22:02:56 #startmeeting reddwarf 22:02:57 Meeting started Tue Nov 13 22:02:56 2012 UTC. The chair is hub_cap. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:02:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:02:59 The meeting name has been set to 'reddwarf' 22:03:09 hub_cap: sorry was 2 min late 22:03:17 ttx: your on my list 22:03:18 ;) 22:03:25 *youre 22:03:28 so that's how it's done 22:03:45 yup then we do things like 22:04:04 jcooley here as well. 22:04:09 lets wait a few minutes for everyone to join up 22:04:19 agenda is here http://wiki.openstack.org/Meetings/RedDwarfMeeting 22:04:21 #link http://wiki.openstack.org/Meetings/RedDwarfMeeting 22:05:04 Nikhil here as well. 22:05:34 ok seems like we have a quorom lets get started based on the link above 22:05:46 #topic Meet and greet 22:06:03 lets spend a few seconds each saying hi and what we do, where we are, who w work for etc.. 22:06:22 im michael basnight, lead of the opensource portion of the project from rackspace 22:06:33 feel free to start typing everyone :) 22:06:51 Hey! Vipul from HP - lead for Reddwarf @ HP 22:06:59 lets add tags to them 22:07:01 justin hopper, working with the dbaas team at HP. Pretty new to all of this :) 22:07:02 #info michael basnight, lead of the opensource portion of the project from rackspace 22:07:07 Hi. Kevin from HP 22:07:13 Josh Dorothy w/ HP 22:07:14 if we add tags itll show in the notes 22:07:25 I sit next to Kevin (rather unpleasant :) 22:07:28 Hi! I'm Thiago and I'm doing a POC of OpenStack for my own start up company (public cloud computing provider). I'm using today the OpenStack Alamo, from Rackspace for my tests... 22:07:32 #info Josh Dorothy w/ HP 22:07:41 #info Clare Springer w/ HP 22:07:51 #info Justin Hopper w/ HP 22:07:53 #info Vipul Sabhya @ hp 22:07:59 ThiagoCMC: welcome! a 3rd party so to speak (figured we would all be from hp or rax) 22:08:01 #info Nikhil Manchanda, with HP. 22:08:06 Hello. #info Dan Nguyen @HP 22:08:14 #info jim cooley, director of database/reddwarf @ HP 22:08:16 hub_cap, awesome! tks! 22:08:21 #info Saurabh, w/ HP 22:08:23 Still wrapping my head around the way OpenStack does things, so thanks for being helpful everyone. 22:08:39 nice to see you all HP'ers, i believe the rax peoplez are in sprint planning, soooo...... they likley wont be on today, but im holding down the fort w/ a crying baby in the background :) 22:08:44 where's all the rackers 22:09:07 grapex2: when grapex and grapex1 dont work 22:09:29 #info Kevin Mansel @ HP 22:10:02 #info Nirmal Ranganathan @ Rackspace.. just eavesdropping 22:10:07 #info grapex is tim simpson, sr dev on the project at rax, cp16net is craig vyvial sr dev at rax, rnirmal is nirmal ranganathan former reddwarfer 22:10:13 lol 22:10:19 ok cool lets keep a-goin 22:10:29 #topic Discuss launchpad participation 22:10:42 just to let everyone know, ive been mucking w/ the launchpad page today 22:10:54 #link https://launchpad.net/reddwarf 22:11:00 #info Aaron Crickenberger w/ HP 22:11:11 hehe 22:11:22 thats gonna show up under the launchpad participation lol 22:11:37 #info added the grizzly series to the page 22:11:55 i figured we should follow the same release schedule as the rest of the openstackers, any objections? 22:12:19 Sounds good to me 22:12:38 Sounds reasonable. 22:12:42 easier to track 22:12:44 thanks for fixing the links 22:13:04 vipul: no problemo, if u see anythign else, and 1) dont have access to fix, tell me, or 2) ask me to fix 22:13:25 id lke ot make it so im not the only person who can do this, i dont want it to be a Autocracy 22:13:49 id prefer the #1 from above over teh #2 ;) 22:14:17 cool, I may have the correct permissions, since i do see 'Change details' 22:14:26 looks like rackers are coming in slowly 22:14:31 vipul: good 22:15:04 ok so does anyone have any info/links/etc for that topic? or we move on? 22:15:05 re: release schedule, are we talking about something more general than http://wiki.openstack.org/GrizzlyReleaseSchedule? 22:15:07 I had a quick clarification. 22:15:15 SlickNik: go for it 22:15:38 #link https://launchpad.net/reddwarf/grizzly 22:15:41 spiffxp: ^ ^ 22:15:58 If I had a bug/issue that I wanted to open against redstack, can I use the reddwarf lauchpad site to do it? 22:16:12 spiffxp: ive added the grizzly milestones to the series as well 22:16:24 got it 22:16:30 SlickNik: thats a good question, we might need to add a separate page for that, lets model it after devstack 22:16:44 looks like devstack has thier own 22:16:59 #action grapex to make a launchpad site for redstack 22:17:11 Awesome, that would be ideal. Thanks! 22:17:27 hub_cap: Sounds cool. 22:17:31 cool any more Qs or moving on? 22:17:44 Long term though, are we thinking of migrating Redstack into Devstack? 22:17:59 so this may be the interim solution, until we get there 22:18:02 vipul: as much as possible i think so 22:18:19 i talked w/ anotherjesse a long while ago about booting special services in devstack 22:19:05 hub_cap: yea there seems to be precendence with non-incubated projects already being part of devstack 22:19:06 they were on board w/ making a configurable system to allow us (and others) to use our own configure logic within devstack, so i think that in some form of fashion we will have our own particulars 22:19:15 lol ya vipul 22:19:27 hub_cap: devstack now is updated to have loadable libs 22:19:37 so could potentially add reddwarf 22:19:49 libs as in bunch of bash scripts 22:20:14 "libs" 22:20:37 ya #topic devstack integration 22:20:52 #info we need a way to configure reddwarf easier within devstack 22:21:00 who wants the action item to talk w/ devstack about it? 22:21:18 #topic devstack integration 22:21:28 lol if the hashtag isint at the begin it does nothing 22:21:41 just noticed that. 22:21:53 #info we need a way to configure reddwarf easier within devstack 22:21:57 these meeting notes are going to be a work in progress :) 22:22:02 vipul: fo sure 22:22:17 so as for ownership any volunteers? if not ill take it 22:22:35 #action hub_cap to discuss devstack integration 22:22:48 we can talk with the devstack/ci folks, some of them are right here :) 22:22:55 #info consider using the libs in devstack 22:23:03 jcooley: u want the action? 22:23:08 sure 22:23:22 #action jcooley (not hub_cap) to discuss devstack integration 22:23:27 #info non-incubated precendence already in devstack, see 'ryu' integration 22:23:49 cool, time to move on? 22:24:39 yeah 22:24:51 #topic Discuss initial core team 22:25:08 so id like ot keep core very small at first, like 2 from each of our teams 22:25:21 #info current reddwarf-core team: hub_cap, grapex, vipul 22:25:23 and add more as we have code reviews / contributions 22:25:35 hub_cap, we'd like to propose one more from our side to get things stated 22:25:43 spiffxp: ? 22:26:00 sure 22:26:10 actually… 22:26:13 vipul: did u have someone in mind? 22:26:18 * spiffxp nominates SlickNik 22:26:23 spiffxp or SlickNik 22:26:36 ok ill let you decide vipul 22:26:39 and ill add to the team 22:27:17 spiffxp 22:27:30 so the _only_ thing we need to be cognisant of is that we make sure the "other" company has a stake in each review 22:27:44 #action hub_cap to add spiffxp to core team 22:27:58 Sounds good. I recommend spiffxp. 22:28:00 Guys, will be a "beta testers team"? If yes, I'm on it! 22:28:02 ;-) 22:28:08 So it seems like we need +4 total on a commit to merge it. Does anyone want it to be higher? 22:28:13 ThiagoCMC: you are the beta tester team! 22:28:19 Cool! 22:28:21 hub_cap: I wasn't sure if we wanted to try encoding that w/ some crazy ci rules at some point, I think it's out of openstack-ci's scope right now 22:28:55 grapex: id say we need a +3 and that should suffice right spiffxp grapex? 22:29:02 +2 from one team and then +1 from the other 22:29:05 grapex: I think you need a minimum of 2 +2's 22:29:19 2 +2 22:29:20 can that be done? can we change that "rule" for our project? 22:29:27 it turns out they are in layers, we can have at least one +2 or do we want 2 x +2? 22:29:41 I propose 2 x +2's 22:29:50 vipul: i think thats not enough tho 22:30:00 and mordred did mention that this is configurable 22:30:02 cuz grapex and i could easily shuffle code past you :) 22:30:10 i think that makes sense. 2 x +2 22:30:22 if we have 3*+2 it woudl require someone from both companies reviewing 22:30:25 hub_cap: Yeah, that's what I was thinking. We could inadvertently both look at something and merge it. 22:30:49 hub_cap: I don't think it requires 2 +2 22:30:52 yes grapex. i think until we have a larger team and better rules around blueprints/bugs/ci etc... 22:30:58 one +2 and +1 approve 22:31:04 rnirmal: you are correct thats how it works now 22:31:11 you could always have more as a rule 22:31:18 id prefer 3 core members need to review it 22:31:23 hmm, good point, just don't want things to be stacked up pending reviews.. 22:31:34 vipul: lets cross that bridge when we come to it 22:31:46 ditto, I am concerned emphasis on +2 downplays +1 participation 22:31:48 rnirmal, yep those are the default rules. 22:32:06 yeah, requiring 3 +2 essentially means the entire core team needs to be reviewing every single commit 22:32:17 yidclare: there are 4 of us now :) 22:32:28 so its 3/4 of the core team 22:32:47 no 2 must go on vacay at the same time :) 22:32:59 vipul: ;) im on vaca right now dude 22:33:07 lol 22:33:26 all i worry about is that 3 x +2 means 3/4s of the core team needs to approve. 22:33:28 What about if instead of three mandatory +2's, we just have a total of 6. That could be one core and four other members. 22:33:30 I think 2 +2 would be better for now while we work on getting folks onboarded through the gerrit process 22:33:45 since we're moving pretty fast, that might slow down changes. if folks are cool with that... 22:33:57 well i dont think speed should be our only motivation 22:33:58 we could do a hybrid approach like grapex suggests 22:34:09 i think our motiviation should be participation from both teams 22:34:17 I'd prefer a quicker number. 22:34:24 I like grapex's ideas as well. 22:34:32 if you guys just push stuff all day long then we will feel like we dont have any skin in the game 22:34:34 and verse visa 22:34:42 standard core rules are 2 x +2 and no concern for how many +1s. 22:34:45 based on what we were told yesterday these numbers are not additive 22:34:46 I second spiffxp - 2 +2's and try to informally makes sure someone from each group sees it 22:34:54 2 x +1 doesn't equal +2 by the way. 22:35:05 4 x +1 doesn't equal 2 x +2 either. 22:35:17 so lets do this, add a reviewer from the other company in your gerrit review 22:35:23 that was what I was trying say - thanks for the examples jcooley 22:35:36 hub_cap, that works 22:35:49 excellent! 22:35:57 #agreed 22:36:02 That works as well. 22:36:06 ok so now the question 22:36:15 do we know for sure that the group has been changed in gerrit? 22:36:32 hub_cap, it has not yet 22:36:35 which group? 22:36:47 the one i just added u to spiffxp 22:36:49 #link https://launchpad.net/~reddwarf-core 22:36:50 :) 22:37:03 ok who is gonna pester mordred then? 22:37:26 to change from -drivers to -core for the +2s 22:37:30 #action Vipul to follow up on reddwarf-core with mordred 22:37:34 i'll do it. 22:37:34 <3 22:37:40 he's sitting next to me :) 22:37:49 feedback loop is very small 22:37:51 :D 22:38:01 #action jcooley to follow up for Vipul on reddwarf-core with mordred :) 22:38:02 lol,n00b question alert. who's mordred? 22:38:12 mordred is Monty 22:38:21 ohhh, gotcha 22:38:34 #info make sure one of the other company (rax/hp) looks at each review (add a particular reviewer from core) 22:38:35 hub_cap: any chance vipuls could also be made an admin of reddwarf-core? 22:38:59 err… vipul that is 22:39:04 sure spiffxp 22:39:24 done 22:39:29 awesome 22:39:37 thx 22:39:41 now dont go adding people willy nilly ;) 22:40:00 ok moving on 22:40:04 just as an FYI, getting more people added to core team: 22:40:04 #topic Proper blueprinting/bug practices 22:40:06 argh 22:40:13 i think we'll have to do the "make sure one of the other company (rax/hp) ..." by unformal convention. not sure they can inforce this in gerrit. 22:40:13 crap sry vipul go head 22:40:25 jcooley: its not enforcable 22:40:38 just goinig to say, it needs to be pushed through the mailing list 22:40:44 DEF 22:40:45 and needs some +1 22:40:58 but its a convention now, we can fix it by having a ptl and doing proper launchpad bug/blueprinting 22:41:10 once we get to that ^ ^ we dont need any rules for who has to review 22:41:35 whick gets to the next topic, does everyone know how to add bugs/blueprints and link them in commits? 22:41:50 nope 22:41:52 agreed. also we should be trying to do that -- but we missed the last openstack blueprint review period. 22:41:53 #action everyone to make sure that reviews are linked with bugs or blueprints 22:42:11 it was discussed yesterday briefly 22:42:22 mordred said yesterday that we could just mention it in the commit message. 22:42:22 i thought it was just having the word 'blueprint xxx' 22:42:25 something about putting bug anywhere in the comment 22:42:33 Yeah, I was under the same impression. 22:42:37 yes, they have regex scanning 22:42:37 yup SlickNik you mention it in the commit 22:42:48 fixes lp#bugID 22:42:55 blueprint BLUEPRINT 22:42:55 bug ####### 22:42:55 "Bug #1234" or "Blueprint blah" 22:42:56 Launchpad bug 1234 in launchpad "Gina is an unmaintainable mess of command line options, environment variables and shell scripts" [Medium,Fix released] https://launchpad.net/bugs/1234 22:43:02 #link http://wiki.openstack.org/GerritWorkflow 22:43:02 or Blueprint: blah, Blueprint-implemented: blah 22:43:08 create your branch as bp/ 22:43:10 thx vipul 22:43:26 can do the same for bugs bug/ 22:43:41 rnirmal: good to know 22:43:42 also fixes in the commit msg 22:43:52 and implements bp/ 22:43:54 wow, there a bot in IRC that apparently brings it up as well. 22:43:55 Nice. 22:43:57 for blueprints 22:44:09 ya SlickNik cool eh? 22:44:20 good to know gina is an unmaintainable mess 22:44:27 #info http://wiki.openstack.org/GerritWorkflow 22:44:31 lol 22:44:33 lol double info 22:44:34 heh 22:44:48 ok any other question on that topic? 22:45:20 too bad was not paying attention :) 22:45:24 :P 22:45:57 you snooze, ya lose. 22:45:59 hub_cap: yeah, one more Q 22:46:01 ok i had roadmap on the chopping block next but im not sure that a meeting would be best for that, lets skip to the heated topic 22:46:04 #topic PTL? 22:46:20 so, 1) do we need one now? 22:46:29 if yes, 2) who? 22:46:45 PTL? 22:47:01 project technical lead juice 22:47:11 I was under the impression PTLs are chosen after Incubation? 22:47:27 are they? cool then we can skip it for now 22:47:49 #info not now, wait for more participation and incubation 22:48:05 good by everyone? 22:48:18 Sounds good to me. 22:48:19 at least we don't need an official one... is my understanding 22:48:21 good 22:48:44 cool we know vipul and i are the technical leads for the projects so we will do our best to discuss things between the teams 22:48:48 sound good vipul? 22:48:57 yep, works 22:49:00 #info vipul and hub_cap to keep communication open between hp/rax 22:49:11 again: good :) 22:49:16 :) 22:49:25 ok now the end of the meeting, ... 22:49:28 #topic Open discussion 22:49:48 does anyone have anything that was not discussed on the meeting we shoudl bring up or make sure we bring up nxt wk? 22:49:48 any update on getting reddwarf-integration to stackforge? 22:50:09 vipul: i think soneone mentioned that earlier? but they called it redstack 22:50:17 and there is a action item for grapex to do that 22:50:22 vipul: I talked to clarkb about it yesterday. Apparently its almost in, but mordred is using it as a guinea pig for a new way of putting projects into StackForge. 22:50:28 That's probably fine, for now. 22:50:43 grapex: thanks 22:51:06 To me, a bigger issue is that 1. the public version of RedStack and the RDLI tests have some bugs that have crept in, and 2. we're not running anything in CI atm. 22:51:14 #info there is likely a decent bit of work to make rdli (reddwarf-integration) work for the public project 22:51:23 I have a few ideas on that. 22:51:27 #info we _need_ _need_ _need_ some CI soon 22:51:39 #action grapex to own CI and make sure it gets accomplished 22:51:50 hub_cap, grapex: are you may have been following, we're having some issues with redstack -- anything (like disable boot from volume) we shoudl be aware of 22:51:56 I think it would be possible to get at least all the fake mode stuff working on a publicly accessible Jenkins node soon. That way we could at least plug into the Gerrit stream and gate on it. 22:52:21 vipul: lets talk them out in irc, im sure there are cobwebs in the public code that we dont know about 22:52:37 my focus (mainly after i get back to work) is to work on the public openstack codebase 22:52:52 as opposed to anything internal / feature driven by rax :) 22:53:17 we have a bunch of folks looking to do their first commit for reddwarf, should we just use dummy bugs to add TODO's, or is there low hanging fruit we could go after? 22:53:18 so ill be able to help a lot and devote a decent ammt of time to helping yall out 22:53:35 spiffxp: sure just add/remove the TODO that was pushed earlier 22:53:56 k 22:54:03 #info we need to integrate the client in to stackforge as well 22:54:23 who wants to own that? 22:54:31 hub_cap, it's already in stackforge 22:54:35 oh it is!?! 22:54:39 yes 22:54:44 NICE 22:54:55 #info hub_cap is a moron, its already there 22:55:08 that was easy :) 22:55:19 hub_cap: nice to know :P 22:55:36 #action hub_cap make sure the client launchpad page is up to date w/ the series like the reddwarf one 22:55:41 we've also been talking with mordred about the road to getting this in openstack CI 22:55:45 I believe a couple of patches went through as well 22:56:14 yes - had a great chat with vipul and jcooley yesterday 22:56:27 I think I understand redstack and reddwarf_integration better now, and am thrilled 22:56:31 hub_cap: folks here @ hp are now working on tip/reference implementation instead of some internal HP-y thing 22:56:39 mordred: sweet!! 22:56:43 jcooley: VERY SWEET!! 22:56:56 id like to get our team there soon too jcooley 22:57:04 hub_cap: I want you to talk to lifeless or devananda at some point about getting your image creation to use the stuff they're working on 22:57:08 very cool, i think we're aligned 22:57:09 in redstack 22:57:17 Awesome. 22:57:20 mordred: fine by me sir. 22:57:31 hub_cap: hi 22:57:36 #action hub_cap talk to lifeless or devananda about the image creation 22:57:43 lifeless: howdy 22:57:46 I need to run up the street for ~ 15m, after that I'll be around again 22:57:53 lifeless: i wont be ;) 22:58:13 #info hub_cap is on paternity leave for 2 more weeks and will be sparse 22:58:31 im on paternity leave lifeless ill hit u up via email 22:58:34 grapex: I really really want to merge the change today. That will allow you to propose a dependent change that creates your new project 22:58:44 grapex: I am actively working on that change now. 22:58:57 clarkb: Sounds interesting. Let's talk in #reddwarf soon. 22:59:13 hub_cap: cool - rbtcollins hp dot com 22:59:24 ok so we are at 1 hr not sure if there is another group after but lets try to wrap up 22:59:26 lifeless: aye 22:59:58 indeed. think we got a lot accomplished. 23:00:03 thanks for moderating, hub_cap 23:00:08 yup very good meeting indeed 23:00:15 good start :) 23:00:15 yidclare: np! 23:00:17 ya thx hub_cap 23:00:34 vipul: thx for getting the ball rolling on the ML too!! 23:00:35 Sweet, thanks all. 23:00:35 thx folks! look forward to working with you guys closer. 23:00:59 hub_cap: np 23:01:04 def. we are a team now for real! 23:01:15 fo sho 23:01:16 ok if no one has anythign else to chat about im gonna end meeting 23:01:18 thumbs up 23:01:29 Awesome work guys. :) 23:01:54 #endmeeting