21:00:34 <flaper87> #startmeeting Zaqar
21:00:35 <openstack> Meeting started Mon Aug  4 21:00:34 2014 UTC and is due to finish in 60 minutes.  The chair is flaper87. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:36 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:38 <openstack> The meeting name has been set to 'zaqar'
21:00:44 <flaper87> #topic roll call
21:00:46 <flaper87> o/
21:00:47 <vkmc> o/
21:00:54 <ametts> o/
21:01:00 * flaper87 wonders where flwang is >.>
21:01:07 <flaper87> He's suppose to be awake and here
21:01:36 <flaper87> ok, lets move on, I guess :D
21:01:42 <flaper87> #link https://wiki.openstack.org/wiki/Meetings/Marconi#Agenda
21:01:48 <flaper87> That's the agenda for today's meeting
21:02:09 <flaper87> there are some topics kgriffs wanted to talk about so I hope he'll be back soon enough
21:02:17 <flaper87> #topic Graduation Status
21:02:23 <flaper87> lets go straight there
21:02:35 <flaper87> So far, we've renamed the project internally
21:02:54 <flaper87> by internally I mean in the repo
21:03:01 <flaper87> but we're still waiting for the repo to be renamed
21:03:11 <flaper87> it'll be schedule sometime during this/next week
21:03:13 <flaper87> I hope
21:03:25 <flaper87> we need to follow-up with the -infra guys on that
21:03:28 <flaper87> #link https://etherpad.openstack.org/p/marconi-graduation
21:03:47 <flaper87> There're many things to write
21:04:14 <flaper87> There's a documentation day scheduled this Thursday but, as kgriffs said in the channel, we shouldn't wait 'til that day comes
21:04:19 <flaper87> vkmc: what are your feelings on this?
21:04:47 <flaper87> vkmc: ?
21:04:57 <vkmc> flaper87, well the thing is that the user guide should be written with docbook
21:05:09 <vkmc> and we are not familiar (at least not me nor kgriffs) with it
21:05:22 <flaper87> vkmc: ok, who can provide some guidance there? Anne?
21:05:42 <vkmc> flaper87, yes, someone in docs
21:05:50 <flaper87> There are other things to write down. For example, the use cases: :Get concrete integration use cases, maybe even POC from 2-3 other projects."
21:05:58 <vkmc> kgriffs already asked Catherine, I heard she contributed with docs once
21:06:05 <flaper87> vkmc: awesome, good call
21:06:09 <vkmc> yes there is a lot to do
21:06:09 <flaper87> yeah, she did
21:06:14 <vkmc> but I'd focus on the basics
21:06:25 <flaper87> Agreed
21:06:26 <vkmc> at least to get to the graduation on time
21:06:39 <vkmc> then we can polish it and add more useful
21:06:45 <flaper87> There are some patches we need to keep an eye on, for example: https://review.openstack.org/#/c/111817/
21:06:59 <vkmc> we should add this ideas to the docs bp so we can remember then on K
21:07:07 * vkmc clicks
21:07:18 <flaper87> for some reason that one failed, the py27 tests ran on a trusty box but others ran on a precise
21:07:20 <flaper87> AFAICS
21:07:26 <flaper87> vkmc: +!
21:07:27 <flaper87> vkmc: +1
21:08:12 <flaper87> Lets make sure enough reviews are provided, ok?
21:08:15 <flaper87> #link https://review.openstack.org/#/q/status:open+project:openstack/marconi,n,z
21:08:29 <flaper87> but lets focus on patches addressing blueprints for juno
21:08:39 <flaper87> Speaking of blueprints:
21:08:41 <flaper87> #link https://blueprints.launchpad.net/marconi/juno
21:08:52 <flaper87> I cleaned up some of them but I think there's still some clean-up to do
21:09:09 <flaper87> if you think there's a blueprint that shouldn't be there, lemme know
21:09:29 <flaper87> Same thing applies if you think a blueprint *should* be scheduled for Juno
21:09:51 <flaper87> Blueprints that need more attention are the ones marked as "Essential" or "High"
21:09:58 <flaper87> lets focus on getting those done asap
21:10:25 <flaper87> we're in good shape
21:10:30 <flaper87> or well, at least it's not terrible
21:10:31 <flaper87> :P
21:11:09 <flaper87> ok, lets move on
21:11:40 <flaper87> #topic Project Rename
21:11:54 <flaper87> I already mentioned the status code-wise but there are other things to do here
21:12:09 <flaper87> I'd like to list them now and see if something else comes to your mind
21:12:11 <flaper87> We need to:
21:12:21 <flaper87> 1. Update all the wiki pages (PAINFUL)
21:12:29 <flaper87> 2. Update documentation
21:12:41 <flaper87> 3. Release a new version of the client on pypi (with the new name)
21:12:53 <flaper87> 4. Rename the project in our brains
21:13:08 <flaper87> 5. Make sure ppl that gets into #openstack-marconi are redirected to #openstack-zaqar
21:13:26 <flaper87> when the infra patch lands, that'll probably be done automagically
21:13:29 <flwang> o/
21:13:32 <malini> o/
21:13:37 <flwang> the meeting time has been changed, is it?
21:13:38 <flaper87> malini: flwang hey hey :)
21:13:39 <vkmc> 6. Add Zaqar tag to the Marconi questions in ask.openstack.org
21:13:48 <flaper87> flwang: yup, we're in the meeting now
21:13:57 <flaper87> it started like 13min ago
21:14:00 <flwang> cool
21:14:03 <flaper87> vkmc: +1, good one
21:14:17 <flaper87> 7. Make a final announcement on the mailing list
21:14:54 <flaper87> malini: flwang we just walked through the status of the graduation process and the things left to do for the rename
21:15:11 <flwang> flaper87: nice
21:15:11 <flaper87> malini: flwang is there anything concerning you?
21:15:25 <malini> no :)
21:15:34 <malini> do we have the pending items list somewhere?
21:16:03 <flaper87> malini: I'll start cleaning these up after the meeting: https://etherpad.openstack.org/p/marconi-graduation
21:16:07 <flaper87> or well, tomorrow on the plane
21:16:09 <flaper87> :P
21:16:31 <flaper87> malini: we need eyes on this: https://review.openstack.org/#/c/111817/
21:16:32 <flwang> based on the patches in reviewing, seems we're in a good shape
21:16:53 <flaper87> py27 passed but the others failed... I think they ran on precise for some reason
21:17:04 <flaper87> although the regex seems correct and it worked for py27
21:17:15 <flaper87> flwang: we are
21:17:29 <flaper87> I mean, we're not in a terrible shape but there's lot to do on the docs side
21:17:43 <flaper87> malini: could you follow-up with infra on that?
21:17:53 <malini> on py26 failure?
21:18:36 <malini> hmm..py33 is failing too
21:18:56 <flaper87> malini: but py33 raised the version exception
21:19:00 <flaper87> py26 raised some other exceptions
21:19:13 <malini> py33 still on precise
21:19:28 <flaper87> malini: right but that's weird, the regex should've caught py33 too
21:19:39 <flaper87> matched*
21:19:43 <malini> & py26 is on centis6
21:19:58 <malini> See http://logs.openstack.org/17/111817/1/check/gate-marconi-python26/2cd857d/console.html
21:19:58 <flaper87> oh, mmh
21:20:16 <flaper87> ah lol, I was grepping on ubuntu
21:20:18 <flaper87> ok
21:20:48 <flaper87> based on the errors it looks like centos box doesn't install mongodb by default
21:21:13 <malini> but we set our default servers to be trusty
21:21:14 <flaper87> it basically means we can't do this switch just yet
21:21:22 <malini> guess we need to follow up on tht patch
21:21:31 <flaper87> malini: exactly, could you ping infra ?
21:21:48 <malini> sure..can you give me the review link if you have it handy?
21:21:51 <flaper87> malini: this is the patch: https://review.openstack.org/#/c/111784/
21:21:53 <flaper87> lol
21:22:24 <malini> thx! I will follow up on tht
21:22:42 <flaper87> awesome, thanks
21:23:25 <flaper87> all, if you have time, go through this list and check whether it makes sense: https://blueprints.launchpad.net/marconi/juno
21:24:16 <flaper87> also, flwang's patch could use some reviews
21:24:22 <flaper87> as soon as he pushes the update
21:24:34 <flaper87> lets focus on getting Essential/High blueprints done
21:24:41 <flwang> flaper87: I will post a new PS
21:24:43 <flwang> today
21:24:47 <vkmc> https://blueprints.launchpad.net/marconi/+spec/api-v1.1-user-guide this one should be retargeted to K
21:24:48 <flaper87> flwang: awesome, thanks!
21:24:50 <flwang> my today I mean :)
21:25:06 <flaper87> vkmc: I meant to ask you that, thanks!
21:25:35 <vkmc> flaper87, oh sorry, I could do that :)
21:25:55 <flaper87> vkmc: npp
21:25:56 <flaper87> done
21:26:37 <vkmc> thx
21:26:46 <flaper87> ok, anything else on this topic?
21:27:19 <malini> I will probably need somebody else to help with https://blueprints.launchpad.net/marconi/+spec/devstack-marconi-tests
21:27:31 <flaper87> malini: what's missing there?
21:27:34 <malini> this is to point our functional tests against devstack
21:27:43 <malini> This is not needed for graduation
21:27:51 <malini> But will be good to have eventually
21:28:07 <flaper87> malini: ok, I'll lower the priority then
21:28:23 <flaper87> Anything that is not needed for graduation shouldn't be considered essential
21:28:57 <flaper87> malini: if you get a chance, pls, explain in the blueprint what the bp is about and what's missing
21:29:06 <malini> ok
21:29:49 <flaper87> malini: I just read what clarkb said in -infra
21:30:17 <flaper87> in that case, we need another patch that sets the ZAQAR_ENABLE_MONGODB flag on the py27 gate
21:30:27 <flaper87> instead of changing the default
21:30:45 <malini> for the benefit of others "that job can only run on centos6 because ubuntu does not have python 2.6"
21:31:19 <flaper87> malini: could you ask how to set an env bar for a get job?
21:31:20 <malini> We are py34 compilant, rt?
21:31:25 <flaper87> malini: no idea :P
21:31:35 <flaper87> I think I saw how to do that but I don't recall where
21:32:17 <flaper87> ok
21:32:19 <flaper87> lets move on
21:32:21 <malini> flaper87: I dont understand this "how to set an env bar for a get job?"
21:32:22 <flaper87> kgriffs: there?
21:32:28 <kgriffs> yep
21:32:30 <flaper87> malini: sorry, lemme explain
21:32:36 <kgriffs> sorry, got stuck in a meatspace meeting that went way over.
21:32:37 <flaper87> since *just* py27 gate can run on trusty
21:33:09 <flaper87> we should then configure the gate to set the MONGODB environment var so that the mongodb tests will be enabled on the py27 gate
21:33:20 <malini> yes
21:33:26 <flaper87> kgriffs: FYI, turns out that just py27 can run on trusty
21:33:27 <malini> I figured tht much
21:33:36 <kgriffs> ah
21:33:42 <flaper87> malini: ok, there's a way to do that in the `config` repo
21:33:47 <flaper87> I don't recall how
21:33:48 <kgriffs> so we set ZAQAR_TEST_MONGODB=1 on py27
21:33:49 <kgriffs> and
21:33:53 <kgriffs> ZAQAR_TEST_MONGODB=0 on the others?
21:33:53 <flaper87> kgriffs: yup
21:33:58 <malini> I'll find tht one
21:34:06 <flaper87> kgriffs: or just ZAQAR_TEST_MONGODB=1
21:34:11 <flaper87> and we abandon your patch
21:34:12 <flaper87> :P
21:34:14 <kgriffs> book
21:34:15 <kgriffs> booh
21:34:20 <kgriffs> :)
21:34:52 <kgriffs> I guess if we have it running on py27 in the gate bugs won't get merged on accident
21:34:56 <flaper87> I'd prefer to merge your patch since I'd like everyone to test on MongoDB
21:35:03 <flaper87> kgriffs: yeah
21:35:09 <flaper87> it should be enough
21:35:11 <flaper87> I hope
21:35:36 <malini> but if we merge kgriffs patch, gating will become complex, rt?
21:35:54 <malini> or we set it to false in py26 & py33
21:35:56 <flaper87> malini: right, we just need to enable mongodb tests on py27
21:36:05 <flaper87> we won't merge kgriffs patch, for now.
21:36:09 <kgriffs> kk
21:36:34 <flaper87> (unless there are better ideas)
21:36:35 <flaper87> :D
21:36:41 <flaper87> anything else?
21:37:04 * flaper87 is getting blinder everyday
21:37:18 * flaper87 wonders where his glasses are
21:37:23 <flaper87> ok moving on
21:37:32 <flaper87> kgriffs: you put some things in the agenda: https://wiki.openstack.org/wiki/Meetings/Marconi#Agenda
21:37:42 <flaper87> we covered the roadmap and the rename process
21:37:44 <kgriffs> yeah, I can run through those real quick
21:37:47 <flaper87> and also mongodb
21:37:50 <flaper87> ok cool
21:37:58 <flaper87> #topic Getting in the habit of using DocImpact, SecurityImpact, and UpgradeImpact keywords in commit messages
21:38:03 <flaper87> kgriffs: floor is yours
21:38:49 <kgriffs> OK, I just wanted to make everyone aware of these keywords and encourage everyone to start using them. I think reviewers should also check for them if they are needed.
21:39:09 <kgriffs> #link https://wiki.openstack.org/wiki/GitCommitMessages#Including_external_references
21:39:31 * flaper87 clicks
21:39:37 <flaper87> kgriffs: +1
21:39:49 <kgriffs> This will help us track relevant doc, security, and upgrade/migration work
21:39:50 <flaper87> someone asked me what DocImpact is for the other day
21:40:05 <flaper87> I didn't know there was a wiki page with that
21:40:09 <flaper87> awesome
21:40:13 <malini> tht was mw
21:40:15 <kgriffs> yep
21:40:27 <flaper87> malini: I should've known (kidding)
21:40:28 <flaper87> :D
21:41:04 <flaper87> ok, any questions?
21:41:05 <kgriffs> so if you add a new feature, api change, config changes, etc. that's when these tags come into play
21:41:29 * kgriffs realizes he just about described every patch, ever. :p
21:41:35 <flaper87> reviewers should pay extra attention to this point
21:41:41 <kgriffs> +1
21:41:43 <flaper87> I know flwang is very good at remembering these things
21:41:56 <flaper87> I'm not but I'll put extra attention
21:42:00 <flwang> flaper87: are you kidding ? :D
21:42:21 <vkmc> omitting this tags are enough reason for a -1?
21:42:26 <flaper87> flwang: no, I've counted how many -1s you've sent on my patches with a: "Should this have DocImpact?" comment
21:42:27 <flwang> flaper87: but I will add the docImpact to my /health patch for sure :)
21:42:28 <flaper87> >.>
21:42:47 <flwang> flaper87: in Glance, right? haha
21:42:51 <flaper87> vkmc: yes, they are
21:42:57 <flaper87> flwang: yes >.>
21:43:01 <flaper87> :D
21:43:11 <vkmc> cool
21:43:24 <flaper87> anything else?
21:43:35 <flwang> flaper87: I'm shame a glance guy for now, seems I put more effort on Zaqar recently
21:43:49 <flaper87> flwang: oh, you're a great guy
21:43:51 <flaper87> that's the way to go
21:43:57 <malini> flaper87: be sure to chk openstack-zaqar, after the meeting & before you sleep :)
21:43:58 <flaper87> that's what the whole community should do
21:44:01 * flaper87 ducks
21:44:27 <flaper87> malini: +!
21:44:28 <flaper87> malini: +1
21:44:32 <flaper87> ok, moving on
21:44:35 <flaper87> #topic Move or copy _TRANSPORT_LIMITS_OPTIONS to pool catalog?
21:45:14 <kgriffs> ok, so as part of the work to do flavors, I think it makes sense to have different limits per pool/flavor
21:45:35 <kgriffs> for example, a Redis pool may want a smaller max message size since everything goes into RAM
21:46:03 <kgriffs> we talked about making pools "always on"
21:46:20 <kgriffs> in which case you would configure a single pool and set config options, such as limits, in there
21:46:37 <kgriffs> otherwise, we will have a DRY problem with configuration
21:46:49 <flaper87> +1
21:47:04 <flaper87> I'm leaning a lot towards requiring to have a pool, always
21:47:24 <flaper87> The only benefit of not using pools is that it's easier to deploy a single pool zaqar instance
21:47:42 <flaper87> but, I don't think it's too much of a trouble to create a pool
21:47:49 <flaper87> it's a 1 command config
21:47:57 <kgriffs> right. on the other hand, we can probably write some setup/deploy scripts to automate most of that pain away
21:48:03 <flaper87> kgriffs: +!
21:48:05 <flaper87> kgriffs: +1
21:48:07 <kgriffs> ok
21:48:11 <kgriffs> cool
21:48:17 <kgriffs> so, I guess we talked about two things
21:48:28 <kgriffs> first, making pools always on
21:48:29 <flaper87> (as usual we can't just talk about 1)
21:48:33 <kgriffs> (lol)
21:48:37 <flaper87> (what's wrong with us?)
21:48:43 <kgriffs> second, using that to provide per-pool configs like limits
21:49:03 <flaper87> They can happen in parallel
21:49:12 <flaper87> My concern about the first one is that it'll break devstack
21:49:32 <flaper87> so, here's an idea
21:49:41 <flaper87> what if we update devstack first to use pools
21:49:52 <flaper87> and then we make the "always use pools" change in marconi ?
21:50:31 <flaper87> kgriffs: malini ?
21:50:36 <malini> just make sure to fix devstack if you break it
21:50:45 <kgriffs> makes sense. BTW, we will need to figure out a migration plan for non-pooled deployments
21:50:58 <flaper87> kgriffs: right
21:51:10 <flaper87> do we have a blueprint for this?
21:51:14 <flaper87> I don't think so
21:51:21 <malini> do we need this in j-3?
21:51:24 <kgriffs> I don't think we do for either item
21:51:36 <flaper87> Lets start with devstack and then we'll see where we are at and decide whether we should make this in j-3 or not
21:51:36 <kgriffs> malini: no, this would be kilo i guess
21:52:08 <flaper87> actually, you are right, it's too late in the game
21:52:23 <flaper87> But I'd sill want to see the devstack patch in Juno
21:52:28 <malini> FYI...we have experimental job in devstack-gate, make sure to run tht before any change
21:52:51 <kgriffs> flaper87: great, you just added a third item. :p
21:53:00 <flaper87> :D
21:53:07 * flaper87 is getting better at this
21:53:16 * kgriffs chuckles
21:53:35 <flaper87> ok, we need blueprints for those 2 items
21:53:45 <flaper87> kgriffs: can you create them?
21:53:49 <kgriffs> sure
21:53:50 <flaper87> otherwise I'll do it tomorrow
21:53:53 <flaper87> cool beans
21:54:07 <flaper87> on the flavors note
21:54:16 <flaper87> I still think we should get that work done in Juno
21:54:27 <flaper87> is almost done and we just merged metadata back into the queue endpoint
21:54:42 <flaper87> I'll complete the work there asap so we can review it
21:55:02 <flaper87> #topic Open Discussion
21:55:06 <flaper87> 5min left
21:55:13 <flaper87> vkmc: welcome to the core team ;)
21:55:22 <flaper87> thank you for your hard work
21:55:22 <malini> woot!
21:55:36 <flaper87> and lemme tell you something, you're doomed now ;)
21:56:12 <flaper87> anything you guys want to bring up?
21:56:13 <vkmc> :D
21:56:17 <vkmc> thanks guys
21:56:44 <vkmc> not for the doomed part... but for being so cool with me
21:56:45 <flaper87> You better write some docs on Thursday or there won't be a Como-Lake meetup during Kilo >.>
21:57:01 <flaper87> (and that's for everyone)
21:57:04 <flaper87> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
21:57:09 <kgriffs> heh
21:57:24 <kgriffs> and read up on docbook
21:57:31 <flaper87> +1
21:57:33 <kgriffs> :-o
21:57:40 * vkmc hides
21:58:18 <vkmc> btw, in case somebody didn't saw the mail list... we are having an all-hands doc day for Zaqar on Thursday
21:58:30 <vkmc> any help is more than welcome
21:58:35 <flaper87> vkmc: +1 +1 +1 +1
21:58:55 <flaper87> #link https://review.openstack.org/#/c/111061/ <- coool
21:58:57 <vkmc> flaper87 has gummybears to share
21:59:06 <flaper87> vkmc: wait what?
21:59:13 <flaper87> you should give me gummy bears
21:59:17 <flaper87> not the other way around
21:59:19 <flaper87> >.>
21:59:21 <kgriffs> lol
21:59:23 <vkmc> lol
21:59:25 <flaper87> I'm the adict, you are the dealer
21:59:37 <flaper87> ok guys, time's up
21:59:40 <flaper87> lets call it
21:59:45 * ametts wonders when this group changed from pop tarts to gummy bears
21:59:46 <vkmc> will code for gummybears
21:59:48 <flaper87> have a great rest of the day
21:59:54 <flaper87> #endmeeting