[00:36] marcoceppi: we were going to look at that yaml thing tonight. I'm exhausted. Another time? [00:36] jrwren_: yes, another time indeed === CyberJacob is now known as CyberJacob|Away === urulama-_ is now known as urulama === CyberJacob|Away is now known as CyberJacob === CyberJacob is now known as CyberJacob|Away [08:55] hello, how do I upgrade-charm from my locally modified copy when I initially deployed from the charm store? === scuttlemonkey is now known as scuttle|afk [10:36] james_w: hello, you can use the --switch flag [11:31] marcoceppi: thanks === gsamfira1 is now known as gsamfira [12:17] dpb1: /quit [12:17] pfft [13:23] mbruzek, the other mp is referenced in the bottom of that first's comments [13:23] hazmat, got it. [16:27] Using vagrant for work on a charm. I have it and postgresql deployed, and add a relation between the two. If I ssh to my charm's machine and try to use psql to connect to postgresql, using the relation's private-address/user/password, it's failing, saying there's no entry in pg_hba.conf. [16:28] I'm connecting from 10.0.3.232, but postgresql is seeing it as 10.0.3.1 [16:28] Should I be able to connect from machine to machine that way? [16:40] well you got to first figure out why the ip is not working- then move on to the config file. [16:40] aisrael, [16:55] thanks html. I'll dig in deeper and see what I can find. [16:56] aisrael, im just a noob that is saying the basics [16:57] Understood :) [16:58] aisrael, luckly you that you got juju working- i dont know or even get it working [16:59] i wish :/ [17:07] I'm still learning myself, but this is a good place to ask questions if you're stuck on something. [17:19] aisrael: are you executing that from the host, the vagrant virtual environment, or from the postgresql lxc container running in the vagrant environment? [17:34] Hi Bruzer [17:34] Hello Flint_ [17:34] how goes the battle? [17:35] ..er work [17:35] round and round. [17:36] do you have bluetooth with your linux box? [17:36] ...looking for a stable setup [17:37] Yes I have bluetooth working on Ubuntu 14.04 [17:37] Flint how would you plan to use it? [17:38] Flint_, although I suspect that depends on the hardware using it. I am using a Thinkpad with an intel wireless + bluetooth card. [17:39] lazyPower: inside vagrant. It fails the same way if I juju ssh mycharm/0 and try to connect, or via debug-hook on db-relation-joined [17:40] aisrael: have you confirmed the postgresql service is running on postgresql/0? [17:40] sorry it lagged [17:41] I'm suspecting its my headset going bad [17:42] although Motorola should be reliable [17:42] Flint__, I use a bluetooth mouse when I travel and it works pretty well. [17:43] Flint__, trouble connecting with your computer or having trouble keeping it connected? [17:43] keeping it connected [17:43] I have seen a decrease in battery life, so it might be the root failure [17:45] I'm setting up Kubuntu 14.04, and had heard that some bluetooth drivers weren't up to speed for a while. [17:45] Flint__, Well that actually is something I noticed. The bluetooth mouse's battery status is not correctly reported in linux. [17:46] Flint__, when I connect the mouse it reads 0% charged, and they are fresh batteries [17:47] lazyPower: Yep. I suspect it might be vagrant at fault, with the way it's routing requests between machines [17:47] ok, thanks [17:47] aisrael: hmm.. its all being routed through a local loopback interface though. [17:48] are you running 2 vagrant boxes, with services spread between them? [17:49] Just one vagrant box for all the services. I pretty much followed the juju vagrant workflow. === html is now known as html5 [17:50] ok, that's the recommended method to use the vagrant box - interesting that you're having an issue connecting though. If you are stuck in a blocked state, file a bug aganist the pgsql charm and assign it to me please. I'll investigate it at my earliest availability [17:51] (i'm lazypower on launchpad as well) [17:52] Will do, thanks. [17:55] aisrael: did you get your psql figured? [17:56] aisrael: you need to vagrant ssh, then you'll probably need to `juju ssh` in to the unit of the charm you're writing [17:56] that should give you the proper access to psql [17:56] there's also a psql charm you could deploy which will give you console access [17:57] marcoceppi: not yet. I did those steps, but it appears that postgresql/0 sees the connect coming from the gateway, not the unit ip [17:57] aisrael: so you're inside your deployed LXC machine? interesting [17:57] sounds like a weird artifact of the local provider [17:57] Correct. [17:58] I'm walking through the config(s) to see if anything sheds some light on it [18:01] sorry it cut em out again === CyberJacob|Away is now known as CyberJacob === BradCrittenden is now known as bac [18:18] getting W: Failed to fetch http://ppa.launchpad.net/ubuntu-audio-dev/ppa/ubuntu/dists/trusty/main/binary-i386/Packages [18:33] Flint_: I think you're in the wrong room [18:34] nah, just tried the laod instructions [18:35] Flint_, what operation generated that error? [18:35] sudo apt-get update && sudo apt-get install juju-core [18:35] it's just a warning [18:36] and the failure of a ubuntu-audio-dev PPA is unlikely to have an influence on how well juju works :) [18:36] Flint_, I found that that ppa does not yet have a trusty release. [18:36] ah ok [18:37] Flint_, I suspect that was installed for an older version of Ubuntu? [18:37] trusty, i think so [18:39] although the documents say it should have been final realease in april [18:48] marcoceppi: i'm having cranial flatulence. If we exit(0) from a hook due to the environment not being ready - how do we re-execute the hook context? we have to relation-set on the initiator dont we? [18:49] lazyPower: well, yes, each relation-set will re-execute the -changed hook for the opposing unit in the relation === roadmr is now known as roadmr_afk === Guest64178 is now known as med_ === med_ is now known as medberry === urulama is now known as uru-afk [20:47] marcoceppi: how do I get help on juju charm subcommands? juju help charm generate, juju charm help generate, juju charm-help generate, charm-help generate, juju-charm help generate all fail [20:48] marcoceppi: nevermind... juju charm generate --help works. === roadmr_afk is now known as roadmr [21:58] marcoceppi, this recent change in mysql http://manage.jujucharms.com/charms/trusty/mysql, pretty much breaks anything that was using juju add-relation nova-compute mysql [21:58] since now it is required to set the interface we want [22:04] stokachu: what recent change? [22:09] jrwren_, https://bazaar.launchpad.net/~charmers/charms/trusty/mysql/trunk/revision/124 [22:09] added an additional interface [22:10] stokachu: don't know about that one, sorry. === CyberJacob is now known as CyberJacob|Away === mup_ is now known as mup [22:32] stokachu, its not about the additional interface surely but the multi-request around shared db [22:33] hazmat, either case you used to be able to do juju add-relation nova-compute mysql [22:33] stokachu, ack, change broke extant [22:33] but diagnosis is the key to recovery [22:34] indeed [22:35] stokachu, what breaks exactly.. i don't see additional interface of note [22:35] it doesn't know whether you want shared-db or nrpe-external-master [22:36] ugh [22:36] fair enough [22:38] stokachu, nova-compute is wrong btw [22:39] the only charms that can require on a container scoped relation are subordinates [22:39] the hosts are all providers [22:39] jamespag`, ^ [22:40] stokachu, he's on vacay this week [22:40] ah [22:40] since its broken anyways the simplest fix is remove that line from nova-compute [22:41] * hazmat runs blame out of curosity [22:43] about a year ago bundled in an ancilliary change [23:10] is it ok to mix juju 1.20.1 and 1.20.5 ? the state-server is 1.20.1, the allocated nodes (maas env) are getting 1.20.5 --