[01:20] <wallyworld> axw: i'm missing somthing for sure - doesn't assigning time.Unix(0,0) to a globalEpoch var mean that the var's value will be stale when it comes to be used?
[01:21] <axw> wallyworld: it's meant to be constant. it's a fixed point in time (jan 1, 1970)
[01:21] <wallyworld> ah, doh, ok
[01:34] <wallyworld> axw: if it is changed to a model where each controller has a clock update worker, is the recovery for the mgo session dying (for whatever reason) just to restart the worker and it picks up a new session?
[01:34] <axw> wallyworld: that already happens, it'll continue to do that
[01:35] <wallyworld> ta, just wanted to be sure
[01:43] <babbageclunk> wallyworld: I'm still having trouble making a local datasource directory - it never seems to pick up the tarball I have under devel/
[01:43] <wallyworld> i'll have to look at the code to see where it searches for the tarballs
[01:45] <wallyworld> babbageclunk: it looks in the top level metadata dir
[01:45] <wallyworld> line 167
[01:46] <babbageclunk> file?
[01:48] <wallyworld> toolsmetadata.go
[01:48] <wallyworld> that's what you are running i assume
[01:49] <wallyworld> that's my guess anyway without debugging
[01:51] <babbageclunk> wallyworld: hmm, actually I think they need to be in tools/<stream>/.
[01:51] <babbageclunk> Thanks!
[01:51] <wallyworld> ah ok, i'll take your word for it
[01:51] <wallyworld> been ages since i looked at this stuff
[02:01] <babbageclunk> wallyworld: Ok, I think that worked - it says it found the file and I can see a new entry for it in com.ubuntu.juju-devel-tools.json
[02:01] <babbageclunk> wallyworld: why does it have all the entries for other tarballs that I don't have, though?
[02:14]  * babbageclunk goes for a run in the beautiful sunshine
[02:58] <thumper> wallyworld: the invite is on jam's calendar, so you could tell him I'll be a little late :)
[03:59] <jam> wallyworld: how late?
[05:21] <wallyworld> axw: swap you reviews? https://github.com/juju/juju/pull/7955
[05:23] <axw> wallyworld: yup, in a little while after lunch
[05:23] <wallyworld> no worries
[07:51] <wallyworld> axw: thanks for those pick ups, sad that i missed them first time
[07:52] <axw> wallyworld: no worries, looking now
[08:27] <axw> jam: if you're still looking at the lease PR, I've finished up the code and it's now at https://github.com/juju/juju/pull/7957
[08:27] <axw> jam: wallyworld's already +1'd, but I'll wait till tomorrow before landing. need to do some more heavy testing anyway.
[08:28] <jam> axw: k
[08:29] <axw> jam: and thanks for your comments so far, I'm happier with how it turned out as a result
[08:56] <mup> Bug #1680787 opened: openstack: add support for neutron networks where port security is disabled <serverstack> <sts> <OpenStack Charm Test Infra:Fix Released by 1chb1n> <juju:Fix Released by hmlanigan> <juju-core:New> <https://launchpad.net/bugs/1680787>
[15:11] <balloons> jam, you are thinking 2.2.6 with force upgrade?
[17:03] <mup> Bug #1726945 opened: [2.3] consume feature hard to manage <docteam> <juju-core:New> <https://launchpad.net/bugs/1726945>
[17:05] <jam> balloons: I'm not sure that 2.2.6 would be needed for just this. More that it is trivial to do in a 2.2 so we might as well target it, in case there is something that *does* warrant 2.2.6
[17:07] <jam> balloons: as for you're version question, i'm not positive, I'll try to give it a look. I thought what I saw was someone trying to go to 2.2.5 using a 2.2~rc2 binary, and there 2.2~rc2 is definitely older than 2.2.5
[17:08] <balloons> jam, right. I seem to remember it was saying 2.2~rc2  was greater than 2.2.5.. probably useful just to include a unit test for beta and rc versions so we know it behaves properly
[17:08] <balloons> not a big deal in the sense this will skip that check, heh
[17:09] <jam> balloons: IIRC we had a bug in the past where the "greater than" was given the wrong way around
[17:09] <jam> the check was correct, but the message was backward
[17:09] <balloons> jam, lol, hah. Ok, that may have been all it was then. Either way, that stuff is easy to mix uop
[21:34] <wallyworld_> balloons: thumper: release call?
[22:23] <wallyworld_> babbageclunk: hey! at some time today, would appreciate a fairly simple review. a bit of boiler plate since a new facade method is added https://github.com/juju/juju/pull/7960
[22:23] <babbageclunk> wallyworld_: sure!
[22:23] <babbageclunk> wallyworld_: looking nwo
[22:23] <babbageclunk> now
[22:23] <wallyworld_> yay, ty
[22:31] <babbageclunk> wallyworld_: looks good - kind of depressing how much boilerplate we need to expose one state method though.
[22:32] <babbageclunk> (This is not a new observation, obvs)
[22:32] <wallyworld_> babbageclunk: yeah, agreed
[23:48] <babbageclunk> Browser's hung - I can hear you but not see or unmute to talk
[23:48] <babbageclunk> wallyworld: anyway, I'll try to grab you in a bit.
[23:48] <wallyworld> ok