[00:18] <thumper> wallyworld: would like a quick chat re: intermittent failing provisioner test when you have a moment
[00:36] <thumper> veebers, wallyworld: https://github.com/juju/juju/pull/8434
[00:36] <thumper> or anyone really
[00:51] <anastasiamac> thumper: lgtm
[00:51] <thumper> thanks
[00:51] <thumper> anastasiamac: I have another good bug for you
[00:51] <thumper> I'm just writing a comment
[00:52] <anastasiamac> :( i was goign to keep going with cred ux (my bug-meter was appeased with 2 for the day) \o/
[00:52] <anastasiamac> thumper: but i'll look (curiosity and the cats as u know...)
[00:54] <thumper> https://bugs.launchpad.net/juju/+bug/1638714
[00:54] <mup> Bug #1638714: Upgrading a juju model before the controller gives unhelpful error message <usability> <juju:Triaged> <https://launchpad.net/bugs/1638714>
[00:54] <thumper> anastasiamac: this relates to the issue someone brought to the mailing list earlier today as well
[00:54] <thumper> we aren't good at helping people when they do the wrong thing
[00:55] <thumper> anastasiamac: btw, the setting to trace gives more output from the actual provisioner worker
[00:55] <thumper> it says more about what it is doing
[00:55] <thumper> the info in the test was just for extra logging
[00:56] <thumper> and yes, it works :)
[00:56] <thumper> I did test it
[01:06] <thumper> oh ffs
[01:06]  * thumper doesn't know vi
[01:43] <anastasiamac> thumper: looked and commented... since it's coming from state, the re-phrasing may not be as simple... i think that this is the case of where we are lacking some guidance as the user confusion is because they need to upgrade 'controller' model first
[02:39] <thumper> anastasiamac: but we can do extra checks... it isn't that hard, it just may be a few more api calls, but a better user experience
[04:57] <thumper> anastasiamac: did you want to have a quick chat before I EOD?
[05:00] <jam> just checking that other people are failing to connect to canonical IRC, right?
[05:00] <jam> (I'm assuming it is the restart-for-spectre issue)
[05:00] <veebers> jam: I am
[05:01] <veebers> wgrant: I imagine related to a firewall reboot? (I saw a comment to that affect just before connection dropped :-))
[05:01] <thumper> jam: yes IS is rebooting machines
[05:01]  * thumper EODs
[05:02] <jam> veebers: just reconnected
[05:05] <wgrant> Also see your email :)
[05:06] <wgrant> An announcement of utter chaos was sent last night.
[07:12] <jam> can I get a review on https://github.com/juju/juju/pull/8432 ?
[07:12] <jam> anastasiamac: https://github.com/juju/juju/pull/8436 reviewed
=== frankban|afk is now known as frankban
[10:37] <anastasiamac> jam: thnx but i believe it's not ready - tim and i will talk tomorrow. whilst re-phrasing the message is good, the UX needs to b slightly adjusted for better experience.... fwiw, the wording was almost verbatum to what was in the bug
[11:29] <jam> anastasiamac: I'd argue that your fix is still better than no fix, even if we ultimately want to do more
[11:31] <frankban> jam: re your comment at https://github.com/juju/juju/pull/8427 (which is ready for review) develop will be 2.4?
[11:31] <jam> frankban: yes. develop == 2.4
[11:31] <frankban> jam: I guess we can backport the patch later when required?
[11:32] <jam> frankban: IME its usually easier to start at the older code base, but its up to you.
[11:32] <jam> frankban: and since you guys are the ones that want the patch, its more a question of whether you're ok waiting until we do a 2.4 release
[11:32] <jam> which 2.4 final is at least a month away
[11:34] <frankban> jam: well, the patch against develop is already there, I;ll talk with uros about that. who could I ping to get a review of that part of the code?
=== kwmonroe_ is now known as kwmonroe
=== rharper` is now known as rharper
=== frankban is now known as frankban|afk
=== tasdomas` is now known as tasdomas
[20:25] <thumper> babbageclunk: ping
[20:25] <thumper> https://github.com/juju/juju/pull/8438
[20:51] <babbageclunk> thumper: yeah, saw that - looking at it now
[20:53] <thumper> babbageclunk: did you want to talk through it or are you happy?
[20:53] <babbageclunk> thumper: Oh, I thought you'd put in a skip and I'd need to fix it. No, that seems good to me!
[20:53] <thumper> no... I fixed it
[20:54] <thumper> :)
[20:54] <babbageclunk> Thanks! Sorry!
[20:54] <thumper> that's fine
[20:54] <thumper> babbageclunk: want to approve the pr? and I'll land it
[20:55] <thumper> babbageclunk: got 5 minutes?
[20:55] <thumper> babbageclunk: hangout?
[20:55] <babbageclunk> thumper: done. Ah, I had seen 8434 but not the fix
[20:55] <babbageclunk> yup
[20:55] <babbageclunk> 1:1
[20:55] <thumper> omw
[22:06] <thumper> ugh...
[22:07] <thumper> I'm sure my local test failures are probably due to me running go 1.10, and the merge and test bots using 1.9
[22:37] <thumper> anyone https://github.com/juju/juju/pull/8439
[22:54] <thumper> thanks wpk
[22:54] <thumper> gym time
[22:54]  * thumper heads off
[23:35] <kurt_> Do you guys know if anyone watches the maas channel anymore?
[23:35] <kurt_> Seems like they don't