[12:14] frankban: were you able to grab my branch? [12:15] bac: yes [12:15] frankban: good. i hope it is helpful. [12:16] see you tomorrow [12:16] bac: it is, thank you, have a good day [12:17] * frankban lunches [13:26] teknico, re-reviewed just for the heck of it. Looks good. I had an important thought for an upcoming branch. Are tests running now? Do you just need a second review? [13:52] * gary_poster cheers bcsaller1 [13:53] gary_poster: :) There may still be some issues but its a real improvement and Makyo already has a start on the next branch stacked on that one [13:53] * gary_poster also cheers for a GUI branch that does not touch the Makefile :-P [13:53] great bcsaller1 [13:53] indeed [13:54] bcsaller1, for kanban board, which one is Makyo working on? Will drag it into the coding lane [13:54] relations [13:55] ok thanks [13:57] bcsaller1, if you review hazmat's annotations branch then he will have two reviews and can land, and will clear up space in secondary story [13:57] gary_poster: I'll do that this morning then === bcsaller1 is now known as bcsaller [13:57] thanks [14:01] * teknico is officially back to work :-) [14:01] frankban, benji or Makyo if you could give teknico his second review that would help move things along also [14:02] hey teknico :-) [14:02] * benji looks [14:02] gary_poster, I did not yet run tests, only had time to make changes and push them this morning [14:02] ack teknico [14:02] I'd like to test that code a little more, at least manually [14:03] probably a unit test on the "build" function would also be appropriate [14:03] but would delay things more [14:04] gary_poster, I'd also like to clear my head about something that puzzles me quite a bit, this: https://code.launchpad.net/~bac/juju-gui/1091787/+merge/140528 [14:05] teknico, do what you think is necessary for this to be acceptable, pushing things to another branch where possible. If you want to talk through that with someone, me or anyone else, feel free, but I leave that decision to you. [14:06] teknico, MP: which part? [14:06] gary_poster, all of it. Can we have a quick call on it, with bac and possibly you and/or benji? [14:06] teknico, bac is out today [14:07] oh, great :-/ [14:08] teknico, I think bac did it because of some issues that frankban and he encountered. frankban does that sound right? anyway teknico, why don't you and I talk it through and then we can figure out if we need more people in on it. let's see if juju-ui is open [14:08] gary_poster, anyway, you approved that branch, so you likely have some knowledge/opinion on it :-) [14:08] yes [14:08] I'm in juju-ui teknico [14:08] gary_poster, sure, maybe frankban could have a say too, if bac discussed it with him [14:08] teknico, I'd like the two of us to talk about it first [14:09] then we can bring others in if necessary [14:42] Quick question: if I can download an installer for Juju to my Mac and run it form my desktop... why can't I do this in Ubuntu? [15:20] goodspud, juju is in software center/debs; would be the equivalent of being in the Mac app store [15:21] bcsaller, I got sidetracked. Would a daily call be helpful today, or is the current happy state relatively obvious? [15:21] gary_poster: I think we are doing fine today [15:22] cool bcsaller, thanks. I'll get back to review...at 11:30. If you get someone else to beat me to it that's fine [15:28] bac bcsaller benji frankban goodspud hazmat Makyo teknico call in 2 [15:29] bcsaller, also IMO and in yellow's past, pairing counted as a review. If we all agree with that then your branch already has a review from Makyo, yeah? [15:31] bcsaller, starting without you [15:38] bcsaller, want to tell me the issues on hangout for your speed, or here? [15:39] back to hangout :) [15:42] bac, so apparently its golang 1.0.3 which has the bug for http client.. sadly that s the latest version.. 1.0.2 should work [15:52] gary_poster: to create a stable release without needing to change trunk, it is ok to just modify CHANGES.yaml (replacing "unreleased" with a version number, and then "BRANCH_IS_CLEAN=1 FINAL=1 PROD=1 make dist"? [15:54] frankban, I want to be helpful, but I don't like it as a precedent or as part of history (we want to be able to relate a release to a revision). Maybe there's another approach we could have to make this fast. Quick hangout in juju-ui? [15:54] gary_poster: sure [16:13] gary_poster: could you please review another Makefile quick fix? https://codereview.appspot.com/6965048 [16:16] frankban, approved and I'm ok with you just landing [16:17] thanks gary_poster [16:29] gary_poster, blame canada ;-) [16:29] :-) [16:40] gary_poster: release failed :-( the error seems to be related to BRANCH_IS_CLEAN, which executes "bzr missing --this". the latter raises a "bzr: ERROR: No peer location known or specified" failure. [16:40] :-( hm [16:41] frankban, use BRANCH_IS_CLEAN=1 and I will file a bug [16:41] gary_poster: ok [16:45] gary_poster: trying BRANCH_IS_GOOD, because also IS_TRUNK_BRANCH does not seem to work with checkouts [16:45] (it tries to grep for the parent branch) [16:46] frankban, :-( it worked at one point. benji ^^ I'll file bugs for this stuff [16:46] hmm [16:46] frankban, I'm guessing when you make a temporary release in the charm you already are saying BRANCH_IS_GOOD? [16:49] gary_poster: no, I only use NO_BZR=1. [16:49] frankban: the release machinery is meant to be used with a branch of trunk, not a checkout [16:49] benji, oh, that's not the way I had it [16:49] gary_poster: and it works... [16:49] benji, sorry [16:49] frankban, sorry [16:50] gary_poster: we discussed this and decided branches would be at least as good as a checkout and arguably better [16:50] frankban, if it works then that's all I care about :-) [16:51] gary_poster: no. it works in the charm, when I create a local distfile from a lightweight checkout [16:51] benji, ok. Maybe you could connect with frankban after this is done to run though the releae process [16:51] that we have documented [16:51] and figure out what could be documented better [16:51] sure [16:52] thanks [16:52] gary_poster: ok. proceeding with BRANCH_IS_GOOD? [16:53] frankban, yes. If we have a revision that is distinctly associated with a release, I'm happy for now. [16:54] teknico, I filed low priority bug 1092199 and bug 1092204 per our discussions [16:54] <_mup_> Bug #1092199: Makefile build target is misnamed < https://launchpad.net/bugs/1092199 > [16:54] <_mup_> Bug #1092204: We copy YUI assets rather than symlinking them < https://launchpad.net/bugs/1092204 > [16:54] gary_poster, great, thanks! [16:54] welcome, thank you [17:00] lp2kanban appears to be dead [17:00] I think bac is the one with the keys [17:01] bcsaller, I filed bug 1092208, and put both it and Nick's bug 1091616 in the secondary story tasks [17:01] <_mup_> Bug #1092208: Pending relation line is not getting events < https://launchpad.net/bugs/1092208 > [17:01] <_mup_> Bug #1091616: Unusual drag/zoom behaviour

< https://launchpad.net/bugs/1091616 > [17:01] gary_poster: thanks [17:02] welcome [17:03] gary_poster: done! https://launchpad.net/juju-gui/stable [17:03] benji, "Checklist for Making a Release" in trunk begins with Get a checkout of the trunk:: ``bzr co lp:juju-gui``. :-) I am making a bug to update this, and will ask you to do it when you get a chance next [17:03] yay frankban! L0( [17:03] :-) [17:03] one-off smiley [17:04] :-) [17:08] benji, filed bug 1092216. I asked whoever tackles it to consult with you; alternatively you can clarify the process in the bug; alternatively you can consult with yourself :-) [17:08] <_mup_> Bug #1092216: Release documentation is incorrect < https://launchpad.net/bugs/1092216 > [17:09] oh lp2kanban, where are thou [17:09] heh [17:10] * gary_poster completes filing bugs and goes to have a bit of lunch before reviews [17:12] gary_poster, quick questio, how long did it take from deploy/expose commands to finish? [17:12] I can't deploy a working environment, for some reason :-/ [17:29] teknico, about 20 minutes [17:29] teknico, what does juju status say? [17:30] gary_poster, I need to wait a little more then [17:30] gary_poster, always this: http://pastebin.ubuntu.com/1450398/ [17:31] teknico, it's been less than 20 minutes? If not, I would ssh to the machine and do a top or ps and see what's going on [17:31] look at the log too of course [17:32] which might show that it is doeing something in the Makefile, which is another file you can tail [17:32] doing [17:32] gary_poster, using juju-ssh juju-gui/0 , as you mentioned? [17:32] teknico ``juju ssh 1`` [17:32] it's the machine number. maybe unit works as well [17:32] ok [17:33] * gary_poster steps back to lunch break, trying to fix disposal :-) [18:09] yay frankban :-) [18:10] teknico, if you are heading to your EoD soon and you are still seeing problems with your branch, consider trying to pass it off to me to see if I can move it along any [18:10] I'm willing, and might have some time [18:11] gary_poster, that's probably a good idea [18:11] gary_poster, 40m now, not yet started, and "juju ssh 1" does not work either :-( [18:11] maybe something wrong in my environments.yaml file? [18:11] gary_poster: heh, the diff is a bit long, I am sorry, but there are a lot of tests [18:12] at least [18:12] teknico, weird :-/ if you want to have a call let me know [18:12] gary_poster, I need to fix this problem in short order, but you're right, no need to hold this up any more [18:12] frankban, yeah, I think you are forgiven for the diff length ;-) [18:12] gary_poster, so consider the branch passed off to you :-) [18:13] gary_poster, I added the docs and fixed the other "build" invocation as discussed [18:13] :-) [18:17] gary_poster, yes, if you have time maybe a (hopefully short :-) ) call could help [18:27] teknico, join me in juju-ui whenever you are ready [18:39] bcsaller, for some reason every one of the rietveld pages for your branch shows "error: old chunk mismatch." Example: https://codereview.appspot.com/6971045/diff/1/app/app.js [18:39] Any idea why? Makes it difficult to review :-) [18:39] gary_poster, i suspect it needs a merge from trunk.. i noticed that as well [18:40] gary_poster, so its kinda of wierd [18:40] hazmat, ah ok. bcsaller could you try and fix? [18:40] gary_poster, you can got at it from the backside and it works.. [18:40] or redo or something [18:40] hazmat, backside do you mean MP or bzr branch or someth? [18:40] gary_poster, i think it might an issue in reitveld [18:41] ah ok [18:41] if you start from the index in the middle its okay https://codereview.appspot.com/6971045/patch/1/6 [18:41] but if you start at the top and use j/k to navigate they all show that error [18:41] oh, it is the side-by-side that is broken [18:41] gary_poster, yeah [18:42] still not so good for reviews :-P [18:52] hazmat, fyi, the eu-west-1 region does not seem to work with juju, at least for the juju-gui charm [18:52] and now I'll go [20:03] hey jimbaker. I asked jitsu test for verbose logging with -v and it's not been as verbose as I hoped for. This is all I've got, almost 40 minutes into the test: http://pastebin.ubuntu.com/1450714/ [20:03] Did I do something wrong, or could I get better logs, or is this a place where jitsu test could do a better job in the future [20:14] has anyone else seen this? I am running my tests by requesting "http://localhost:8084/test/?grep=login%20mechanism". It only works intermittently. Most of the time I get a completely blank page. [20:14] I have not benji, but I haven't used grep in a while. bcsaller or Makyo would be people to ask [20:15] benji, also a problem in trunk? [20:15] I haven't tried on trunk. [20:51] jimbaker, also, it looks like we don't have stdout/stdin from the test so we can't run debuggers, like pdb, which is a drag. I think the debugging story right now is the most difficult [20:51] given the logging and break point issues [21:11] I need to run. I'm still trying to figure out the charm test situation with nicola's branch. I hope frankban has some good tips