| timClicks | Is there a minimum Kubernetes version required by Juju (not charms)? | 00:14 |
|---|---|---|
| tlm | timClicks: we don't have one defined but here would be based on our API usage | 00:16 |
| tlm | would have to go through the code and work it out | 00:16 |
| timClicks | I can probably do that myself | 00:17 |
| kelvinliu | wallyworld: got this pr for adding startupProb https://github.com/juju/juju/pull/11830 +1 plz | 02:29 |
| === balloons4 is now known as balloons | ||
| SpecialK|Canon | What's the "opposite" of `juju deploy` please? I'm trying to clean up after `juju deploy cs:~juju/mariadb-k8s` and, well, it's not `destroy` or `rm` or `undeploy`... | 10:29 |
| stickupkid | juju remove-application | 10:33 |
| SpecialK|Canon | stickupkid: thanks! | 10:34 |
| achilleasa | manadart_: can you take a look at https://github.com/juju/juju/pull/11831? Try to see if you can break it while QAing | 10:35 |
| manadart_ | achilleasa: Yep. | 10:35 |
| stickupkid | manadart_, added tests https://github.com/juju/python-libjuju/pull/430 | 11:12 |
| manadart_ | stickupkid: Thanks. It's already approved. | 11:16 |
| manadart_ | achilleasa: https://github.com/juju/juju/pull/11832 | 12:38 |
| achilleasa | manadart_: since subnets are stored per model, the upgrade step for the ports collection would have to do N space reloads, followed by a per-machine operation... that seems wasteful for large controllers | 15:12 |
| manadart_ | achilleasa: Yes, and we want to change it to avoid per-model repetition. One of many things we'll get to one day. | 15:15 |
| manadart_ | Not sure what we can do here though. Need to think about it. | 15:16 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!