[00:57] <babbageclunk> axw: hold off - it turns out you can't sent ProvisioningState in the update - trying out my fix for the fix now.
[01:09] <axw> babbageclunk: ok
[01:09] <axw> let me know when
[01:36] <wallyworld> axw: you free to catch up now?
[01:36] <axw> wallyworld: sure
[02:22] <axw> babbageclunk: I see you pushed an update, ready for review now?
[02:32] <babbageclunk> axw: I think so - doing one last test now
[02:33] <babbageclunk> axw: (sorry, went for a run)
[02:33] <axw> babbageclunk: ok, I'll review optimistically
[02:37] <babbageclunk> axw: testing looks good
[02:41] <axw> babbageclunk: LGTM
[02:42] <babbageclunk> axw: great, thanks
[02:57] <axw> veebers: is http://people.canonical.com/~leecj2/perfscalemem/ from running perfscale_longrunning.py ?
[02:58]  * veebers looks
[02:59] <veebers> axw: yes, from ages ago. (Note, the perfscale tests no longer generate standalone report html pages like the one shown here)
[02:59] <axw> veebers: ah right that's from october last year :)  ok, thanks
[03:00] <axw> veebers: are the long running ones not part of normal QA? don't see it on qa.jujucharms.com/perfscale
[03:01] <axw> normal CI I mean...
[03:02] <veebers> axw: there are no long running tests that run at the moment. We should be getting something in place in the near future
[03:03] <veebers> the parts are largely there , just need to set things up
[03:03] <axw> veebers: okey dokey
[03:03]  * axw nods
[04:36]  * menn0 is EOD
[04:36] <menn0> have a good weekend all
[07:02] <wallyworld> axw: oh btw, i thought i had, but i hadn't. deleting the certs in ~/.local/share/juju/lxd fixed the issue with certificate validation
[07:03] <axw> wallyworld: hrm. that's a bit odd. the client should upload them to LXD
[07:03] <wallyworld> yeah, not sure what's happening there
[07:03] <axw> wallyworld: will look into it next week
[07:03] <wallyworld> sure, it appears it's just me anyway
[07:04] <axw> wallyworld: may well affect anyone that removes/reinstalls LXD though
[07:04] <wallyworld> i'm also having weird lxd snap issue, so might be related
[07:04] <axw> hm, ok
[07:25] <wallyworld> axw: i think this explains things a bit. i juju bootstrapped and the snap lxd was used unexpectedly. http://pastebin.ubuntu.com/23965584/
[07:25] <wallyworld> not sure what's going on
[07:25] <wallyworld> the snap juju was also used unexpectedly
[07:26] <axw> wallyworld: juju was connecting to the wrong lxd?
[07:27] <wallyworld> axw: that's my guess. but i have NFI why the wrong juju binary s being run
[07:28] <wallyworld> which juju should indicate what binary is used right
[07:28] <wallyworld> doesn't seem to be for some reason
[07:28] <axw> wallyworld: ah, I bet the snapped lxd is being told to run on the same HTTPS port. juju will open the unix socket owned by the non-snapped lxd
[07:28] <wallyworld> right, but why is the snapped juju being used
[07:28] <axw> wallyworld: so we'll upload the cert to the non-snapped lxd, then try to connect to the snapped one?
[07:28] <axw> hrm, doesn't explain why it starts working after removing the certs
[07:28] <wallyworld> i didn't ask for it and my juju is in the path first
[07:29] <axw> I don't know
[07:29] <wallyworld> i'll remove the snap and that will "fix" it
[07:29] <wallyworld> i'll ask adam and nicolas for advice
[07:37] <wallyworld> axw: yeah, something is screwed http://pastebin.ubuntu.com/23965626/
[07:37] <wallyworld> awesome
[10:05] <axw> jam: I think it might be a bit risky to land this now, but wanted to get it in front of your eyes for review: https://github.com/juju/juju/pull/6962
[11:11] <jam> axw: reviewed
[11:27] <axw> jam: thanks. would you like me to change `"bond-" in` to startswith? it did seem odd, but didn't seem worth changing
[12:14] <anastasiamac_> axw: if it fixes refered bug, why do you think it'd be risky to land it?
[12:51] <axw> anastasiamac_: there was enough change that it might break something else
[12:52] <axw> anastasiamac_: but it's rc1 for a reason, to catch bugs
[12:52] <axw> so landed it
[12:52] <anastasiamac_> axw: \o/
[12:52] <anastasiamac_> axw: so shall i move the bug to rc1 milestone and mark at Fix committed?
[12:52] <anastasiamac_> axw: or is ther more changes anticipated?
[12:53] <axw> anastasiamac_: I know that it fixes the issue on rackspace, but I don't know about MAAS - and that was the reported one
[12:53] <axw> anastasiamac_: so without knowing that they're the root cause I can't say that it fixes the issue with any certainty
[12:53] <anastasiamac_> axw: k, will keep it as is for now
[12:53]  * axw disappears into the night
[13:01] <perrito666> morning all
[13:03] <frobware> last PR for me folks - https://github.com/juju/juju/pull/6963
[13:04] <perrito666> frobware: it has been a pleasure working with you man
[13:04] <frobware> perrito666: you too!!!
[14:12] <rick_h> frobware: <3
[14:15] <frobware> rick_h: adios dude!
[15:30] <mup> Bug #1663633 opened: Juju 1.2.10: Conversation scope is missing sometimes - causes relation hook failure <juju-core:New> <https://launchpad.net/bugs/1663633>
[15:42] <redir> adios frobware
[15:43] <frobware> redir: adios && take care! Hope all goes well for you over the next few months and enjoy the break before all hell breaks loose. :-D
[15:43] <redir> frobware: :) you too
[15:44] <redir> morning juju-dev
[16:21] <alexisb> morning redir
[16:21] <redir> alexisb: o/
[16:27] <perrito666> elho redir
[16:28] <redir> ¡Buenas! perrito666