[07:14] babbageclunk: hiya === frankban|afk is now known as frankban [08:25] Bug #1639170 opened: openstack trusty:liberty deployment fails due to no relations and insufficient peer units [12:28] dooferlad, afternoon to you [12:29] balloons: hi [12:29] balloons: I was just about to get lunch. Will this be quick? [12:29] so dooferlad, mgz and I spent some time debugging why the triggers weren't working yesterday, and made some changes along with updating the plugin we use [12:30] dooferlad, yea, don't skip lunch for me ;) I was just letting you know that it seems the troubles with some people not being able to trigger a rebuild aren't fixed -- since you still can't [12:30] balloons: thanks for the update. I wouldn't skip lunch, only delay it. === linaras is now known as deanman [13:49] babbageclunk, voidspace, mgz, katco, natefinch: I have two little PRs up still: https://github.com/juju/juju/pull/6539 https://github.com/juju/juju/pull/6538 if someone could take a look [13:49] dooferlad: grabbing coffee first, then will grab one [13:49] voidspace: thanks! [13:59] dooferlad: reviewed 6538 [13:59] katco: just in time for the standup too! [14:00] dooferlad: I'm looking at 6539, but standup time [14:03] mgz: standup time [14:03] katco: ta, I'll hop over [14:59] perrito666, you around? [15:01] alexisb: just recovering from a computer crash [15:01] alexisb: what can I do for you? [15:11] katco: can you please join https://hangouts.google.com/hangouts/_/canonical.com/rick?authuser=1 as soon as you can ? [15:11] rick_h___: brt [15:16] going to pick up daughter from school [15:16] brb [16:17] whoever decided to standardize on ctxt as an abbreviation for context should be ashamed of themselves :/ [16:18] natefinch: why? [16:18] because ctx is a much more common abbreviation, which is why I constantly get errors like this: undefined: ctx in ctx.Stdout [16:21] natefinch: ah! [16:21] fair enough [16:21] right, I'm off to London [16:21] see you all on Monday [16:40] katco, https://bugs.launchpad.net/juju/+bug/1639291 has the trace log attached if you still want it [16:40] Bug #1639291: upgrade-juju fails after bootstrapping with agent-version [16:41] balloons: ta [16:42] katco, note.. I see different behavoirs under 2.0.0 than 2.0.1. I will redo the tests to confirm [16:42] both logs are attached [16:42] balloons: k, i'll treat this as central place for info [16:53] Should juju 2 be looking at kvm support? juju.cmd.jujud determining kvm support: INFO: Your CPU does not support KVM extensions [16:54] balloons: i thought we had foregone kvm support for 2.x [16:54] that was my thought.. Another bug I believe [16:55] No isn't redir working on that? [16:55] rick_h___, what does juju 2 use kvm for? [16:55] yup [16:56] ah, you are solving https://bugs.launchpad.net/juju/+bug/1547665 [16:56] Bug #1547665: juju 2.0 no longer supports KVM for local provider <2.0-count> [16:57] balloons: currently I am working on removing our uvtool-libvirt dependency so kvm will work on ppc64 (and arm64) === frankban is now known as frankban|afk [18:45] redir: made all your suggested code review changes: https://github.com/juju/juju/pull/6498 [19:11] who can tell me everything I need to know about spaces and AWS [19:13] not sure there are many networking people on at this time of day. I think they're all european timezones [19:14] marcoceppi: as natefinch said, most networking people is afaik in europe and some even traveling for a sprint [19:16] perrito666 natefinch :( [19:30] marcoceppi: make sure your account is vpc enabled [19:31] rick_h___: it is [19:31] rick_h___: I want to bind relations on different spaces [19:31] marcoceppi: though the code isn't complete and not everything will work, like endpoint binding I believe. [19:32] rick_h___: so we need juju to be updated, otherwise it's all flat VPC? [19:32] i see the vpc-id in model config, and stuff [19:32] marcoceppi: so you can deploy to different spaces but not get the binding. The charm has to be grabbing the right nic I think. [19:33] marcoceppi: try it? Just deploy with --bind for the spaces you've got. [19:35] https://jujucharms.com/docs/2.0/network-spaces and https://jujucharms.com/docs/2.0/charms-deploying [19:36] marcoceppi: ^ the deploying with binding section [19:42] rick_h___: are spaces supported in bundles? [19:43] rick_h___: cool, I'll try bind when I land [19:44] marcoceppi: so spaces constraints are, I'll have to check the code if bindings with endpoints made the spec. [19:45] rick_h___: awesome, bundles would be nice, but if it has to be a script that runs deploy a few times with bind for this guy, all the better [19:47] Marcoceppi yea, only in constraints sorry [19:47] cheers, no worries [19:59] sinzui: FYI github-check-merge doesn't seem to rerun if you git commit --amend && git push -f on an existing PR [19:59] (which is something that I do all the time) [20:00] e.g. https://github.com/juju/juju/pull/6498 [20:00] natefinch: interesting [20:00] natefinch: even if you !!something!!? [20:00] perrito666: well, I could yes, but I shouldn't have to, is the point [20:01] this happens fairly often - I push something up, realize I forgot to git add a file, so I do so, ammend the previous commit and push up. The checks fail on the original push (because missing file), but then it never reruns.