[03:02] very quiet monday here [03:29] now you've done it [04:50] wallyworld: for the grade field in the snapcraft.yaml, should this be stable or devel depending on if it's a release snap? see https://snapcraft.io/docs/snapcraft-yaml-reference and also your pr https://github.com/juju/juju/pull/11138/files [04:51] do we currently just manually build the dev snaps? [04:52] current release scripts just use default grade: to stable [04:52] I guess I could sed the snapcraft.yaml to change it [09:18] can I get a quick CR on this tiny PR? https://github.com/juju/charmrepo/pull/160 [09:56] achilleasa: can take a look [09:57] stickupkid: up for a 5 min hangout regarding the race conditions a mentioned? May need a pointer or an idea [09:57] nammn_de, daily [09:58] achilleasa, we should ping wgrant about this [09:59] stickupkid: I will when the change lands in juju 2.7 branch [10:04] stickupkid: have you brought in your charmrepo changes from last week? [10:08] stickupkid: arghh.. 2.7 still uses charmrepo.v3; gotta backport the change [10:35] stickupkid: https://github.com/juju/juju/pull/11198 [10:46] nammn_de, done [10:51] stickupkid: ta! I see you are working on the not finding artefacts thing on lxd. Any updates on that one? This one seems to be a great help to me. Need a helping hand? [10:52] nammn_de, ah, not actively atm [10:52] stickupkid: small pointers for me, so that I can take that over? [10:53] nammn_de, when the test suite bombs, we need to work out why we don't create a output tar.gz file [10:54] stickupkid: got it, lookin [11:07] nammn_de: or stickupkid quick CR for bringing charmrepo changes into juju: https://github.com/juju/juju/pull/11199 and https://github.com/juju/juju/pull/11200 [11:08] achilleasa: approved [11:11] stickupkid: have you seen this error before? https://github.com/juju/juju/pull/11200/checks?check_run_id=435919954 [11:13] is charmrepo.v4 gomod-only now? [11:14] achilleasa, why does dep care about gomod though [11:14] achilleasa, shouldn't it just be a code repo? [11:17] stickupkid: yeah... don't get it. The juju .lock file has been updated but the digest looks empty: https://github.com/juju/juju/pull/11200/files#diff-bd247e83efc3c45ae9e8c47233249f18R1436 [11:17] haha, lol wat [11:18] can you rebuild it? [11:18] Let me try with modules off... [11:30] stickupkid: second time worked... go figure [13:15] morning party folks [13:24] Morning rick_h [13:29] nammn_de: I think we have a flaky test (TestConstraintsOpsForSpaceNameChange). Was that added by the recent work for space renaming? If that's yours you may want to throw in a strings.Sort prior to comparing https://paste.ubuntu.com/p/pnmnxrdBpQ/ [13:30] achilleasa: yeah thanks! Added that in the next spaces patch [13:30] achilleasa nammn_de: Or use jc.SameContents. [13:31] manadart: thanks for the tip; didn't know that was a thing! [13:31] manadart achilleasa: sorted it was put in the bigger patch here: https://github.com/juju/juju/pull/11186/files if pressuring I can exclude and put in seperate patch [13:31] manadart: ohh did not know that one === narindergupta is now known as narinderguptamac === icey_ is now known as icey [15:23] rick_h, so the schema tests won't work because both juju 2.7 and develop target the pylib master branch and we don't know if there are breaking changes [15:24] rick_h, the easy fix is to add branches back to pylib [15:24] stickupkid: boooooo, maybe [16:17] manadart, nammn_de maybe we should just push all artifacts to s3 [16:17] be done with it [16:22] stickupkid: Sounds like a plan. [16:22] stickupkid: that sounds like something great. I can look into that next ci day (may need here and there some consideration points) [16:23] nammn_de, we'll still need to get it out of the container though... [16:23] stickupkid: yeah thats something i am looking today, if not next week =D [16:56] rick_h, ho? [17:21] stickupkid: /me jumps in [17:21] oh wait, that was 30min ago sorry [17:21] stickupkid: what's up? [17:21] in ho:) [20:11] morning juju [20:11] morning timClicks [20:12] howdy timClicks [20:15] it looks like we're starting to see more and more lively discussions on discourse [20:15] chit-chat [21:30] where is the source code for the k8s charms in https://jaas.ai/u/juju/#charms? we should flesh out their readmes [22:03] easy review for someone? https://github.com/juju/juju/pull/11202 [22:03] babbageclunk: love those [22:04] babbageclunk: the old shortwait to longwait switcheroo [22:08] timClicks: ha, thanks! [23:28] remind me again why there's no default model provided with k8s clouds? I did know once [23:29] default is a namespace that exists in a kube cluster [23:29] and may be used by stuff already deployed in a cluster [23:29] so we [23:29] would need a new name [23:30] easier to just not create one atm [23:31] or if we did, the default model would need special behaviour so it doesn't create the namespace nor delete it. It would also have to be able to play nice with what is already in there/share with another juju controller etc [23:31] just too much to deal with. [23:31] hpidcock: I thought it was to do with the namespace issue [23:32] yeah, thats the root of the issue [23:32] but whatever way you try to work around it isn [23:32] 't nice [23:33] a user could delete the namespace then create a model called default [23:33] but that still has issues