[00:19] <thumper> axw: thanks for that controller login branch, I'm sure it'll help some 2.2.4 controllers going to 2.3
[00:53] <axw> thumper: np
[00:53] <axw> thumper: I think it's a little tidier now anyway, but I'm biased
[01:08]  * babbageclunk steps out to look at a house
[01:20] <axw> jam: I think I'm going to have to back out the change to stop running a pinger for other controllers
[01:20] <axw> jam: it seems there's no guarantee that the controller workers will connect to their own apiserver. that seems wrong, but I don't know that I can fix it in time for 2.3
[01:21] <axw> jam: post 2.3, each controller machine can just run a pinger for itself in its dependency engine
[01:40] <axw> thumper: can you please review https://github.com/juju/juju/pull/8122?
[01:43] <thumper> axw: sure
[01:48] <thumper> axw: we should talk about this change, because I don't think it is right
[01:48] <axw> thumper: it's reverting a change to how it was, but ok
[01:48] <thumper> oh...
[01:49] <thumper> right you are
[01:49] <thumper> this is fine
[01:49] <axw> thumper: you wanna HO?
[01:49] <thumper> although, just 5m to confirm
[01:49] <thumper> lets 1:1 HO
[01:49] <axw> sure
[02:41] <babbageclunk> thumper: are we going to have something that's going to collect and amalgamate audit logs from different controller machines?
[04:19] <thumper> babbageclunk: um...
[04:19] <babbageclunk> ?
[04:19] <thumper> we aren't looking at writing a tool at this stage
[04:19] <thumper> they are aware that an amalgamated view would be needed
[04:20] <babbageclunk> thumper: just occurred to me as I was ripping out the DB storage pieces.
[04:20]  * thumper nods
[04:55] <babbageclunk> thumper: since this is a stream of either calls or facade methods, it should probably include a type indicator on each record. Maybe just have a top-level item struct with an omitempty field for each of the two records?
=== frankban|afk is now known as frankban
=== mup_ is now known as mup
=== frankban is now known as frankban|afk
[20:20] <babbageclunk> thumper: I remember you saying something about introspecting the types of API results to find errors - what was that about?
[21:32] <thumper> babbageclunk: otp right now
[21:32] <thumper> with you after
[21:39] <babbageclunk> tsk you're always on the phone
[21:40] <thumper> that's what managers do
[21:41] <babbageclunk> :(
[23:41] <axw> veebers: can https://github.com/juju/juju/pull/7715 be closed?
[23:45] <axw> veebers: and is https://github.com/juju/juju/pull/7858 still something you intend to land?
[23:49]  * veebers looks
[23:50] <veebers> axw: ah right, ci-run is work in progress I need to get back to it and land it, same with the 2nd, I can bring down the PR for that one if it's distracting as there is a little more work needed there (was put up as a PR for the discussion)
[23:50] <axw> veebers: no worries, just checking
[23:50] <axw> sometimes PRs just hang around forever
[23:51] <veebers> axw: axk, thanks for the heads up; I do need to move them forward