18:00:01 <cp16net> #startmeeting trove
18:00:02 <openstack> Meeting started Wed Sep  2 18:00:01 2015 UTC and is due to finish in 60 minutes.  The chair is cp16net. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:03 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
18:00:05 <openstack> The meeting name has been set to 'trove'
18:00:16 <amrith> all hail cp16net
18:00:17 <mvandijk> o/
18:00:19 <vgnbkr> o/
18:00:30 <cp16net> hey yall
18:00:36 <peterstac> \o
18:00:41 <schang> o/
18:00:47 <cp16net> i hope everyone had a good trip back home from the midcycle
18:00:56 <amrith> cp16net, the agenda has a link to midcycle
18:01:00 <amrith> not bp priorities ;)
18:01:01 <ashleighfarnham> o/
18:01:03 <dloi> o/
18:01:04 <atomic77> 0/
18:01:07 <peterstac> yep, who needs more than two hours of sleep anyways! ;)
18:01:15 <cp16net> lolz
18:01:36 <vkmc> o/
18:01:51 <mat-lowery> o/
18:02:00 <pmalik> *\0/*
18:02:12 <amrith> thx cp16net
18:02:14 <amrith> got it
18:02:24 <cp16net> #link https://wiki.openstack.org/wiki/Meetings/TroveMeeting
18:02:48 <cp16net> #topic pulse update
18:02:58 <cp16net> #link https://etherpad.openstack.org/p/trove-pulse-update
18:03:26 <cp16net> looks like there are a more reviews built up
18:03:40 <shayneburgess> it looks like nothing is making it through the gate
18:03:42 <cp16net> i think thats partly due to the fate
18:03:54 <cp16net> heh the gate is the fate
18:03:56 <cp16net> :-P
18:04:09 <cp16net> shayneburgess: yea speaking of that...
18:04:33 <amrith> what, the gate is hosed again?
18:04:34 <cp16net> so we had a patch that resolevd the image building
18:04:53 <peterstac> cp16net, it still hasn't started to run (last time I checked)
18:05:05 <pmackinn> 12 hours and counting
18:05:06 <cp16net> but... then the image name changed again
18:05:16 <cp16net> hence the gate being broken again
18:05:19 <vkmc> did the second patch got merged?
18:05:25 <shayneburgess> is it this? https://review.openstack.org/#/c/219548/
18:05:39 <peterstac> nope, as pmackinn said 12.5 hours and still not running
18:05:47 <peterstac> shayneburgess, yep, that's it
18:05:49 <cp16net> #link https://review.openstack.org/#/c/219548/
18:05:52 <vkmc> yup
18:05:58 <cp16net> this is the patch that will fix our woes
18:06:02 <amrith> same here
18:06:09 <amrith> I'm hosed on that; i had a hack fr that.
18:06:44 <cp16net> so if you run in to this issue locally testing or building an image just remove that line and it will work now
18:06:46 <peterstac> I'm sensing there will be an inordinate amount of FFE's this cycle ...
18:06:49 <peterstac> ;)
18:07:17 <amrith> peterstac, I think these issues only impact us. not rest of openstack
18:07:26 <cp16net> peterstac: lets try and get as much done as possible
18:07:41 <shayneburgess> when is the FF date now?
18:07:48 <amrith> tomorrow morning
18:07:49 <vkmc> early thursday
18:07:51 <amrith> thursday
18:07:58 <amrith> early AM eastern time.
18:07:59 <cp16net> yeah early thurs
18:08:13 <cp16net> so what that means with a gate thats taking 12+ hours
18:08:20 <peterstac> That's when they cut the image, right?
18:08:24 <shayneburgess> yeah it's already too late
18:08:35 <peterstac> There's no way all our stuff can make it in
18:09:03 <amrith> I suggest we WF+1 all the things which are ready to go and let them wind their way through the queue
18:09:08 <amrith> they won't have merged in time
18:09:31 <vkmc> how much time is the FFE?
18:09:35 <amrith> let's not change our actions based on the deadline, we have to still review and get the code right, merged, etc.,
18:09:46 <shayneburgess> +1
18:09:48 <cp16net> we may have to have a convo about this to ttx but lets try and narrow it down if we can
18:09:56 <vkmc> we need enough time to get the ready patches to go through the gate
18:10:11 <amrith> vkmc, the time to get through the gate is out of our control.
18:10:22 <amrith> reviewing, merging, getting to code right; those are in our control.
18:10:27 <amrith> let's try and do the things that we can control.
18:10:32 <pmackinn-afk> merged? how so?
18:10:36 <amrith> we can't control the rest of it ...
18:10:43 <cp16net> so this is bleeding into the next topic so i'll switch to it
18:10:51 <amrith> pmackinn-afk, sorry; merging as in getting conflicts merged.
18:10:55 <amrith> not merged upstream.
18:11:07 <pmackinn-afk> amrith, ack updating patch sets
18:11:12 <cp16net> #topic liberty-3 blueprint priorities
18:11:23 <cp16net> #link https://etherpad.openstack.org/p/TroveLibertyPriorities
18:12:18 <cp16net> we have done a good job getting eyes on most of the reviews
18:13:19 <amrith> cp16net, I'm running into some issues while testing. I've been working through some of them and resulted in https://review.openstack.org/#/c/219544/
18:13:24 <amrith> that change isn't needed for liberty.
18:13:38 <amrith> but some issues running tests locally
18:14:02 <cp16net> ah yeah that will be helpful cleaning up the output
18:14:52 <cp16net> looks like the mariadb support needs another set of eyes
18:15:10 <amrith> will get to it next.
18:15:14 <vkmc> and the datastore registration api as well
18:15:22 <vkmc> I'm reviewing that one now :)
18:15:30 <amrith> don't know if datastore registration api was on the list. is it?
18:15:49 <cp16net> alright nice
18:16:10 <amrith> vkmc, yes it is.
18:16:11 <vkmc> amrith, yeah, just before horizon patches
18:16:21 <peterstac> amrith, yep at the bottom
18:16:22 <cp16net> can someone step up and on the cluster root enable?
18:16:26 <saurabhs> these two patches also need some help from reviewers
18:16:27 <saurabhs> https://review.openstack.org/#/q/topic:bp/vertica-cluster-user-features,n,z
18:16:40 <peterstac> I'll take a look at MariaDB
18:17:18 <amrith> saurabhs, are these changes on the list?
18:17:22 <cp16net> i'm looking at the cluster scaling for mongo
18:17:28 <amrith> some of them are not.
18:17:40 <saurabhs> - https://review.openstack.org/#/c/218548/ (python-troveclient)(ready for review)
18:17:40 <saurabhs> - https://review.openstack.org/#/c/196913/ (ready for review)
18:18:06 <amrith> like https://review.openstack.org/#/c/219544/
18:18:13 <amrith> which I'm not looking at for now.
18:18:45 <cp16net> amrith: its the cluster root enabled
18:19:13 <amrith> wrong paste
18:19:16 <amrith> https://review.openstack.org/#/c/213354/
18:19:33 <saurabhs> amrith the blueprint url was wronng. the review links are correct though for root-enable on cluster
18:20:08 <saurabhs> amrith yeh the reason I think its not in in the page is the spec has not merged for it
18:20:43 <amrith> saurabhs, what about https://review.openstack.org/#/c/207578/2?
18:20:52 <amrith> it is listed as related to the ones you want to merge
18:21:35 <cp16net> i see that as well
18:21:52 <amrith> hence my confusion.
18:22:04 <saurabhs> amrith the new gerrit UI is bit misleading it doesn't show the dependancies very clearly. let me grab the links for you
18:22:32 <saurabhs> So these two are the parent patches for root-enable on cluster
18:22:33 <saurabhs> - https://review.openstack.org/#/c/218548/ (python-troveclient)(ready for review)
18:22:33 <saurabhs> - https://review.openstack.org/#/c/196913/ (ready for review)
18:22:47 <amrith> OK, thanks
18:22:48 <saurabhs> we need these to get merged for liberty-3
18:23:03 <amrith> will review those two.
18:23:35 <shayneburgess> thanks Amrith
18:23:39 <cp16net> ok i think we cleared up the confusion
18:23:41 <cp16net> :)
18:24:11 <saurabhs> amrith thanks
18:24:47 <amrith> isn't root enable the thing that brought us to an interesting place in the CI ;)
18:24:56 <amrith> did that change (that I reverted) get added back?
18:25:17 <amrith> https://review.openstack.org/#/c/217943/
18:25:20 <cp16net> ok so i'm putting names under the reviews on that page so that we can split up the load
18:25:26 <saurabhs> amrith yup it was the same patch. we fixed it through now its backwards compatible
18:26:21 <amrith> I propose that we should revert https://review.openstack.org/#/c/217881/ as the first changein the next release.
18:26:21 <saurabhs> the API and cli both were fixed so that for the case when password is not passed in no body is sent on the POST call
18:26:40 <shayneburgess> and we fixed the python-troveclient gate so the patch will get tested this time
18:26:47 <shayneburgess> right?
18:26:56 <shayneburgess> +1 Amrith
18:27:17 <cp16net> amrith: yeah that wasnt the issue we had so i'd agree that its worth reverting after
18:27:20 <amrith> shayneburgess, when you say "we fixed", I assume you are asking saurabhs ?
18:27:25 <saurabhs> shayneburgess I think so, :) . I am sure its fixed because now Depends-On in the commit message actually works
18:27:46 <cp16net> nice work saurabhs
18:28:08 <cp16net> so taking another look over the list
18:28:10 <shayneburgess> Amrith - if I remember, the gate wasn't actually testing patches (was pinned to a particular build) so that's how the break got through the first time? Or did I misunderstand?
18:28:26 <amrith> I see no depends-on in https://review.openstack.org/#/c/217881/ and https://review.openstack.org/#/c/218548/
18:28:29 <amrith> am I missing something?
18:28:50 <amrith> shayneburgess, I don't think so.
18:29:47 <cp16net> the tests were using the version of the client in pypu
18:29:53 <saurabhs> amrith so when the CLI patch for root-enable merged the dvsm gate job was not puliing the trove CLI changes from the patch instead it was installing the CLI from the pypi server and thats why the issue was not caught in the first place
18:30:15 <cp16net> yeah that was it
18:30:52 <shayneburgess> yeah that's what I was refering to
18:31:13 <saurabhs> amrith cp16net Nikhil fixed that issue with this patch : https://review.openstack.org/#/c/216915 but this one landed after the CLI was merged
18:31:50 <cp16net> ahh that makes sense
18:32:12 <cp16net> ok we should not have that issue again
18:32:28 <shayneburgess> cp16net yeah that was my point
18:32:35 <saurabhs> and in fact we figured out the issue with dvsm gate only because CLI patch merged and that tests on the API patch were still failing the code form CLI was not available in the API tests because the gate alwasy took cli from pypi
18:33:21 <cp16net> ok we have a few patches for redis
18:34:31 <cp16net> looks like the redis cluster is still being worked on
18:34:55 <saurabhs> cp16net this one also need some core votes
18:34:56 <saurabhs> Add databag ('extended-properties') to cluster-create
18:34:56 <saurabhs> - https://review.openstack.org/#/c/206931/
18:35:35 <cp16net> saurabhs: working our way down
18:35:58 <mvandijk> saurabhs this is the base of a couple of features being added that was not bp'd
18:36:19 <mvandijk> i have beef with these client changes going through without being vetted by the community
18:36:41 <cp16net> who can look over the redis replication?
18:37:35 <amrith> mvandijk, https://review.openstack.org/#/c/206931/ has a +1 from you ...
18:38:07 <mvandijk> i know, im taking it off after having thought more on it
18:38:16 <saurabhs> mvandijk Nikhil mentioned that this is already discussed.
18:38:21 <pmackinn> https://review.openstack.org/219548 moves into 4th place at the turn...
18:38:49 <mvandijk> saurabhs, this was discussed and we agreed it was a problem that needed to be planned out better
18:39:19 <mvandijk> once these go in we cannot revert or change the api
18:39:34 <mvandijk> so we need to get the experience right the first time
18:39:40 <mvandijk> cluster create is already loooking bad
18:40:06 <mvandijk> i think i we have more pressing issues this week
18:40:56 <cp16net> mvandijk: i think there is some valid concerns although i'm not sure what SlickNik has said before since he was authoring this patch
18:41:52 <saurabhs> mvandijk just so that I understand this, given a choice you would do it differently, can you post comments on that patch what/why/how will you do it differently? and what are the issues with current approach?
18:41:58 <amrith> saurabhs, please take a look at https://review.openstack.org/#/c/218548/3
18:43:11 <cp16net> alright so we have a few more patches here
18:43:46 <cp16net> there are a couple cassandra patches there
18:44:07 <amrith> would someone more python literate (than me) please review my comment on https://review.openstack.org/#/c/218548/3
18:44:16 <cp16net> they dont look like anyone has looked at them
18:44:50 <cp16net> they may be candidates to be pushed out if we dont get to them
18:45:52 <cp16net> alright so it looks like we have a plan on the given patches
18:46:33 <peterstac> amrith, you're right, mock.patch decorators are applied in reverse order in the arguments list
18:46:33 <cp16net> lets work through them and if they are good lets get them in the gate
18:47:10 <cp16net> so moving to open discussion
18:47:18 <cp16net> #topic open discussion
18:47:27 <cp16net> anyone have anything else?
18:48:03 * amrith runs back to look at second review for saurabhs ;)
18:48:10 <cp16net> 1.
18:48:21 <cp16net> ...2...
18:48:29 <saurabhs> amrith will ping you as soon as I have the review up again
18:48:56 <cp16net> alright i guess thats all folks
18:48:59 <amrith> thanks cp16net
18:49:03 <cp16net> #stopmeeting
18:49:09 <saurabhs> thanks cp16net
18:49:09 <amrith> #endmeeting
18:49:14 <shayneburgess> thanks
18:49:14 <cp16net> #endmeeting