18:00:13 <amrith> #startmeeting trove
18:00:14 <openstack> Meeting started Wed Mar  9 18:00:13 2016 UTC and is due to finish in 60 minutes.  The chair is amrith. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:15 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:00:18 <openstack> The meeting name has been set to 'trove'
18:00:47 <dougshelley66> o/
18:00:55 <amrith> reminder for vkmc johnma peterstac slicknik cp16net
18:01:03 <johnma> o/
18:01:12 <vkmc> o/
18:01:48 <amrith> Meeting agenda at https://wiki.openstack.org/wiki/Meetings/TroveMeeting
18:01:52 <amrith> #link https://wiki.openstack.org/wiki/Meetings/TroveMeeting
18:01:53 <peterstac> o/
18:02:04 <amrith> hello folks
18:02:13 <amrith> let's wait a couple of minutes and get started
18:02:28 <amrith> peterstac, anyone else from the frozen cold climates joining us today
18:02:33 <atomic77> /O\\
18:02:43 <peterstac> yep, just cleared the lunch room
18:03:03 <schang> o/
18:03:03 <peterstac> s/just/we all just/
18:03:32 <vgnbkr> o/
18:03:39 <amrith> vkmc, johnma anyone else joining from your teams?
18:04:17 <amrith> #topic Trove pulse update
18:04:28 <amrith> #link http://bit.ly/1VQyg00
18:04:33 <johnma> imandhan will be joining
18:04:38 <imandhan> o/
18:04:42 <amrith> thx to dougshelley66 for filling out the paperwork and getting the pretty graph updated.
18:04:57 <amrith> the number of reviews in the week dropped a bit
18:05:14 <amrith> would be good to get some reviews of the things we want merged in mitaka
18:05:55 <amrith> there's still a reasonable list of things that we would like to get merged, and there are potentially bugs you'd like to see merged as well.
18:06:39 <amrith> If you look at the "Other" tab on the google sheet, you'll notice that the number of reviews waiting on reviewer has been going up.
18:06:43 <johnma> will do amrith.
18:06:54 <amrith> this means (surprise, surprise) we need to get those reviews done ...
18:06:58 <johnma> do we have a starred list for bugs as well
18:07:08 <amrith> yes we do, it's later in the agenda
18:07:19 <amrith> and the link is
18:07:22 <amrith> #link  https://review.openstack.org/#/q/starredby:amrith+status:open
18:07:25 <johnma> has the FFE been accepted. not sure what the process is for that
18:07:49 <peterstac> johnma, the PTL is supposed to do that ...
18:08:00 <amrith> that's a great question. we sent the list out, I'm waiting to hear back. I haven't seen 'acceptances' for any other projects as well.
18:08:05 <johnma> oh ok
18:08:10 <peterstac> I suspect that cp16net is still out of action
18:08:11 <amrith> In cp16net's absence, I'll find out what the process is
18:08:24 <amrith> and if he is the one who has to send out the acceptances, I'll do that
18:08:28 <johnma> sounds good
18:08:34 <amrith> #action [amrith] figure out who has to send FFE acceptances
18:09:03 <johnma> so amrith, once FFE is accepted how much time do we have to get the remaining features approved
18:09:10 <vkmc> good question johnma
18:09:23 <amrith> johnma, yes
18:09:27 <amrith> I'm just reading the process
18:09:29 <amrith> #link https://wiki.openstack.org/wiki/FeatureFreeze
18:09:44 <pmackinn> o/
18:09:46 <amrith> in pertinent part, the process seems to be "The PTL, with the assistance of the core developers of the associated product, will evaluate the request and grant or deny the exception. The farther we are in the release cycle, the less likely it is for the exception to be granted. Remember that the next cycle is just a month away :) "
18:09:59 <amrith> So, I guess we're all here (except cp16net)
18:10:00 <peterstac> I believe there's still a hard-cutoff that you have to make
18:10:31 <amrith> Is there any FFE that any one of you (johnma, vkmc, peterstac) believe should *Not* be granted the FFE exception?
18:10:53 <peterstac> I think we should do our best to get it all in ;)
18:11:03 <amrith> peterstac, that wasn't the question ...
18:11:12 <amrith> it says we should evaluate and grant or deny the exception.
18:11:20 <amrith> is there any FFE request that you would like to deny?
18:11:38 <peterstac> let me restate: I believe all the outstanding FFE requests should be granted
18:11:55 <amrith> vkmc?
18:11:57 <amrith> johnma?
18:11:58 <johnma> +1 peterstac
18:12:21 <vkmc> how many FFE do we have? and how much time do we have to review?
18:12:33 <amrith> let me get you the list
18:13:03 <amrith> http://openstack.markmail.org/thread/66t3kgifagphxvdf
18:13:10 <vkmc> I feel confident with the features that has been around for a while and need some manual testing, but not sure if all the FFE's are in the same stage
18:13:13 <vkmc> thx amrith
18:13:16 <johnma> granting FFE and getting them approved are different things, right
18:13:23 <amrith> http://markmail.org/message/yt46gdj7ystburjl
18:13:28 <vkmc> yeah I think so johnma
18:13:28 <amrith> johnma, yes
18:13:35 <amrith> I'm thinking now only about the granting part
18:13:38 <amrith> See above
18:13:45 <amrith> the process indicates that we have to grant or deny them.
18:13:55 <amrith> http://markmail.org/message/pwzfxpevnhrsderp
18:14:02 <johnma> ok, makes sense. I think all should be granted
18:14:06 <peterstac> We should probably shoot to get them in by Friday - that's my guess
18:14:07 <amrith> http://markmail.org/message/opbljf72drrkyia2
18:14:20 <peterstac> they're targeting next week for RC1
18:14:29 <amrith> #agreed all FFE's requested should be granted
18:14:41 <amrith> Ok, now on to the next question, how long do we have
18:15:20 <amrith> #topic Announcements
18:15:26 <vkmc> cool :)
18:15:31 <amrith> #link  http://releases.openstack.org/mitaka/schedule.html
18:15:35 <amrith> We have until RC-1
18:15:42 <amrith> Mar 14-18
18:16:10 <peterstac> Yeah, but that includes bug fixes (which we have some as well)
18:16:11 <amrith> In other words, all the code that we want in M release should be in the RC-1
18:16:19 <amrith> which includes bug fixes and features.
18:16:34 <amrith> so the answer to your question(s) is that we have till March 14-18 to get these reviews done.
18:16:37 <amrith> the sooner, the better.
18:16:55 <johnma> ok, that sounds good.
18:17:05 <amrith> At this point, the reviews we need by RC-1 are https://review.openstack.org/#/q/starredby:amrith+status:open
18:18:16 <amrith> Is there other stuff that should be on the list?
18:18:31 <amrith> I've added a couple of things (we'll talk about them later in the meeting).
18:18:36 <peterstac> Yes, and probably stuff there that doesn't belong ...
18:18:46 <amrith> but, these are the reviews that I think should be done by RC-1
18:18:49 <amrith> peterstac, what
18:18:50 <johnma> the requirements one
18:18:57 <amrith> which ones?
18:19:26 <peterstac> Yeah, I'm not convinced that Matt R. has grasped the problem entirely (I'm still trying to grock it)
18:19:32 <johnma> well the ones Matt has are for the requirements project. We can review it but not approve them
18:19:49 <amrith> johnma, they are on the list because they are a package. Let's table that conversation till later
18:19:57 <johnma> ok
18:20:12 <amrith> as stipulated, the 4 things related to requirements and backward compatibility, we'll talk about them.
18:20:15 <amrith> what about the others.
18:20:19 <amrith> are we missing anything?
18:20:24 <amrith> Are there extraneous things?
18:20:50 <peterstac> #link https://review.openstack.org/#/c/290177/
18:21:00 <peterstac> #link https://review.openstack.org/#/c/290275/
18:21:23 <amrith> ok, these are the module management ones
18:21:29 <johnma> I need to open a defect for the db2 install issue
18:21:31 <amrith> which we got FFE's for
18:21:32 <peterstac> still have to fix a py27 issue in the second one - should have that done this afternoon
18:21:57 <dougshelley66> i assume the starred list should only contain features that have FFEs at this point?
18:22:09 <amrith> johnma, peterstac please send me the reviews so I can star them.
18:22:11 <amrith> dougshelley66, yes
18:22:17 <johnma> I thought the starred list was the complete list for bugs and features
18:22:18 <amrith> are there things that shouldn't be on there?
18:22:52 <amrith> yes johnma, if there are things that are not there, let me know.
18:22:53 <dougshelley66> i don't think there is an FFE for https://review.openstack.org/259167
18:22:57 <amrith> vkmc, is the list complete?
18:23:04 <peterstac> well, I'm not convince the requirements stuff is being handled correctly
18:23:38 <vkmc> amrith, checking out
18:23:40 <amrith> peterstac, yes this is stuff we need to review.
18:24:14 <vkmc> there is no FFE for #259167
18:24:18 <amrith> johnma, do you have FFE's for all the thing there (CouchDB B&R, DB2 B&R, CouchDB user/db functions, couchDB replication)
18:24:43 <imandhan> couchdb replication is for newton I believe
18:24:50 <johnma> we are not doing couchdb replication this release amrith. We pushed that to next release
18:25:06 <johnma> we need to remove that from your list
18:25:08 <amrith> yes vkmc, I removed the star
18:25:12 <amrith> refresh your screen
18:25:19 <vkmc> amrith, cool, thx
18:25:43 <amrith> johnma, so https://review.openstack.org/255437 comes off the list
18:25:51 <amrith> done, please refresh
18:25:59 <johnma> yes Sir
18:26:08 <amrith> :)
18:26:27 <amrith> TBD's for the bugs that johnma and peterstac will send in
18:26:37 <amrith> as soon as the reviews are in, I'll star them.
18:26:48 <amrith> so the link is https://review.openstack.org/#/q/starredby:amrith+status:open
18:27:15 <amrith> if we're good with the list, let's move on (we'll get back to the requirements stuff)
18:27:26 <amrith> #topic Summit planning
18:27:33 <amrith> #link  https://etherpad.openstack.org/p/trove-newton-proposed-sessions
18:27:33 <peterstac> amrith, my reviews are in - I included links above
18:27:51 <amrith> please add sessions you'd like for summit at that etherpad
18:28:03 <amrith> I requested rooms based on a shortlist we'd made at the mid-cycle
18:28:19 <amrith> peterstac, I've starred your two reviews.
18:28:43 <peterstac> ok, see them now, thx
18:29:37 <amrith> Remember to add your name to the sessions (session leader)
18:31:52 <amrith> ok ...
18:31:58 <amrith> so we can all do that as our homework!
18:32:24 <amrith> #topic Open Discussion
18:32:32 <amrith> peterstac, ... go ahead
18:32:41 <amrith> I can give you a synopsis of the requirements issue to start.
18:32:43 <amrith> if you'd like
18:32:52 <amrith> so everyone is caught up on the story so far.
18:33:18 <amrith> ok, here goes.
18:33:55 <amrith> if a person wrote a python client application that used the python-troveclient and used the slave_of parameter to create(), then version 2.1.0 of the client will break the application.
18:34:09 <amrith> this is the 'backwards compatibility' issue that Matt is describing.
18:34:51 <amrith> this email (a part of a long thread) will tell you where we're at now.
18:34:51 <amrith> http://markmail.org/message/njtfjbbryyxfmgna
18:35:06 <amrith> I believe that the changes being proposed [1] https://review.openstack.org/290048/ [2] https://review.openstack.org/290023/ [3] https://review.openstack.org/290021/ [4] https://review.openstack.org/290168/
18:35:11 <amrith> are at best incomplete.
18:35:31 <amrith> if we do want to ensure backward compatibility the way Matt is describing, I think we should also back out the API change.
18:35:43 <amrith> I'll let y'all read the email and get caught up.
18:36:10 <amrith> This link (http://markmail.org/message/bm3dwtgec3bz6ewc) was the email I sent this morning
18:36:26 <amrith> the link earlier was Matt's response ...
18:37:11 <amrith> everyone caught up?
18:41:24 <amrith> yes ... someone say something ;)
18:41:30 <dougshelley66> i'm caught up
18:42:55 <amrith> peterstac, vkmc, johnma, anyone else ...
18:43:05 <amrith> what are your thoughts?
18:43:05 <vkmc> nothing from my side
18:43:23 <amrith> vkmc, what are your thoughts on this backwards compatibility issue?
18:43:34 <dougshelley66> seems like the cleanest way forward is to back out the client and back end changes and blacklist the 2.1.0 client
18:43:42 <dougshelley66> which is close to what matt's patches contemplate
18:43:44 <peterstac> I need to do some more research - I did some last night, but I've been focused on getting my stuff up for review
18:44:12 <vkmc> I have to catch up with that, that's why I cannot share my thoughts yet :)
18:44:23 <amrith> ok, we'll have to wait
18:44:29 <amrith> we need to decide something about this
18:46:52 <peterstac> If we do revert it, we should make sure that our tests use the correct (i.e. new) interface
18:47:11 <peterstac> I think that was part of the problem - our tests kept using the deprecated APIs
18:48:27 <amrith> well peterstac ... I asked that in email and Matt specifically said no, since the slave_of is a valid parameter he is still using that.
18:48:49 <dougshelley66> i think peterstac is saying that we should change the tests in master
18:48:55 <dougshelley66> to use "replica_of"
18:49:02 <amrith> they already do
18:49:12 <amrith> they have to because the 2.1.0 client has no slave_of
18:49:19 <amrith> and the tests are in trove repository
18:49:21 <dougshelley66> right but if we revert the back end change it will put the test code back to?
18:49:25 <amrith> and the clietn is in troveclient ;)
18:49:44 <dougshelley66> ok
18:49:56 <amrith> dougshelley66, yes, I guess revert* is what I meant.
18:50:09 <amrith> johnma, any thoughts on this?
18:50:49 <pmackinn> amrith, so this is actually impacting users, right? as opposed to a technical transgression
18:51:16 <amrith> pmackinn, hmm.
18:51:21 <amrith> let me see.
18:51:28 <dougshelley66> pmackinn it was noticed by the tests on stable/liberty
18:51:58 <peterstac> amrith, my point is that as soon as we deprecated 'slave_of' we should have immediately changed the tests to reflect that
18:52:08 <amrith> to the best of my knowledge, this is a testing discovery.
18:52:14 <peterstac> (should have been done 2 cycles ago)
18:52:20 <amrith> pmackinn, ^^
18:52:50 <amrith> peterstac, we never did deprecate slave_of the way Matt is doing now
18:52:52 <peterstac> I think if we would have changed our tests way back when, nobody would have noticed this
18:52:54 <amrith> with a warning and all that.
18:53:04 <amrith> peterstac, that's Matt's point, I guess
18:53:05 <peterstac> yeah, that was also a mistake
18:53:22 <amrith> that this test did in fact find something which shouldn't have happened.
18:53:32 <dougshelley66> peterstac, you are correct but merely fixing the tests in stable branches isn't appropriate now
18:54:17 <peterstac> sure, but I want to make sure that if we roll back the change that we *don't* roll back the test changes
18:54:30 <amrith> ok, we're getting close to time. I'll look for any input y'all may have either on #openstack-trove or by email.
18:54:43 <peterstac> the tests should be using replica_of, even if slave_of is to be supported again (briefly)
18:55:00 <dougshelley66> peterstac agreed
18:55:01 <amrith> peterstac, I agree. But not everyone agrees with that point of view.
18:56:38 <amrith> ok, so let's move on.
18:56:46 <amrith> anything else for open discussion in the last 3 minutes
18:58:27 <amrith> 2 minutes.
18:58:46 <amrith> ok, going once.
18:59:07 <amrith> #endmeeting trove