[08:48] <jamespage> gnuoy, https://code.launchpad.net/~james-page/charm-helpers/keystone-v3-support/+merge/272355
[08:52] <jamespage> gnuoy, http://www.florentflament.com/blog/setting-keystone-v3-domains.html
[10:09] <urulama> mgz: hey ... could we rung "chicago-cubs" branch against CI please? last run was 2 days ago and it's been there from 9/11 ... now, hm, that date might be a sign by its own :)
[10:09] <urulama> s/rung/run
[10:10] <mgz_> urulama: sure
[10:10] <urulama> mgz: ty
[10:10] <mgz_> master is being tested at present, I can make cubs happen next
[10:11] <urulama> mgz: np
[15:16] <beisner> jamespage, updated n-g tests ready for review/landing @ https://code.launchpad.net/~1chb1n/charms/trusty/neutron-gateway/amulet-update-1508/+merge/271960    plz & thx!
[15:54] <beisner> jamespage, fyi n-ovs tests complete @ https://code.launchpad.net/~james-page/charms/trusty/neutron-openvswitch/dkms-12.04/+merge/272390
[16:13] <bdx> beisner: whats up man!? Any thoughts on the new MR? --> https://code.launchpad.net/~jamesbeedy/charms/trusty/nova-compute/next/+merge/272431
[16:14] <beisner> hey bdx !   just wrapping up, going eow soon.   there are functional tests that are automatically queued up to run and report back with any breakage.  once we have lint + unit + amulet results passing, we can turn a human reviewer to it.
[16:15] <beisner> bdx, thanks for retargeting that mp
[16:15] <bdx> sweet. totally. thanks for the heads up!
[16:32] <beisner> np sir!   o/  nice weekend, all
[16:32] <lazypower> cheers beisner, enjoy your weekend
[16:33] <beisner> ditto lazypower !
[16:33] <lazypower> make sure you feed osci some snacks before you go :D
[16:33] <lazypower> !oscisnack
[16:35] <beisner> ha!  i've joked before that it sounds like someone's little puppy name.  but it's eating everyone's lunch already, and has an unhealthy appetite.  #thatqueue
[16:35] <beisner> down, osci, down.
[16:35] <beisner> \o
[16:36]  * lazypower plays rimshot.wav
[17:50] <kwmonroe> service A has 2 units.. when i add a relation between A and B, which private-address does B see?  is it the lowest unit num for A?
[17:55] <marcoceppi> kwmonroe: both private addresses
[17:55] <marcoceppi> kwmonroe: you get a relation-joined on B for each unit of A
[17:56] <kwmonroe> oh - neat!  thx marcoceppi
[17:56] <marcoceppi> realtions are per service, but events are modeled per unit
[19:15] <arosales> may I just say
[19:15] <arosales> watch juju status --format=tabular
[19:15] <arosales> is awesome!
[19:15] <arosales> espeicially when the charms have status
[19:16] <arosales> thank you big data folks
[19:19] <asanjar> you are welcome arosales
[19:20] <arosales> asanjar, :-)
[19:20] <arosales> asanjar, I'll see you next week
[19:21] <asanjar> arosales: definitely ..
[19:22] <arosales> asanjar, excellent. Looking forward to some interesting spark solutions
[19:23] <asanjar> how about some sparkling solution
[21:35] <cholcombe> i'm a bit confused how the charm subordinate works
[21:36] <cholcombe> i declared my charm to be a subordinate of ceph and it says when i add relation that none exists
[21:46] <cholcombe> i think i got it but i'm not sure it's right
[21:51] <marcoceppi> cholcombe: you need to have an interface and scope defined for the subordinate relation
[21:51] <cholcombe> is that defined in the super charm or the subordinate?
[21:52] <marcoceppi> cholcombe: subordinate
[21:52] <cholcombe> ok
[21:52] <marcoceppi> as an example
[21:52] <cholcombe> is it always a container scope?
[21:52] <marcoceppi> yes
[21:52] <cholcombe> so it doesn't matter if it's bare metal, vm, etc.
[21:53] <marcoceppi> cholcombe: scope: container is just a thing that made sense when juju was created before the concept of containers was popular
[21:53] <marcoceppi> scope: unit might be a better name for this now
[21:53] <cholcombe> yeah i agree
[21:53]  * marcoceppi files bug
[21:53] <cholcombe> haha
[21:55] <cholcombe> marcoceppi, if my hook fails to install when i add the subordinate relation can i still do the upgrade force+retry bit to try again?
[21:55] <marcoceppi> cholcombe: I don't see why not
[21:56] <cholcombe> i think juju is confused then
[21:56] <cholcombe> upgrade works but retry fails saying it's not in an error state
[21:56] <marcoceppi> cholcombe: what are you retrying? the subordinate or the primary?
[21:56] <cholcombe> the primary
[21:57] <marcoceppi> cholcombe: what's in error? the subordiante or the primary?
[21:57] <cholcombe> the primary is saying hook failed 'install'
[21:57] <cholcombe> so i tried to attach a subordinate to ceph and the install hook failed
[21:57] <marcoceppi> huh, could you pastebin your juju status output and the commands you're running?
[21:57] <cholcombe> now i'm stuck.  i could trash it but reinstalling ceph is quite slow
[21:57] <cholcombe> yup
[21:58] <cholcombe> marcoceppi, https://pastebin.canonical.com/140643/
[21:59] <marcoceppi> cholcombe: ceph-cephdash isn't in error, ceph-metrics-collector/0 is, run resolved --retry on it
[22:00] <marcoceppi> cholcombe: if you run `juju status --format=tabular` it's easier to consume the output
[22:00] <cholcombe> ah there we go :)
[22:00] <cholcombe> marcoceppi, thanks :)
[22:01] <cholcombe> happy friday dude
[22:01] <marcoceppi> cholcombe: cheers, np!