[12:49] <beisner> o/ good morning
[12:49] <beisner> jamespage, ack on trusty icehouse, juno, kilo @ proposed, will kick those shortly
[12:50] <beisner> gnuoy, woot re: initial tempest charm!
[12:50] <gnuoy> beisner, I have some questions............... :)
[12:51] <beisner> gnuoy, sure.  let me look at it real quick.
[12:51] <gnuoy> beisner, when you have a sec have you seen http://paste.ubuntu.com/12013477/ before ?
[12:52] <gnuoy> It looks like it's looking up the 'orchestration' service in keystone, which doesn't exist
[12:52] <gnuoy> do we normally disable those tests somehow?
[12:53] <gnuoy> :q
[12:54] <beisner> gnuoy, no we run all of the tests for all of the services.  we just did some tuning to the template surrounding heat issues.  one remains:
[12:54] <beisner> bug 1431013  ... appears to be a heat package issue.  the /etc/heat/templates/ dir is awol.
[12:54] <mup> Bug #1431013: Resource type AWS::RDS::DBInstance errors <amulet> <canonical-bootstack> <openstack> <uosci> <heat (Ubuntu):Fix Committed> <heat (Ubuntu Vivid):New> <heat (Ubuntu Wily):Fix Committed> <heat (Juju Charms Collection):Invalid> <https://launchpad.net/bugs/1431013>
[12:55] <beisner> gnuoy, this template, gives us just 1 tempest fail, and it's that ^ bug.  http://bazaar.launchpad.net/~ost-maintainers/openstack-charm-testing/trunk/view/head:/templates/tempest/tempest.conf.template
[12:55] <gnuoy> beisner, fantastic, thanks
[12:56] <beisner> gnuoy, can we please add charm config options for two installation use case scenarios:
[12:59] <beisner> gnuoy, http://paste.ubuntu.com/12013518/
[12:59] <gnuoy> beisner, hmm, that template you pointed me at is the one I'm using
[13:00] <beisner> gnuoy, ^ that is in the interest of making installation options flexible
[13:01] <beisner> gnuoy, here is a t-k/next tempest result from yesterday:  http://10.245.162.77:8080/view/Dashboards/view/OpenStack%20Deploy/job/deploy_with_deployer/10381/artifact/tempest-allresults.10381/*view*/
[13:04] <gnuoy> beisner, I need to nip out for ~1 hour but I think scenario one is already covered
[13:04] <beisner> gnuoy, ok thanks again
[14:28] <gnuoy> beisner, I've added pip proxy config option. I don't know where to go with my tempest failure though.
[14:28] <beisner> gnuoy, i suspect tempest-lib - i may be on an older version of that
[14:45] <marcoceppi> lazyPower: I think it's all the nulls
[14:45] <beisner> jamespage, trusty icehouse|juno|kilo @ proposed deploy tests complete;  tempest results== distro/updates runs (ie. looks OK to me).
[14:47] <jamespage> beisner, awesome-o
[14:48] <gnuoy> beisner, I'm cloning tempest rather than tempest-lib is that wrong?
[14:50] <beisner> gnuoy, there may be more than one way to work this.  i've generally had to pip install tempest-lib and keep it updated, as well as gitting tempest itself.
[14:52] <gnuoy> beisner, I'm happy to encapsulate stuff in the charm but I don't think I'm a hardened enough tempest warrior to work out what the charm should be doing. It just seems like a horrific pip mess
[14:52] <beisner> gnuoy, it will be
[14:52] <beisner> gnuoy, all sorts of opportunity for package vs pip version conflicts
[14:54] <beisner> gnuoy, tldr, what uosci does:  install all of the clients as listed in o-c-t's configure script;   then pip install tempest-lib;   then git tempest and do the things in ./configure.
[14:54] <gnuoy> beisner, so I'm running the run_tempest with the please-create-a-venv flag and it's pulling in tempest-lib into that venv by the looks of it
[14:54] <beisner> gnuoy, that's great.  we want that instead i think.
[14:54] <beisner> so prob tempest-lib 0.7.0
[14:55] <gnuoy> yep, it is
[15:08] <redelmann> marcoceppi, hi
[15:08] <redelmann> marcoceppi, if i run somthing like this in charm "hookenv.log('--------')"
[15:08] <redelmann> marcoceppi, debuglog show this: "logger.go:40 error: flag provided but not defined: --------"
[15:09] <redelmann> marcoceppi, could be a bug?
[15:11] <marcoceppi> redelmann: what version of Juju are you using?
[15:11] <marcoceppi> OH
[15:11] <marcoceppi> I know why
[15:11] <redelmann> marcoceppi, 1.24.3
[15:11] <redelmann> marcoceppi, client and agent
[15:12] <marcoceppi> it's being literally passed to juju-log as `juju-log ----------` so juju log thinks it a paramter
[15:12] <marcoceppi> if you used anything else other than a -, like =
[15:12] <redelmann> marcoceppi, y suppose that
[15:12] <marcoceppi> you wouldn't get that error
[15:12] <redelmann> y = i
[15:12] <marcoceppi> because the error message says "flag provided but not defined"
[15:13] <redelmann> marcoceppi, yes it's
[15:13] <redelmann> marcoceppi, I assumed that
[15:13] <marcoceppi> redelmann: so if you use any other character other than '--------', say '[15:13] <marcoceppi> ' it'll work
[15:13] <redelmann> marcoceppi, ok, thank for you time
[15:13] <marcoceppi> we can look into patching charmhelpers to escape this properly
[15:14] <marcoceppi> but to unblock you I'd just change it for now
[16:50] <mwenning> jamespage, also https://bugs.launchpad.net/charms/+bug/1431445 , thx for any help
[16:50] <mup> Bug #1431445: Support for Calico in OpenStack: bird <Juju Charms Collection:Fix Committed> <https://launchpad.net/bugs/1431445>
[20:05] <pmatulis> with maas provider and 'juju add-unit --to <blah>' , is <blah> always a host ID or can it be a hostname too?
[23:44] <apuimedo> is it already possible to do quickstart of a bundle with charms on launchpad?