15:06:01 <kamil_> #startmeeting monasca
15:06:01 <openstack> Meeting started Wed Mar 15 15:06:01 2017 UTC and is due to finish in 60 minutes.  The chair is kamil_. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:06:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:06:05 <openstack> The meeting name has been set to 'monasca'
15:06:19 <kamil_> Agenda for Wednesday March 15 2017 (15:00 UTC) Migration to StoryBoard - 20th March Remove Cassandra code from monasca repositories ? Reviews https://review.openstack.org/#/c/396596/ https://review.openstack.org/#/c/442280/ https://review.openstack.org/#/c/356815/ https://review.openstack.org/#/c/442280/ https://review.openstack.org/#/c/409252/ ( https://review.openstack.org/#/c/440326/ ) refactor - https://review.openst
15:06:31 <kamil_> hmm... this doesn't looks well formated
15:06:42 <kamil_> but at least I started the meeting
15:06:58 <kornica> guess, next command is topic
15:06:59 <kamil_> #topic Migration to StoryBoard - 20th March
15:07:07 <kamil_> kornica?
15:07:08 <kornica> ok, I can take that
15:07:20 <sc> stupid question: what's StoryBoard?
15:07:34 <kornica> sc: it's equivalent of launchpad
15:07:36 <diablo_rojo_phon> It's the task tracker that is replacing storyboard :)
15:07:45 <kornica> oh, hello diablo_rojo_phon
15:07:46 <diablo_rojo_phon> I mean lp
15:07:49 <diablo_rojo_phon> Lol
15:08:11 <sc> kornica: I see
15:08:13 <diablo_rojo_phon> It will be a Queens release goal to get all of the projects over to SB from lp
15:08:28 <kornica> well apart from telling propososed date (next monday) + notyfying you guys that we should cease from using LP during migration, that's about that
15:08:38 <kornica> diablo_rojo_phon: would you explain bit more ?
15:08:49 <diablo_rojo_phon> I've sent out a few emails with blog posts and whatnot talking about the decision.
15:09:11 <kornica> sc: storyboard.openstack.org
15:09:16 <diablo_rojo_phon> It's API first which is different than LP. And it's the final step we need to make to get away from Ubuntu one.
15:09:31 <diablo_rojo_phon> Infra has been using it for quite a while now.
15:10:06 <diablo_rojo_phon> https://storyboard-blog.io/
15:10:43 <kamil_> okay. So from monday everybody of use can or should use storyboard?
15:10:50 <diablo_rojo_phon> We are migrating refstack the same day as you :)
15:11:16 <diablo_rojo_phon> Starting Tuesday. We can migrate everyone Monday night if that works.
15:11:35 <kornica> diablo_rojo_phon: what about migration status (BPs, Bugs, existing changes and their refs to launchpad) [not sure if everyone here are assigned to monasca's mailing list]
15:11:36 <kamil_> okay
15:11:53 <kamil_> At the moment the storyboard (https://storyboard.openstack.org/#!/) is loading and loading aaaand loading
15:12:14 <diablo_rojo_phon> All of the existing stuff in LP will get moved to SB.
15:12:20 <witek> Lauchpad will be offline for the time of migration?
15:12:25 <diablo_rojo_phon> We have migration scripts we will run.
15:12:53 <diablo_rojo_phon> No it won't be offline, but if new things are added or changes are made during the migration that info won't get carried over.
15:13:08 <witek> ok, thanks
15:13:16 <stevejims> Is storyboard down? I'm getting this: 400: GET /api/v1/users/preferences: Invalid input for field/attribute user_id. Value: 'preferences'. unable to convert to int
15:13:41 <diablo_rojo_phon> Not that I know of.
15:13:46 <kamil_> I had the same error
15:13:49 <diablo_rojo_phon> Huh.
15:13:53 <sc> I'm in
15:14:04 <diablo_rojo_phon> fungi: any idea what's up?
15:14:06 <kamil_> But now it looks like it's up again
15:14:10 <witek> #info Launchpad -> Storyboard migration on 20th March, please do not use LP during migration
15:14:21 <diablo_rojo_phon> fungi: nvm :)
15:14:39 <kornica> it works for me too ;-)
15:14:57 <diablo_rojo_phon> So I'd say refrain using it after like 20 UTC on Monday and we will do the migration then
15:15:13 <kornica> +1
15:15:42 <diablo_rojo_phon> Thanks guys!
15:15:49 <kamil_> Okay. I think we can jump to the next topic. Thanks kornica and diablo_rojo_phon
15:15:58 <kamil_> #topic Remove Cassandra code from monasca repositories ?
15:16:35 <kornica> well, that actually comes from us, our PM has been thinking if maybe since cassandra seems a dead topic, we ought to propose removing that
15:16:39 <kornica> + all dependencies
15:17:05 <kornica> for instance we need to make sure cassandra-driver is provided just to run unit-tests
15:18:04 <witek> I cannot see Shinya, perhaps we should ask in the mailing list
15:18:08 <kamil_> Hmm. Maybe we should wait with this question for roland next week?
15:18:15 <witek> but for me there are no reasons to keep it
15:18:18 <kornica> witek: +1
15:18:23 <kornica> kamil_: +1
15:18:31 <rbrndt__> I think Roland would be ok with removing, provided no one else wants it
15:18:41 <rbrndt__> Personally, I would love to get rid of it
15:18:47 <kornica> rbrndt__: there are still changes opened that nobody takes care of
15:18:57 <kornica> rbrndt__: my opinion is that is not even completed
15:18:59 <fungi> diablo_rojo_phon: stevejims: kamil_: that brief sb blip was _probably_ a trove socket timeout, due to the current low activity level that service sees
15:19:16 <rbrndt__> konica: that is my understanding. We certainly have never seen it really work
15:19:34 <diablo_rojo_phon> Thanks fungi
15:19:35 <stevejims> thanks fungi - seems to be working now :)
15:19:45 <kamil_> Okay. Sounds like we could remove it
15:19:51 <kornica> rbrndt__: well I had, briefly during testing changes, but that's about that
15:20:12 <fungi> yeah, it intermittently seems to cause user tokens to get discarded, resulting in that attribute error you saw... i know the sb devs have been having trouble tracking down the real cause
15:20:30 <kornica> kamil_: I think that fastest way might be removing that simply (that won't be hard) and adding Shinya + Roland to reviews
15:20:39 <kornica> maybe mentioning it during next meeting
15:20:40 <kornica> too
15:20:41 <kornica> ?
15:20:47 <kamil_> Sounds good
15:21:01 <rbrndt__> +1
15:21:42 <witek> I'm fine with removing in gerrit and adding Shinya as reviewer
15:22:01 <kornica> cool, we have that covered
15:22:07 <kamil_> Let's switch to the reviews
15:22:10 <kamil_> #topic https://review.openstack.org/#/c/396596/
15:22:49 <kornica> heah, that's from Shinya
15:23:03 <kornica> but I took it over, rebased and fixed all tests
15:23:10 <kornica> so might be ready to final review and marge
15:23:14 <kornica> *merge
15:23:57 <kamil_> I think we need more reviewers on that
15:24:09 <kornica> yeah, but that was pointless before gates got green
15:25:02 <kamil_> okay. folks review please
15:26:05 <kamil_> kornica: do you want to say something more to that patch, or can we go to the next review?
15:26:18 <kornica> we can go further
15:26:28 <kamil_> #topic https://review.openstack.org/#/c/442280/
15:26:30 <kornica> any questions we may answer in review
15:26:50 <kornica> oh...that is merged by Artur already
15:26:57 <kamil_> I see
15:27:07 <kamil_> #topic https://review.openstack.org/#/c/356815/
15:27:08 <kornica> that was cherry-pick that blocked setting monasca-common Py35 gate to voting one
15:27:41 <kornica> I just wanted to ping you guys about that....I've done review here but would be nice to get another pair of eys
15:27:49 <kamil_> Is Li here?
15:28:02 <yingjun> yeah
15:28:22 <yingjun> do we need to keep backward compatible with keystone v2?
15:28:35 <kornica> rbrndt__: what do you think ?
15:29:00 <rbrndt__> At the moment, I would say yes we need compatibility with keystone v2
15:29:20 <rbrndt__> but I haven't seen that review yet, so I will take a closer look
15:29:37 <kamil_> thanks rbrndt__
15:29:51 <kornica> yingjun: I'd expect that client should be smart enough to know the difference between v2 and v3 but apparently it knows how to talk with v3 only
15:29:57 <witek> have to go, bye everyone
15:30:02 <kornica> witek: cya
15:30:07 <kamil_> bye witek
15:30:10 <sc> I would prefer to keep both
15:30:53 <kornica> yingjun: maybe there's an different API you missed/we're not aware that automatically recognizes which Keystone API it deals with
15:30:54 <sc> IIRC we still have v2 in Helion OpenStack
15:30:58 <kornica> or maybe an extra param
15:31:21 <yingjun> ok, i'll check that
15:31:25 <dhague> v2 is only Deprecated right now, it's not yet Unsupported
15:31:51 <dhague> much as I hate v2, I think we need to support it for the time being
15:31:55 <kornica> yingjun: thx
15:32:04 <kornica> dhague: thx for that info
15:32:26 <kamil_> Thank you yingjun
15:32:39 <yingjun> you're welcome
15:32:49 <kamil_> I think we can go further
15:32:52 <kamil_> #topic https://review.openstack.org/#/c/442280/
15:32:54 <kornica> yingjun: sorry for blocking that :(
15:33:07 <yingjun> shouldn't be a hard thing to support both
15:33:24 <kornica> hmm...
15:33:25 <kamil_> kornica... it's your turn... again
15:33:39 <kornica> I think that we already covered that, did I paste same review twice ?
15:33:50 <kamil_> yes you did
15:33:55 <kornica> kamil_: sorry, been through all relevant for me
15:33:58 <kamil_> okay next topic
15:34:06 <kornica> kamil_: sorry for dup
15:34:10 <kamil_> topic: https://review.openstack.org/#/c/409252/ ( https://review.openstack.org/#/c/440326/ )
15:34:27 <kamil_> no problem
15:34:36 <kornica> yeah those two, are quite imporant for me
15:34:41 <kornica> maybe not in context of the code
15:34:49 <kornica> but deployment procedure is always welcome
15:35:05 <kamil_> steve are you there?
15:35:09 <kornica> however during testing that I discovered that monasca-persister does not copy configuration file over to known location
15:35:21 <kornica> the 2nd change between ( )
15:35:23 <stevejims> here
15:35:36 <kornica> stevejims: did you look at the 2nd change ?
15:36:04 <stevejims> yes briefly, happy to change it back
15:36:13 <stevejims> figured the way i left it was less hassle
15:37:06 <kornica> sc: could you ping someone from HPE to take a look into those ?
15:37:29 <sc> kornica: yes
15:37:40 <kornica> sc: thx, that'd be great
15:38:26 <kornica> kamil_: I think that's all, just wanted to bring some more attention to those
15:38:27 <stevejims> we would love to help a documentation effort, but feels like there isn't too much interest in reviewing (except kornica :)
15:38:41 <kornica> stevejims: I am review-junky :P
15:39:13 <kamil_> maybe I can take a look on this tomorrow
15:39:21 <stevejims> not a bad thing!
15:39:26 <kamil_> thank stevejims and kornica
15:39:50 <kamil_> #topic https://review.openstack.org/#/c/440847
15:40:38 <kornica> I remember the time when monasca devstack got broken due to changes in keystone client
15:40:52 <kornica> that approach put another abstraction layer that should protect from that
15:41:07 <kornica> rbrndt__: would you take a look ?
15:41:20 <rbrndt__> Looks good to me, just haven't had a chance to run it yet
15:41:26 <sc> I love this review. ;-)
15:41:35 <kornica> sc: blushing....
15:41:54 <kornica> sc: if you like it so much, why there's no +1 :P
15:42:03 <sc> I'm trying on devstack next 10-15 minutes
15:42:15 <kornica> FYI: devstack for log-api is already using that
15:42:28 <kornica> and it's quite happy last time I asked ;-)
15:42:44 <sc> kornica: I need to add log-api in my devstack config
15:42:50 <kornica> nope
15:43:04 <kornica> devstack api can exist on its own
15:43:21 <kornica> but log-api devstack requires api devstack, because of some of its components
15:43:31 <kornica> like kafka, zookeeper for instance
15:44:28 <kornica> I think that's all about this change from my side
15:45:41 <kamil_> #topic https://review.openstack.org/#/c/442958/
15:45:54 <kamil_> last topic for today
15:46:07 <kornica> that's just FYI
15:46:34 <kornica> WSGI + kamil_ bug fix are included in there
15:47:37 <kamil_> thats great
15:48:05 <kamil_> I think it's worth to get this merged
15:48:12 <kornica> ahm...it's merged :)
15:48:18 <kamil_> even better
15:48:22 <kamil_> haven't seen it
15:48:23 <kornica> heh
15:49:09 <kamil_> okay. Any other topics? We have 12 minutes left
15:49:26 <kornica> also FYI: https://review.openstack.org/#/c/445433/
15:49:35 <kornica> we talked about that while ago
15:49:46 <kornica> that change makes devstack use Grafana from SAP fork
15:50:35 <kornica> since twc-openstack repo won't be used anymore, that seemed fitting to just get this merged
15:50:45 <dhague> Sorry it took a while to push that commit - thought I'd done it until I tried a "git pull -r" the other day and it mentioned uncommitted changes...
15:51:11 <dhague> Also, just to mention that I have been playing with Ceilometer, and with my recently-merged commits there it can now send its metric/event notifications via Kafka instead of RabbitMQ.
15:51:22 <dhague> which means nice things for Monasca :)
15:51:49 <dhague> it was always possible in theory, but a couple of bugfixes were needed in practice
15:51:56 <kamil_> Funny. I tried to install devstack today, and it failed while downloading this repository. I got a timeout
15:52:18 <kamil_> I need to check our proxy tomorrow
15:52:35 <kornica> dhague: nice, I have this change : https://review.openstack.org/#/c/400070/ that I posted long time ago (don't have time to work on though), taking your comment about metrics it might be that it won't be needed at all
15:52:46 <kornica> kamil_: it worked without any issue in my case
15:52:53 <dhague> I had a devstack fail yesterday, but due to a 403 installed the influxdb schema - everything else was OK. Needed to install the vagrant proxy plugin though
15:52:57 <kornica> kamil_: and I think we share the proxy pain
15:53:20 <kamil_> kornica: yeah, i think it is a network trouble
15:53:35 <sc> about proxy I think we need to add to devstack plugin some code to do the configurations Vagrant file did
15:54:13 <kornica> I always thought that vagrant-proxy plugin takes nice care of it by passing system settings
15:54:19 <kornica> anyway that always worked for me
15:55:00 <kamil_> for me the vagrant-proxy plugin works well
15:55:39 <kamil_> But in case that someone is not using vagrant he needs to setup the proxy manually?
15:56:22 <dhague> Seems likely, but I think the Vagrantfile gives some pretty big clues what to do
15:56:30 <sc> kamil_: I'm not using vagrant, I use KVM
15:56:51 <sc> dhague: it's better than docs
15:58:40 <kamil_> two minutes left. Anything else?
15:59:17 <kamil_> So, thanks for participating and bye
15:59:26 <yingjun> bye
15:59:28 <rbrndt__> o/
15:59:32 <kamil_> #endmeeting