| manadart_ | achilleasa: I did you see my latest change on that patch? | 09:25 |
|---|---|---|
| achilleasa | manadart_: approved; thanks for the change. Can you take a look at https://github.com/juju/juju/pull/11788? (subnets on lxd) | 09:58 |
| manadart_ | achilleasa: Yep. | 09:58 |
| manadart_ | achilleasa: If we could populate parent interfaces for KVM, we could remove the provisioner logic for setting link-layer devices prior to container creation. | 10:19 |
| achilleasa | manadart_: I believe that's doable (that would be the next card on my todo list btw) but I am not entirely sure where this is supposed to happen | 10:21 |
| achilleasa | so in the meantime, I am working on the firewaller spike | 10:22 |
| manadart_ | achilleasa, stickupkid: Forward merge: https://github.com/juju/juju/pull/11808 | 11:00 |
| achilleasa | manadart_: looking | 11:00 |
| achilleasa | manadart_: missed a VirtualPort field ;-) | 11:04 |
| stickupkid | achilleasa, https://discourse.juju.is/t/multichecker-for-unit-tests/3320 | 11:05 |
| manadart_ | achilleasa: Damn. Thought I picked it up in the 2 places you mentioned... | 11:06 |
| achilleasa | there were 3 of them | 11:06 |
| manadart_ | achilleasa: Stand by. | 11:07 |
| manadart_ | achilleasa: Amended it. | 11:10 |
| === skay is now known as skay_ | ||
| === tinwood is now known as tinwood-afk | ||
| stickupkid | hml, CR for find https://github.com/juju/juju/pull/11809 | 13:32 |
| hml | stickupkid: ack | 13:34 |
| === tinwood-afk is now known as tinwood | ||
| hml | stickupkid: reviewed | 14:44 |
| stickupkid | hml, so, this is what I wanted to chat about, I'm unsure we should be doing so much in the client | 14:56 |
| hml | :-) | 14:56 |
| stickupkid | hml, thinking about this stuff https://github.com/juju/juju/blob/develop/cmd/juju/charmhub/infowriter.go#L100-L112 | 14:58 |
| hml | stickupkid: me too | 14:58 |
| stickupkid | hml, I'm going to make some more dramatic changes then | 15:00 |
| hml | stickupkid: daily? | 15:00 |
| stickupkid | sure | 15:01 |
| achilleasa | wallyworld: any particular (performance?) reason why we use https://github.com/juju/juju/blob/develop/worker/firewaller/firewaller.go#L798 instead of collecting all CIDRs and attempting to merge at the end? | 15:45 |
| stickupkid | hml, in relation to "Output from not found charm is confusing, there is no mention that the search term failed, nor that the list is of interesting charms?" | 15:48 |
| stickupkid | hml, that's because the find endpoint is canned response | 15:49 |
| stickupkid | hml, I update my PR | 15:49 |
| hml | stickupkid: right, so it’s not even doing not found? bleh | 15:49 |
| stickupkid | nope | 15:49 |
| stickupkid | yeah, love this, moving target | 15:50 |
| hml | stickupkid: we’ll have to watch that when things go live | 15:50 |
| stickupkid | yeah | 15:50 |
| hml | w00t! | 15:50 |
| === vern_ is now known as vern | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!