Wednesday, 2014-04-09

*** rgbkrk has joined #openstack-sdks00:19
*** wchrisj has quit IRC00:50
*** jamie_h has quit IRC01:03
*** rgbkrk has quit IRC01:05
*** wchrisj has joined #openstack-sdks01:09
*** wchrisj has quit IRC01:20
*** Klumben has joined #openstack-sdks01:22
*** rgbkrk has joined #openstack-sdks01:28
*** rgbkrk has quit IRC01:56
*** rgbkrk has joined #openstack-sdks02:10
*** Guest98185 is now known as redrobot02:36
*** wchrisj has joined #openstack-sdks04:37
*** wchrisj has quit IRC04:52
*** HenryG has quit IRC04:59
*** rgbkrk has quit IRC05:11
*** rgbkrk has joined #openstack-sdks05:11
*** terrylhowe has quit IRC05:22
*** jamielennox is now known as jamielennox|away06:35
*** rgbkrk has quit IRC07:52
*** jamie_h has joined #openstack-sdks08:29
*** samchoi_ has joined #openstack-sdks10:26
*** samchoi has quit IRC10:26
*** jamie_h has quit IRC10:31
*** jamie_h has joined #openstack-sdks10:31
*** jamie_h has quit IRC10:41
*** jamie_h has joined #openstack-sdks10:46
*** terrylhowe has joined #openstack-sdks11:02
*** HenryG has joined #openstack-sdks11:07
*** jamielennox|away is now known as jamielennox11:30
*** jamielennox is now known as jamielennox|away11:33
*** terrylhowe has quit IRC11:54
*** terrylhowe has joined #openstack-sdks11:58
*** samchoi_ has quit IRC12:00
*** wchrisj has joined #openstack-sdks12:44
*** krames has joined #openstack-sdks13:17
wchrisjmhagedorn elight krames - I just goofed - mistakenly pushed two commits to master13:35
wchrisjelight krames will fix13:36
wchrisjrookie error13:43
wchrisjBeautiful thing is, in about 10 mins I was able to do this:<params>) and get a compute service instance back...13:44
wchrisjleveraged our identity work13:44
wchrisjscrapped about 80% of existing code, if not more13:44
wchrisjin the compute class13:44
*** rgbkrk has joined #openstack-sdks13:55
elightwchrisj: đź‘Ťđź‘Ť14:05
wchrisjjust did my smallest PR to date - you'll be happy elight !!!14:05
elightSo I can start working on a Swift client now for tng??14:06
wchrisjyep ;-)14:06
elightI guess we finally made it!14:06
wchrisjshould be able to follow that PR and get yours bootstrapped14:06
wchrisj15 mins14:06
wchrisjI considered doing it all in one, but backed off14:07
*** terrylhowe has left #openstack-sdks14:19
wchrisjkrames elight Just realized that I patched my setup to hit a local version of fog-core so the compute.rb code would work correctly... just in case you try to do anything with that PR14:37
*** TravT has quit IRC14:46
*** mfer has joined #openstack-sdks15:03
*** mhagedorn has joined #openstack-sdks15:14
mferPHP SDK meeting in #openstack-meeting-3 in 13 minutes15:17
*** etoews has joined #openstack-sdks15:17
*** samchoi has joined #openstack-sdks15:24
glencGood morning, mfer15:25
mfergood morning15:25
*** ycombinator has joined #openstack-sdks15:26
*** dolphm has quit IRC15:29
*** dolphm has joined #openstack-sdks15:30
*** wchrisj has quit IRC15:45
*** etoews has quit IRC16:12
*** etoews_ has joined #openstack-sdks16:12
elightmhagedorn krames: niiiice
elightTrying this now16:15
mhagedornelight been meaning to play with vagrant/OS..16:15
mhagedorntell me how you like it :16:16
elightmhagedorn: krames and I work on the plugin for rackspace16:16
*** etoews_ has quit IRC16:16
elightmhagedorn: That server we paired on?
elightmhagedorn: Most of the work in the VagrantFile is provisioning the machine. Just shell calls.16:17
elightmhagedorn: Doing a couple of conf talks on it soon.16:18
mhagedornelight I actually learned chef from vagrant :)16:18
elightno shit?16:18
mhagedorndid my last LSRC talk on vagrant.. pre HP16:18
elightI still haven’t gotten around to suffering through the chef learning curve. Krames has been playing with it16:18
mhagedornbut havent touched that stuff in awhile16:18
elightMitchell keeps changing it. For the better, from what I can see.16:18
elightGoing to get started on TNG Storage this afternoon hence local devstack16:19
mhagedornelight, thats what mr johnson does as well.  I use a big honking HP machine they gave me, cause dont want the storage space hit on my MacBook Air16:20
mhagedornelight, he likes the portability16:21
mhagedornas for me havent needed that (yet)16:21
elightI’d figure RAM would be the biggest restriction on the Air16:21
elightIt’s why I caved and went for bigger iron16:21
mhagedornelight for my personal use.. am up for an upgrade my MBP is like circa 2007.  I wind it up16:28
*** mordred has joined #openstack-sdks16:31
ycombinatorhey php folks, continuing from #openstack-meeting-3...16:31
jamie_hso the question was: how is a fresh project more inclusive for new contributors?16:31
mferjamie_h samchoi ycombinator glenc sorry i lost track of time. unfortunately, i've got a meeting right now. i'm likely to be slow in responding.16:33
glencLet's see if we can schedule a followup discussion16:33
glencCan someone send out the survey thing to find a time?16:33
glencRather resolve this sooner than later16:33
ycombinatorglenc: I'll do it16:33
mordredtwo cents from the peanut gallery - all of the current openstack codebases (with one exception) started with a possibly-sub-optimal done-somewhere-else repo  and then iterated on it once it was in the community16:34
mordredthe one exception is keystone, which also started that way, and then had a forklift-rewrite behind the scenes by like one guy16:34
jamie_hi believe the python SDK is starting with a fresh repo16:35
mordredthe main stackforge example I know of to the contrary is solum16:35
samchoipeanut gallery comments are quite interesting :)16:35
mordredyah - I'm merely saying that the existing repos in openstack/ land all did a "Start with code, add contributors" model16:35
mordredI'm not saying it's better or worse16:35
jamie_hit's good to have the discussion though :)16:36
ycombinatoras a potential contributor I know I'd find it easy for me to start off with a clean slate when we have the opportunity to do so16:36
mordredmerely that it certainly hasn't caused problems with the maddening onslaught of contributors16:36
edleafejamie_h: correct. We chose not to use any one company's existing code base in the Python SDK16:36
* mordred ducks and covers to hide from the hoards of angry contributors ... runs screaming16:36
mferin any discussion we have around this i'd like to keep 1) the needs of php app developer using openstack first. 2) to avoid bike shedding on major architectural issues. I love discussion. just the nuanced difference.16:37
ycombinatoragreed on both points, mfer16:37
mordredI think glenc is right though - there should be another meeting where folks can dive in to the topic16:37
mferi'm happy to chair that.16:39
mfersorry, in a meeting and a little slow. and now i have to interact so i'll be gettign slower to repond16:39
ycombinatormfer: I'll send out a doodle but when is your first next free time slot? trying to pre-optimize a bit here16:39
mferycombinator 2pm - 4pm ET today16:41
samchoiOut of curiosity, where is the python SDK at in terms of progress? There's been a lot of mention of drawing from the experiences of that team.16:42
mferdon't take the whole 2 hour slot :)16:42
mfersamchoi there isn't a lot of code... and two things out for review...,n,z16:43
edleafesamchoi: The Python SDK has code for one proposed design, and is waiting on a second. Once that choice is made, we will start implementation16:43
jamie_hsamchoi i think they're focusing on agreeing consensus first before writing code16:43
*** edleafe has left #openstack-sdks16:44
*** edleafe has joined #openstack-sdks16:44
edleafejamie_h: exactly16:44
edleafeI have a complete implementation ready16:44
jamie_hwhich is where i'd like our project to be16:44
edleafebut others may have their own ideas16:44
samchoithank you all for the updates. I'll keep tabs on the repo.16:44
ycombinatorsamchoi, mfer, jamie_h, glenc:
glenclove the picture16:48
mferycombinator is there an open meeting room or are you thinking it's in here?16:49
ycombinatormfer: default to here but I'm working on finding a room as well16:50
elightWhew this Virtualbox devstack is taking a while to build. stack started about 45 minutes ago…..16:59
ycombinatoralright, mfer jamie_h samchoi glenc: after sacrificing a few animals I was able to get us an openstack meeting room: see you all in #openstack-meeting-alt at 1900 UTC today (that's noon for Sam and me, 2PM for Glen, 3PM for Matt and 9PM for Jamie, assuming my math is good)17:12
mferycombinator i'm ok with that17:21
*** wchrisj has joined #openstack-sdks17:34
*** wchrisj has quit IRC17:45
*** WhatTheFoley has joined #openstack-sdks17:53
*** rgbkrk has quit IRC18:17
*** jamie_h has quit IRC18:22
mhagedornhey elight… two PRs relevant to todays discussion and… FYI18:27
*** mhagedorn has quit IRC18:29
*** wchrisj_ has joined #openstack-sdks18:30
elightThis is mixing the fix I made to fog-core Identity?18:38
elightOh. Yeah. Nevermind. Definitely.18:39
*** jamie_h has joined #openstack-sdks18:54
elightwchrisj_: So I just noticed that Fog::OpenStackCommon::Identity doesn’t inherit from Fog::Service.18:58
ycombinatorsamchoi, we are in #openstack-meeting-alt19:00
*** mhagedorn has joined #openstack-sdks19:31
*** mhagedorn_ has joined #openstack-sdks19:35
*** mhagedorn has quit IRC19:36
*** mhagedorn_ is now known as mhagedorn19:36
elightwchrisj_ mhagedorn: So.. yeah. Doesn’t look like we’re quite done with Identity yet. :-(19:42
mhagedornelight… work in progress :019:43
*** mhagedorn has quit IRC19:51
*** mhagedorn has joined #openstack-sdks19:53
*** openstackstatus has quit IRC20:15
*** openstackstatus has joined #openstack-sdks20:16
*** mhagedorn has quit IRC20:27
*** mhagedorn has joined #openstack-sdks20:27
elightArgh now I can’t clone github repos!!!!20:30
mhagedornelight… its a monday on a weds :)20:36
mferelight it's git man... it's all distributed n stuff20:40
*** jamielennox|away is now known as jamielennox20:43
jamielennoxdtroyer: i put up a basic Resource orientated SDK base class,
jamielennoxhave a look for me and see if you disagree anywhere and i'll fix it up for next week20:50
dtroyerjamielennox: cool…I hope I have time tonight to look at it.20:51
*** edleafe has left #openstack-sdks20:52
*** edleafe has joined #openstack-sdks20:52
*** edleafe has left #openstack-sdks20:52
*** edleafe has joined #openstack-sdks20:52
jamielennoxdtroyer: thanks20:52
jamielennoxdtroyer: this is something i'm playing with around sessions as well that you might be interested in:
*** etoews has joined #openstack-sdks21:06
*** etoews has quit IRC21:08
*** etoews has joined #openstack-sdks21:08
*** etoews has quit IRC21:09
*** openstackstatus has quit IRC21:14
mhagedornelight… seeing some wierdness in the Fog::Service hierarchy when I try to make the requests collection work across my subclass and the parent class (i.e. OSC)21:20
elightI just tried creating a Tenant and I’m getting an error21:21
elightmhagedorn: more specific?21:22
*** openstackstatus has joined #openstack-sdks21:22
mhagedornelight sharing your pain… tryting to make my subclassed IdentityV2 class respond with all the requests it knows about (two) and all the requestst OSC::V2 knows about (lots)21:22
mhagedornno luck so fare21:22
mhagedornelight as it turns out here… the subclassing trick does not extend to the collections on service.. models, requests etc21:23
elightmhagedorn: dumb question: what do you use for region for your devstack?21:27
mhagedornthats not dumb21:28
elightOn another note, I didn’t supply an opentack_region and no error was raised until I went to save a new Tenant.21:28
mhagedornits probably really dependant on your setup scripts21:28
mhagedornfor me it was RegionOne21:28
mhagedornand I got that by looking at the JSON retured from a successful login21:29
mhagedornbut I didnt know that before hand21:29
dtroyerelight: the default DevStack region is RegionOne, it's hard-coded all over the place...21:29
elightThat seemed to work here21:29
elightdtroyer: thanks. :D21:29
elightmhagedorn: I’m not having any problems but then I’m not using the hp_tng. I’m just using openstack_tng right now21:30
*** mfer has quit IRC21:31
elightwchrisj_ mhagedorn:
mhagedornelight  it works from ruby perspective, from a fog api perspective the requests collection only reflects the subclass not the superclass21:31
elightmhagedorn: Hmmm maybe Service#setup_requirements was never called on the superclass21:32
elightFog was never designed to support full blown class families under Fog::Service21:32
mhagedornelight so even though the subclass understands a plethora of requests… it only shows the subclass requests21:32
mhagedornwill look into that21:33
elightOnce again, we’re the oddballs because of OpenStack.  Only Fog providers who care about code reuse!21:33
elightmhagedorn: I can look at writing a complete enough Fog::Rackspace::Identity tomorrow. Delegation should solve that problem whereas I can see how inheritance wouldn't21:34
elightmhagedorn: Wanna bet Fog::Service#initialize doesn’t call super? :D21:35
mhagedornelight yeah21:35
mhagedornsounds about right21:35
*** krames has quit IRC21:36
elightHrm… it doesn’t call super but it does something less idiomatic:
elightIf your HP::Identity < OSC::Identity then that *should* do the trick….21:38
wchrisj_LGTM elight21:39
*** jamie_h has quit IRC22:15
*** wchrisj_ has quit IRC22:28
*** openstack has joined #openstack-sdks22:34
*** rgbkrk has joined #openstack-sdks22:54
*** rgbkrk has quit IRC23:17
*** wchrisj has joined #openstack-sdks23:28
wchrisjelight - you still around?23:29
*** wchrisj has quit IRC23:47

Generated by 2.14.0 by Marius Gedminas - find it at!