[01:05] <wallyworld> vino: did you see my email and did it make sense? we should chat when you are free so i can clarify various aspects
[01:05] <vino>  wallyworld: i am waiting for u
[01:05] <wallyworld> vino: ok, let's chat now
[01:06] <wallyworld> am in standup HO
[01:06] <thumper> veebers: ping
[01:07] <veebers> thumper: pong, hows things?
[01:07] <thumper> veebers: not too bad, can you chat now?
[01:07] <veebers> thumper: I can in like 3 minutes?
[01:07] <thumper> sure, ping when you're ready
[01:11] <veebers> thumper: ready when you are
[01:11] <thumper> veebers: coming
[04:02] <kelvinliu_> veebers, would u mind take a look tiny PR when u have time, thx  -> https://github.com/juju/juju/pull/8733
[04:02]  * veebers looks
[04:04] <veebers> kelvinliu_: have you tested that change?
[04:05] <veebers> kelvinliu_: if not I can kick of a run on a jenkins node for completeness sake
[04:05] <kelvinliu_> it's running now, will take a round 30mins
[04:06] <veebers> kelvinliu_: ah, ack. Change LGTM, will comment on the PR
[04:06] <veebers> kelvinliu_: essentially if the add-k8s command works thats good enough (nothing else has changed here
[04:07] <kelvinliu_> the cmd should just work.
[04:20] <kelvinliu_> veebers, wow tests failed at a different failure.
[04:20] <veebers> kelvinliu_: pastebin?
[04:25] <kelvinliu_> https://pastebin.ubuntu.com/p/Cv6N8Cbvqz/
[04:26] <kelvinliu_> there was a different error which was my fault, but now running into controller not found(event can't bootstrap)
[04:28] <veebers> kelvinliu_: looks like a juju error: ERROR detecting credentials for "localhost" cloud provider: adding certificate "juju": Unknown request type
[04:29] <veebers> kelvinliu_: the controller not found errors are cleanup attempts, unfort they're not aways clear if it's a "this is a normal cleanup attempt, it's ok if they're not found" or "I expected to find this but it's not there"
[04:36] <kelvinliu_> if self.existing_controller is None:
[04:36] <kelvinliu_>             raise NoExistingController()
[04:36] <kelvinliu_> is it because controller not found, then raise this error?
[04:38] <veebers> kelvinliu_: no, that's a bad trace not sure why it's mentioning the NoExistingController. You can see that it's attempting to bootstrap something (well after jujupy attempts to use an existing controller) (the line: ERROR Command '('juju', '--debug', 'bootstrap', '--constraints',. . . .)
[04:41] <veebers> kelvinliu_: ah I see why it mentions it, it's crappy but makes sense. There is an "except NoExistingController:" and within that we "with self.new_bootstrap(...) yield" it's within new_bootstrap that the actual error occurs. That's why we see "During handling of the above exception, another exception occurred:"
[04:41] <veebers> kelvinliu_: the issue is the "juju bootstrap" command failed, as per the ERROR seen there
[04:49] <kelvinliu_> weird, i can bootstrap manually.
[04:57] <babbageclunk> Mechanical review for removing tags from machine server IDs, anyone? https://github.com/juju/juju/pull/8734
[05:05] <veebers> kelvinliu_: what command are you using to bootstrap? (it would be interesting as well to see what's in --config/tmp/xxx.yaml, passed into bootstrap command)
[05:07] <babbageclunk> wallyworld: Could you take a look at https://github.com/juju/juju/pull/8734 plz?
[05:07] <wallyworld> babbageclunk: sure, meeting will be done soon
[05:08] <kelvinliu_> veebers, is this config file created by bootstrap py script?
[05:10] <veebers> kelvinliu_: aye it is
[05:11] <babbageclunk> wallyworld: thanks
[05:23] <wallyworld> babbageclunk: done!
[05:23] <babbageclunk> wallyworld: awse
[05:34] <kelvinliu_> hi veebers u r right, the error is related with detecting credentials for "localhost" cloud provider: adding certificate "juju": Unknown request type
[05:36] <veebers> kelvinliu_: ack, not sure why you're seeing it though
[05:40] <kelvinliu_> can't remember which branch my current juju was built from. I re-installed it from snap, it works. need to investigate later about this issue
[07:13] <manadart> Can I get a review of: https://github.com/juju/juju/pull/8730 ?
[12:02] <manadart> Small one for review: https://github.com/juju/juju/pull/8736
[13:04] <srihas> hi, how can we tell a charm to look for custom apt key while installing apt packages on the nodes?
[13:13] <acwork> hello can anyone tell me how I can change system setting after reboot.  For example each time I reboot a ceph monitor  the process comes up on the wrong listener.  I have changed the ceph.conf files but when I reboot it get overwritten.
[13:15] <acwork> Where is juju storing the system settings for a charm?  I changed the /etc/interfaces.d/50-cloud-init.cfg setting but this has no effect on reboot.
[14:37] <bdx> srihas: layer-apt
[14:39] <srihas> bdx: thanks
[14:41] <manadart> Another one for review: https://github.com/juju/juju/pull/8738
[14:42] <manadart> Mostly a backport of 2.4 changes for the same bug.
[14:50] <frankban> hey all, is anyone available for reviewing https://github.com/juju/juju/pull/8735 ?
[15:03] <manadart> frankban: Approved.
[15:03] <frankban> manadart: ty!
[15:03] <manadart> frankban: NP.
[17:00] <bdx> rick_h_: ping
[17:00] <rick_h_> bdx: pong