=== gary_poster is now known as gary_poster|away [12:22] jujugui: i'm looking at huw's problem [12:22] k === gary_poster|away is now known as gary_poster [13:01] thanks bac [14:05] morning all [14:05] morning [14:08] gary_poster https://twitter.com/clarler/status/405565740700274688 schedule wise looks like I'm available [14:09] woot, hatch on tv [14:10] lol [14:10] like a baws [14:10] :-) hatch will lookin a sec; on call [14:20] bac are you planning on fixing huws branch or are we just leaving it to him? [14:20] I could also take over if needed [14:21] hatch: if i had time this afternoon i was going to look at it [14:21] cool, lemme know if you run out of time [14:34] * frankban run the charm tests in ~25mins \o/ , but the legacy server header test fails :-/ [14:48] hatch cool about roundtable. +1, for you and whoever wants to show up. [14:49] eggcellent, now I just need to figure out what to talk about [15:50] jujugui call in 10 [15:50] ty [15:51] * rick_h_ runs [15:51] hey frankban, I'm about to qa the quickstart change to support the authtokens. Could you take a quick look over the diff to make sure it looks alright before I do so? http://paste.ubuntu.com/6484972/ [15:51] gary_poster: sure [15:51] ty [15:54] maybe I should tell the user if we have a token or not...they will see effect, but maybe that will ease surprise? [15:54] s/ease/reduce/ [15:55] gary_poster: can we stay for five minutes after the call? I have a comment on the diff and an unrelated question [15:55] frankban, sure [15:55] gary_poster: thank you [15:58] jujugui call in 2 [16:01] benji yoo hoo? [16:01] gary_poster: coming [16:11] gary_poster: i have no commercial vouchers available for private projects on LP [16:13] bac, darn, but thanks. rick_h_ ^^^ wgrant maybe? [16:13] gary_poster: yea, will have to catch him one night [16:13] if only there were an asynchronous messaging system... [16:24] frankban, Would like to go over tasks when you get a chance. Can make all the cards and see about starting a small thing, but I'm out thurs/fri, don't want to leave too much hanging. [16:41] frankban, https://plus.google.com/hangouts/_/calendar/Z2FyeS5wb3N0ZXJAY2Fub25pY2FsLmNvbQ.j0rk5d371ph8331ijtf48t2uj0 [16:41] Makyo, frankban would be great if frankban can work on it on Thurs/Fri [16:41] Yeah [16:41] sounds like that's what you mean [16:41] cool [16:47] gary_poster, Makyo: sounds good, I can start prototyping something [16:48] cool [16:48] gary_poster or Makyo: call? [16:48] frankban, is ready for you :-) [16:48] cool joining [16:48] https://plus.google.com/hangouts/_/calendar/Z2FyeS5wb3N0ZXJAY2Fub25pY2FsLmNvbQ.j0rk5d371ph8331ijtf48t2uj0 [16:50] benji any idea how I do a multi check conditional in bash? https://gist.github.com/hatched/e7d882ee100f52ca1eb0 this is throwing an error [16:51] * benji looks [16:52] line 62: syntax error in conditional expression: unexpected token `;' [16:52] that;s the error for that line [16:53] hatch: I don't see a problem with that code, what error is it generating? [16:53] 2013-11-27 16:45:18 INFO juju.worker.uniter context.go:255 HOOK /var/lib/juju/agents/unit-failtesterp-0/charm/hooks/install: line 62: syntax error in conditional expression: unexpected token `;' [16:53] the conditional in that gist is the line in question [16:54] oh shoot I know what it is [16:54] missing the space after the final True [16:54] damn I hate bash [16:56] thanks for the help :) [16:57] hatch: heh, I just noticed the same thing, and yes, shell scripting is nasty [16:58] * benji pines for MPE job control language [17:01] Makyo: available when you want [17:01] frankban, https://plus.google.com/hangouts/_/7acpir7e3t3ummvqu4iun0imac?hl=en [17:40] guihelp: I need two reviews for https://codereview.appspot.com/34130043 , no rush, thank you! [17:40] frankban, I'll take one. [17:40] frankban, I'll take a no-qa one :-) [17:41] yay 120 -> 40 ! [17:41] :-) [17:41] thank you both [17:47] mm, fresh Thanksgivikkuh latke. day early, but too good to pass up. [17:52] frankban, am I right in understanding that you removed test_local_release because TestBuiltinServerLocalRelease uses the local release for all of its tests? [17:53] gary_poster: yes, that scenario is already tested there [17:53] ok cool. [17:56] hatch, Makyo I might be late for our calls. Is that alright? I'd png you when I was available. [17:56] gary_poster, sounds good [17:56] ty [17:57] frankban, "LGTM with trivial. Thank you!" [17:58] gary_poster: great thanks! [18:04] gary_poster no rush, take your time [18:04] thanks hatch [18:15] ok, afk for awhile [19:01] jujugui looking for a really quick review https://codereview.appspot.com/34120044/ [19:02] hatch: looking [19:02] thanks [19:09] benji thanks, and oops I forgot to commit the tests, will run them and submit [19:09] cool [19:26] benji interesting bug I am guessing because the element is never focused because it's generated on the fly [19:27] yeah, I figure it will be something like that (assuming we ever try to fix it given its importance and the likelyhood that we're ripping that code out anyway) [19:35] ok, how do we build this into the gui? http://david.li/waves/ [19:36] benji: when we get back I'd like to see if you and I could sit down and plot out the best way to setup the new project https://github.com/mitechie/jenkins-github-lander [19:36] benji: my brain is fried for the day, but if you get a chance to peek appreciate it. The makefile setps to setup and such are a bit of a mess [19:36] rick_h_: sounds good [19:37] gary_poster: ^ pushed up to a person branch a working start that does what the thing did yesterday with a couple of tests (far from complete) and docs. Calling that a day. Will have notes for the email for monday. [19:37] rick_h_, fantastic! thanks and happy Thanksgiving [19:41] jujugui, anyone object to me making a bzr release and putting it in the ~juju-gui charm (not the official ~juju-gui-charmers one)? I want to make it available for testing quickstart [19:41] sounds fine to me [19:44] k [19:44] gary_poster: a-ok. [19:44] thx === BradCrittenden is now known as bac [19:46] benji: your new random service name + 'make deploy' in the charm make for some fun [19:47] bac: hmm, maybe there should be an optional service name which __main__ would pass [19:47] jujugui: please don't look at my review just yet. i'm doing QA and writing instructions [19:48] benji: that'd be nice. [19:48] bac: I'll make a card [19:52] gary_poster: flurries in CH. you got anything? [20:07] bac. just rain [20:11] jujugui https://codereview.appspot.com/34260044/ quick quickstart reviews/qa, please! [20:12] Makyo: i'll do it as soon as i can [20:12] bac, cool, thanks. [20:18] gary_poster just FYI a large data transfer down a websocket will block because the sockets are sent in order. Other have gotten around this by splitting up the large files but....no [20:18] :) [20:18] hatch, great data, thanks :-) [20:18] and I'll be ready for call soon [20:27] gary_poster sure thing, whenever you're ready [20:29] Makyo: code reviewed. can't do QA until i get my current run sorted out. [20:29] bac, sounds good. [20:33] bac, overloaded var; qaing frankban's branch, but will fix in a few. [20:39] Hah, roommate (who works for a grocery store) said his friend was raving about this weird cloud product called 'Juju'... :) [20:40] w00t w00t [20:40] we are famous! [20:41] heh [20:41] rick_h_ you still around? [20:41] hatch zero rush but https://plus.google.com/hangouts/_/calendar/Z2FyeS5wb3N0ZXJAY2Fub25pY2FsLmNvbQ.s6mt389qadeobe2a9tm9d8nm40 [20:41] hatch: what's up? [20:42] rick_h_ any experience writing hooks for apache2 and haproxy? [20:42] hatch: no, would just have to check out what things like charmworld does, or wordpress [20:45] ok I'll check there [21:12] oh netflix you silly silly people https://www.evernote.com/shard/s219/sh/cd7f6ae5-8cdc-4282-ada4-d97e4e695ebd/99a3294013dea17dfb24c22e80cf8db2 [21:12] hatch: i'm not going to have time to debug huw's branch [21:13] bac ok no problem [21:13] bac so shall I get him to do it or does it look like ther eis going to be a lot of work for me to do? [21:14] just trying to decide what to start next? [21:15] hatch: the error is in new js code that he wrote. so, i figure he can figure it out but help would be appreciated. [21:15] hatch: i didn't dig in enough to even know what he's trying to accomplish [21:15] ok then maybe I'll wait until he gets back and assist with it [21:15] he should be starting pretty soon I think [21:16] hatch: it may be as simple as he's looking for the wrong selector. or, he may need to be defensive about checking for null if the selector isn't there [21:17] alright cool thanks [21:17] have a good one [21:18] jujugui: i'm running locally and did a 'make deploy' followed by a set of juju-gui-source to use lp:juju-gui. the latter gets stuck and never recovers. i'm a bit perplexed as to why. i've been chasing this for a couple of hours trying to QA my branch. thoughts/ [21:18] s///?/ [21:18] wow, how'd that happen? [21:18] lol [21:18] bac: check the /var/log/juju/unit..... log [21:18] just tried to change the / to a ? [21:18] bac: that should have info if it failed to change source [21:19] rick_h_: yeah i have but don't see anything good [21:19] bac: is the hook change event in there? [21:19] errr config change hook fire in there [21:19] rick_h_: i've taken a new tack and just made a new release and stuffed it in the charm [21:19] rick_h_: no err. it does say it is preparing a change event and then goes away [21:19] jujugui small but cool quickstart review request: adds last step of autologin/authtoken work. https://codereview.appspot.com/34420043 [21:20] bac: machine log? [21:20] Makyo, call now? [21:20] gary_poster, sure, same as on cal? [21:21] Makyo yes [21:21] rick_h_: stuffing the release into the charm seems to be the fastest way to deploy. it'll have to remain a mystery [21:28] jujugui: small charm review and QA: https://codereview.appspot.com/34270044/ [21:33] bac on it. [21:34] gary_poster: i should rephrase "small review, pita QA" [21:34] heh [21:34] np [21:34] * gary_poster wishes local envs worked for him [21:46] gary_poster, what's the issue? [21:47] hazmat, same one that I sent out over a month ago (I think you replied then). still not working. Tim could not dupe. was told to go talk to Serge and lxc folks, and haven't gotten around to it [21:49] bac LGTM and QA good, no change [21:50] gary_poster: great [21:54] gary_poster, another option is using manual provider with ubuntu-cloud lxc containers. [21:54] needs trunk juju-core [21:55] albeit assuming you have a working lxc [21:55] hazmat, ah right. can help with some use cases, yes. thanks. will put on list to try. I need to actually verify I have lxc working on this machine. done lxc stuff on another [22:12] Makyo: qa was fine. [22:12] * bac eod. bye. [22:16] bye bac. have fun [22:17] Makyo, stepping away, but will return later to see how review/qa went [22:17] gary_poster, review okay, QAing now. frankban's branch doesn't work for me, but unrelated to his changes, so qaing yours now. [22:17] cool ty [22:27] Now that I have internet again... [22:27] Morning [22:42] hey huw [23:02] huwshimi if you need help getting your branch up and running lemme know [23:02] I should be around for a while yet [23:03] hatch: Yeah, trying to get it working at the moment... [23:21] Ah, fixed. [23:26] great [23:27] huwshimi I see in your email you added checks just for the tests - that's what we call a 'code smell' :) [23:28] adding code simply for a test is sometimes not a good thing [23:28] just keep that in mind :) [23:29] hatch: I realise, but we seem to do that a bunch for nodes that don't exist in tests. [23:29] yeah it's always a close call between requiring it for a test or making a check [23:29] we have the utils.makeContainer('id') method for just this purpose fyi [23:31] I have no idea what you're doing though so it could be a lot easier to just check in the app :) [23:31] hatch: Just proposing this branch. Would you mind having a look and see if you think I need to clean it up? [23:31] sure [23:33] hatch: https://codereview.appspot.com/34520043 [23:33] cool on it [23:35] thanks! [23:39] huwshimi which bundle/charm did you use for testing? [23:40] hatch: Most of them should have content long enough to scroll.. [23:41] hatch: You could look at the Readme tab on the Liferay charm [23:41] cool almost done [23:44] huwshimi done and qa done [23:45] little more work required - maybe I shouldn't have LGTM'd :P [23:45] I'm hoping that the 'jumping' mentioned in the QA happens for you as well [23:49] Thanks I'll take a look [23:55] hatch: Not sure if I can stop that jump [23:56] huwshimi well you just add top and padding, can you maybe transition them with the header transition? [23:56] at the same 0.2s? [23:56] But then instead of a jump you'll have the text sliding around... isn't that worse for trying to read it? [23:57] hatch: The text would jump up and then slide down...