=== mhall119_ is now known as mhall119 === rogpeppe2 is now known as rogpeppe === jcsackett_ is now known as jcsackett [12:30] * frankban lunches [12:38] hi rick_h_ [12:38] morning bac [12:38] good time while away? [12:38] rick_h_: hey what is the card re: comingsoon about? you just want me to add ssh keys for more people? [12:39] bac: yea, there's a css issue atm and no one could fix it [12:39] ouch [12:39] bac: so I was going to ask for hte details on where it's at and do we need to move it or get access for more folks [12:39] rick_h_: just a matter of adding more ssh keys from lp [12:39] I never really followed where that server is at [12:39] rick_h_: you just ssh to comingsoon.jujucharms.com [12:40] rick_h_: so tell me who... [12:40] bac: ok cool, yea me please, hatch and makyo. [12:41] bac: maybe add a doc section in https://wiki.canonical.com/CDO/Juju/GUI/CI please [12:43] cool google doodle alert [12:45] no doodle here :( [12:47] bac: can you ping when you have time for a quick call please? [12:47] rick_h_: now is good [12:48] bac https://plus.google.com/hangouts/_/7ecpjkem9ldms8f9qpbj5pois8?authuser=1&hl=en === lazypower_travel is now known as lazyPower [13:08] rick_h_: huw's changes are in the branch on comingsoon. no conflicts are shown. i've cleaned and rebuilt. still busted. [13:08] bac: working here [13:08] bac: thanks for the update, stale cache maybe on your end? [13:08] oh. cache problem for me [13:08] very good then [13:38] welcome back bac [13:39] thanks frankban [13:39] hey rick_h_ or frankban [13:39] which version of quickstart would be missing MAAS support? [13:40] sabdfl tried it on his garage maas and could not find an option for maas [13:42] jcastro: could you please file a bug? we can try to add maas to the next release. is --upload-tools still required when using maas? [13:42] ok [13:44] jujugui juju-core 1.17.6 deploys the 1.17.7 tools and will throw an error about instance-id. If you get this error update your juju client (I ran into this on the weekend) [13:44] frankban, aha! He's on vanilla trusty with 1.1 instead of 1.3, which is in the PPA [13:45] anyone have the status for -quickstart in main? Are we still shooting for that? [13:46] jcastro: it's shot down for 14.04 [13:46] jcastro: it'll be in universe [13:46] ack, I assume that will get bumped to 1.13 though? [13:47] jcastro: quickstart in main is blocked by core in main. 1.3 yeah, same question here, is quickstart universe auto-updating? [13:47] it's past freeze now right? I believe we need to have it manually synced [13:47] we should check [13:48] right, they have to upload still I believe [13:48] because they update from the pypi [13:48] jcastro: FWIW 1.3 does not include maas support, we still need to implement that, it wasn't planned [13:48] that's the official 'release' that packaging is built off of [13:48] so my quickstart has an entry for maas [13:49] jcastro: did you have that in your environments.yaml? [13:49] I mean, it has the fields and stuff, do those not work? [13:49] jcastro: I don't think it's a provider we suport [13:50] huh, well, I see it in there, and I certainly didn't add it [13:50] jcastro: in the 'create a new environment' or from your existing environment list? [13:50] jcastro: quickstart works like the following: if you have an environment in your envs.yaml file, which quickstart does not recognize/support, quickstart will still let you use that and even edit the fields it found, but do not expect validation/creation. [13:51] it's in my environments.yaml, but I didn't create it, anyway, that's not as important right now as to what I should tell Mark about MAAS/quickstart, heh [13:52] maas/quickstart doesn't work and wasn't planned. It was deemed a quick 'juju helper' and maas didn't come into the conversation [13:52] so it's all news to us now that you ask abou it. We didn't consider how it would work. Does the deployer/bundles work on maas/ [13:52] ? [13:52] we'd have a lot of questions to ask/figure out to see how quickstart work work in that world [13:57] I don't know if deployer/bundles work on MAAS [13:57] I am not supposed to care about my provider remember? :) [13:58] jcastro: right, that's from an internal point of view. I'll have to ask/figure out that to see what a quckstart tool looks like in a maas world [14:02] yeah [14:02] so I am wondering if we ask you to do X steps for MAAS does quickstart really make a difference? [14:03] "I know how to set up DNS and DHCP, but adding a value to a yaml file? Now you ask too much!" [14:03] jcastro: yea, that's my thing. If we can't do bundles to maas I'm not sure what all we'll be doing for them [14:03] IMO we should be doing bundles in MAAS [14:04] I'm in calls but will try to find some answers. If we can make it near close feature-wise we can schedule some work on quickstart [14:04] provider-specific limitations totally ruin Juju's promise [14:04] rick_h_, yeah I don't think it's big enough to stop what you're doing, I'm just thinking we should have an idea of discussion for vegas [14:05] jcastro: +1 [14:05] like, I think it missing from quickstart isn't a big deal, these people are doing bare metal installs, and a little yaml file is no big deal [14:05] but non-working bundles would be big I think [14:05] right, but I agree bundles on maas should be something we look into [14:05] rick_h_, do we know it doesn't work or do you speculate it doesn't work because we never tried it? [14:06] jcastro: never tried it so just not sure [14:06] so probably doesn't work [14:06] I'm trying to find some report of people doing that [14:06] let me ask [14:06] * rick_h_ hasn't used maas but an excuse to pick up some hardware is always welcome [14:10] * hatch is reviewing huws branch [14:14] jcsackett did you get a chance to try the new http interface on the ghost charm? [14:46] I found a charm this weekend which has multiple maintainers in the maintainer field https://bazaar.launchpad.net/~charmers/charms/precise/haproxy/trunk/view/head:/metadata.yaml [14:46] re our charmworld discussion about multiple maintainers [14:47] atm it only shows the first one [14:48] hatch__: right, so we've got a card to move to a maintainers field and the charm should be updated to use it [14:48] rick_h_: just filed expenses for AWS for last four months [14:48] bac: k, thanks [14:49] rick_h_: fwiw, on canonicaladmin "Sign off by" still shows gary === hatch__ is now known as hatch [14:49] bac: hmm, will look. I thought I was setup for everything [14:50] oops maybe that was an historic one [14:50] if I could log in I'd tell you if I can see it or not [14:50] rick_h_: been diving into charmworld this morning, but I don't have any cards to work on right now. Was wondering if you had any suggestions… [14:51] If not, then this looked like something I could tackle: https://launchpad.net/bugs/1290323 [14:52] kadams54: hop on the stand up a couple min early and we can run through [14:52] k [14:54] jujugui call in 7 kanban please [15:07] Makyo did you see they released a vim plugin for Go? [15:08] hatch, not yet, though I'm intrigued. I don't mind Sublime for Go, but it takes a bit to switch editor modes [15:08] yeah that's my thoughts - right now I switch from sublime > tmux > osx > ubuntu [15:08] I end up hitting about 5 shortcuts before I find the right one [15:08] lol [15:09] This came up on Friday: here's what I currently use for window management/position in OS X: https://github.com/fikovnik/ShiftIt [15:09] hatch, have a link? I already have go.vim [15:09] Pretty basic, not as powerful as Amethyst, but does the stuff I want. [15:10] Makyo umm one sec [15:10] http://blog.gopheracademy.com/vimgo-development-environment [15:10] kadams54 yeah I've been trying moom because I want custom (saved) layouts [15:11] layouts are a big requirement for me [15:13] Makyo: got a sec to chat? [15:13] rick_h_, sure [15:30] rick_h_ I'm going to bench this delta card and work on it as slack so we can keep pushing forward....anything you'd like me on next? [15:30] hatch: state :) [15:30] lol [15:30] hatch: drink your coffee today? [15:30] I just might implement the API I want then :P [15:30] haha drinking tea now [15:30] that'll do [15:31] yea, I mean we need someone to pick up state work and it's blocking a lot of things right now. [15:31] hatch: if you don't want to do that we can find something else [15:31] no no it's fine [15:31] :) [15:32] pow wow time then? [15:32] * hatch puts his politically correct hat on [15:32] lol [15:32] yeah I'm ready whenever [15:32] kadams54: ^ [15:32] link? [15:32] ok, will get a link [15:33] https://plus.google.com/hangouts/_/72cpigoqi36bc4apugurhfskv8?hl=en [15:33] kadams54: feel free to join in the fun ^ as well [16:31] * hatch changes state object name to KylesState [16:31] let's just get it all the way there KylesFault :P [16:31] rofl [16:32] that would be epic [16:32] welcome to the club, it can go next to the class RicksFault, JeffFault, and MattAlwaysWins [16:33] hahaha [16:33] oh man that would make for a funny codebase [16:40] * rick_h_ goes for lunchables [16:53] Makyo there are a bunch of warnings now "Dropping unused function..." from the d3 stuff [16:53] anything to be worried about? [16:55] hatch, no, just some poorly managed dependency tree stuff on D3s side. That's just uglify getting rid of stuff that is never used. [16:55] ahh ok cool [16:55] Gotta give uglify credit for being pretty thorough. [16:58] yup [16:58] little scary if those methods were called by this[callthismethod] [16:58] heh [16:59] rick_h_ I have a question about the ghost inspector and viewNavigate/state lemme know when you're back [17:01] Forgot to mention this in standup, but I'm going to be out in an hour for an appointment. Will be back in the saddle around 3 PM Eastern. [17:02] hatch: k, back [17:02] Also: KylesFault: less scary than SupportedNowAndAlwaysByKyle [17:02] I've got no problem with it being my fault as long as someone else has to support it ;-) [17:02] oooh, I like that one [17:03] https://plus.google.com/hangouts/_/7acpiciu7u09dga8ifaaamclss?hl=en [17:03] FindKyleAt12480004444 [17:03] lol [17:10] jujugui anyone running in OSX with vagrant? Having trouble getting the tests to pass natively with that, though the pass in CI and on other systems. [17:11] Makyo: i am [17:11] Makyo running fine here [17:12] hatch, rick_h_ PR updated, btw. [17:12] Makyo: but i haven't tried the tests ina while [17:12] Hmm, okay. I'm on the old vagrant, will update. [17:14] I have 2-3 tests that fail on a semi-regular basis [17:22] Makyo: cool thanks looking [17:25] kadams54 can you make note of them - they may be because of a cascading state issue (we used to have a lot of those) [17:25] Makyo oh yeah I'm on the new one, sorry try updating [17:26] Working on it. NFS is unhappy [17:26] :( I love the nfs it's so fast [17:26] Oh, nevermind. /etc/exports was messed up [17:27] Had some stuff in there left over from old vbox setup [17:27] Conflicting mount points [17:31] ohh that'll do [17:44] hatch, so the vagrant/nfs has continued to work well? === hatch__ is now known as hatch [17:46] bac yep works like a charm, so fast [17:47] it's virtually instant to load up the site now when running locally [17:47] it used to take a few seconds before [17:47] total time to run the tests has also dramatically sped up [17:50] yay, i'm reading my email backwards and just got to the part where allhands has been bludgeoned to death. whee. [17:51] haha yeah I cheered for that too [17:52] rick_h_ can the filter stuff at the top of getUrl in state.js go? [17:52] hatch: not until we fix the url to not need to worry about 'search' in the url [17:52] we don't -do- filters any longer I don't think [17:52] ohh ok [17:53] hatch: oh, but we do use filters in state. the search term, etc [17:54] hatch: clear and such for updating. and supposedly we'll get to 'tag' like search where you can enter 'precise' and we'll have to detect it as a filter of series [17:54] hatch: but yea, not used to the full extent right now [17:55] 14 mooms remaining :0 [17:55] mooms? [17:56] I thought you were trying moom [17:56] every time you use it it counts down from 100 [17:56] yea, not sure what '14 more' fit into [17:56] huh? missed that [17:57] oh you've bought it already> [17:57] is that a trial one or something? [17:57] oh yea, I just bought it [17:57] oh lol ok [17:57] * hatch is cheap [17:57] lol [17:57] didn't realize it had a trial, just found it in the store [17:57] ohh yeah I went to their website [17:57] you should go register on their site - there are some limitations to the store version [18:39] hatch, mind giving the PR a +/-? [18:39] oops sorry checking [18:40] Makyo done - I thought it had already landed :) [18:40] hatch, sorry, just wanted to make sure! [19:01] jujugui going afk for a little bit before the interview. [19:54] hatch__: ping, hop into the call === hatch__ is now known as hatch [20:59] Makyo: can you come back for a min? [21:00] Ack, sorry [21:04] Makyo lol that system 76 [21:04] I KNOW [21:04] Ugh [21:04] MOAR FANS! [21:05] rofl [21:05] Need to get a separate mic for it. I think th emic is on top of one of the fans. [21:05] Stupid. [21:05] lol [21:05] Oh well. [21:05] Looking at Thinkpads to replace it. [21:06] ugh sorry [21:06] It makes an okay "desktop", but the Air is a little less than ideal for sprint work, since it's not linux on metal. Makes Go work tough. [21:06] kadams ran away? [21:06] Makyo: hatch looks like his branch is up for review, if either of you get time to check it out before EOD appreciate it. [21:07] on it [21:07] Makyo: yea, I've got $$ in my pocket but nothing I'm happy to spend $$ on. [21:07] Or, well, let me walk dogs first. Will be free after. [21:07] rick_h_, yeah. You still pro thinkpad? [21:07] The air is a nice form factor, but no metal and crappy keyboard :( [21:07] I love my 230, but want a more air form factor, better display, HD [21:07] the 240 does that but changes the buttons to a 'click pad' which reviews say is awful [21:08] Makyo I write go in OSX [21:08] Makyo: so I'm thinkpad cranky atm. They botched the carbon, messed up the 240 trackpoint [21:08] I haven't found a time when it doesn't work in OSX but does in Ubuntu [21:08] oh boy did they ever botch the carbon [21:08] hatch, oh, cool. Was curious about the juju project. [21:08] wth did they do to that keyboard lol [21:08] Makyo: honestly I was looking at the xps13, some people seem to like it, but a friend I trust hated it [21:09] Makyo: so I'm hoping refresh season might hit middle of the year and something good comes out of it [21:09] Makyo juju probably won't work as it relies on Ubuntu specific dependencies [21:09] hatch, yeah, that was my concern with upcoming Gophercon [21:09] Makyo do you have the haswell equipped air? [21:09] I'll get a VM up and running. [21:09] Um. [21:10] I think as long as it's -not- haswell then 14.04 will run on metal [21:10] Makyo: http://discourse.ubuntu.com/t/laptops-for-ubuntu-cant-make-up-my-mind/1541 [21:10] ok all, EOD have fun. Thanks again. [21:11] cya rick_h_ [21:13] I'm going to try fusion at some point here to try and get a good vm running === hatch__ is now known as hatch [21:31] http://macaw.co/ [22:07] Morning [22:08] morning [22:14] Big huge +1 on NFS mount./ [22:14] haha see! [22:14] lol [22:18] bac: looks like that's a winning change. :) [22:18] morning huwshimi [22:18] rick_h_: well, it was a bit sleazy [22:18] rick_h_: you talking about ceph? [22:19] hi huwshimi! [22:19] Morning all [22:19] bac nfs [22:19] bac how was your vacation? [22:20] ah, nfs not sleazy at all [22:20] hatch: it was fantastic! beautiful, deserted beaches [22:20] bac: oh I missed an nfs change [22:20] hatch: only downside is the dog is now mopey. he had such a good time he's pissed to be home [22:20] bac: I mean I missed a ceph change [22:21] bac what was the result of that then? [22:21] rick_h_: after lots of experimenting on staging, i decided the best course was just to delete ceph-22 and ceph-20. when ingested again they were fine. [22:21] bac: ah cool thanks [22:21] i can't explain what happened [22:21] that's where the sleaze comes in [22:22] bac: yea works for me. I think it fell into an exception. We'll see if we hit issues again [22:22] let's blame benji and move on [22:22] hah, good call [22:24] rick_h_: I see you've played some poker. Anything in particular you'd like me to move on to next? [22:25] hatch: Thanks for landing that branch. [22:25] huwshimi: so it's not on here, but you can get the inspector to show up (non ghost) [22:25] and we need to fit it into the full size of the sidebar [22:25] huwshimi: so go to comingsoon, deploy a charm, hit the close button, then click on the service block to get the normal inspector [22:26] oh, we can't now that the MV panel is there heh [22:26] ok, guess not that one [22:27] oh try without the mv flag [22:27] just il [22:27] rick_h_: Yep, got it. [22:27] huwshimi: ok, so the normal inspector on the left on a tall display doesn't fill the height properly [22:28] rick_h_: Ah yeah, looks like another candidate for flexbox. [22:28] huwshimi: so that'd be some good updates around current stuff. [22:28] rick_h_: Would save all our resize calculations too. [22:28] huwshimi: and aside from that, we can keep moving forward on the machine view parts. We need the deployed unit tokens. Those I think should be widgets and done like the charm token [22:28] huwshimi: cool, if it works across our supported browsers I'm all for it [22:29] rick_h_: Yep, it's how the machine view stuff is working [22:29] huwshimi: cool [22:29] huwshimi: so the 'create a deployed service token' or the 'create machine control in unplaced units panel' would be next UI components to start on [22:30] huwshimi: that help? [22:31] rick_h_: Yep, that's great. Just poking around at the wireframes to see if we have details for the expanded units etc. [22:32] I think there's enough there to get started. [22:32] huwshimi: yea, there should be an expanded view when you place a unit token, but we can start out with just a small widget that can represent a unit. Assume it gets the data from the unit model as a raw JS object for now [22:32] huwshimi: much like the charm/bundle tokens get their model, but as a raw JS object [22:33] rick_h_: OK [22:37] yay....ghost state done [22:37] I should add "// I hate this" everywhere :P [22:37] lol [22:39] rick_h_ are you up for a post-implementation review? [22:47] hatch: let's do it in the morning? [22:48] sounds good [22:48] hatch: and maybe not "I hate this" but "XXX this should go YY in refactor card" [22:48] :) [22:48] it's in my github fork if you get curious [22:48] lol yeah [22:48] hatch: k, just create a WIP pull request please [22:49] I can comment on that but not just looking at the code [22:49] https://github.com/juju/juju-gui/pull/214 [22:49] ty [22:49] will try to look after the boy goes to bed [22:49] yep no rush [22:50] stepping away for a bit [22:50] bbiab [23:55] kadams54 you're back [23:55] you kinda disappeared lol [23:55] Makyo I made some notes in the interview doc, feel free to comment/change as you see fit [23:55] Comcast: "Your area is experiencing an outtage. Estimated time for repair is 8 PM Eastern." [23:56] noice [23:56] someone drive into the internet pole? [23:56] :) [23:57] *sigh* [23:57] the challenge of working from home sometimes :) [23:58] My cell provider doesn't charge extra for tethering so I try to jump on that if my home network goes bonkers