever?
15:18:36 <Kevin_Zheng> Yeah
15:20:14 <sjmc7> ok. so it’s just a case of setting that flag rather than deleting, and by defaulting filtering those out
15:20:36 <sjmc7> hopefully will be able to get to that pretty soon; it’s a bit fiddly but hopefully not too time consuming
15:21:39 <Kevin_Zheng> We could also write a coomand to delete them?
15:22:42 <sjmc7> yeah, maybe a management command i guess
15:23:03 <sjmc7> unless you create and delete a billion VMs it’s not going to matter too much to elasticsearch
15:24:35 <Kevin_Zheng> Yeah
15:24:44 <sjmc7> ok, that was all i had to discuss
15:24:47 <sjmc7> #topic open discussion
15:25:00 <sjmc7> since you’re the only person here Kevin_Zheng , anything else you had? :)
15:25:24 <Kevin_Zheng> No
15:25:39 <sjmc7> ok. well, i’ll end it here i guess. thanks all!
15:26:05 <Kevin_Zheng> Thanks
15:35:07 <sjmc7> #endmeeting