[00:10] hatch: :/ [00:11] rick_h_ yeah but the good news is that they have accepted the bug report [00:12] so it should at some point get fixed haha [03:12] jujugui review/qa for the AM - https://github.com/juju/juju-gui/pull/118 Dealing with dead/dying fridge, so out for the night. [03:49] Makyo looks like you forgot to lint :) [03:50] hatch, It's running now, sorry. [03:51] :) np [03:51] And pushed [03:51] just playing hearthstone [03:52] I'm thawing a freezer. Your night sounds better. [03:55] agreed haha [03:56] although im not very good at this game [04:00] i'll review your branch [04:06] Makyo I found a QA issue so I'm going to hold off on reviewing the code until resolved [04:07] Later [04:07] yeah no problem :) [04:14] See the issue, but this isn't data-bound and outside of scope. Will discuss tomorrow. [04:21] sure thing [04:21] I'm out have a good night [12:50] trusty update-ageddon [12:55] bac: yea, I get nervous [12:56] jcastro had bad timing and blew up during a trusty upgrade at cape town [12:56] jujugui I'm taking today as swap to catch up around here. I'll be in/out if anyone needs anything through the day [12:56] i've been having issues with X on the vm so i get jittery when i see a bunch of x updates. but, perhaps, they are the fixes i've been waiting for. yay, irrational exhuberance. [12:57] hey rick_h_ could you spare 5 minutes for a chat ? [12:58] bac: sure thing [12:59] rick_h_: https://plus.google.com/hangouts/_/calendar/Z2FyeS5wb3N0ZXJAY2Fub25pY2FsLmNvbQ.t3m5giuddiv9epub48d9skdaso === rogpeppe1 is now known as rogpeppe [14:32] hi abentley, i have some questions about charmworld migrations and exoduses. [14:32] bac: OTP. [14:32] currently in migrations/versions we have 019 - 023 [14:32] abentley: ok, no rush [14:41] rick_h_ hey I found something for you http://hackaday.com/2014/02/10/python-with-braces/ lol [14:50] hatch: :P [14:50] Makyo: hatch I'm swapping today but added a card with links to the twitter account info and azure stuff for ci. [14:50] Makyo: hatch so if you guys can note that in the call appreciate it [14:50] got it! [14:50] thanks [14:50] swap away [14:54] bac: Okay, I can talk now. [15:02] abentley: ok, currently we have 019-023 in migrations/versions. 019 is an exodus. i need to add a new exodus as 024. so, if i'm reading everything correctly, adding a new exodus requires that i delete the existing exodus (019) and the migrations (020-023). is that correct? [15:03] bac: That is correct. [15:04] ty [15:04] bac: If you really want to keep the previous migrations/exoduses, you can rewrite the migrations a new migration 025 to happen after the exodus. [15:05] err, "can rewrite the old migrations into a new migration 025" [15:06] the debug server and app server running on the same port is irritating [15:33] jujugui I need someone to test a bug for me who's running stable chrome on Ubuntu [15:33] https://code.google.com/p/chromium/issues/detail?id=342554 click the fiddle and let me know the outcome please [15:36] hatch: the zip file was right but i saw application/x-yaml for the yaml file [15:36] ok so it looks like it works properly on Ubuntu but not OSX or windows [15:36] thanks for confirming [15:37] hatch: here's my chromium from trusty [15:37] ii chromium-browser 31.0.1650.63-0ubuntu1~20131204.1 [15:37] yeah thats an older version than is on OSX too [15:50] jujugui call in 10 [15:50] jcsackett: thanks for your review last night. i've made fixes you can look at if you want otherwise i'll land with your previous review. [15:59] jujugui call in 1 [16:07] let's shoot for five tomorrow [16:08] haha [16:43] rick_h_, hello [16:43] rick_h_, I know you got some more work on your plate so I wasn't sure if you would have time for writing a jenkins test for the charm audit [16:43] rick_h_, but I also didn't want to steal that one from you if you were going to work on so I thought I would ping here :-) [16:43] arosales, I think rick_h_ is taking a swap day today, but I can make a card for that. [16:54] hey rick, I've got a few merges that need review for amulet, got a few seconds? [16:54] rick_h_: * [16:55] https://github.com/marcoceppi/amulet/pull/19 and https://github.com/marcoceppi/amulet/pull/22 [16:59] Makyo, ok thanks, that would be good. Just didn't want to steal that from rick_h_ if he was working on it [17:03] arosales, alright, card on the board in Maint. - High for the time being. [17:03] marcoceppi, ^^^ re rick_h_ [17:04] Makyo, thanks [17:04] Makyo: ah, thanks [17:04] jujugui added @jujuui to http://jujugui.wordpress.com/ so that posts are autotweeted, and the user is @mentioned when anyone shares one of our posts through the widget. [17:11] OOooOOoo [17:11] Makyo +1 [17:12] [17:12] done your good deed for the month? Time to relax! [17:22] I'm out. I'll be over here petting these dogs. [17:26] haha [17:26] Makyo review done [17:27] hatch, cheers, thanks. [17:28] no prolbemo [17:31] marcoceppi: I'll take a look in a bit. Thanks for the links [17:32] arosales: honestly, probably not room on the plate with picking up all the knowledge in gary's head and getting things here ready for machine view asap. sorry [17:32] arosales: I'll see if we can get a slack card in case someone has some blocked time or something if it'll help [17:32] rick_h_, ack and thats what I figured just wanted to confirm, now get back to your swap day [17:33] any help is appreciated [17:33] benji: sorry on the port issue. I hit that the other day and thought the same thing. I liked it in theory when I updated it all. [17:33] benji: can you add a slack card split that into two ports again? I'll see if I can update it and get CI updated to use the right one. [17:34] Makyo: woot on auto tweeting robots [17:37] Makyo: hatch have you guys seen the email thread about allowing a null/none type? I want to chat/check how that effects us as we don't way to use 'unset' from the Gui and I worry we'll be stuck with an impossible workflow [17:38] marcoceppi: maybe we can setup a quick time tomorrow mid-day to chat on it? ^ [17:38] rick_h_ I was under the impression that was only for proof and testing [17:39] but I'd participate in a call to learn more [17:39] hatch: well if it passes proof then it means the default value going into the charm config will be None. That effects us for 'use default values' and allowing a user to get back to a default once changed right? [17:40] right now proof marks it as an error if it's the wrong type and won't ingest/process I believe. [17:40] so I think this effects us on both charmworld and gui config UX [17:40] well we have 'real' defaults though [17:41] I think they are saying when people have defaults as 'empty' [17:42] hatch: right "" vs None. "" is a string, None is not. In JS it's the equiv of undefined. [17:42] How do we allow a user to get back to undefined from an input field [17:42] ohh now I see what you're getting at [17:42] and how to we get that sent via the api, or is there a magic "use default" we use instead or something? [17:43] we can solve this with a checkbox 'reset to default' on each field.... bleh [17:43] hatch: right, I think this effects us. I'd like to talk about it romorrow [17:43] tomorrow [17:43] if you can Makyo and maybe bac can peek at it and think on the repercussions so we can talk to marcoceppi about it from our end. [17:43] yup, I'm available, just invite me to whenever it is [17:44] * hatch swears at the damn power key [17:44] hatch: ok cool. Maybe we'll chat and talk it through on our end first. Anyway, ponder it please. [17:45] we need to determine if "" can always be undefined [17:45] I'm thinking no [17:46] hatch: right [17:46] I'd like a concrete case where that's true though [17:46] where "" !== undefined [17:46] there is a distinction between a blog title with an empty name and a undefined name. [17:46] maybe undefined is invalid, but "" is ok? [17:46] yea blah [17:46] exactly :) [17:46] I can't think of a real case where it's not the same [17:46] but in reality it is [17:53] rick_h_: let me know, I haven't changed proof yet, but several charm authors cried when proof was updated to be strict about formatting [17:54] marcoceppi: yes, I think it brings a few issues for us we need to work through for that change to take place. I understand your desire for at least the undefined case, but we need to verify that core will accept it and figure out how to represent that in the GUI [17:56] rick_h_: whatever unset does is what gui needs to do, I don't care either way, I'm just tired of fighting a few of the authors who are opposed to the current way it works. Before it was "well, you can't set a None type via juju set AND the GUI can't handle it" now that unset exists (which is revert to default, FYI) GUI is the only reason not to have the switch back [17:57] marcoceppi: right, we'll have to explore some way of exposing 'unset' in the gui which we don't currently have is what I think we're heading towards [17:57] marcoceppi: and room is tight to add an "unset" button to every field in config [17:58] rick_h_: yeah, I understand that, cheers ;) [17:58] marcoceppi: I agree it's something that needs to happen, just going to take a little bit to figure out the right way to do it. [17:58] rick_h_: how does the reset to default work at the bottom? Or is that only reset the changes since last commit? [17:59] so I believe that just walks and sets to the default value. It's not using unset [17:59] and that's across all fields, not just one in particular you might want to 'unset' [17:59] right, jw [18:51] jujugui could I get a review on https://github.com/juju/juju-gui/pull/119 no qa needed [18:51] thanks [18:57] rick_h_ how the heck do you have so many commits on the GUI? https://github.com/juju/juju-gui/graphs/contributors [18:58] do you just move code around to increase your ranking? lol! [18:58] * hatch taking lunch [19:00] hatch: :P guess it'll start to slow down a bit now [19:00] haha good I need to catch up! [19:05] benji: would you have a moment to look at the exodus change? https://code.launchpad.net/~bac/charmworld/1278222-2/+merge/205835 [20:24] hazmat, rick_h_: I cannot find the deployer bug that's "ignore broken units" and keep going [20:35] jcastro, its framed by gui notion of bundle after bundle [20:36] yeah I am just looking for the bug report. [20:37] jujugui anyone available for a pre-imp discussion on https://bugs.launchpad.net/juju-gui/+bug/1278559 I think I need to move the inspector creation code [20:37] <_mup_> Bug #1278559: dropping charm zip while inspector is open opens series select inspector over current inspector [20:39] jcastro, 1252301 [20:39] bug 1252301 [20:39] <_mup_> Bug #1252301: guiserver reports second bundle as failing after the first fails

[20:40] ^ easy link :) [20:40] that's a new bug, I filed one when I originally found the race condition in the mongo charm [20:48] hey so hazmat [20:49] you kind of mention this is two bugs [20:49] basically I want to communicate that this is the bug eco wants fixed [20:49] I don't care too much where it lives as long as it's on someone's radar [20:51] jcastro, could you file a separate bug for the unit failure mode.. hopefully its easier to track that way [20:51] I had filed it [20:51] but now I can't find it for the life of me [20:52] jcastro, perhaps against distro? [20:52] or gui charm? [20:52] it was against the gui initially [20:52] but was "also affects" deployer [20:53] ok I will file a new one [20:57] heh yeah the bug search is kind of poor [20:57] https://bugs.launchpad.net/juju-gui/+bug/1276107 this one? [20:57] <_mup_> Bug #1276107: gui fails to deploy a bundle due to "Invalid charm id: undefined" [21:03] hatch: still need a pre-imp? [21:03] I do, are you not swapping today? [21:03] hatch: meh, semi swapping? e.g. destroyer of things blocking folks :) [21:03] haha ok creating hangout [21:04] https://plus.google.com/hangouts/_/76cpjlpbspqf8fkeidl096n1jg?authuser=1&hl=en [21:15] * bac dog walk. bye. [21:48] rick_h_ lol wow this is embarrassing... https://gist.github.com/hatched/8944860 [22:01] Morning [22:01] hey huwshimi [22:04] huwshimi you're blocking the high priority lane with two cards btw [22:06] Well, I think the only course of action is for everyone to stop working until I've moved those cards! :) [22:15] I'm personally not a massive fan of my laptop hard locking every day. [22:17] haha [22:17] hard locking? Are you running different types of vm's? [22:18] when I run parallels and virtual box it hard locks instantly [22:18] hatch: Nothing like that... no idea what's causing it [22:19] hatch: Do you know how I can get a branch I created under OSX in Ubuntu? [22:21] jujugui looking for a review and qa for https://github.com/juju/juju-gui/pull/121 [22:21] are you using vagrant? [22:22] or are you running both on metal? [22:24] hatch: Maybe I should rephrase the question. I have a branch on GitHub that I want to play with locally, how do I do that? [22:24] git clone your fork [22:24] git checkout your branch [22:25] git clone [22:25] git checkout [22:25] thaz it [22:26] heh what was the last message you got? [22:26] hatch: Crashed again :( [22:26] hatch: when I checkout I get "error: pathspec 'safari-bugfixes' did not match any file(s) known to git." [22:27] looking at your repo one sec [22:27] This is on an existing clone [22:27] run `git branch -a` is the branch there? [22:28] if not .... then `git fetch` [22:28] nope [22:29] after the git fetch? [22:29] ah, yes! [22:29] hatch: Thanks! [22:29] :) np [22:30] huwshimi_ looks like you have some conflicts in your PR [22:30] hatch: Yeah, that's what I'm trying to fix :( [22:30] alright well let me know if you need any help [22:32] hatch: Thanks, I think I just need to merge develop [22:33] hmph, somehow I hit a keyboard shortcut to quit the channel [22:47] You are just having all sorts of problems. [22:52] haha nah [23:04] hatch: Fixed the build issues [23:04] cool [23:06] hatch: lol, coolio [23:06] huwshimi_ think you could rebase it down to a single commit? [23:07] sure [23:07] you can keep the commit message with all the fix details [23:07] but just to keep the develop branch clean [23:09] hatch: So what process should I follow to rebase it? [23:09] rick_h_ is probably the best guy to ask because you have a merge in there [23:12] huwshimi_: which branch is this? [23:12] rick_h_: Mine: https://github.com/juju/juju-gui/pull/120 [23:13] looking [23:13] huwshimi_: yea, don't merge develop if you don't need to before pushing up [23:13] rick_h_: There was a conflict with develop [23:13] since that's your last commit I'd just rebase that away [23:14] huwshimi_: ah, cool then [23:14] then I'd have changed that commit message be 'resolve conflicts' [23:14] so the goal is to rebase away the lint commit, change the message on the last one to "Resolve conflicts with develop" [23:14] git rebase -i HEAD~~~~ [23:14] and change the lint to a `s` to squash sit [23:15] then `git ammend` should let you change the commit message of the last point [23:16] rick_h_: Oh, don't we want this as one commit? [23:18] huwshimi_: naw, I think it's safe [23:18] huwshimi_: I mean you're fixing a series of bugs [23:18] ok cool [23:18] as long as it's sane it doesn't have to be one commit [23:18] just don't need noise like 'lint' [23:18] or 'prototype step 1' and such [23:23] rick_h_: http://dpaste.com/1606090/ [23:25] huwshimi_: so should be a conflict to resolve [23:25] git status [23:25] look for the files that are modified and noted [23:27] ah, completed [23:27] huwshimi_: cool [23:27] rick_h_: "git: 'ammend' is not a git command." [23:27] bah, old version of git. [23:27] git commit --ammend? [23:28] nope [23:29] this is just whatever version comes with Ubuntu I think [23:29] 13.10 [23:29] huwshimi_: git commit --amend -m "New commit message" ? [23:29] git version 1.8.3.2 [23:29] * rick_h_ is looking [23:29] What do I want that new commit message to be? [23:30] * huwshimi_ doesn't know what we're trying to achieve here [23:30] "Resolve conflicts with trunk" or the like vs "merge" [23:32] OK, that worked. Pushing up changes. [23:34] rick_h_: Pushing now fails: http://dpaste.com/1606119/ [23:34] git push origin safari-bugfixes -f [23:38] rick_h_: Is this correct? https://github.com/juju/juju-gui/pull/120 [23:38] huwshimi_: looking, looks crazy [23:39] huwshimi_: give me a minute [23:42] holy schmoly [23:43] hmm, git clone takes a while at 74k/s [23:44] ugh, 77% [23:44] go go go [23:49] rick_h_: Are you on your phone? [23:49] huwshimi_: on my hotspot [23:49] huwshimi_: going really slow atm :/ [23:49] I'm at a LUG meetup atm [23:50] at 81% [23:50] sorry, trying to pull down your branch and to see/verify that the extra commits don't dupe/etc [23:50] but need to clone first and guess 50MB is heavy for my current signal strength [23:59] yay, downloaded...working on the branches now