15:05:51 <kgriffs> #startmeeting marconi
15:05:52 <openstack> Meeting started Mon Dec  2 15:05:51 2013 UTC and is due to finish in 60 minutes.  The chair is kgriffs. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:05:53 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:05:55 <openstack> The meeting name has been set to 'marconi'
15:06:13 <kgriffs> #topic review actions from last time
15:06:43 <kgriffs> #link http://eavesdrop.openstack.org/meetings/marconi/2013/marconi.2013-11-18-16.09.html
15:07:06 <kgriffs> flaper87: you're up first
15:07:12 <flaper87> damnit
15:07:28 <kgriffs> I can't remember what you were supposed to talk to him about...
15:07:56 <flaper87> kgriffs: you shouldn't be surprised by the fact that I don't remember it either
15:07:59 <flaper87> :D
15:07:59 * kgriffs vows to write better action items in the future
15:08:24 <flaper87> kgriffs: ahhh
15:08:27 <flaper87> I think I remember
15:08:31 <flaper87> it's about the API
15:08:42 <flaper87> and the specs, extensions and versioning
15:08:48 * flaper87 reads the logs
15:08:54 <flaper87> anyway, I haven't so, we can move on
15:08:59 <flaper87> lets re-add the action item
15:09:12 <flaper87> no, it was about the sdk
15:09:13 <kgriffs> maybe JMS?
15:09:13 <flaper87> and the client
15:09:15 <flaper87> :D
15:09:17 <kgriffs> ah
15:09:19 <kgriffs> that's right
15:09:26 <kgriffs> sharing ideas about client design
15:09:34 <flaper87> so, my memory works after all
15:09:40 <flaper87> I just need to fix my brain's indexes
15:09:47 <kgriffs> #action flaper87 to get in touch with Everett Toews and share ideas about client design
15:10:06 <kgriffs> Alej isn't here
15:10:12 <flaper87> :(
15:10:13 <kgriffs> amitgandhi: ?
15:10:36 <kgriffs> do you know if Alej is going to be available this week?
15:10:38 <amitgandhi> informed him
15:10:41 <amitgandhi> he is signing in
15:10:44 <kgriffs> ah
15:10:54 <amitgandhi> he didnt know the 10am change (EST)
15:11:15 <kgriffs> sorry about the calendar invite - I updated an instance, not the recurring, so that is probably why he didn't know
15:11:20 <alcabrera> o/
15:11:34 <kgriffs> alcabrera: ni hao!
15:11:44 <alcabrera> I miscalculated the UTCs while catching up on 300+ emails. :P
15:11:50 <alcabrera> Thought I had one more hour, heh.
15:11:57 <kgriffs> no worries
15:12:03 <kgriffs> have you had a chance to talk with the heat team?
15:12:29 <alcabrera> no
15:12:54 <kgriffs> alcabrera: do you have bandwidth to do that this week?
15:13:05 * alcabrera considers this
15:13:06 * kgriffs upgrades alcabrera to 10g
15:13:16 <flaper87> alcabrera: HELLOOOOO
15:13:17 * alcabrera considers with greater bandwidth
15:13:20 <alcabrera> :D
15:13:23 <alcabrera> flaper87: heeeey!
15:13:34 <amitgandhi> kgriffs: probably not
15:13:39 <kgriffs> kk
15:13:55 <amitgandhi> we can maybe look at adding it to the next sprint
15:13:56 <kgriffs> anyone else interested in taking this action item?
15:14:10 <kgriffs> amitgandhi: when does that start?
15:14:13 <alcabrera> yeah, I'm pretty full up on wrapping some things up
15:14:18 <flaper87> kgriffs: I can re-take it and dig a bit more
15:14:32 <flaper87> I'm not sure I'll be able to do anything significant on it, though
15:14:37 <flaper87> (like real coding)
15:14:43 <kgriffs> flaper87: ok. You have some Heat contacts in Red Hat, right?
15:14:43 * alcabrera gives flaper87 the heatz and a pop tart
15:14:49 <amitgandhi> i think it should be starting on thursday (meeting invites are messed up due to holiday)
15:14:58 <flaper87> but at least I can take some time to figure out what we need to do and write an etherpad
15:15:03 <kgriffs> flaper87: we just need someone to spec out the work
15:15:09 <flaper87> kgriffs: I do
15:15:14 <kgriffs> kk
15:15:19 <flaper87> kgriffs: PTL + several devs
15:15:23 <kgriffs> sweet
15:15:29 <alcabrera> thanks, flaper87!
15:15:39 <flaper87> alcabrera: my pleasure :)
15:16:15 <kgriffs> #action flaper87 to spec out heat integration blueprint
15:16:31 <kgriffs> #action kgriffs to play with swob
15:16:41 <kgriffs> I still need to do that
15:16:54 * flaper87 wonders why is it that kgriffs is always playing
15:17:03 * flaper87 hides
15:17:16 <alcabrera> lol
15:17:31 <flwang> I can do some work about heat as well maybe
15:17:43 <flaper87> flwang: awesome, thanks a lot!
15:17:44 <flwang> my colleague is heat core member
15:18:08 <flaper87> flwang: if you've some bandwidth then I can hand that action over to you
15:18:12 <flaper87> :D
15:18:30 <flwang> I'm waiting for some low-hanging-fruit :)
15:19:04 <flaper87> flwang: no time for that, lets attack dinosaurs
15:19:05 <kgriffs> #action flwang to work with flaper87 to figure out what templates we want to do for marconi
15:19:10 <flaper87> :D
15:19:17 <flwang> so it would be nice if there is any good man tell me how big it's
15:19:36 <flaper87> it's like this big: | |
15:19:53 <flwang> =-O
15:20:05 <kgriffs> flwang: I imagine we will want a template to create some queues, also one to deploy Marconi itself
15:20:11 <flwang> i'm just a little boy :)
15:20:22 <flwang> kgriffs: i see.
15:20:27 <flwang> thanks for the clarification
15:20:27 <flaper87> flwang: jokes apart, it'll require some work because we need to create templates for it and we need to make them sqs compliant as well
15:20:32 <kgriffs> re deploying marconi, I will put you in touch with Oz who architected Rackspace's deployment for some tips
15:20:58 <flaper87> Oz++
15:21:42 <kgriffs> k
15:21:56 <kgriffs> let's see what else
15:22:16 <kgriffs> alcabrera to work on sharding feedback and get it approved
15:22:53 <alcabrera> that's still in progress
15:23:05 <alcabrera> I'll give it some proper attention today, now that I'm refreshed with a much needed break.
15:23:07 <kgriffs> the main patch got merged, tho, right?
15:23:08 <alcabrera> :)
15:23:15 <flaper87> remaining patches are about to be merged
15:23:18 <alcabrera> yeah, patch #1 in that sequence
15:23:23 <kgriffs> ok
15:23:28 <flaper87> I just had one last tiny comment in one of Zyuan patches
15:23:28 <alcabrera> afaik, queue listing and caching still need to get in
15:23:37 <flaper87> and I just saw he worked on that
15:23:40 <kgriffs> let's get the queue listing and caching ones merged
15:23:44 <flaper87> so, I guess we can get them in today
15:23:50 <kgriffs> I need to rebase the cache ones
15:24:23 <alcabrera> I'll take a look at those patches after this meeting
15:24:27 <kgriffs> amitgandhi: is Zhihao in teh house?
15:24:38 <kgriffs> alcabrera: thanks!
15:25:26 <alcabrera> kgriffs: zyuan is here
15:25:30 <kgriffs> #action everyone to get remaining sharding patches merged
15:25:33 <alcabrera> he's occupied at the moment, though.
15:25:36 <kgriffs> kk
15:26:01 <kgriffs> moving on
15:26:04 <kgriffs> #topic Future of the home doc
15:26:21 <kgriffs> So, json-home RFC is alive and well, according to mnot
15:26:35 <kgriffs> he is working on a new draft to replace 03, which has expired
15:26:40 <flaper87> kgriffs: btw, dunno if you went through openstack-dev emails
15:26:47 <flaper87> I raised this issue upstream: http://lists.openstack.org/pipermail/openstack-dev/2013-November/020289.html
15:26:49 <kgriffs> flaper87: not yet since I got back
15:27:33 <kgriffs> ah, cool
15:27:33 <flaper87> and since keystone guys are using a similar approach, I asked them what would be the effort needed to use json-home there as well
15:27:37 <flaper87> instead of what they're using
15:27:57 <flaper87> they agreed on shipping it along with what they have for now and see how it works
15:28:12 <flaper87> obviously, I don't think it'll have a priority higher than 'underground'
15:28:20 <flaper87> but at least we got json-home 'out there'
15:28:28 <kgriffs> ok, can you nudge them towards Mark N. ?
15:28:32 <flaper87> and some folks agree that it's the right thing to use
15:28:37 <kgriffs> I think we he would love the feedback
15:28:43 <flaper87> kgriffs: I did
15:28:47 <kgriffs> rock on
15:28:55 <flaper87> and also the other way around
15:28:59 <flaper87> Mark -> openstack-dev thread
15:29:04 <kgriffs> cool
15:29:18 <flaper87> so, json-home FTW
15:29:21 <kgriffs> The standard will be more likely to succeed if it is based on real-world needs/usage
15:29:38 <kgriffs> kk
15:29:42 <kgriffs> related: bug
15:29:49 <flaper87> also, I noticed we're defining query params in our home document
15:29:50 * kgriffs looks that up
15:30:03 <flaper87> the RFC explicitly says that query params are optional and shouldn't be part of the url
15:30:11 <flaper87> at least I thik I read that
15:30:27 <flaper87> I'd like to have a way to specify custom query params and headers, though.
15:30:45 <kgriffs> flaper87: hmm. Let me check on that. I probably just read that is was a URI template and assumed any part of URI templates was fair game
15:31:12 <kgriffs> flaper87: we may want to add extensions to json-home if we can't convince Mark to add them to the spec
15:31:30 <kgriffs> flaper87: as long as they are extra fields, not changes to fields, should be OK
15:31:50 <kgriffs> but, if we need them, other people probably do to, so we should lobby for them to go into the spec
15:32:23 <flaper87> yeah, exactly
15:32:28 <kgriffs> #action kgriffs to confirm whether or not query params are ok in json-home URI templates
15:32:48 <flaper87> I got the feeling that the rFC has a strong opinion about query params
15:33:02 <flaper87> and the fact that they shouldn't be there
15:33:08 <flaper87> We could aregue that
15:33:10 <flaper87> argue
15:33:12 <kgriffs> ok, I will follow up
15:33:52 <kgriffs> so, related: https://bugs.launchpad.net/marconi/+bug/1245656
15:34:09 <kgriffs> I added a comment to the bottom of the bug
15:34:09 <flaper87> kgriffs: cool, thanks
15:34:37 <kgriffs> Mr. Vollero in teh house?
15:34:53 <flaper87> kgriffs: don't think so
15:35:12 <kgriffs> flaper87: do you think he would cry if we reassign?
15:35:21 * kgriffs doesn't like to make people cry
15:35:23 <flaper87> kgriffs: don't think so
15:35:26 <kgriffs> ok
15:35:31 <kgriffs> flwang: interested in picking this one up?
15:35:47 <flwang> yep, sure
15:35:52 <kgriffs> w00t
15:35:56 <kgriffs> thanks, man
15:36:07 <kgriffs> #link http://paste.openstack.org/show/53559/
15:36:11 <flwang> thank you :)
15:36:23 <kgriffs> flwang: ^^^ That should explain the problem and solution pretty well
15:36:59 <kgriffs> hmmm. Paste seems to be slow today
15:37:01 <flwang> kgriffs: got, but seems it's owned by Francesco Vollero already
15:37:32 <kgriffs> reassigned. :D
15:38:08 <kgriffs> #action kgriffs to re-target bps and bugs to Icehouse milestones
15:38:13 <flwang> I assume i'm not hijacking
15:38:24 <kgriffs> nope
15:38:30 <flwang> :P
15:38:37 <kgriffs> no worries
15:39:04 <kgriffs> #topic Doc generation
15:39:37 <kgriffs> so, just wanted to get everyones thoughts on the priority for getting at least basic marconi docs going
15:39:46 <flaper87> +1
15:39:48 <alcabrera> +1
15:40:01 <alcabrera> I'd love to get those going.
15:40:09 <flaper87> so, I think we should document base classes extensibly and build docs out of that
15:40:12 <alcabrera> For whatever tiny amount of bandwidth I have available this week.
15:40:32 <flaper87> Mark did that with oslo.messaging and docs came out pretty cool
15:40:37 <kgriffs> ok
15:40:47 <flaper87> also, re auto-generate sample config files
15:40:53 <flaper87> that's blocked by this bp: https://blueprints.launchpad.net/oslo/+spec/improve-config-discovery-for-docs
15:41:02 <flaper87> because we're not using global configs
15:41:05 <kgriffs> alcabrera: can you create a blueprint and assign it to a milestone and then I will prioritize it?
15:41:13 <alcabrera> flaper87: I'd like to take it a step further than that. At least a front page introducing the marconi docs, something I'd feel good about reading on RTD if I knew nothing about marconi. :D
15:41:18 <alcabrera> kgriffs: will do
15:41:22 <kgriffs> there may already be a bp, but it is probably really old and could use some TLC
15:41:41 <kgriffs> alcabrera: there is an "open stack way" to generating docs iirc
15:41:44 <flaper87> alcabrera: oh, cool!
15:41:50 <kgriffs> I'm not familiar with the details
15:41:50 <flaper87> kgriffs: correct
15:42:03 <flaper87> kgriffs: classes + sphix and upload it to docs.openstack.org
15:42:10 <alcabrera> there's three doc-related BPs atm: docstring-catchup, docstring-add-types, http-redirects-for-docs
15:42:13 <alcabrera> I'll add one more
15:42:21 <kgriffs> kk
15:42:41 <kgriffs> alcabrera: you can ping Anne Gentle - she is docs PTL, isn't she?
15:42:41 <flwang> any plan to add the api details to api site?
15:42:51 <alcabrera> kgriffs: that's correct
15:43:00 <alcabrera> I'll check with her before I run off and start anything. :)
15:43:04 <kgriffs> alcabrera: kk, she's a Racker so should be easy to lookup her contact info
15:43:17 <alcabrera> cool
15:43:17 <flwang> kgriffs: like http://api.openstack.org/api-ref.html
15:43:19 <kgriffs> alcabrera: excellent, thanks!
15:43:47 <kgriffs> flwang: it would be great to get on that page; we may not be allowed to do that until we graduate, however
15:43:58 <kgriffs> alcabrera: ^^^ I guess you can ask about that
15:44:02 <flwang> kgriffs: got
15:44:27 <kgriffs> we can put some stuff on readthedocs in the meantime
15:44:44 <alcabrera> kgriffs: adding a note on the BP I'm drafting to address the API page. :)
15:44:50 <alcabrera> flwang: thanks for bringing that up!
15:44:51 <flaper87> kgriffs: mmh, I'm not sure about RTD
15:44:53 <flaper87> tbh
15:45:12 <flaper87> I'd prefer to wait a bit more before uploading it there
15:45:33 <flaper87> I don't want people to be confused as to where the docs are and stuff
15:46:12 <kgriffs> flaper87: good point. I just thought of it because ceilometer did that
15:46:28 <flaper87> kgriffs: oh, mmh, interesting
15:46:30 <flaper87> I didn't know that
15:46:31 <kgriffs> well, first step is to get them generating
15:46:35 <flaper87> yeah
15:46:40 <kgriffs> we can worry about publishing later
15:46:40 <flaper87> and then we'll figure out what to do
15:46:48 <alcabrera> #link https://blueprints.launchpad.net/marconi/+spec/basic-docs
15:46:53 <flwang> may be it's stupid, is the rst dev doc including?
15:46:54 <alcabrera> drafted
15:47:42 <flaper87> flwang: not sure what you mean :(
15:47:57 <kgriffs> amitgandhi: can you see if there is time in your next sprint to work this doc stuff in?
15:48:09 <kgriffs> if not, we can shoot for first thing in january
15:48:12 <flaper87> kgriffs: I'd leave the remaining topics for next week, we've got 10mins left
15:48:17 <flwang> the dev docs associated with source code
15:48:24 <flaper87> and I don't want to rush those discussions
15:48:27 <flaper87> :D
15:48:27 <amitgandhi> its unlikely to get in
15:48:33 <kgriffs> flaper87: ah, come on. Don't you want to have a 3 hour meeting?
15:48:36 <flaper87> concurrency and notifications, that sounds like liters of coffee
15:48:48 <flaper87> kgriffs: ah sure, in that case, lets keep it going
15:48:50 <flwang> flaper87: just like https://github.com/openstack/glance/tree/master/doc
15:48:51 <flaper87> :D
15:49:05 <amitgandhi> but we can discuss at planning...
15:49:09 <flaper87> flwang: ah yeah, that's what this bp is about
15:49:23 <flaper87> writing RSTs
15:49:25 <flaper87> code docs
15:49:28 <flaper87> and generating those
15:49:39 <kgriffs> amitgandhi: ok, the docs stuff isn't super high priority, but I would like to have it done sometime by the end of Jan
15:49:46 <kgriffs> alcabrera: ^^^
15:49:48 <flwang> flaper87: cool, got it
15:49:57 <amitgandhi> ok sure , we'll make sure its in the backlog
15:50:01 <alcabrera> cool
15:50:03 <kgriffs> thanks!
15:50:27 <alcabrera> It's critical for the sake of wide-spread marconi advocation efforts, but I know other things have higher priority atm
15:50:46 <alcabrera> (It's -> basic docs)
15:51:35 <flwang> alcabrera: +1
15:52:34 <kgriffs> alcabrera: I will make it a high priority for I3, If we get it done in I2, great!
15:53:02 <alcabrera> sounds good, kgriffs.
15:53:57 <kgriffs> amitgandhi: notifications discussion, is that something you are OK waiting on until next week?
15:54:22 <amitgandhi> i would like to discuss this week on #marconi if possible
15:54:26 <kgriffs> ok
15:54:33 <flaper87> amitgandhi: lets set a day and hour
15:54:38 <flaper87> kgriffs: alcabrera ^
15:54:42 <amitgandhi> we need to start nailing down some parts of the api so that some internal stuff mirrors it
15:54:48 <kgriffs> this channel isn't scheduled for several hours, so we could do it now if everyone has time
15:54:57 <kgriffs> otherwise, we will need to reschedule
15:55:11 <amitgandhi> i have time
15:55:15 <flaper87> kgriffs: lets reschedule, I've got a call in 5
15:55:18 <flaper87> :(
15:55:22 <kgriffs> ok
15:55:35 <flaper87> tomorrow
15:55:35 <kgriffs> flaper87: do you have time today or should we try for tomorrow?
15:55:44 <flaper87> same time ?
15:55:49 <alcabrera> +1 for tomorrow
15:56:30 <alcabrera> same time works for me
15:56:41 <alcabrera> that being.... 1500 UTC?
15:56:47 <flaper87> alcabrera: correct
15:57:07 <alcabrera> yup, works for me
15:57:11 <alcabrera> kgriffs, amitgandhi: ^
15:57:30 <amitgandhi> sure
15:57:41 <flaper87> awesome
15:58:02 <kgriffs> ok, let's do that
15:58:37 <kgriffs> #action everyone to participate in notifications breakout tomorrow, Dec 3, at 1500 UTC in #openstack-meeting-alt
15:58:46 <kgriffs> #topic open discussion
15:58:52 <flaper87> o/
15:59:14 * kgriffs calls on flaper87
15:59:16 <flaper87> I just wanted to share this marconi_ws hack I worked on yday: https://github.com/FlaPer87/marconi-websocket
15:59:37 <flaper87> It's a POC for websocket and it currently just proxies our public wsgi API
15:59:46 <alcabrera> I've had it sitting in a browser tab all this morning. Great work, flaper87!
16:00:08 <amitgandhi> nice!
16:00:16 <flaper87> I'll now clean it up and it'll be ready to be used as another transport for Marconi
16:00:24 <flaper87> :)
16:01:09 <flaper87> gtg guys, I'll share more details in a bit!
16:01:11 <kgriffs> groovy
16:01:13 <flaper87> great meeting
16:01:21 <kgriffs> #endmeeting