[01:23] With a default install on Centos 7, if I pass user-data through Cloudformation, will those directives get stacked with could.cfg or will they preempt it? [01:29] s/could.cfg/cloud.cfg/g [01:47] I feel like I must be doing something wrong. Can I get a radio check? === shardy_afk is now known as shardy === shardy is now known as shardy_afk === shardy_afk is now known as shardy [14:48] bah timeout on CI [14:48] Error: Cannot find a valid baseurl for repo: base [14:48] Could not retrieve mirrorlist http://mirrorlist.centos.org/?release=6&arch=x86_64&repo=os&infra=stock error was [16:34] blackboxsw: best to just re-run sadly [16:35] yeah I kicked it a bit ago. all green now [18:30] blackboxsw: did we have a trello card for the in-flight sru tests ? [18:58] rharper: kinda.... empty [18:58] https://trello.com/c/wROS4mKT/458-sru-171 [18:58] moved it to doing, I'll start adding a checklist for the bugs related [18:58] I wasn't sure how far you got yesterday [19:00] rharper: did you have a copy of the generated changelog, I can doctor that up to markdown linked bugs etc. [19:01] yeah, one sec [19:02] blackboxsw: http://paste.ubuntu.com/25675009/ [19:02] thanks [19:19] blackboxsw: when you get a chance, https://code.launchpad.net/~raharper/cloud-init/+git/cloud-init/+merge/331755 [19:25] rharper: will do [19:26] blackboxsw: thanks, and your datasource get/crawl branch needs review as well ? [19:26] can you make sure I didn't do anything nasty in my last few checkins https://code.launchpad.net/~chad.smith/cloud-init/+git/cloud-init/+merge/330115 [19:26] yeah was just sayin ' ) [19:27] I tabled surfacing a standardized "platform" key for a followup as this branch is growing to a point it's getting hard to review [19:27] cloud-name would be something akin to "juju clouds" 'cloud' column [19:28] platform would be something like ec2, lxd, configdrive, openstack etc. [19:29] again w/ the instance-data.json. I want to iterate quickly if we can to sort an official 'v1' of our metadata generalized/standardized keys [19:30] do you have a link to that "juju clouds" column ? [19:30] ahh and validated that py2 json can serialiaze bytestrings, so userdata not necessarily always get b64encoded [19:31] rharper: http://pastebin.ubuntu.com/25675178/ [19:31] I wonder if it's worth doing a directory of versioned files [19:31] similar to metadata services [19:31] so someone could always use the first version, but new fields/changes can happen [19:32] that does mean a growing set of file writes if it changes frequently [19:32] blackboxsw: interesting [19:32] the juju clouds output [19:33] here's an official doc https://jujucharms.com/docs/devel/clouds [19:34] there is an interest for us to generally align across our products where it makes sense per https://trello.com/c/CD9xT2nG/45-cloud-id-output-match-juju-cloud-formatting [20:50] rharper: https://bugs.launchpad.net/cloud-init/+bug/1718287 is that actually fix committed? or still in progress [20:50] Ubuntu bug 1718287 in cloud-init "systemd mount targets fail due to device busy or already mounted" [High,In progress] [20:51] Merged into cloud-init:master at revision da6562e21d0b17a0957adc0c5a2c9da076e0d219 [20:51] should be fix committed [20:52] ok tweaked the bg [20:52] ok tweaked the bug [20:52] k [20:55] ok rharper I think this captures the bug list https://trello.com/c/wROS4mKT/458-sru-171 [20:56] y [21:09] can someone remind me how the file /run/cloud-init/enabled gets written ? [21:12] smoser, ^^ by chance ? [21:13] ... looking to be sure [21:14] Looks like systemd/cloud-init-generator maybe? [21:18] powersj, thanks man [21:29] takes me too long. +1 powersj ./systemd/cloud-init-generator [21:29] geez man. I need to tone down the caffeine [21:29] makes me jumpy [21:30] lol [21:30] .... you can imagine how loud my typing is today [21:31] I can almost hear it ;) [21:38] lol [21:57] you know you've had enough when your hands start shaking [21:57] * dpb1 likes coffeee [21:58] sedition <--- knows coffee [21:59] oh? [22:00] i switched to an aeropress a few months ago. truly a wonder purchase [22:00] i can't believe i didn't do it sooner [22:00] wonderful* [22:01] * rharper loves aeropress [22:06] aeropress is good [22:06] * dpb1 used one for a couple years [22:07] but my espresso machine replaced it [22:07] :) [22:07] if i had an espresso machine you'd have to chain my to my desk to keep me from flying away [22:08] yes! [22:26] hm. suddenly my init script doesn't want to add this PPA. interesting [22:28] https://paste.ubuntu.com/25676096/ [22:28] manually running apt-add-repository ppa:x2go/stable seems to work though [22:36] i moved the PPA command above the package command and it worked. black magic [22:36] will do more digging [22:44] I think I've run across a cc_ntp-related bug, but I'm not sure whether it's in the unit tests or the handler itself. [22:44] When I run the unit tests in a chroot that I'm using to build packages, the cc_ntp code can't fine a package manager, so it writes out a systemd-timesyncd config file instead of ntp.conf. [22:45] The unit test then blows up because it can't find ntp.conf. [22:45] s/fine/find/ [23:09] Eh, I'll just file it anyway. [23:10] Can figure out how to fix it later