=== cpaelzer_ is now known as cpaelzer === cpaelzer is now known as cpaelzer_ === cpaelzer_ is now known as cpaelzer === cpaelzer is now known as cpaelzer_ === cpaelzer_ is now known as cpaelzer [17:00] smoser: rharper I'm no longer op in this channel. can either of you change topic to Next status meeting Monday: 12/11 [17:01] blackboxsw: /topic all yours === blackboxsw changed the topic of #cloud-init to: Reviews: http://bit.ly/ci-reviews | Meeting minutes: https://goo.gl/mrHdaj | Next status meeting: Monday 12/11 16:00 UTC | cloud-init 17.1 released [17:02] teach a man to fish .... [17:02] tjx [17:02] thx [17:20] someoow i can make it auto-op you. let me find [18:23] Hi all. Having problems partitioning a device: https://paste.ubuntu.com/26113667/ Wondering if it is me or an actual bug… [18:24] Do I need to explicitly wipefs the device beforehand perhaps? [18:25] blkadder: well, ideally you would not need to. [18:25] That’s what I would hope. :-) [18:25] i'd consider it a bug in cloud-init, but yeah. thats what the issue is. [18:25] maybe there is some way to say that ... let me check [18:25] Gracias. [18:28] what is yit looks like if you can live with a gpt it might work [18:28] table_type: <'mbr'/'gpt'> [18:28] k [18:29] I’m trying to move /var so hope so. [18:29] go ahead and file a bug though please [18:29] ubuntu-bug cloud-init [18:29] Will do, thanks. [18:29] will do it [18:39] Done. [18:47] gpt works, thanks. [20:24] hi [20:24] can somebody help me find an error in a cloud-init python config script? http://paste.ubuntu.com/26114214/ the error is at the bottom [20:25] * rharper takes a peek [20:25] for there were the default path variables wrong on centos [20:25] s/for/first [20:25] i changed those [20:25] and then this error comes up [20:26] I wonder if there's something up with the python2.7 puppet in centos then ? [20:26] 6? 7 ? [20:26] its 7 the latest patches [20:27] i already wrote a post to the mailling list but no one replied yet. and i need that puppet working [20:27] with six.StringIO() as outputstream... [20:27] perhaps something wrong with six? [20:27] there was a BytesIO instead of stringIO [20:27] I think I already submitted a patch for that [20:27] google is telling me python2.7 and with and StringIO may be problematic [20:27] * blackboxsw lookgs [20:27] f831a874021f3d6d24cbe5639a176f416b5436a6 in master [20:28] blackboxsw: ah, fixed upstream but maybe not yet in centos's cloud-init [20:28] not sure if that's in the centos version that is being run agains [20:28] there are some fedora/rhel folks in here [20:28] ybaumy: you may need to file a centos/rhel issue to get the upstream commit merged into the distro cloud-init [20:29] i had trouble with django issues with katello/foreman and i installed the latest django version. would something like that be possible to fix this? [20:29] i installed that with pip [20:30] rharper: that will take forever i guess [20:30] yeah given the comment on line 77 of the pastebin, my patch for BytesIO isn't in that version [20:30] yeah, I dunno what the general frequency of updates [20:30] I had a look at debian/sid, and python2 doesn't like with six.StringIO() either. [20:31] ok then i will write an update to my mail to the mailling list [20:31] hrm, we have a copr daily build of master at https://copr.fedorainfracloud.org/coprs/g/cloud-init/cloud-init-dev/ that might help in this case. :/ at least it'd have the cloud-init fix [20:33] ah it might be in epel already [20:33] let me check that [20:34] hmm no not in the official [20:42] yea [20:42] :D [20:42] works [20:42] you still have to change pathes in py and compile to pyc [20:44] thanks guys [20:48] good to hear [20:48] at least we don't have an upstream bug to work ;) [20:50] ybaumy: what's the path changes ? [20:50] that sounds like possible upstream bug or is this related to your pip ? [20:50] ok last question before i quit for today. the certname is nocloud.kent-hanchett.1000.local but my hostname is kent-hanchett.1000.local. why is that? i added certname: "%i.%f" to the cloud-init config [20:51] rharper: i dont think so. on the client i havent updates django yet only on the server [20:58] i removed the %f [20:58] i mean i [20:58] now the hostname is correct [21:02] @blackboxsw I've addressed your comments, when you get a chance can you look again? https://code.launchpad.net/~dojordan/cloud-init/+git/cloud-init/+merge/334341 [21:06] dojordan: thanks, will do today. trying to unstick myself at the moment ;) [21:29] rharper: path changes to /etc/puppetlabs/puppet and the /var/lib/puppet/ssl to /etc/puppetlabs/puppet/ssl [21:29] rharper: this is the client from the official repo [21:29] ybaumy: does that match what distro clients use? [21:30] rharper: i dont know... the official katello install docu says i should use this client. so i did [21:30] I think it could make sense to support both ; [21:30] rharper: think so too. [21:31] ybaumy: can you file a bug ? https://bugs.launchpad.net/cloud-init/+filebug and attach your changes to the path ? we can pick that up from there [21:31] or make the path configurable [21:31] in the cloud-init.cfg [21:32] yeah [21:32] rharper: will open a bug [21:32] thanks [21:32] np [21:33] does puppet.cfg announce that configuration path difference? [21:33] * blackboxsw wonders if we can inspect that config change from puppet itself (either through cmdline or config file) [21:35] https://bugs.launchpad.net/cloud-init/+bug/1736261 hope thats enough [21:35] Launchpad bug 1736261 in cloud-init "puppet.conf and puppet ssl path changed in official client" [Undecided,New] [21:36] im off for today. thanks for your fast help guys. really brought me forward [21:38] ybaumy: thanks again for the bug [22:40] blackboxsw: http://paste.ubuntu.com/26114924/ [22:40] that seems to function for laujnch-softlayer [22:40] lots of improvements still to do, but generally gets near there. and makes me more familiar with launch-ec2 also [22:44] smoser: good deal, I was just pushing up changes to launch-ec2 for specific vpc/subnet/route_table fixes [22:44] checking it out [22:45] just pushed to qa-scripts everything I have ( just finished testing --proposed ec2 for the dhcpclient sanbox fixes) [22:45] smoser: can you push a launch-softlayer to qa-scripts too. we can consolidate on next iteration [23:06] blackboxsw: i'll push tomorrow