15:05:09 #startmeeting Marconi 15:05:10 Meeting started Tue Dec 10 15:05:09 2013 UTC and is due to finish in 60 minutes. The chair is flaper87. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:05:11 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:05:14 The meeting name has been set to 'marconi' 15:05:20 #link https://wiki.openstack.org/wiki/Meetings/Marconi#Agenda 15:05:28 first of all, I want to make this part of the meeting logs 15:05:37 cpallares: Welcome to the team! 15:05:46 welcome!! 15:05:50 cpallares: yes! welcome! :D 15:06:10 #topic Bug triage 15:06:26 we'll come back to the action-review later? 15:06:29 actually, lets start with actions 15:06:39 #topic Review homeworks 15:06:43 thanks alcabrera, flaper87! 15:07:02 o/ 15:07:03 kgriffs: o/ 15:07:21 We just started reviewing actions. 15:07:29 * kgriffs tries to think of an excuse for being late and fails 15:07:35 alcabrera: kk 15:07:39 * flaper87 didn't get in touch with Everett 15:07:48 #action flaper87 to get in touch with Everett Toews and share ideas about client design 15:08:03 flaper87 to spec out heat integration blueprint 15:08:18 I didn't do the work myself but flwang point me to someone that will take care of it 15:08:30 sweet 15:08:30 Which is definitely good 15:08:38 there's some work going on there already 15:08:57 the only bit missing to officially make that patch land is a release of marocni client 15:09:04 we'll get there before i-2 15:09:07 kk 15:09:12 kgriffs to play with swob 15:09:13 what kinds of templates? 15:09:16 how close would you say we are with marconiclient? 15:09:33 (what will it/they do?) 15:10:20 alcabrera: so, I'm working on the Message object now 15:10:35 which is almost ready but required me to change some things in tests and the code that already landed 15:10:43 (patches in gerrit already) 15:10:46 alright. :) 15:10:50 I'll have plenty to review later. 15:10:51 after that, the only piece missing is claims 15:10:58 which flwang wanted to tackle 15:11:04 flwang: are you still up for that? 15:11:19 btw 15:11:21 #chair kgriffs 15:11:22 Current chairs: flaper87 kgriffs 15:12:08 kgriffs: if you want to take over, feel free. 15:12:10 flaper87: thanks for getting things started 15:12:17 * kgriffs blushes 15:12:21 kgriffs: my pleasure :) 15:12:33 kgriffs to play with swob 15:12:40 so, i didn't get to that yet 15:12:52 but am still planning on doing it soon 15:12:56 #action kgriffs to play with swob 15:13:10 > everyone to get remaining sharding patches merged 15:13:17 kudos to everyone! 15:13:21 poptarts on the house! 15:13:24 #info everyone got the remaining shard patches merged 15:13:25 w000t 15:13:32 :) 15:13:37 * flaper87 wonders who this everyone d00d is 15:13:52 very productive, this "everyone" 15:13:53 > kgriffs to confirm whether or not query params are ok in json-home URI templates 15:13:54 crap 15:13:56 I forgot about that one 15:14:01 * kgriffs blushes again 15:14:03 lol 15:14:07 looool @ crap 15:14:33 #action kgriffs to confirm whether or not query params are ok in json-home URI templates 15:14:45 * flaper87 ran out of gummy bear 15:14:47 :( 15:14:55 kgriffs to re-target bps and bugs to Icehouse milestones 15:14:59 #info flaper87 ran out of gummy bear 15:15:04 lol 15:15:11 heh 15:15:15 #info kgriffs retargeted bps and bugs to icehouse 15:15:15 re: the retargeting 15:15:18 hi 15:15:26 you should all be worried about that, my productivity is in danger 15:15:27 it's pretty much what we're doing in this meeting, isn't it? 15:15:30 #info kgriffs updated milestones with due dates 15:15:32 oz_akan_: hey! 15:15:38 oz_akan_: O/ 15:15:41 mpanetta: o/ 15:15:42 oz_akan_: o/ 15:15:45 hi :) 15:15:45 megan_w: o/ 15:15:48 mpanetta: o/ 15:16:15 alcabrera: somewhat, but i did an initial pass to get everything off havana 15:16:29 gotcha 15:16:32 > everyone to participate in notifications breakout 15:16:40 agreed 15:16:41 we had a great brainstorming meeting 15:16:46 #info many participated in notifications breakout 15:16:48 #link http://eavesdrop.openstack.org/meetings/marconi_notifications_brainstorm/2013/marconi_notifications_brainstorm.2013-12-03-15.05.html 15:16:51 and there's this 'everyone' d00d again 15:16:57 #link https://etherpad.openstack.org/p/marconi-notifications-brainstorm 15:17:07 that was a real good meeting 15:17:07 #action kgriffs to create bp for notifications based on brainstorming session 15:17:27 one more action that wasn't listed - 15:17:42 > alcabrera to look into documentation process 15:17:51 #action alcabrera to look into documentation process 15:18:00 ^^^ what he said 15:18:06 I started looking through the openstack documentation pages yesterday. I also started reaching out to annegentle 15:18:19 I'm awaiting a response while I read through things. 15:18:29 ok, cool. You can give us an update next week 15:18:33 yup! 15:18:39 that's all the actions from last time. 15:19:08 alcabrera: and it better include these words: "I got it all figured out, wrote 100 pages of online documentation and even published a book!" 15:19:18 LOOOOOOOOOOOOOOL 15:19:27 especially the 500 pages part 15:19:28 kgriffs: ooohh, very tempting. I *have* been meaning to publish by the end of this year. Something. :D 15:19:30 * kgriffs is channeling Anne 15:19:33 I'm not sure if y'all want to know, but I have moved my "marconi-simple" docker container to a trusted build. It is building as we speak. 15:19:35 hahaha 15:19:51 mpanetta: gtk! 15:19:59 mpanetta: awesome! 15:20:05 #info mpanetta has a marconi-simple docker container available 15:20:13 Link? 15:20:13 mpanetta: I'd like to play with that 15:20:20 mpanetta: do you have a template for it or something? 15:20:24 mpanetta: how can we get our hands on said dockerfile? 15:20:24 yes 15:20:25 * flaper87 is not a docker expert 15:20:26 sec 15:20:38 https://github.com/bentwire/marconi-simple.git 15:20:45 #link https://github.com/bentwire/marconi-simple.git 15:21:09 And as soon as the build is done I should have a docker repo link as well 15:21:23 The automated build system must have quite the backlog... 15:21:25 mpanetta: +1 15:21:30 cool, cool 15:22:13 kudos to mpanetta 15:22:29 triage, triage, triage, triage 15:22:32 #info marconi is officially cool now that it has a docker file 15:22:41 mooooving on 15:22:41 LOL 15:22:45 LOLOLOLOLO 15:23:07 #topic Review Graduation BPs/Bugs 15:23:22 #link https://wiki.openstack.org/wiki/Marconi/Incubation/Graduation 15:23:43 ah crap. I need to do the 1.1 spec thing 15:23:56 #action kgriffs to write v1.1 api spec 15:24:12 we have almost exactly two months worth of time to wrap up all pending items 15:24:20 yes, not much time 15:24:25 indeed! 15:24:41 the most critical one that is still pending is the sqlalchemy thing 15:24:54 definitely 15:24:54 ykaplan already submitted the first patch 15:25:12 and she started to work on the controllers 15:25:15 I wonder... 15:25:24 I guess we should help her a bit 15:25:28 #link https://blueprints.launchpad.net/marconi/+spec/sql-storage-driver 15:25:29 by splitting the work there 15:25:32 If I wrap up marconi-redis, would that count as having finished that requirement? 15:25:54 don't think so! 15:26:00 heh. :P 15:26:03 :D 15:26:12 but you should wrap it up anyway 15:26:18 anyway, +1 towards helping sqlalchemy happen. :D 15:26:20 yes, you should. ;) 15:26:20 will do 15:26:43 #info sqlalchemy is of critical priority 15:26:48 alcabrera: will you be able to enforce FIFO + once-and-only-once delivery? 15:27:02 oaoo delivery 15:27:07 (we had talked about relaxing that requirement to be specific to the storage driver) 15:27:11 kgriffs: yes. 15:27:19 I think I can even make it a configurable thing. 15:27:24 interesting 15:27:27 [...:redis] fifo = True 15:27:29 can't wait to see the implementation 15:27:45 I wouldn't add that config option yet 15:27:47 so, back to sqlalchemy 15:27:52 lets wait until the API layer is done 15:28:02 and we can expose what features are enabled 15:28:04 and what not 15:28:10 next steps to get SQL support? 15:28:20 review the tables patch 15:28:25 start working on the controller 15:28:32 #link https://review.openstack.org/#/c/60223/ 15:28:34 controllers* 15:28:42 queue, message, claims 15:28:46 needs 3 controllers ^^ 15:28:50 ok, if we have one more person volunteer to help 15:28:51 and two more for sharding 15:28:54 alcabrera: you could work on the sharding part f it 15:28:58 :D 15:29:06 then they can divide up controllers with Yeela 15:29:06 I volunteered alcabrera 15:29:08 :D 15:29:13 I'm tied up 'til January, but after that, I'd love to. :) 15:29:29 alcabrera: i was hoping you'd say that :) 15:29:31 sounds good! 15:29:36 sharding should "just work" shouldn't it? 15:29:36 megan_w: LOL 15:29:41 #info alcabrera volunteers to work on sqlalchemy/sharding controllers for January 15:29:56 kgriffs: yes, with one caveat 15:30:03 As it stands now 15:30:12 People would still have to deploy a mongodb-based catalogue 15:30:24 so~ 15:30:41 sqlalchemy/{shards,catalogue}.py need to be implemented to meet the grad req 15:30:49 but the sharding is optional 15:31:02 hmmm 15:31:07 so, as far as the requirement is concerned, having support for the storage controllers it's enough 15:31:16 IMHO 15:31:22 if sharding is optional, then those controllers are optional. I'm worried about the wording of the req "real production" 15:31:45 alcabrera: don't worry, we can get rid of that 15:31:48 :D 15:31:49 kk 15:31:49 :D 15:31:50 lol 15:31:51 :D 15:31:54 I'm joking 15:32:14 mmh, that's an interesting point though 15:32:41 because the "real production" environemnt needs you'd have with mongo are not necesarilly the same you'd have with mysql 15:32:50 you may actually not need on or the other 15:33:11 "Real production" doesn't necessarily mean "public cloud" or "massive deployment" 15:33:21 which is where sharding comes in handy 15:33:22 exactly 15:33:41 alright 15:33:41 we should try to make it happen, but I think it could slip into i-2 if need be 15:33:49 that's why I think that having support for the storage controllers would indeed cover that graduation requirement 15:33:59 alcabrera: can you create a blueprint for doing that for sharding? 15:34:15 "that" == add sqlalchemy sharding controllers? 15:34:17 kgriffs: ^^ 15:34:22 alcabrera: yes 15:34:27 will do! 15:34:30 thanks! 15:34:40 #action alcabrera to draft BP for sqlalchemy sharding controllers 15:34:45 ok, so everyone go review that table patch! 15:34:45 awesome 15:35:12 alcabrera: just assign it to graduation milestone for now. I'll set priority and mark as "accepted" 15:35:22 alcabrera: ping me when you have it ready. Thanks! 15:35:25 kk :) 15:35:51 moving on 15:35:53 #link https://blueprints.launchpad.net/python-marconiclient/+spec/python-marconiclient-v1 15:36:07 let me set that milestone to graduation 15:36:24 I don't think there's one 15:36:25 heh 15:36:27 :D 15:36:33 looks like the client doesn't have teh milestones 15:37:08 flaper87: would you mine creating an icehouse release and some milestones for the client? 15:37:14 s/mine/mind 15:37:35 Yes sir 15:38:05 btw, I've an off-topic thing to say about the client milestones. Will do that offline 15:38:13 kk 15:38:22 kgriffs: https://blueprints.launchpad.net/marconi/+spec/sql-sharding-controllers 15:38:26 also 15:38:27 #link https://blueprints.launchpad.net/marconi/+spec/sql-sharding-controllers 15:39:43 alcabrera: thanks! updated. 15:39:53 awesome 15:40:00 https://blueprints.launchpad.net/marconi/+spec/heat-template 15:40:10 should we change the status to "good progress"? 15:40:26 #link https://blueprints.launchpad.net/marconi/+spec/heat-template 15:40:34 do we need to reassign? 15:40:36 kgriffs: slow progress 15:40:46 or blocked? 15:40:53 blocked by client lib? 15:41:01 Yup, that sounds about right 15:41:37 +1 for Blocked 15:42:07 kgriffs: I reassigned it to liang 15:42:10 kk 15:42:27 I guess the assignee doesn't have to actually do the work, but is responsible for reporting status 15:42:36 anyway, not a big deal 15:42:50 #link https://blueprints.launchpad.net/marconi/+spec/devstack-support 15:43:07 implemented 15:43:13 extra-fresh box of pop-tarts for flaper87! 15:43:18 w00000t 15:43:27 and two tickets to go see that band he loves 15:43:39 * flaper87 shares with everybody 15:43:54 srsly, thanks for getting that done 15:44:05 we are getting official-er all the time! 15:44:11 btw 15:44:27 something raised in that review, which I agree with, is that we picked a weird port number 15:44:29 :) 15:44:39 it's actually gunicorn's default port 15:44:52 There are weird port #'s? 15:45:08 port 8000 seems like the most normal port in the world to me. ;) 15:45:13 * alcabrera uses gunicorn all the time 15:45:14 so, if we'll ever think about changing it, we better do it before graduation 15:45:18 oh 15:45:31 were they expecting, e.g., 80? 15:45:50 I think 53 15:45:54 because we hat DNS 15:45:56 hate 15:46:03 lol 15:46:03 heh 15:46:10 why not 42? 15:46:13 * kgriffs likes 42 15:46:19 * cpallares likes 42 too 15:46:19 solution to everything 15:46:21 :D 15:46:23 the answer to everything, kgriffs? :P 15:46:26 heh 15:46:31 haha 15:46:34 erm, answer is the right word 15:46:36 :D 15:46:44 anyway, I'm happy with 8888 15:46:56 ok, whatever they like 15:46:56 it's weird but you know, we're all weird after all 15:47:06 we are not using pecan 15:47:07 * kgriffs thinks people are wierd 15:47:12 we're doing QaaS 15:47:17 we've the funiest channel ever 15:47:18 :D 15:47:28 :D 15:47:29 We couldn't be weirder than this 15:47:37 what? 15:47:39 Oh surely we could. 15:47:40 * kgriffs looks innocent 15:47:51 LOOL 15:47:53 haha 15:48:05 moving on 15:48:09 ;) 15:48:12 #info marconi-team has no shortage of weirdness 15:48:19 * mpanetta looks at /etc/services and sees an unused port at 70 :P 15:48:23 yup - triaging things next? :) 15:48:25 #link https://blueprints.launchpad.net/marconi/+spec/tempest-integration 15:48:29 Nobody uses gopher anymore right? :P 15:48:30 tempest 15:48:43 malini isn't here yet today 15:48:47 booh 15:48:50 I heard that she's ramping up on that 15:48:59 started? 15:49:06 I believe so. 15:49:12 * kgriffs updates status 15:49:32 #link https://blueprints.launchpad.net/marconi/+spec/pecan-framework 15:49:39 still on my todo list 15:49:48 anyone is welcome to help 15:50:07 * kgriffs is biased, after all ;) 15:50:11 :P 15:50:27 What is wrong with falcon? 15:50:27 I'll see what I can do come January. 15:50:30 srsly. Nice guys. Met the maintainer and original author in HK 15:50:49 * mpanetta is curious 15:50:50 but... we have some philosophical differences, I must admit 15:50:58 mpanetta: pecan is the current OS-blessed wsgi framework, so it's preferred tha projects implement their wsgi layer using pecan. 15:51:02 * kgriffs is a minimalist 15:51:03 *that 15:51:13 oic 15:51:17 thanks alcabrera 15:51:20 np. :) 15:51:41 for the record, WSME is not as blessed 15:51:54 mainly because it falls on it's face when it comes to API extensions 15:52:03 15:52:46 8 mins left 15:52:49 next? :) 15:53:00 hmmm 15:53:14 https://blueprints.launchpad.net/marconi/+spec/pecan-framework 15:53:18 LOL 15:53:19 oops sorry 15:53:27 I thought you were trolling us 15:53:29 * kgriffs freudian slip 15:53:29 :D 15:53:31 lol 15:53:34 #link https://blueprints.launchpad.net/marconi/+spec/marconi-horizon-integration 15:53:54 mmh 15:53:55 who is working on that? who would work on that? 15:53:57 that's compeltely stalled 15:54:03 yeah 15:54:29 It'd be nice to have someone from the community to work on that 15:54:40 i don't see it happening soon unless we get a new contributor 15:54:41 FWIW, I don't think it's a strong requirement for graduation 15:54:50 yeah, hence i-3 15:54:59 the wording says "nice to have", on the grad reqs page 15:55:01 so I agree 15:55:32 plus, it can go into horizon project, so as long as it is done before the freeze, we are good 15:55:38 #info Marconi needs eager Horizon volunteer to make Horizon-integration happen 15:55:48 and this is assuming we become integrated this cycle 15:56:11 true 15:56:20 I'm not totally confident that infra has the bandwidth to add us in 15:56:21 what's the freeze?? 15:56:23 (this time around) 15:56:32 cpallares: there is a feature freeze just before each release 15:56:40 #link https://wiki.openstack.org/wiki/Icehouse_Release_Schedule 15:56:42 If we don't get integrated this cycle, it's fine as well 15:56:43 cpallares: ^^^ 15:57:02 flaper87: yeah 15:57:08 but, I want us to still strive for it 15:57:18 you know, student syndrome 15:57:24 I'd rather make sure we're full-featured before being integrated than graduating without having a good support 15:57:29 let's get a "1.0" thing done sometimes this decade. :D 15:57:33 kgriffs: +1 15:57:43 heh, yeah... 1.0. :D 15:57:48 we'll get there 15:58:06 1.5 minutes 15:58:12 ok, we hit the essential bps 15:58:26 #topic open discussion 15:58:30 any closing thoughts? 15:58:32 first thing from me, real quick 15:58:41 * alcabrera gives mic to kgriffs 15:58:46 each mtg I'd like to quickly review in-flight bps 15:58:52 seems to be a pattern followed by other teams 15:58:57 +1 15:58:58 second thing 15:59:04 wait, back on first thing 15:59:11 * kgriffs confuses self 15:59:31 so, if you are assigned a bp or bug, be prepared to defend yourself. I mean, explain youself. 15:59:35 self = self.confused_self 15:59:38 i mean, give an update 15:59:41 :D 15:59:44 lol 15:59:45 second thing 16:00:06 if you register a bp I am going to ignore it until you set a milestone. 16:00:09 kgriffs: all done? 16:00:19 adrian_otto: wrapping up 16:00:36 ok folks, let's let solum folks do their thing 16:00:42 o/ 16:00:43 cool 16:00:50 #endmeeting