bachi rick_h_14:47
rick_h_bac: howdcy14:47
bacrick_h_: you got a sec for a quick call wrt charm proof and subordinate charms?14:48
rick_h_sure thing14:48
rick_h_bac: your link or mine?14:49
bacrick_h_: daily hangout?14:49
rick_h_buh bye15:02
bacrick_h_: that's ok.  i think we were done.  thanks a ton.15:03
rick_h_np, all good15:03
rick_h_frankban: how does the file:/// support in the charm work? Do we use that for anything?15:17
rick_h_frankban: usually we drop the file in locally and build the charm correct? this would be for manually putting the file on a deployed charm and then specifying the config and picking it up?15:17
rick_h_frankban: and if you can copy the file to the deployed charm then you can just copy it to the releases directory and set the version string?15:18
frankbanrick_h_: I think that options was used to deploy the charm behind a firewall. Now the charm is able to do that without workarounds15:20
rick_h_frankban: cool15:20
frankbanrick_h_: see "Deploying behind a firewall" in the readme15:20
Makyojujugui call in 1015:50
Makyojujugui call in 115:58
MakyoPick a number.15:58
MakyoWe have a winner!15:58
bacMakyo: yes, you should sleep16:04
rick_h_frankban: can you join back on the call for a couple of min if you have a sec?16:04
rick_h_sleep is gooooood16:04
frankbanrick_h_: url?16:10
frankbanMakyo: could you please take a look at https://codereview.appspot.com/51520043 ?16:27
Makyofrankban, on it16:27
rick_h_frankban: can you look at the log fix really quick please? https://codereview.appspot.com/5128004716:43
frankbanrick_h_: sure16:44
frankbanrick_h_: done16:49
rick_h_frankban: thanks16:49
Makyofrankban, can't log into rietveld for some reason.  Code lgtm, but maybe 'test_runs_once' or something instead of 'test_runned_once'?  Will QA now.16:49
frankbanMakyo: aargh, sounds good, thanks16:50
Makyofrankban, QA okay.  Will figure out why I can't get to RV soon.17:13
frankbanMakyo: cool17:14
benjirick_h_: can you easily reproduce the "Exodus index "charms_pending_019" does not exist." issue?  I want to to try something if so.  If not you have to watch this: http://www.youtube.com/watch?v=VBmCJEehYtU17:36
rick_h_benji: sec, getting into that lxc17:38
rick_h_benji: ok, yes. If I ./scripts/reset-db and then bin/migrations upgrade I get the error17:40
rick_h_and the video was interesting17:40
benjirick_h_: great, try this:17:40
benjibin/migrations prepare-upgrade --init && bin/migrations upgrade17:41
rick_h_benji: Updated the datastore to version: 2317:41
benjirick_h_: I guess I'm done then. :)17:41
rick_h_benji: heh, I'd suggest we add back the migrations make file targets then to help with this stuff17:41
benji(other than writing this down in a place that will mean we don't waste hours figuring out next time)17:41
rick_h_benji: it used to be that we used to have a make upgrade and such 17:42
rick_h_I'm not sure why they were removed, but I think that's part of what led to this. As you noted, not much to help figure out the 'right' way to run this exodus stuff 17:42
benjirick_h_: as your reward you get this video: http://www.youtube.com/watch?v=QgD5p1XiVT017:42
* rick_h_ completely doesn't know what's up with this royals thing17:43
* benji predicts youtube links will be the debits/credits of the future17:43
jcastrorick_h_, aha!17:55
jcastroso it ends up you pretty format markdown bullets in the readme but not in the description17:55
jcastroI'll file a bug, it's not like the end of the world, more like a papercut17:55
rick_h_jcastro: ah, the readme is a file. We process that whole file17:55
rick_h_there's no way in the descrip;tion to know if it's markdown or not17:55
jcastrothe description is in metadata.yaml17:55
jcastrooh I see17:55
jcastrook, so nm then I guess17:55
rick_h_with the readme we check the file extension17:55
* jcastro nods17:55
rick_h_we can see if it does any 'harm' to render markdown. I'm not sure tbh. I'd have to test it out on a bunch of descriptions17:57
rick_h_even though, summary I think would be bigger/more markdown-y17:57
rick_h_actually, nvm, that's not true any more is it17:57
jcastroyeah, but like I am sure you have way more important stuff17:57
jcastrolet's leave this for "some minor point release to be determined at some point in the future."17:58
jcastrodon't tell alejandra, maybe she won't notice it. :p17:58
jcastrorick_h_, now as far as quickstart/deployer not doing relations because some of the units didn't come up (from friday), who do I talk to about that?17:59
rick_h_that's how deployer works? so hazmat maybe?17:59
rick_h_I'm not sure if there's some good reason it does that down into juju that can be addressed by someone else18:00
bacrick_h_: remember when we had to fork charmtools to get it to use proof on 'localhost'?  ugh, just bit me.20:00
bacrick_h_: i may have a way around having to fork, now.20:01
rick_h_bac: yay?20:01
hazmatjcastro, yeah.. deployer will stop if things can't even install/start without error20:41
bacrick_h_ or benji: at your convenience could one of you review https://codereview.appspot.com/5165004321:37
rick_h_bac: can look in the morning23:04

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!