[00:05] thumper, wallyworld: can I get a review of a description change plz? https://github.com/juju/description/pull/61 [00:05] sure [00:06] thanks! [00:07] looks like there's a connectivity problem from CI to github though? [00:08] (for the description repo bot at least) [00:11] babbageclunk: yes m getting fatal: unable to access 'https://github.com/juju/juju.git/': Failed to connect to github.com port 443: Connection timed out [00:17] https://github.com/juju/juju/pull/10544 for the mongo memory profile restart problem [00:17] * thumper going to walk the dog [00:17] * thumper wonders if it is the https proxy problem [00:18] it is possible we are now expected to go via the proxy [00:18] whereas before we weren't [00:18] just thinking of what it could be [00:21] thumper: yeah, it does sound like that [09:52] achilleasa: With a freshly built develop, I created a nested container. I am now trying after apt purging LXD from a parent. [09:52] manadart: I am getting the same error with 2.6.6 from snap [09:53] Hi all, if you have a few minutes spare - you might like to look at our docs page. I've updated the navigation structure. What do you think? https://jaas.ai/docs [09:58] achilleasa: And it worked after deleting LXD, but this is all on Bionic. [10:00] manadart: this makes it break on my dev box: https://pastebin.canonical.com/p/N9H3PBkdjY/ [10:01] achilleasa: Is it the same if you just add-machine lxd:0? [10:02] manadart: one sec; I am trying to deploy apache2 (everything is using bionic now). I will try the above command next [10:15] manadart: same result with the above command [10:15] should I try to clean up all lxd profiles on the host and bootstrap again? [10:17] achilleasa, what's the issue? [10:18] stickupkid: running https://pastebin.canonical.com/p/N9H3PBkdjY/ makes the lxd container instance error with "Create LXC container: LXD doesn't have a uid/gid allocation. In this mode, only privileged containers are supported" [10:19] i've seen this in lxd issues somewhere [10:21] achilleasa, locally or inside a container? [10:21] stickupkid: so basically it's bootstrap, add machine and create container in that machine (last bit fails) [10:36] achilleasa, those commands worked for me [10:43] Need a review: https://github.com/juju/juju/pull/10545 [10:52] stickupkid: I still get the same error after cleaning up everything and removing all lxd profiles from the host [10:53] achilleasa, including default? [10:53] stickupkid: yes [10:56] achilleasa, can you try this and see it works "lxc profile set default raw.lxc lxc.apparmor.profile=unconfined" [11:11] manadart, done, took some time. [11:11] stickupkid: Thanks. [11:12] stickupkid: bootstrap cannot complete after running ^^^ [11:13] achilleasa, that's interesting, as that removes apparmor, so i'm unsure - now [11:13] it's currently stuck trying to ssh [11:14] i'm not sure how disabling apparmor can prevent that [11:14] let me try again [11:15] if you deleted default profile, i'm unsure how it knows about the device - i.e. lxdbr0 [11:16] this is how the default profile looks like on the host machine: https://pastebin.canonical.com/p/HyrrRvRQFC/ [11:17] everything else looks legit except for the raw.lxc, which we've just added [11:18] stickupkid: oh wait... it seems to proceed now... odd [11:24] stickupkid: same error :-( [11:30] while I am trying to figure out what's going on can someone please take a look at https://github.com/juju/packaging/pull/4? [12:27] achilleasa: I can look in a mo'. [12:28] manadart: I am trying the same experiment on aws now [12:49] manadart: it works with aws so it must be something broken with my local lxd setup... [12:50] stickupkid: did you find out anything with the lxd issue you were hitting? [12:54] rick_h, yeah, so my ineffassign caught a bug, which we needed to fix, both me and manadart located and landed [12:55] stickupkid: ah, gotcha [14:20] Hi. I'm installing openstack-on-lxd and got stuck for a while with rabbitmq endlessly trying to install. SSHing into the container and adding 127.0.1.1 with the container's hostname seems to have fixed it. Looks similar to https://bugs.launchpad.net/charms/+source/rabbitmq-server/+bug/1401830 although that's apparently fixed. Is this a regressi [14:20] on, or have I missed something? [14:20] Bug #1401830: rabbitmq hostname resolution error while deploying

[14:21] * adding to /etc/hosts, that is [16:28] rick_h, 2.6.1 release notes https://github.com/juju/python-libjuju/pull/343 [21:00] timClicks: quick pr review? https://github.com/juju/juju/pull/10546 just had to rerun mockgen for a bunch of files [21:15] thanks timClicks [21:15] hml: sorry for the delay, missed your message at first [21:15] not a problem [23:12] babbageclunk: easy review? https://github.com/juju/juju/pull/10547 [23:13] thumper: ha, got confused seeing the python changes first [23:15] kelvinliu: i've responded to review comments and made updates, had to resolve merge conflicts due to names.v3 landing to develop [23:15] thumper: approved [23:15] babbageclunk: thanks [23:36] which is the earliest version of MAAS that we support? [23:37] timClicks: I think 1.9? [23:48] babbageclunk: timClicks i thought we just recently said that we do not support 1.9? [23:49] thumper: did we not remove it from our ci? [23:49] timClicks: there is the technically, and really [23:49] anastasiamac: that's what I recall (vaguely) [23:49] as of 2.7 we no longer say supporting 1.9 [23:50] woot woot \o/ [23:50] thumper: okay, well I've leave the note in the docs [23:50] but m guessing since th code is still there, we "kind" of support it [23:50] (which commits to supporting MAAS 1.x for the whole of Juju 2.x fwiw) [23:50] wallyworld: thanks!