/srv/irclogs.ubuntu.com/2020/07/13/#juju.txt

wallyworldkelvinliu: trvial PR - we've had an outage on 2.7 due to  a nil pointer https://github.com/juju/juju/pull/1182303:55
kelvinliuwallyworld:  just went to get some food, looking now04:10
=== arif-ali_ is now known as arif-ali
wallyworldkelvinliu: trivial forward port of the previous 2.7 fix https://github.com/juju/juju/pull/1182506:13
kelvinliuapproved06:14
wallyworldty06:14
kelvinliunp06:14
manadart_achilleasa stickupkid: need a tick on https://github.com/juju/juju/pull/1182008:32
stickupkidfyi: I never think empty is good for a unknown value https://github.com/juju/juju/pull/11820/files#diff-78b22e8095fbb037f2aea902e906cfb5R4108:33
achilleasastickupkid: it's probably the same if you are using iota. I've also seen the empty value for enum-like definitions being unexported and only used in the package for sanity checking08:45
stickupkidI just think it should be a value08:48
stickupkiderror if it's not08:48
achilleasaunknown is a value ;-)08:49
stickupkid"" isn't ;p08:49
achilleasatechnically, it's a typed empty value :D08:49
stickupkidha08:49
stickupkidachilleasa, this fixes the 2.8 ci integration failures https://github.com/juju/juju/pull/1182610:27
achilleasastickupkid: looking10:28
achilleasaCan I get a CR and QA on the manual cleanup script backport? https://github.com/juju/juju/pull/11827?14:54
achilleasaand a cherry-pick for 2.8 https://github.com/juju/juju/pull/1182815:07
achilleasamanadart_: I think I found a (hacky) workaround for getting the info we need for the subnets on bionic; as the controller instance will be already up, we can query its state (that api works on bionic) and extrapolate the subnet data. I think we won't be able to figure out the parent bridge though which is probably fine I guess15:30
petevghml: I'm working on QA for https://github.com/juju/juju/pull/11822, btw.19:32
hmlpetevg: ty19:33
petevgnp!19:33

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!