15:01:08 <witek> #startmeeting monasca
15:01:08 <openstack> Meeting started Wed May 15 15:01:08 2019 UTC and is due to finish in 60 minutes.  The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:01:12 <openstack> The meeting name has been set to 'monasca'
15:01:17 <witek> hello everyone
15:01:24 <witek> the agenda for today:
15:01:29 <witek> https://etherpad.openstack.org/p/monasca-team-meeting-agenda
15:01:53 <witek> #topic Train priorities
15:02:07 <witek> thanks for your votes
15:02:18 <witek> https://docs.google.com/spreadsheets/d/1MWp8sCTeoJPnAysFzRKkxhvd8MR4AaJT04LxoJiE2QM/edit#gid=0
15:02:50 <witek> the clear winner in Middleware upgrade
15:03:24 <joadavis> this is always a good exercise. It makes you realize the tradeoffs in time
15:03:27 <sc> not a surprise :-)
15:03:49 <witek> after that it's kind of difficult to make a clear cut, what should be high and low prio
15:04:31 <witek> joadavis: yes, I would prefer to give 100 to almost all of them :)
15:05:01 <sc> @witek: clearly nobody has interest in gnocchi
15:05:57 <witek> I wouldn't say it doesn't make sense
15:06:40 <witek> but it's a lot of effort without clear added value
15:07:10 <witek> and the future of the project is not clear
15:07:58 <witek> how many items do we want as high prio?
15:08:31 <witek> 4?
15:08:37 <sc> no more then 4
15:08:47 <dougsz> Was going to say 3 including the ones in progress
15:09:39 <witek> let's take 4 including the old ones
15:09:48 <sc> and my suggestion is to drop all our preferences and focus on main priority first
15:11:28 <dougsz> sc: Agree it makes sense to focus our review efforts
15:12:37 <witek> OK, I've created the groups of 4 in the spreadsheet
15:13:10 <witek> hm, prio3 should probably have one more item
15:13:47 <witek> now we'll also need owners for these tasks
15:14:05 <sc> that the difficult part, for me
15:15:12 <witek> Doug has started already, thanks
15:15:34 <witek> please take your time and volunteer for tasks
15:15:50 <witek> I'll put them in the spec page in the next days
15:17:18 <witek> for monasca-thresh replacement and query language, we wanted to get back to decision makers to check the readiness for commitment
15:17:49 <witek> I myself don't have answer yet, hope to give some answer next week
15:18:47 <chaconpiza> Similar situation here
15:19:05 <witek> yes, that needs time
15:19:49 <witek> OK, any comments or questions on prioritization?
15:19:50 <sc> mine are desiderata
15:20:24 <witek> sc: ?
15:21:27 <witek> #topic Migration from Ceilometer
15:21:30 <sc> I marked two tasks but I have to confirm them, till now read them as : I would like to be the owner
15:21:44 <witek> that's OK, thanks
15:21:49 <joadavis> sc: to some extent, that applies to all of us. :)
15:22:33 <joadavis> I started a wiki page for migration to capture a few thoughts.  Please feel free to add any ideas or wants you have
15:22:42 <joadavis> We can fill them out as we go
15:22:58 <witek> very nice, thank you
15:23:42 <witek> the section `comparison of features`, what could it include?
15:23:53 <joadavis> Once we have a rough idea, we can try defining a reference example, maybe using devstack and cloudkitty or something
15:24:55 <witek> I think billing is not the only interesting use case, there was e.g. a lot of interest about auto-scaling
15:24:56 <joadavis> for comparision of features, I was thinking it could just be a short list or table.  items would be 'does it handle events' 'does it monitor cpu/storage/network' 'can it export to billing'
15:25:11 <joadavis> yes, we should call out auto scaling
15:26:50 <witek> perhaps identify and list important metrics for the use cases like billing, auto-scaling
15:27:42 <joadavis> for someone who is actually going to migrate, it might be interesting to list the actual metric name equivalents
15:27:53 <witek> right
15:29:20 <witek> on the related topics, there was an email from Public Cloud SIG
15:29:24 <witek> http://lists.openstack.org/pipermail/openstack-discuss/2019-May/006215.html
15:31:08 <joadavis> ah, yes, I was going to mention that
15:31:30 <witek> they want to host a meeting and discuss possible implementations for billing
15:31:30 <joadavis> I added a few notes, but was trying not to dig myself in too deep :*)
15:32:45 <witek> I guess we can move on to reviews?
15:33:03 <joadavis> It might be an interesting question for the OpenStack survey - how many users use billing, and how many of those roll their own solution
15:33:09 <joadavis> yes, we can move on
15:33:50 <witek> #topic reviews
15:34:02 <witek> dropping py35
15:34:16 <witek> I think these were merged already
15:35:05 <witek> I cannot connect to gerrit right now
15:35:30 <joadavis> The one on monasca-common is still hitting intermittent failures https://review.opendev.org/#/c/652418/
15:35:49 <witek> eh :(
15:37:07 <witek> I've added blacklisting bandit 1.6 to pass the pep jobs in some repos
15:37:13 <joadavis> I think I'd missed reviewing the one for monasca-ui as well - https://review.opendev.org/#/c/652419/
15:37:25 <joadavis> thanks for that bandit work
15:37:48 <witek> in monasca-ui I think we have other problems
15:38:13 <piotrowskim> if its about failing tests, its in agenda for today :)
15:39:17 <witek> thanks piotrowskim, yes, let's leave it for the end
15:39:29 <witek> another bumper was falcon 2.0
15:39:38 <witek> https://review.opendev.org/#/c/659264/
15:40:45 <dougsz> What do we do about old releases? Do we backport this patch, or try to limit Falcon to <2.0.0
15:41:21 <witek> is falcon in global requirements?
15:41:49 <witek> yes, so we have upper constraints in stable branches
15:41:57 <dougsz> Ah yes, of course
15:42:17 <witek> 1.4.1 for Stein
15:42:25 <witek> we should be good
15:42:43 <dougsz> Cool, thanks
15:43:12 <witek> I see lower-constraints job failing on your change
15:43:38 <witek> with many tests failing
15:44:20 <dougsz> Still investigating how backwards compatible it is going to be
15:45:24 <witek> thanks for your work dougsz, you're on the good way to solve it
15:45:57 <witek> any other reviews apart from monasca-ui?
15:45:58 <dougsz> you're welcome
15:46:31 <witek> #topic monasca-ui tests
15:46:56 <piotrowskim> ok so I think mu turn
15:47:37 <witek> do you have some findings, what the problems are about?
15:47:52 <piotrowskim> yes
15:48:13 <piotrowskim> in one commit in horizon, author changed the debug flag from django setting from false to true
15:48:58 <piotrowskim> it displays full stack of errors right now, he did it to fix tests on their side
15:49:38 <piotrowskim> with flag set to false its ok, but with set to true it raises ValueError
15:50:08 <witek> can we easily update our code or is it complicated?
15:50:32 <piotrowskim> not sure how I can handle it further, I can in all 22 tests that we have on monasca-ui just try:except the ValueError which we have from the stack
15:50:49 <piotrowskim> or we need to downgrade horizon, which I belive is not possible
15:51:18 <witek> no, the second is definitely not the right path
15:51:46 <piotrowskim> exactly, maybe there is a possibility to do it "right" instead of just try:except
15:51:53 <witek> we could ask horizon team for some hints how to handle it
15:52:21 <dougsz> sounds like a good idea to me
15:52:35 <witek> chaconpiza: do you think you could help on it?
15:52:46 <chaconpiza> Yes
15:53:07 <witek> I can think of one person :)
15:53:50 <chaconpiza> :D
15:54:18 <witek> OK, I hope we can move forward on this, the repo has been blocked for some time already
15:54:30 <witek> thanks piotrowskim for taking this task
15:54:45 <piotrowskim> not a problem, thanks for help
15:55:49 <witek> #topic make Doug angry
15:55:57 <witek> https://www.amazon.com/photos/shared/O7OapGM9RzWmk43r2auwbg.eReetOnraZdLGOztzcAVxm
15:55:59 <witek> :)
15:56:20 <witek> they had very nice beer
15:56:38 <chaconpiza> and thanks for the Sponsor
15:56:47 <witek> SUSE
15:57:28 <joadavis> mmm, that was good buffalo meatloaf
15:57:47 <witek> :)
15:57:48 <chaconpiza> And bison burgers
15:57:53 <joadavis> I didn't include the picture of me standing next to the life size metal gorilla statue
15:58:18 <witek> that's all from agenda and time is also almost over
15:58:31 <witek> thanks for joining
15:58:36 <dougsz> Thanks all
15:58:53 <joadavis> thanks
15:59:00 <witek> see you next week
15:59:05 <chaconpiza> thanks
15:59:08 <witek> #endmeeting