[14:52]  * jcastro taps the mic
[14:54] <med_> no stream yet jcastro, tapping the mic won't help
[14:54] <med_> your stream leader will need to click the on-air button to go live.
[14:55] <med_> and so he has
[14:56] <med_> morning arosales.
[14:57] <med_> it went live
[14:57] <med_> You are good in the page
[14:57] <robin-gloster> jcastro you're live
[14:58] <med_> it went live almost immediately after I told arosales to hit the button.
[14:58] <negronjl> hello
[14:58] <med_> jcastro, yes, you can be in the hangout before it streams
[14:58] <negronjl> jcastro:  How do we join the hangout ?
[14:58] <negronjl> lol arosales:  smile :)
[14:58]  * med_ isn't going to be in this session as per se, likely in the ceph one
[14:58] <negronjl> yes
[14:58] <negronjl> jcastro: yes
[14:58] <jcastro> https://plus.google.com/hangouts/_/3e70c524fdc96a7a3f6b87af69c29d9f6ed5430c?authuser=0&hl=en
[14:59] <negronjl> refreshing, jcastro
[14:59] <robin-gloster> could you add lower third please
[14:59] <arosales> Hello, we are going to get started
[14:59] <arosales> Hangout URL: https://plus.google.com/hangouts/_/3e70c524fdc96a7a3f6b87af69c29d9f6ed5430c?authuser=0&hl=en
[15:01] <arosales> getting started
[15:02] <jcastro> https://blueprints.launchpad.net/ubuntu/+spec/community-r-juju-contributor-onramp-1
[15:02] <jcastro> https://blueprints.launchpad.net/ubuntu/+spec/community-r-juju-contributor-onramp-2
[15:02] <utlemming> jcastro, arosales: in your lower third, why don't you put your IRC nick
[15:02] <negronjl> utlemming, they're too cool for that :)
[15:02] <arosales> https://blueprints.launchpad.net/ubuntu/+spec/servercloud-1303-juju-contributor-onramp
[15:04] <arosales> utlemming: how does that look?
[15:04] <utlemming> aroslaes: looks good :)
[15:06] <mattyw> when is the documentation session?
[15:07] <arosales> mattyw: good point I don't see that on the schedule yet. I'll get it scheduled for tomorrow.
[15:07] <arosales> mattyw: thanks for pointing that out.
[15:07] <mattyw> arosales, thanks, I'll keep an eye out
[15:08] <evilnickveitch> evilnick!
[15:11] <niemeyer> Where is Jorge writing?
[15:11] <niemeyer> The Etherpad isn't showing anything for me
[15:13] <jcastro> I'm writing in the blueprint
[15:14] <jcastro> right intot he work items
[15:15] <niemeyer> That means we can't see the meeting notes
[15:15] <arosales> niemeyer: jcastro:  is editing the work items in https://blueprints.launchpad.net/ubuntu/+spec/servercloud-1303-juju-contributor-onramp
[15:16] <arosales> he hasn't saved it yet, so you probably don't see the update just yet.
[15:16] <niemeyer> arosales: I get it.. but there's an Etherpad everybody in the meeting has in their face
[15:17] <arosales> I think we are moving their next
[15:17] <niemeyer> Cheers
[15:18] <arosales> any comments in-line is appreicated in the pad :-)
[15:19] <mariusko_> +1 for Git. Bzr sucks at merging unmerged topic branches..
[15:20] <niemeyer> We *are* there
[15:20] <niemeyer> People can use github today
[15:20] <niemeyer> Or anything else
[15:20] <arosales> niemeyer: would you like to join the hangout?
[15:20] <arosales> we have room
[15:20] <niemeyer> Sure
[15:21] <arosales> niemeyer: try: https://plus.google.com/hangouts/_/3e70c524fdc96a7a3f6b87af69c29d9f6ed5430c?authuser=0&hl=en
[15:21] <arosales> and any other folks that are interested in talking please join
[15:21] <arosales> we have some open room
[15:21] <arosales> we can accommodate 10 spots.
[15:22] <Daviey> (15)
[15:22] <med_> woot.
[15:22]  * med_ hadn't heard 15
[15:22] <med_> I'd seen one go to 11
[15:22] <Daviey> we have special accounts :)
[15:25] <mariusko_> How to publish an updated charm from Github?
[15:25] <arosales> ah nice, so if folks would like to speak we can accommodate 15 :-)
[15:25] <arosales> thanks for the info Daviey
[15:28] <niemeyer> mariusko_: It'll be something like "juju publish"
[15:28] <arosales> mariusko_: we also have a work item to make those docs clearer :-)
[15:29] <mariusko_> That would be nice. Or publish from whatever git/bzr repo you want (possibly locally)
[15:29] <niemeyer> mariusko_: But today, you can just include the .bzr in your github repo, and keep it there
[15:30] <niemeyer> mariusko_: When you want to publish a new version, bzr commit, bzr push
[15:31] <mariusko_> Also maybe compare with how it is done with nodejs: "npm help publish"
[15:31] <niemeyer> mariusko_: Cool
[15:32] <niemeyer> mariusko_: Sounds like a good idea
[15:37] <arosales> utlemming: the bug file is @ https://bugs.launchpad.net/juju-core/+filebug
[15:39] <robin-gloster> https://juju.ubuntu.com/get-started/maas/ first link is dead
[15:39] <udsbotu> uds-servercloud-2: 5 minutes left in this session!
[15:39] <arosales> robin-gloster: 404?
[15:39] <arosales> robin-gloster: works for me
[15:40] <robin-gloster> arosales: the first link on that site
[15:40] <arosales> https://juju.ubuntu.com/get-started/maas/
[15:40] <arosales> dead link ^
[15:40] <udsbotu> uds-servercloud-2: 4 minutes left in this session!
[15:40] <arosales> robin-gloster: gotcha thank you
[15:41] <udsbotu> uds-servercloud-2: 3 minutes left in this session!
[15:42]  * arosales updated work item to address dead link
[15:42] <davidm> will the local provider allow you to use juju-gui and command line?
[15:42] <davidm> jcastro, ^^
[15:42] <udsbotu> uds-servercloud-2: 2 minutes left in this session!
[15:42] <jcastro> I don't see why it wouldn't
[15:42] <jcastro> but I'll check
[15:42] <davidm> you can't now
[15:42] <jcastro> ok I'll ask about that.
[15:43] <davidm> thanks
[15:43] <utlemming> Filed Bug: #1147136
[15:43] <udsbotu> Ubuntu bug 1147136 in juju-core "Support for different environments files" [Undecided,New] https://launchpad.net/bugs/1147136
[15:43] <davidm> it won't work du to database issues
[15:43] <davidm> you can see the gui but have to install charms via cmd line
[15:45] <udsbotu> uds-servercloud-2: This session has ended.
[15:45] <Nordm> oh hey!
[17:20] <arosales> mattyw: fyi the juju doc sessions is scheduled for tomorrow @ 16:00 - 16:55 UTC
[17:21] <arosales> mattyw: feel free to add your self to the participation essential to https://blueprints.launchpad.net/ubuntu/+spec/servercloud-1303-juju-docs if you are interested in participating.
[17:21] <arosales> mattyw: thanks aging for pointing out that session was missing.
[17:51] <mattyw> arosales, no problem, thanks for letting me know when it's booked for
[17:52] <arosales> mattyw: for sure, hopefully we'll "see" you there
[17:54] <mattyw> arosales, I've definately got some idea for documentation we could add, I'll add myself to the participants list
[17:55] <arosales> mattyw: great, I look forward to hearing them.
[18:08] <arosales> marcoceppi: Hello
[18:08] <arosales> jcastro: marcoceppi also had some good ideas about charm testing.
[18:09] <jcastro> rock and roll!
[18:09] <arosales> marcoceppi: were you planning on joining the Juju Charm Testing session at 19:05 (1pm central)?
[18:10] <arosales> marcoceppi: I subscribed you to https://blueprints.launchpad.net/ubuntu/+spec/servercloud-r-juju-charm-testing just in case you were interested.
[18:16] <arosales> http://summit.ubuntu.com/uds-1303/meeting/21645/servercloud-1303-juju-training-events/
[18:16] <arosales> Lets get started :-)
[18:17] <pixelpapst_> jup
[18:17] <robbiew> I'm here...heh
[18:20] <mattyw> arosales, jcastro if you can save it so that people can watch it back that's a great bonus, both for teaching people, but also they can help advertise future events
[18:20] <mattyw> ^^ webinars
[18:21] <arosales> mattyw: +1 on saving
[18:22] <evilnickveitch> jcastro, the trick is to choose the harder things for webinars - things that people are likely to have questions on
[18:23] <mattyw> arosales, I'd love to see a screen cast of someone making a simple charm - from scratch to deployed (and working). I've found the documentation to be ok for learning about how to write charms. But it would be great to see a recommended workflow for charm development
[18:25] <arosales> mattyw: agreed we would like to have course/road map including making a simple charm
[18:26] <mattyw> arosales, it's the workflow which is important I guess - or at least one of the important things to learn
[18:26] <evilnickveitch> arosales, there should be a good correlation between the screencasts and our user journeys. It would be handy to link them in the docs too.
[18:26] <evilnickveitch> aroslaes, I also don't mind recording some.
[18:27] <mattyw> evilnickveitch, arosales I'm going to regret this but I'd be ok to record some as well
[18:29] <mattyw> arosales, jcastro I think it would be useful, I used juju set
[18:30] <mattyw> arosales, +1 everything you just said
[18:31] <mattyw> arosales, jcastro I'm not sure what the plan is for juju-core - but we might want to think about a video about the differences????
[18:31] <arosales> evilnickveitch: agreed on the correlation  between screen casts and user journeys.
[18:33] <mattyw> jcastro, when is that lightning talk?
[18:40] <arosales> http://jujucharms.com/charms/precise/mongodb
[18:42] <jcastro> http://jujucharms.com/charms/precise/mongodb/config
[18:49] <mattyw> jcastro, I think he left
[18:49] <jcastro> bummer
[18:49] <mattyw> jcastro, I agree if that counts ;)
[18:49] <jcastro> \o/
[18:53] <jcastro> evilnickveitch: you get that? relations and interfaces
[18:54] <evilnickveitch> jcastro, I am listening :)
[18:54] <mattyw> jcastro, arosales forgive me for not joining the hangout - I'm eating
[18:54] <evilnickveitch> heheh
[18:55] <evilnickveitch> arosales, yes, and vice-versa, hopefully
[18:56] <evilnickveitch> screencasts should reference the docs
[18:56] <mattyw> I'm done
[18:57] <evilnickveitch> aroslaes, okay :)
[18:57] <arosales> :-)
[18:57] <arosales> empowered
[18:58] <arosales> jcastro: you updated the work itesm we had, correct?
[18:58] <jcastro> yep
[18:58] <mattyw> arosales, jcastro not sure if you saw it but I'd be happy to help out with some of those screencasts
[18:58] <jcastro> all pruned and updated
[18:59] <jcastro> mattyw: yeah, feel free to add a workitem for yourself
[18:59] <jcastro> so like
[18:59] <jcastro> [mattyw] Do a screencast on foo
[18:59] <marcoceppi> arosales: yes, I do plan on attending
[18:59] <arosales> jcastro: 	ok, I will update the whiteboard now then
[18:59] <arosales> marcoceppi: cool, thanks :-)
[18:59] <mattyw> jcastro, on foo?
[18:59] <jcastro> mattyw: or I can just chase you down when we start, whatever floats your boat
[18:59] <arosales> mattyw: +1 on helping with the screen casts :-)
[18:59] <mattyw> jcastro, either way is fine
[19:00] <arosales> jcastro: is always looking for volunteers :-)
[19:00] <jcastro> mreed: more for the docs sessions, but we could use a hand reviewing when we redo the docs parts
[19:00] <jcastro> "does this suck less than when you tried it last?" basically
[19:01] <arosales> evilnickveitch: I am going to port the pad to the white board now
[19:02] <jamespage> arosales, can I have an invite to charm testing please
[19:02] <evilnickveitch> arosales, ok, cool
[19:02] <arosales> I'll capture your recent adds, but be sureto update the whiteboard for any other additions.
[19:05] <arosales> jamespage: https://plus.google.com/hangouts/_/3c1f41916d419fbe0b2b5e31c350ab7e8b3ced47?authuser=0&hl=en
[19:05] <mreed> jcastro:  I would be happy to help review the docs
[19:08] <arosales> marcoceppi: feel free to join the hangout too if you would like ^
[19:08] <marcoceppi> I would need to...put myself together first
[19:08]  * medberry is listening in.
[19:08]  * med_ is listening in.
[19:09] <arosales> https://blueprints.launchpad.net/ubuntu/+spec/servercloud-r-juju-charm-testing
[19:16] <med_> thyat died abruptly
[19:17] <marco-summit> Shouldn't that be a blocker for the charm to not be in the store?
[19:19] <marco-summit> nevermind, I guess you can't fault the charm for the provider not being fully supportive.
[19:19] <marco-summit> Right
[19:19] <wedgwood> how well can you test relations if the interfaces are not well-defined?
[19:22] <brunogirin> does it mean that the charm should be able to specify what capabilities it expects from providers?
[19:23] <wedgwood> jamespage: the u1 guys have been adding lots of unit tests to the charms they're touching
[19:23] <jcastro> http://jujucharms.com/~ju-jistics-hackers/precise/phoronix-test-suite
[19:23] <dweaver> I agree the idea of listing provider capabilities that a charm is has as dependency would mean you get immediate feedback when deploying a charm that a provider hasn't got the right capabilities.
[19:27] <wedgwood> arosales: the interface *names* are defined, but not the data/handshake they make
[19:27] <marcoceppi> wedgewood, there's not strict convention for that, and that's a limitation for charmrunner. It only checks if the relations is made without errors, not that the data is properly being sent on the wire
[19:27] <wedgwood> so I suppose it's really the workload testing that might be an issue. "pushing data" as you say
[19:28] <marcoceppi> those are all arbitrary names
[19:28] <marcoceppi> The interface is the real important part
[19:29] <marcoceppi> So you can have a db relation
[19:29] <marcoceppi> actually
[19:29] <marcoceppi> let me just join the hangout
[19:29] <wedgwood> heh
[19:31] <wedgwood> marcoceppi: exactly! but where's that spec?
[19:31] <arosales> wedgwood: we don't have one :-/
[19:32] <arosales> but taking volunteers :-)
[19:32] <wedgwood> marcoceppi: that's right. yep.
[19:32] <marcoceppi> https://juju.ubuntu.com/Interfaces/http
[19:32] <wedgwood> I've run into this a few times, particularly when we are tempted to overload an existing interface.
[19:33] <wedgwood> marcoceppi: ah, yes, all of those are gone
[19:33] <marcoceppi> http://jujucharms.com/interfaces/mysql
[19:34] <wedgwood> I think that would be helpful
[19:34] <brunogirin> it'd be nice to know what parameters each interface provide too
[19:34] <brunogirin> and makes sure that similar interface such as mysql and postgresql behave in the same wya
[19:35] <dweaver> How about create a section in the wiki for documenting interfaces to start people off.
[19:35] <wedgwood> here's the use-case I'm thinking of: we like database interfaces to provide a DML user and a DDL user. the typical example stack, wordpress, doesn't include that.
[19:40] <arosales> hangout url https://plus.google.com/hangouts/_/3c1f41916d419fbe0b2b5e31c350ab7e8b3ced47?authuser=0&hl=en
[19:40] <arosales> wedgwood:  ^ feel free to join
[19:42] <marcoceppi> http://paste.ubuntu.com/5588570/
[19:42] <wedgwood> arosales: ta, but I'd have to get that up on my laptop. little time left, and I'm pretty happy with the discussion so far.
[19:43] <wedgwood> marcoceppi: that looks very much like a deployment tool
[19:44] <wedgwood> marcoceppi: which is something we're trying to get figured out
[19:44] <arosales> wedgwood: ok, np. Just wanted to offer :-)
[19:44] <wedgwood> arosales: ok, and now I'm interested again. jumping in
[19:44] <brunogirin> like cucumber? http://cukes.info/
[19:48] <brunogirin> jenkins is the CI system so would actually call the testing framework
[19:49] <dweaver> https://github.com/cucumber/cucumber/wiki/Jenkins-integration
[19:50] <brunogirin> at the end of the day, there are dozens of testing framework out there so as long as we choose one that is well documented, rather than inventing yet another one, it should make it easier to write tests
[19:50] <elmo> cucumber requires ruby
[19:50] <elmo> as in, requires you, the charm writer, to write ruby to translate from the "pseudo-English" DSL into what it's actually doing
[19:51] <elmo> introducting YAL into the mix isn't a good idea, IMO
[19:51] <elmo> (disclaimer: it's been a while since I looked at cucumuber, I say it requires, but I really meant 'required' - I don't know if it still does)
[19:52] <brunogirin> elmo: fair enough, I didn't think of that
[19:52] <med_> yet another lang
[19:52] <elmo> (YAL == yet another language)
[19:52] <elmo> sorry
[19:53] <med_> brunogirin, was the cuke sponsor
[19:53] <dweaver> Yes, requires ruby
[19:53] <med_> not mew
[19:53] <rbasak> There are other implementations of the same DSL in other languages, I believe. There's a Python implementation IIRC.
[19:53]  * rbasak looks
[19:53] <mattyw> rbasak, lettuce I think?
[19:54] <rbasak> That rings a bell
[19:54] <mattyw> rbasak, ISTR it wasn't too bad
[19:54] <rbasak> But ultimately it's just the general mechanism, right? Do we need to settle on a language at all? LIke we don't for charms.
[19:54] <arosales> rbasak: http://lettuce.it/ correct ?
[19:55] <brunogirin> what about doing something on top of the Go testing framework? I don't know how good it is
[19:55] <rbasak> Yes. Thanks mattyw and arosales
[19:55] <mattyw> brunogirin, go's testing language is fairly minimal but ok
[19:55] <brunogirin> and what about performance and security testing of charms? or is that opening another can of worms?
[19:56] <mattyw> brunogirin, there's also https://launchpad.net/gocheck
[19:56] <arosales> https://groups.google.com/forum/?fromgroups=#!topic/django-users/979lJojyxs0
[19:56] <arosales> good stuff
[19:56] <brunogirin> thanks, it's moving slowly :-)
[19:56] <med_> Gut Stuph, ya vohl.
[19:57] <med_> thanks arosales
[19:58] <mattyw> Thanks guys, see you tomorrow