babbageclunk | wallyworld: application data bag state changes: https://github.com/juju/juju/pull/10539 | 00:27 |
---|---|---|
wallyworld | babbageclunk: why do we need Role in relationApplicationSettingsKey? isn't "r#123#mysql" enough? | 00:43 |
babbageclunk | wallyworld: I wasn't sure it was guaranteed to be unique - can you ever relate an application to itself? | 00:44 |
wallyworld | no, that's what peer relations are for | 00:45 |
babbageclunk | well, peer relations are for relations that can only be for this application. I mean, could you have an application that has a require and provide endpoint with the same interface, where the other end could be a different application, but could also be the same application | 00:47 |
babbageclunk | ? | 00:47 |
babbageclunk | It just seemed like including the role avoids that question | 00:48 |
wallyworld | technically but i would hope we disallow that | 00:48 |
wallyworld | now you've made me want to check | 00:48 |
* babbageclunk makes a charm to see | 00:48 | |
babbageclunk | ha | 00:48 |
wallyworld | babbageclunk: doesn't seem possible, i get "no relations found" | 00:50 |
wallyworld | which is reassuring | 00:50 |
babbageclunk | well still, there's nothing in the data model that prevents it - if anything it would make more sense to remove application, but that seems a bit weird. | 00:54 |
wallyworld | but it is something juju disallows and doesn't make sense semantically | 01:04 |
wallyworld | babbageclunk: lgtm modulo the role in the id | 01:11 |
thumper | ugh... | 01:39 |
thumper | while looking at one intermittent failure in TestCharmProfilingInfoError I found another | 01:39 |
thumper | perhaps it is the same problem in a different guise | 01:39 |
* thumper digs more | 01:39 | |
thumper | quick review for someone: https://github.com/juju/juju/pull/10570 | 02:03 |
wallyworld | thumper: lgtm | 02:05 |
thumper | wallyworld: cheers | 02:06 |
thumper | anastasiamac: speak of the devil | 02:06 |
thumper | anastasiamac: I was wanting to hand you a nice PR, but wallyworld just looked at it instead | 02:06 |
thumper | anastasiamac: was this one... https://github.com/juju/juju/pull/10570 | 02:06 |
anastasiamac | thumper: devil!! DEVIL! maybe at least d'evil :D ? | 02:08 |
thumper | heh | 02:08 |
thumper | Dr Evil? | 02:08 |
anastasiamac | niiice! thnx wallyworld - so delighted u r holding the fort :D | 02:08 |
anastasiamac | nuh.. not a dr and dont want to b one... | 02:08 |
thumper | I had thought about studying for a PhD | 02:09 |
anastasiamac | thumper: but since u r in the talking mood, do u have a sec to talk about that test i mentioned yesterday | 02:09 |
thumper | but I couldn't think of anything that I really wanted to study that badly for three years | 02:09 |
thumper | anastasiamac: sure | 02:09 |
anastasiamac | 1:1? | 02:09 |
thumper | omw | 02:09 |
timClicks | thumper: could you create the hello-juju repo and authorise juju hackers to push to it? | 04:34 |
thumper | timClicks: in a minute | 04:34 |
anastasiamac | wallyworld: loved 'show-action' PR btw :) thank you ! | 04:34 |
thumper | otp | 04:34 |
wallyworld | thanks for reviewing | 04:35 |
anastasiamac | anytime :D | 04:35 |
wallyworld | i have another almost ready, pending tests | 04:35 |
anastasiamac | \o/ | 04:35 |
wallyworld | all part of actions v2 | 04:35 |
anastasiamac | i have one to go too.. jsut doing live tests now to fill out details.. 'll b epic | 04:36 |
wallyworld | anastasiamac: i didn't change the existing test infrastructure for show cmd, as i neede to use what we had already for list action tests | 04:37 |
timClicks | thumper: oh also hello-juju-charm | 04:37 |
anastasiamac | wallyworld: yeah it's fine.. m more curious if u r keen to change to use cmd.Output :) | 04:38 |
wallyworld | probs, otp, will look properly in a bit | 04:38 |
anastasiamac | nws | 04:38 |
thumper | anastasiamac: https://github.com/juju/juju/pull/10571 | 05:14 |
thumper | anastasiamac: a very simple fix in the end | 05:14 |
thumper | jam: https://github.com/juju/juju/pull/10571 for what we talked about before | 05:33 |
* thumper opens up the PR for any takers | 05:45 | |
thumper | hpidcock, kelvinliu, wallyworld, babbageclunk: https://github.com/juju/juju/pull/10571 | 05:45 |
kelvinliu | thumper: lgtm ,thanks! | 05:47 |
thumper | kelvinliu: thanks | 05:47 |
anastasiamac | thumper: \o/ how did i miss the ping? :( | 05:50 |
anastasiamac | well done!!! | 05:50 |
thumper | anastasiamac: no idea | 05:50 |
thumper | I thought you were just ignoring me :-P | 05:50 |
anastasiamac | i would not dare :P | 05:51 |
kelvinliu | np | 06:05 |
* thumper is done | 06:12 | |
thumper | later peeps | 06:12 |
babbageclunk | hey jam, the application data bag prs we were talking about are https://github.com/juju/juju/pull/10539 (state changes) and https://github.com/juju/juju/pull/10572 (apiserver) | 06:22 |
parlos | Good Morning Juju! | 11:42 |
=== narindergupta is now known as narinderguptamac | ||
stickupkid | anyone around for a CR https://github.com/juju/os/pull/11 | 15:36 |
hml | stickupkid: looking | 15:41 |
stickupkid | hml, yeah, good shout, i'll do that now | 15:48 |
hml | stickupkid: i’m not parsing the comments around DefaultSupportedLTS(). … | 15:49 |
hml | stickupkid: the example reads to be what juju 2.3.x shouldn’t have for a defaul tto me | 15:49 |
hml | stickupkid: done | 15:50 |
stickupkid | hml, that was copy and pasted from juju source code | 15:51 |
hml | ha | 15:51 |
hml | stickupkid: from < 2015 | 15:52 |
hml | ? | 15:52 |
stickupkid | yeah | 15:52 |
stickupkid | hml, done | 16:02 |
hml | stickupkid: looking | 16:04 |
hml | stickupkid: approved | 16:05 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!