[00:47] hey huwshimi I can hopefully talk briefly in about 30 min or so. work for you? [00:48] gary_poster: That's fine, if it's too much trouble we can reschedule to another day [00:49] huwshimi, s'ok, talking to you at least once a week seems a bare minimum :-) and I want to talk about the qa thing and notifications [00:49] gary_poster: Sure :) [01:10] hey huwshimi . meet you in https://plus.google.com/hangouts/_/c4cb6fc957b0334b9e26b3d2212f9b164613fbd9 ? [01:10] garyposter: On way [01:10] cool [01:36] ahoy! [01:53] hey hatch. :-) I'm stuck over on the Mac side because OS X demanded a software update. If you think of it, could you beg bcsaller's pardon for me that I couldn't finish his review tonight? He'll need to find someone else to do it tomorry. [01:53] textual is a nice client [01:53] garyposter: still here, its no problem. [01:54] cool thanks bcsaller [01:54] sad that I enjoy the ubuntu side a lot except I can't find an IRC client I like or a Mail client I like [01:54] Mac options suit me better there [01:55] Mail in particular [01:55] but xchat doesn't even have undo in its text entry. :-/ [01:55] OK, I'm running away. [01:55] Have a great weekend, everyone! [01:55] have a good one [02:03] sorry I stepped away :) [02:03] bcsaller: I can do your review in the am, I am just tracking down some of the final bugs (I hope) in my model conversion for the ghost inspector [02:04] hatch: sounds good, thanks. I'm going to head off, have a good one [02:04] you too [12:40] jujugui: has anyone seen this loadFixture failure ?http://paste.ubuntu.com/6132453/ i've been getting it a lot when running lbox. 'make check' by itself does not fail. i think my vm may be memory constrained. i killed some extraneous processes and the next time 'lbox submit' worked. [12:42] jujugui: I have a mostly mechanical branch up for review: https://codereview.appspot.com/13786046 [12:42] benji: ok [12:42] I've not seen that error. [12:43] I don't think the JSON string in question would be so big as to exceed the available memory. [12:58] hey luca__! just so you know, I'm out today--need a day off. back monday [12:59] I'm trying to hand off emails, like the one from Sally, to people on th team before I run away [12:59] gary_poster: no worries, have fun :D [12:59] thx :-) [12:59] benji: sorry i nodded off a bit doing your review. done now. there should be a waiver for big but mechanical branches only requiring one review. [13:00] benji: that failure was repeatable but only using lbox. odd. now the branch landed so i can't poke at it anymore [13:02] bac: thanks for the review. Re. waver: I only intend on getting one review. [13:02] the eye-burning fumes of fabuloso stinking up the house must mean its friday [13:45] jujugui: I may have broken trunk (visually, tests pass). I'll report back here in a few minutes. [13:47] tisk tisk tisk :P [14:01] anyone still need a roomie for SFO? [14:09] bac: did you get past that json error? [14:11] oh rick_h_ you're here today... [14:11] hatch: no, just checking in while loading hte camper :P [14:12] ohhh big mistake - there is a bug with your name alllll over it :P [14:12] hatch: :( which one is that? I see a few going by in the email so far [14:12] unit remove button does nothing [14:12] I was sure you landed that fix [14:13] hatch: https://code.launchpad.net/~rharding/juju-gui/resolve-retry/+merge/185784 [14:14] interesting [14:14] well have a good trip! [14:14] hatch: ty, sorry for the bug/issue :/ [14:15] ahh it happens [14:16] * rick_h_ goes to test locally real quick [14:17] hatch: oh though, this is under the running units [14:17] yeah... [14:20] benji: i'm free to talk whenever you are about next steps [14:20] brb [14:20] BradCrittenden: k, I'll start a hangout [14:21] benji: give me a few minutes. === BradCrittenden is now known as bac [14:21] bac: k [14:23] benji: ok, have headphones now [14:24] bac: https://plus.google.com/hangouts/_/dd1bb934156a9de6972d2df5d8b98227bff12f66?hl=en [14:33] bac: thanks for the qa and the email! [14:34] hatch: are you looking at this bug. I've chased it to a strange place [14:35] rick_h_: I probably will be unless someone else picks it up between now and when I finish my current branch [14:35] hatch: would love to transfer what I found if you've got a sec [14:35] lets do it! [14:50] jujugui call in 10 kanban now [14:59] jujugui call in 1 [15:07] fastest daily and retrospective......ever! [15:12] that meeting was too short. [15:12] i'm conditioned to race off to lunch as soon as i hang up but it is too early now [15:12] oh haha [15:20] jujugui if anyone is looking for a task today plz see the cards under Urgent on the kanban [15:29] that moment you realize the bug you think you caused is in trunk :/ [15:33] jujugui can someone do a fresh checkout of trunk and check a quick bug for me plz? [15:33] hatch: will do [15:33] does it need to be fresh or just a pull? [15:33] bac: thanks, steps to repro: .... [15:33] pull should be fine [15:34] put a breakpoint handlers.js:152 deploy a charm with all defaults, once it's ready click and destroy the service [15:34] ok, i have r1072 [15:34] shortly after unitInfo will be called [15:34] which should throw a console error [15:35] about no service to add units too [15:38] it doesn't happen on comingsoon so that's why I am confused... [15:40] hatch: i see Units added without matching Service [15:40] awesome thanks!! [15:41] creating bug [15:42] np [16:26] jujugui looking for two reviews and a solid qa please https://codereview.appspot.com/13334054/ [16:26] small diff - big difference :) [16:27] hatch: on it [16:27] thanks, and I can now get onto yours [16:27] :) [16:52] hatch: i can review now but can't qa for a while [16:54] * bac bbiab [16:55] bcsaller bac: thanks! [16:55] bcsaller: for your branch is there a QA story? [16:56] hatch: yes, from test/data you should be able to DnD wp-deployer.yaml on the canvas and then you can look at the relations tab, things should be sensible now but they were not doing the same thing with trunk [16:57] nice [17:17] bcsaller: comments for the review have been sent and new code is up on reitveld [17:17] round 2... fight! [17:17] lol [17:17] * hatch holds his finger over the 'remove lgtm' button on bcsallers review [17:20] doh [17:21] does anyone have a juju-core lxc setup to test my latest branch? [17:21] jujugui ^ [17:37] yes [17:38] bac: ok it's landed already so if you could still qa that would be awesome :) [17:38] so i show up for my 1pm haircut, wait 15 minutes and then ask how long it will be. "Maybe you come back, one hour." argh. [17:38] lol [17:38] time for a razer [17:38] hatch: so you want me to qa trunk? [17:40] yeah just deploying a charm and stuff [17:40] nothing too crazy, just want to make sure my changes didn't break anything for core [17:43] hatch: spinning up a local env. will be a while [18:22] hatch: tried to delete services and am getting null pointer exceptions [18:22] relation.js:541 [18:23] bac: sorry I stepped away for lunch [18:23] hmm ok checking [18:23] bac: can you gist up the stack trace? [18:23] hatch: destroyed services are not removed [18:24] hmm [18:24] trying to see if I can dupe on sandbox [18:24] no stack trace [18:26] very odd - no issue beyond the pre-existing units error on sandbox [18:26] what's the actual error you get? [18:29] looks like I'm going to have to setup my other machine so that I can actually work on this and debug the error [19:06] bac: when you set yours up did you go through the guide on juju.ubuntu.com ? [19:06] https://juju.ubuntu.com/docs/config-local.html [19:07] oh nm [19:07] it threw a permissions error but only because the bootstrap node hadn't yet started [19:07] i did the same steps but not by following that guide. [19:08] very confusing error message [19:08] jorge wrote something on askubuntu that is good [19:08] I have it up and running now - but I'm going to file a bug about this error message [19:09] because it acts like it's broken [19:09] even though the boostrap node is still going [19:09] s/going/starting [19:12] hatch i've tried again but cannot replicate [19:13] so the bug could have been a cache issue? [19:14] I'm just pumped I can finally run juju :D [19:14] wow I'm a nerd [19:16] hatch: a real nerd might've gotten to that point a little sooner. :) [19:17] hatch: as to the failure, i'm not sure what is going on. odd things happen on my vm that seem unique. [19:17] bac: lol - ok this nerd has 'real' work to do though :D [19:17] oh right you had another odd failure yesterday right? [19:18] yeah, but i've seen it again. zombie services. [19:19] ohhh I think there was a bug about that [19:19] sec [19:19] oh no this was zombie inspector [19:20] when you juju status the service is gone? [19:20] but it shows in the GUI? [19:21] bac: how long does it take the GUI to come up? [19:21] long time [19:21] mine has been sitting on 'pending' for a good 10 mins [19:21] oh ok [19:22] yeah. you like coffee? maybe you should go make some. [19:22] lol [19:22] deal! [19:28] bac: ok it's up, now when I expose it, its public address is some hyper local ip [19:28] i don't know what that means. what is it? [19:29] 10.0.XXX.XXX style IP [19:30] which just hangs on connecting, I'm guessing this is an lxc thing? [19:30] you trying to connect from osx or inside the vm? [19:30] i do the latter [19:30] the former is asking for pain [19:31] from OSX, the vm doesn't have a GUI, it's a box in my basement [19:32] oh my. so you have a separate box running a vm? [19:32] well it's not a vm [19:32] if you can ssh to it you can set up a tunnel [19:32] oh [19:32] it's running precise [19:32] so it is a machine [19:32] right [19:32] a (v)m [19:33] well that's a little better [19:33] yeah, it also doesn't have a keyboard or monitor so I'm ssh'd into it right now :) [19:39] so is this ip issue something we can improve within our charm? [19:40] it feels to me like a juju-core issue [19:40] because exposing to an internal ip doesn't really work :) [19:53] * benji gets ready to go to the doctor. [19:53] hatch: it isn't a charm thing. if the machine pool has public ip addresses then when exposed it will be public [19:53] bac: ahh ok so this is definitely a juju/lxc issue [19:54] i've been looking online to try and find out how to expose this [19:54] but not having much luck [19:54] hatch: i've been trying to create a tunnel into my vm but can't get it to work. something lik [19:54] ssh -N -p 22 -c 3des bac@saucy64.local -L 8888/10.0.3.162/80 [19:59] yeah I haven't had any luck either [19:59] yay [19:59] this works [19:59] ssh -N -p 22 -c 3des bac@saucy64.local -L 1111/10.0.3.162/443 [20:00] where saucy64.local is my vm and juju-gui is on 10.0.3.162 [20:00] hmm when I do that it just hangs in the console [20:01] it is supposed to. that's what -N says to do [20:01] ok then in your browser you fisit saucy64.local:1111 ? [20:01] run the ssh and then on os x hit http://127.0.0.1:1111 in your favorite browser [20:01] Can anyone verify that the charm store API omits 'peers' from its results? For example cs:precise/hbase-6 should return peers but I'm seeing nothing. My guess is we hardcode provides and requires and its an oversight [20:02] bac: needed https [20:02] but you rock! [20:02] :D [20:02] hatch: no, that's why 443 is on the ssh rule [20:02] oh you're right [20:02] makes sense [20:02] bcsaller: looking [20:04] bcsaller: I see no peer either [20:04] hatch: there is an os x program called 'ssh tunnel manager' which may make it easier... [20:04] hatch: yeah, thats a blocker :) [20:04] bac: now I have no idea what the pw is for the gui lol [20:04] bcsaller: for which one? [20:04] sorry brain fried [20:04] admin-secret in your .juju/environment.yaml local: [20:05] hatch: for me, adding peer support to fakebackend so we can test those better, but it looks like the real API call doesn't return the proper data [20:06] ohh right right [20:06] bac: oh right, woops [20:06] bcsaller: what about on api3? [20:06] looking [20:06] nope [20:06] :) [20:06] this actually seems like a pretty huge issue [20:07] and to verify, its in the actual JSON and in the metadata.yaml http://manage.jujucharms.com/charms/precise/hbase/json [20:08] so yeah... [20:08] I'd guess its a very small fix though, I expect they hard coded provides and requires [20:08] yeah probably right [20:09] holy poop juju rocks [20:10] just started a wordpress blog [20:10] NO BIG DEAL! [21:49] jcastro: kickin around? I have a question about the wordpress charm [22:03] bcsaller: did you land the card that's on the board in review? [22:03] I want to start on one of those bugs (if I can) [22:03] yes [22:03] moved [22:03] rock on [22:03] so the peer relations thing [22:03] what did you decide? [22:04] oh :D [22:06] made a new card and linked a stub branch, but I'd rather pre-impl that with someone that knows the codebase before I take it further, I do expect my patch works but I can't deploy it [22:12] ahh cool cool