| [05:52] <rick_h__> morning | |
| [06:07] <huwshimi> rick_h__: Morning | |
| [06:14] <rick_h__> huwshimi: hey, funny running into you there | |
| [06:14] <huwshimi> rick_h__: Yeah, strange to see you at this time :) | |
| [06:14] <huwshimi> rick_h__: How was the travel? | |
| [06:17] <rick_h__> huwshimi: you know, time in a tin can flying through the sky | |
| [06:17] <huwshimi> hehe | |
| [06:17] <rick_h__> working on catching up on sleep | |
| [07:10] <rogpeppe> mornin' all | |
| [07:14] <huwshimi> Morning | |
| [07:21] <urulama> morning all | |
| [07:35] <rogpeppe> urulama: hiya | |
| [13:19] <rogpeppe> anyone with good http-fu know what's going on at line 1503 here? http://golang.org/src/pkg/net/http/server.go#L1503 | |
| [14:45] <jcsackett> Makyo: do you know if a unit would ever *not* have a state outside of being uncommitted? | |
| [14:46] <Makyo> jcsackett, units should always have a state, I believe. Is something going wrong? | |
| [14:46] <jcsackett> Makyo: no, not at all. that's a good answer. :) | |
| [14:46] <jcsackett> uncommitted units don't have agent_state, which makes sense. | |
| [14:46] <rogpeppe> urulama, jrwren: https://github.com/juju/charmstore/pull/52 | |
| [14:47] <jcsackett> the QA issue you found was b/c the condition for identifying uncommitted units isn't correct once the service is deployed. | |
| [14:47] <rogpeppe> anyone know if frankban is around today? | |
| [14:47] <jcsackett> but since they don't have state, we can use that. | |
| [14:47] <jcsackett> and that's always accurate. | |
| [14:47] <rogpeppe> or bac | |
| [14:47] <Makyo> rogpeppe, frankban is out today on swap. | |
| [14:47] <rogpeppe> Makyo: ah, thanks | |
| [14:47] <Makyo> jcsackett, ah, yeah, that makes sense | |
| [14:48] <jcsackett> rogpeppe: bac is in nuremburg. | |
| [14:48] <Makyo> Boo, forgot about that., | |
| [14:48] <Makyo> CI doesn't seem to be working for me. | |
| [14:48] <rogpeppe> jcsackett: ah, i'd forgotten about that | |
| [14:48] <Makyo> Maybe jenkins needs a kick in the pants | |
| [14:49] <rogpeppe> i guess i'm just not gonna land anything today | |
| [14:49] * rogpeppe goes to the pub :-) | |
| [14:50] * jcsackett laughs | |
| [14:50] <jcsackett> rogpeppe: starved for reviews, or jenkins issues? | |
| [14:50] <rogpeppe> jcsackett: reviews | |
| [14:51] <rogpeppe> jcsackett: i think there's only one person around ( jrwren ) that can review the code, but i need two reviews for landing | |
| [14:52] <jcsackett> rogpeppe: can Makyo or hatch sub in for today, or do you need deep knowledge of the code base? they're both go-literate, if memory serves. | |
| [14:52] <jcsackett> (it's not ideal, but better than blocking) | |
| [14:54] <Makyo> jujugui call in 6 | |
| [14:55] <urulama> rogpeppe: i'll try to do it | |
| [14:55] <urulama> rogpeppe: but it's a huge one and with interruptions, it might take a while :) | |
| [14:55] <rogpeppe> urulama: if you could, that would be great. unfortunately, it's a bit of a bad of bits. | |
| [14:56] <rogpeppe> perhaps i should try to split it up | |
| [14:56] <rogpeppe> anyone know of a good tool for splitting up branches? | |
| [14:57] * urulama suggests an axe | |
| [14:57] <rogpeppe> lol | |
| [15:00] <Makyo> jujugui call now. | |
| [15:18] <jrwren> http://ci.jujugui.org:8080 its up | |
| [15:18] <jrwren> I have no answers for why it wasn't up. | |
| [15:19] <Makyo> jrwren, awesome, thanks | |
| [15:34] <kadams54> jcsackett: finished QAing your PR | |
| [15:35] <kadams54> Makyo: you mentioned running debug logging while setting the juju-gui-source… how do I do that? | |
| [15:35] <Makyo> juju debug-log | |
| [15:38] <kadams54> This is what I get: http://pastie.org/private/8kodvcgaorziu7fq9evxa | |
| [15:39] <kadams54> Seems to be an error right off the bat, maybe in fetching the new source from git? | |
| [15:44] <kadams54> guihelp: ^ | |
| [15:45] <jcsackett> kadams54: thanks. | |
| [16:08] <rick_h__> hey all, anyone know what's the status of hatch's card and potential gui release? | |
| [16:09] <rick_h__> jujugui ^ | |
| [16:14] <Makyo> rick_h__, working on QAing now, then checking on some problems with the charm, then I'll see about starting release prep. | |
| [16:14] <rick_h__> Makyo: awesome thanks. | |
| [16:14] <rick_h__> Makyo: what's up with the charm? | |
| [16:15] <urulama> rogpeppe: i've looked your PR ... it's huge, for last parts /v4/stats.go and stats_test.go i need more time ... | |
| [16:16] <Makyo> rick_h__, some folks are having trouble setting juju-gui-source - looks like it fails to start back up. | |
| [16:16] <Makyo> Going to research that. | |
| [16:16] <rogpeppe> urulama: yes, i'm sorry about that | |
| [16:16] <rogpeppe> urulama: i fixed lots of things in an organic way when trying to make the tests pass | |
| [16:16] <rick_h__> Makyo: note that only works in trusty and not precise due to git changes | |
| [16:16] <urulama> rogpeppe: yes, that's what i'm worried about | |
| [16:17] <urulama> need to go to dinner | |
| [16:17] <rogpeppe> urulama: i could spend a few hours splitting it up | |
| [16:18] <Makyo> rick_h__, Aha! Good to note. kadams54 ^ | |
| [16:19] <rick_h__> Makyo: yea, would love to get details on the series and the string used for that field to make sure it's supposed to work | |
| [16:19] <rick_h__> Makyo: so trusty + 'develop' should be trunk and be good to test with | |
| [16:19] <rick_h__> and should come up, if not there's a log file to check in /var/log/juju/all-? | |
| [17:16] <kadams54> rick_h__, Makyo: alas, I'm in trusty. | |
| [17:16] <kadams54> I didn't see anything obvious in the all-machines.log file, but I'm going to give it another go, this time tailing both that and running debug-log | |
| [17:20] * rogpeppe is done for the day | |
| [17:20] <rogpeppe> g'night all | |
| [17:38] <Makyo> Can't seem to deploy the charm at all, hmm. | |
| [17:40] <kadams54> rick_h__, Makyo: And of course, this time, it appears to have worked for me. FWIW, I confirmed that my card, bug #1341653, seems to be fixed. | |
| [17:40] <Makyo> kadams54, Woo! | |
| [17:40] <Makyo> I might be running into a juju bug, not sure. Trying on other computer. | |
| [17:42] <kadams54> I'm going to see if I can clear a few more cards out of the QA queue. | |
| [17:51] <Makyo> Yeah, seems to be something weird going on in tip, up and QAing juju-gui-source now. | |
| [18:09] <hatch> Makyo juju-core is failing CI atm so you'd best to use a stable release | |
| [18:09] <Makyo> hatch, so I found. | |
| [18:09] <Makyo> Now go back to not working. | |
| [18:10] <Makyo> It's a holiday :) | |
| [18:10] <hatch> haha I am not working :) I have to do html's and css's for my dads website....bleh, I wonder if I could convince huw to do this for me lol | |
| [18:10] <hatch> right Monday Holiday | |
| [18:10] <kadams54> Provincial Day! | |
| [18:10] <tvansteenburgh> anyone in here familiar with the charmhelpers.contrib.jujugui python module? | |
| [18:11] <hatch> jujugui https://twitter.com/FromAnEgg/status/496158671710466049 hahaha | |
| [18:11] <tvansteenburgh> i have a MR to remove it, wondering if anyone here cares | |
| [18:11] <hatch> tvansteenburgh I've never even heard of it :) what does it do? | |
| [18:12] <tvansteenburgh> nothing any more! muwhahaha | |
| [18:12] <hatch> lol w00t | |
| [21:09] <urulama> hi there, jujugui | |
| [21:10] * urulama hears pings go all over the world ;) | |
| [21:10] <hatch> haha hey urulama how goes ze Germans? | |
| [21:10] <hatch> (little Top Gear reference there) | |
| [21:11] <urulama> ze Germanz are fine ... at least their bear and pork products are ;) | |
| [21:11] <urulama> s/bear/beer | |
| [21:12] * urulama now wonders what a bear tastes like | |
| [21:12] <hatch> haha | |
| [21:12] <hatch> tastes pretty gamey | |
| [21:12] <urulama> you had one? | |
| [21:12] <hatch> of course | |
| [21:12] <hatch> elk deer, etc etc | |
| [21:13] <hatch> I live in the North remember | |
| [21:13] <hatch> north of the wall ;) | |
| [21:13] <urulama> survival of the fittest :) | |
| [21:14] <urulama> i had deer and boar and whatnot, but not bears ... and we have a lot of them, just not allowed to eat them | |
| [21:14] <urulama> so how's it going? CI working now? | |
| [21:15] <hatch> no idea, I'm off today | |
| [21:15] * kadams54 is going to start referring to hatch at "that wilder". | |
| [21:16] <rick_h__> howdy all | |
| [21:16] <kadams54> urulama: I just landed a branch for juju-gui, so that CI seems to be working | |
| [21:16] <kadams54> rick_h__: hey there | |
| [21:17] <urulama> hatch: remember the talk about being away from computer ... now's probably the time then :) | |
| [21:18] <urulama> kadams54: ok, good news | |
| [21:19] <rick_h__> Makyo: issues? no release it sounds like? | |
| [21:20] <hatch> hey rick_h__ | |
| [21:21] <hatch> urulama haha, I'm sitting in the back yard on the swingy chair with a beer working on my dads website.....it's pretty day-ofish :) | |
| [21:21] <rick_h__> hey hatch | |
| [21:35] <Makyo> rick_h__, I think release is a go, pulling together changelog now. May actually go out tomorrow. | |
| [21:35] <Makyo> (was on the patio, sorry) | |
| [22:15] <hatch> to submit forms on this vps I have to write a PHP script.... | |
| [22:16] <hatch> I don't even think I remember how to write a form submission script in PHP any longer lol | |
| [22:17] <hatch> I can't even write it in python.... | |
| [22:17] <hatch> what kind of vps restricts to ONLY php | |
| [22:17] <hatch> haha | |
| [22:38] <Makyo> Someone landed a branch while I was writing the change log and now there's a conflict >:/ | |
| [23:02] <huwshimi> Morning | |
| [23:24] <huwshimi> hatch: Morning, in a test I have a beforeEach that sets "placeUnit: utils.makeStubFunction()". How would I go about having one test that has the real function there? | |
| [23:24] <Makyo> jujugui GUI release is done. Going to get dinner prepped, then work on charm release. | |
| [23:24] <huwshimi> Makyo: Yay! | |
| [23:25] <Makyo> On going through the change log, decided on 1.1.1, since much of our work was done behind a feature flag meant for a larger release. | |
| [23:25] <hatch> huwshimi have a code sample? From that one there it looks like you're just passing in a fake stub not the actual ecs object? | |
| [23:25] <hatch> Makyo +1 Thanks for doing the release! | |
| [23:26] <huwshimi> hatch: Yeah, so it's stubbed out for all the tests, but it one test I don't want to stub it out... | |
| [23:26] <hatch> huwshimi well it looks like your not stubbing it out but passing a mock ecs in? | |
| [23:27] <huwshimi> hatch: https://github.com/juju/juju-gui/blob/develop/test/test_machine_view_panel.js#L223 | |
| [23:28] <hatch> huwshimi ok yeah you're passing a mock in, you'll need to set the ecs to a real ecs object if you want to use the real placeUnit | |
| [23:28] <huwshimi> hatch: Ah ok, so any ideas on how to do that? | |
| [23:28] <huwshimi> Do we do this anywhere else? | |
| [23:28] <hatch> you can't really 'just' use placeUnit because it relies on the ecs instance in a lot of places | |
| [23:28] <hatch> well what are you testing? | |
| [23:29] <huwshimi> hatch: OK, so I'm trying to test that a container token is created when placeUnit occurs | |
| [23:30] <huwshimi> I'll push up my code | |
| [23:31] <hatch> well you don't really need to do that | |
| [23:31] <hatch> you just need to test that placeUnit is called with the proper data | |
| [23:31] <hatch> then there should be a test which takes said data and makes sure placeUnit does the proper stuff with it | |
| [23:31] <hatch> ^ huwshimi | |