=== kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away [13:49] mbruzek: when you get settled in this morning can I get a review on https://github.com/chuckbutler/flannel-docker-charm/pull/16 === scuttle|afk is now known as scuttlemonkey === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away [17:55] rick_h_: hey when you do the new bundle spec [17:55] will old bundles stop working? [17:55] jcastro: no [17:55] jcastro: we'll do like we do now and probably go through a transform step [17:56] ok [17:56] jcastro: but atm the new work is around just 'synchronized uncomitted changes in the gui' so we'll work with thumper and his planning work on that [17:56] these bundles will be around a long while [18:32] is there a way to remove a container - which already had the unit which it was running removed - from a machine, other than force remove the machine running the container? [18:34] jrwren: when you destroy the service, it should clean up the container as well as part of the machine reap process. if that's not the case - i'd file a bug. [18:34] jrwren: oh wait i misread that - this is a problem with precedence, as the machine itself went away and the container is still registered on the environment? [18:35] lazyPower: no. I expected exactly what you just said. maybe it is bug. === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away [19:48] wwitzel3: (migrating here as its juju specific) - There's a lot of the OpenStack networking vendors coming aboard the Juju Openstack ecosystem [19:48] see: midonet, plumgrid, etc. [19:49] wwitzel3: it may be worthwhile to link akanda to our ISV Onboarding team to do outreach/contact post ODS [19:50] lazyPower: yeah, that's a good idea [21:47] lazyPower, marcoceppi: who knows most about python django charms? looking for celery stuff [21:47] thumper: it does celery ootb? [21:47] I noticed that django-python relates to rabbitmq-server through amqp [21:47] rick_h_: but it doesn't set up the celery workers, no? [21:48] rick_h_: I noticed the celery config [21:48] * marcoceppi shys away quickly [21:48] thumper: oh, I didn't realize django shipped with that [21:48] but I want the time based checker on master [21:48] thumper: I'd guess that it might write out to settings.py but can't imagine what else it would do [21:48] rick_h_: it does write out settings [21:48] * rick_h_ pokes at the charm hooks [21:48] thumper: iirc django needs a maintainer :) [21:48] thumper: you seem to be really familiar with the codebase [21:49] but I was wondering if someone else has already set up the worker bits [21:49] * lazyPower nominates thumper [21:49] lazyPower: maybe... [21:49] lazyPower: at least I'd learn how to do it properly [21:49] thumper: might ping the other end of the company as they use django more I think. SSO is a django app if I recall. [21:49] I'm +1 on having a core maintainer owning a single charm [21:49] if nothing, it gives you cause/effect to dig your hands in deep in the grits of charming [21:49] lazyPower: greedy aren't we? :P [21:49] lazyPower: perhaps while I'm waiting for lxd to implement the needed functionality, I should look more at django-python [21:50] lazyPower: would really like virtual env support [21:50] and python 3 [21:50] and django 1.8 [21:50] I'm using the current one with django 1.7 and it is fine [21:50] thumper: oh interesting it does install the package for you heh [21:50] pip_install('django-celery') [21:51] yeah... [21:51] which I'd have thought would be in the requirements.txt anyway [21:51] or at least, I was going to do that [21:51] hmm, do yea it uses django-celery, sets up the config, and the relation bits. [21:51] thumper: we have virtualenv support [21:51] lazyPower: not in the python-django charm I'm using [21:51] thumper: we're using that in a few charms, where the charm drops in the .venv and does the bits it needs to do [21:51] oh that, yeah man, maintainer needed :) [21:52] * lazyPower nudges thumper closer to agree'ing to take it on [21:52] lazyPower: let me poke around before you throw it at me fully [21:52] too late, already signed you up [21:52] incoming MP marking you as the active maintainer [21:52] #dealwithit [21:52] lazyPower: what facility does charm helpers give me for writing service scripts? [21:52] thumper: meaning upstart/systemd jobs? [21:52] like: a celery worker [21:53] thumper: we have jinja and cheetah support.... [21:53] lazyPower: ack [21:53] lazyPower: so... write a template file, and put it in the right place... [21:53] charmhelpers.contrib.templating iirc [21:53] ye [21:53] hmm... [21:53] lazyPower: what's the status of a stand alone nginx charm? [21:54] I currently have that smashed into my payload charm [21:54] since python-django is a framework charm, meaning its providing scaffolding, there are some conventions you can follow - like making an implicit templates/contrib/etc. dir in your project that the charm scans and renders. [21:54] thumper: there's one that exists... let me find it. its not ~recommended tho [21:54] it was in progress then other things surfaced [21:54] lazyPower: it didn't exist july last year when I needed it [21:54] https://bugs.launchpad.net/charms/+bug/1356856 [21:54] thumper: ^ [21:54] Bug #1356856: New Charm: NGiNX [21:55] plz don't use that [21:56] rick_h_: who are the main users of django internally? [21:56] rick_h_: who are using charms? [21:56] psh, who isn't using charms! [21:56] I should poke [21:56] marcoceppi: well... maas uses django [21:56] marcoceppi: but I don't think we have a maas charm [21:56] thumper: we do have a vmaas charm iirc [21:57] marcoceppi: for local? [21:57] * marcoceppi searches [21:57] ODS wifi is a bit slow [21:58] thumper: https://jujucharms.com/q/maas [21:58] https://jujucharms.com/q/vmaas [21:58] ninja'd by marcoceppi [21:58] lazyPower: and mine includes all maas charms <3 ;) [21:59] LOL - beat by 1 charm that wasn't even maas related [21:59] you win [22:00] * thumper wanders into a meeting === kadams54 is now known as kadams54-away === mattgrif_ is now known as mattgriffin