17:00:10 <gibi> #startmeeting nova notification
17:00:11 <openstack> Meeting started Tue Nov  8 17:00:10 2016 UTC and is due to finish in 60 minutes.  The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:15 <openstack> The meeting name has been set to 'nova_notification'
17:00:26 <gibi> hi!
17:00:57 <gibi> who we have today for this nice lovely notification meeting?
17:03:27 <mriedem> o/
17:04:15 <gibi> I think there is not point waiting for others
17:04:20 <gibi> short recap for the last weeks
17:04:34 <gibi> I think there is really nice progress in the notification transformation work
17:04:42 <gibi> I hope we can continue with this speed
17:05:06 <gibi> I keeping the nova priority etherpad up to date
17:05:17 <gibi> I also looked that the searchlight bug reports
17:05:25 <gibi> let's say I triaged them
17:05:36 <gibi> One seems to be already solved by the transformation
17:05:50 <gibi> In the other two I asked some questions
17:06:35 <gibi> syjulian continued working on the json schema work for the notification payloads I'm keeping an eye on those patches as well
17:07:08 <gibi> I think that is the current state of the subteam work I'm aware off
17:07:20 <gibi> mriedem: do you have anything to add or ask?
17:07:32 <mriedem> let me read through that
17:08:03 <mriedem> i don't have anything to add, i didn't yet go back on those searchlight bugs to see the latest
17:08:13 <mriedem> did you see my comments in the one about adding the instance BDMs to the instance notifications?
17:08:31 <gibi> I think the hardest part of those bugs is adding the BDM to the notification payloads
17:08:46 <gibi> we have to figure out what level of information searchlight needs
17:08:56 <mriedem> yeah, i'm not entirely sure if searchlight really needs that
17:08:59 <mriedem> or if they can work around it
17:09:19 <gibi> that would be even better
17:09:21 <mriedem> it might add extra db and rpc strain to get the BDMs in all cases
17:09:29 <mriedem> the performance impact is what i'm concerned about
17:09:42 <mriedem> which is related to the dev list email about having a way to tell if the notifications are even going to be sent
17:09:54 <gibi> yes, in case of instance.update we need an extra db query. in case of instance.create the bdm is already available
17:10:06 <mriedem> there is a bug for that now in oslo.messaging btw https://bugs.launchpad.net/oslo.messaging/+bug/1639287
17:10:06 <openstack> Launchpad bug 1639287 in oslo.messaging "oslo.messaging support for nova to query if notifications are setup for a given host/queue" [Undecided,New]
17:10:44 <mriedem> we could make it configurable....like how we have a config option for task_state changes triggering an update notification
17:10:49 <gibi> one way to solve this load is to make sending notifications configurable on the nova side
17:10:56 <gibi> exactly :)
17:10:57 <mriedem> but more config options is never my first idea
17:11:04 <gibi> yeah, I agree
17:11:08 <stewie925> hi gibi
17:11:19 <gibi> stewie925: hi!
17:11:20 <stewie925> sory I thought the meeting is an hour later
17:11:37 <gibi> stewie925: the nice daylight saving time change
17:11:51 <stewie925> thank you for your input on the rescue notification - I am passing functional tests - but still investigating my unit test issue :)
17:12:13 <stewie925> oh  @ time change
17:12:30 <gibi> stewie925: I'm happy to help, actually that notification is not the easiest one due to the new payload type but you did a great job
17:13:02 <stewie925> gibi: thank you :)  I try, and its a learning experience
17:13:21 <gibi> stewie925: I will look back to your patch tomorrow to see if I can help with the unit test too
17:13:57 <gibi> stewie925: do you have anything else we can help with?
17:14:04 <stewie925> gibi: appreciate it - but will do my best to resolve it myself this time
17:14:09 <stewie925> no, I am good for now, thanks!
17:14:14 <gibi> stewie925: OK :)
17:14:35 <gibi> mriedem: anything else on your side?
17:16:44 <mriedem> sorry got sidetracked
17:16:45 <gibi> Oh one thing on my side.
17:16:46 <mriedem> i don't have anything else
17:17:02 <gibi> I changed the meeting from biweekly to weekly as we can make good progress now
17:17:28 <gibi> these are short meetings anyhow most of the time :)
17:17:33 <mriedem> gibi: only thing i'd ask is,
17:17:43 <mriedem> if you can't make the weekly nova meeting to provide highlights from this meeting,
17:17:54 <mriedem> then just post those in the nova meeting agenda prior to the meeting on thursday
17:18:19 <gibi> mriedem: good point. I will do that if I cannot be on the meeting
17:18:44 <gibi> mriedem: the last couple weeks was a bit problematic for me timingwise but I will improve
17:19:04 <mriedem> np
17:20:12 <gibi> mriedem: btw the text and the link contradicts on the nova meeting page November 10th 2016 1400 UTC, #openstack-meeting (http://www.timeanddate.com/worldclock/fixedtime.html?iso=20161110T210000)
17:20:46 <gibi> will it be 1400 or 2100 UTC this week?
17:25:05 <gibi> OK I will chase mriedem on the nova channel about the timing
17:25:09 <gibi> lets wrap this up
17:25:15 <gibi> thanks for joining
17:25:18 <gibi> #endmeeting