15:00:24 #startmeeting Marconi 15:00:26 Meeting started Tue Jan 28 15:00:24 2014 UTC and is due to finish in 60 minutes. The chair is alcabrera. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:27 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:29 The meeting name has been set to 'marconi' 15:01:03 alcabrera: hi 15:01:11 This brings us to the 38th meeting of the marconi team. We've come a long way. Let's keep moving forward! :D 15:01:24 balajiiyer: o/ 15:01:39 #topic Review Actions 15:01:56 o/ 15:02:38 #info flwang to get suggestions for KPIs for health from ops folks 15:02:50 how's that coming along, flwang? 15:03:08 I will do that today, after the meeting, sorry for the late 15:03:16 no worries. :) 15:03:35 I'll just carry it forward 15:03:39 #action flwang to get suggestions for KPIs for health from ops folks 15:03:57 alright, what's the word on the Heat patch, flwang? 15:04:39 Liang had synced with flaper87|afk 15:04:48 ah, okay 15:04:56 I was not in the chat 15:04:56 In that case, I'll have flaper87|afk update us when he returns 15:05:01 Thanks! 15:05:10 #action flaper87 to update us on state of Heat patch 15:05:24 #info alcabrera to work with ykaplan and flaper87 to come up with a plan to reboot the SQL driver work 15:05:45 I've put no time into making this happen this week. :/ 15:06:01 ykaplan and flaper87|afk aren't present at the moment, so I'll carry this forward. 15:06:10 #action alcabrera to work with ykaplan and flaper87 to come up with a plan to reboot the SQL driver work 15:06:16 and last of the actions - 15:06:26 malini to follow up with infra after i-2 lands 15:07:13 malini is not present either - so we'll carry that forward 15:07:20 #action malini to follow up with infra after i-2 lands 15:07:37 kgriffs: hey! 15:07:40 o/ 15:07:41 :D 15:07:48 looks like flaper87|afk is mia 15:08:05 who else is here for marconi? 15:08:41 I've been reviewing action items thus far with at least balajiiyer and flwang. 15:08:44 kgriffs: Im here 15:08:49 :) 15:08:51 o/ im here as a fly on the wall 15:08:52 #topic 15:08:54 oops 15:08:59 #chair kgriffs 15:08:59 Current chairs: alcabrera kgriffs 15:09:05 you now have the power. :) 15:09:08 o/ 15:09:17 ok, thanks guys. I think something is messed up with my calendar - I keep missing the reminder. 15:09:19 aaaaanyway 15:09:37 alcabrera: go ahead and finish up with the current topic 15:09:45 sure thing 15:09:57 that's all for previous action items. 15:10:05 Next up: graduation BPs 15:10:09 kgriffs: would you mind figuring out again the operations guys I can talk to get the PKI of /health? 15:11:29 flwang: sure 15:11:41 let's discuss that right after this mtg 15:11:54 kgriffs: cool 15:12:30 #topic review graduation bps 15:12:53 #link https://blueprints.launchpad.net/marconi/+spec/graduation 15:13:17 first up: heat-template 15:14:03 so, it looks like we have a heat resource now! https://review.openstack.org/#/c/61036/ 15:14:15 nice! 15:14:35 #info Marconi now has a heat resource 15:14:56 Is it still the case that Heat progress is blocked on the client lib? 15:15:08 nope, we can update that bp 15:15:16 cool 15:15:24 * kgriffs does that 15:15:28 thanks! 15:16:03 so, I guess we don't have a heat template to actually deploy a marconi service itself, but i think that is fine for now 15:16:22 i think heat is more geared towards provisioning resources 15:16:36 o/ 15:16:54 what does the heat template actually do then? 15:17:02 looks like it provisions queues 15:17:24 hmm 15:17:32 https://blueprints.launchpad.net/heat/+spec/mqaas-marconi-resource 15:17:56 it sounds like it'd be pretty awesome to expand that at some point to yield a sharded deployment template, or something to create a marconi shard 15:18:24 that's a nice-to-have, though, atm, imho 15:18:28 alcabrera: feel free to register a new bp for that - we can discuss later 15:18:35 sure thing, kgriffs 15:18:38 thanks! 15:18:49 #topic pecan-framework 15:19:00 balajiiyer: can u give us an update? 15:19:19 kgriffs: In progress, starting to document the findings. 15:19:54 I should be able to spend sometime on it this week and should be quick 15:20:10 will meet the i3 deadline 15:21:02 are you just doing pecan or pecan + wsme ? 15:21:28 right now pecan, later pecan + wsme 15:22:26 thanks, balajiiyer! 15:22:40 cool, glad to see some progress there 15:22:42 #info good progress on pecan driver - expecting to meet i3 deadline 15:22:47 btw, I am planning on working on Falcon docs 15:23:13 sweet 15:23:26 also, my talk was well-recieved at the Austin Web Python meetup 15:23:31 that'll make falcon more accessible 15:23:46 awesome! 15:23:55 people definitely see a place for Falcon 15:24:07 for the record, I see a place for Pecan 15:24:26 What I think is silly is saying "either/or" - a common fallancy 15:24:32 s/fallancy/fallacy 15:24:34 aaaanyway 15:24:47 (fallancy - a fallacy of fancy) 15:24:51 yup 15:24:52 next? 15:25:08 would be good to document use cases where to use falcon and where to use pecan 15:25:39 balajiiyer: can you add a section to your wiki notes as a placeholder for that ^^^ 15:26:04 amitgandhi: I did touch on that in a blog post, but yeah, it should be in the falcon docs proper 15:26:13 where, you make it more generic 15:26:19 (not really pecan-specific) 15:27:24 moooving on 15:27:28 yes 15:27:49 #topic sql-sharding-controllers 15:28:01 no progress on this yet, kgriffs 15:28:07 #link https://blueprints.launchpad.net/marconi/+spec/sql-sharding-controllers 15:28:17 ok 15:28:40 #info no progress on sql sharding controller 15:28:40 #topic SQL Storage Driver 15:28:56 Anyone seen Yeela around lately? 15:29:31 I'm getting nervous about this 15:29:36 no progress for several weeks 15:29:57 I haven't spoken to ykaplan yet 15:30:08 I'll make it a priority for this coming week to make progress on this 15:30:15 ok 15:30:30 I may have some time to lend some code to it 15:30:40 that'd be awesome. Let's make it happen! 15:30:44 please let me know as soon as you find anything out from ykaplan 15:31:10 if you don't hear from her in the next couple days, we'll just have to reassign i think 15:31:19 agreed 15:31:24 #action alcabrera to follow up on SQL Storage Driver 15:31:56 #topic tempest-integration 15:32:01 malini: updates? 15:32:32 #link https://review.openstack.org/#/c/65145/ 15:32:47 support for Marconi in the devstack gate looks like it was approved, but blocked by Jenkins 15:33:28 otherwise, it seems like there's two patches being blocked by this one. 15:33:55 the implementation seems complete - it's just a matter of reviews, from what I can tell. 15:33:56 kgriffs: I got some eyes on the devstack patch 15:34:05 ok 15:34:08 Hopefully tht'll be merged soon 15:34:18 one of them got approved. but needs a rebase 15:34:19 that would be FAN-TAS-STIC 15:34:23 I'll get tht done today 15:34:27 thanks! 15:34:30 #info good progress on tempest integration - needs reviews 15:34:48 with the 2 devstack patches merged, the QA folks will be able to merge the tempest as well 15:35:02 awesome! :) 15:35:08 ok, glad that is moving along 15:35:11 the tempest patch has just one review so far 15:35:40 But getting the pre-requisites out of the way will be a big step forward :) 15:35:53 ok, cool 15:36:26 do they need core marconi reviewers to +1 it before they will approve? 15:36:34 malini: ^^ 15:37:04 kgriffs: not necessarily, but lots of +1s do look good on the patch ;) 15:37:20 haha, this is true of +1s 15:37:45 we should channel out patch submissions via g+ :) 15:37:52 *our 15:37:54 lol 15:38:05 Cloud notifications ;) 15:38:33 malini: can you drop the links to all the patches in the bp? 15:38:37 #link https://blueprints.launchpad.net/marconi/+spec/tempest-integration 15:38:43 (in the whiteboard) 15:38:45 +1 15:39:02 kgriffs: I already have it in the whiteboard 15:40:01 ah, see it now 15:40:02 sorry 15:40:05 #link https://review.openstack.org/#/q/topic:bp/add-basic-marconi-tests,n,z 15:40:14 thts easy to miss, I'll add a heading for tht link 15:40:14 ah yes, the abridged form 15:40:19 malini: +1 15:40:27 rock on 15:40:40 let's get some more +1's on those 15:40:54 will do 15:40:57 last graduation topic is: docs 15:41:01 #topic docs 15:41:27 so, I haven't heard of any progress on that 15:41:27 kgriffs: reg docs, did you see the email I sent you yesterday? 15:41:40 heh 15:41:54 We may have Rose's team help us with graduation docs 15:41:55 sorry, I am still catching up from being sick 15:42:17 balajiiyer: that would be really cool 15:42:19 Catherine is in Rose's team, so Catherine might be able to help her with the docs 15:42:40 *us 15:42:45 can't type today, very cold 15:43:23 balajiiyer: mind if I assign the bp to you? Not that you will do all the work, but that you can spearhead the effort and report progress 15:43:30 #link https://blueprints.launchpad.net/marconi/+spec/docs 15:43:50 kgriffs: will be happy to take that on 15:43:56 thanks! 15:44:47 ok, so docs and SQL are the two main ones I am nervous about 15:45:03 #action balajiiyer to lead initiative on docs 15:45:06 we need to start seeing concrete progress on those ASAP 15:45:10 agreed 15:45:29 docs we have a couple extra weeks, but let's pretend we don't. ;) 15:46:05 yup. even basic docs would be a huge plus. 15:46:11 demystified: marconi for everyone 15:47:13 shall we move on to bugs, kgriffs? 15:47:35 yes, let's take a quick pass and then have some time for open discussion 15:47:39 kk 15:47:46 #topic updates on bugs 15:48:15 so, a couple to triage 15:48:18 #link https://bugs.launchpad.net/marconi/+bug/1273376 15:48:34 I think this is critical. Broken tests === badness 15:48:43 agreed 15:48:52 three test suites fail as a result 15:49:18 I have a fix in with... 15:49:20 #link https://review.openstack.org/#/c/68267/ 15:49:32 mmm 15:49:36 is the bug referenced? 15:49:41 yes 15:49:44 which is... 15:49:46 weird 15:49:47 I don't see it linked from the bug 15:49:50 yeah. :/ 15:49:59 Can you add a comment there? 15:50:02 will do 15:50:05 and assign yourself? 15:50:11 yes. :) 15:50:17 thanks! 15:50:39 this one also addressed by that patch? 15:50:39 https://bugs.launchpad.net/marconi/+bug/1273377 15:51:00 yup 15:51:12 ok, I guess that needs a comment/assignment as well 15:51:28 assigned and commented 15:51:36 importance/status? 15:51:37 thanks! 15:51:40 updated 15:51:43 cool 15:51:49 #link https://launchpad.net/marconi/+milestone/icehouse-3 15:52:25 flwang: "[queues] Sharding resource PUT should replace, not ignore existing" 15:52:35 malini: can you confirm this is invalid? 15:52:42 #link https://bugs.launchpad.net/marconi/+bug/1243899 15:52:57 kgriffs: I remember seeing this one 15:52:58 yeah, flwang and I were looking into it yesterday. There's even unit tests that check for the overwriting behavior. :) 15:53:20 ah, ok 15:53:25 #link https://github.com/openstack/marconi/blob/master/tests/unit/queues/transport/wsgi/test_shards.py#L132 15:53:29 #1264853 Driver load fails if config options are registered beforehand 15:53:57 what's the status on this? 15:54:07 #link https://bugs.launchpad.net/marconi/+bug/1264853 15:54:29 I'm not sure, but flaper87|afk may have done some work on this already 15:54:35 it makes me think of the recent config patch 15:54:38 mmm ok 15:54:45 we'll have to follow up with him 15:55:03 namely 15:55:04 #link https://review.openstack.org/#/c/68441/ 15:55:07 but yes 15:55:12 flaper87|afk: can tell us the status later 15:55:33 ok 15:55:35 #action check up with flaper87|afk on status of #1264853 15:55:47 so, i think we have a bunch of patches pending review for bugs 15:55:52 let's get those merged 15:56:07 +1 15:56:08 and then concentrate on nocking out the remaining bugs in priority-order 15:56:17 I will dedicate some time myself to that 15:56:26 need a few others to do the same 15:56:53 #topic open discussion 15:57:10 our first official release went out last week! 15:57:15 (yay) 15:57:18 hurray! 15:57:20 kgriffs: YAY 15:57:27 #link http://tarballs.openstack.org/marconi/ 15:57:50 thanks everyone for making that happen 15:58:07 hopefully zuul won't be so...broken for i-3 15:58:10 quite a mighty release, at ~200K all tar/bz2'd up 15:58:17 just in case, we should be planning on getting stuff in early 15:58:45 any other thoughts, questions, concerns... 15:58:51 rude remarks... 15:59:00 so over the next month, let's: wrap up the client, finish up sql storage, get some docs, finalize pecan, get hooked into temptest, and be integrated with heat 15:59:02 so busy 15:59:09 lots of fun. :) 15:59:11 yep, for sure 15:59:14 (to be had by all) 15:59:34 that's it from me - thanks for joining us, everyone! 15:59:38 ok team, thanks for working so hard and lets nail i-3 15:59:42 #endmeeting