=== harlowja_ is now known as harlowja_away [08:04] harlowja_away: thanks, I see you fiddled with the distro code as well :p Ill try to merge it tonight, thanks [08:04] my problem was just the unittest, but this is fine as well :) [12:52] Hey! Is there a way to disable cloud-init after first boot? [12:53] Once the host is provisioned, I don't expect cloud-init to modify anything [14:22] vbernat: c-i will configure the lot at first boot, and will check at consequtive boots if it needs to do something [14:22] so there is no real reason to disable it [14:23] harmw: I am using NoCloud provider and if it is not able to detect its CDROM, the machine will try to get meta-data using the EC2 way [14:24] ah, and you umount the cd drive causing the instance to fallback to ec2 [14:24] I don't want cloud-init to fight with puppet on each boot, I would feel more confident if it wasn't there. [14:24] which slows it down [14:24] yes [14:24] hm, I believe there is some option to have it just stop [14:25] I have tried an empty /etc/cloud/cloud.cfg or an almost empty one, no luck [14:25] hm, specifying an empty datasource perhaps [14:26] I have tried datasource_list: [] but no luck [14:26] ah ok [14:26] in /etc/cloud/cloud.cfg [14:26] yea [14:27] smoser: this would be where you fly into the scene and make some clever one-liner :p [14:28] I could just uninstall cloud-init but I don't know how to uninstall it while ensuring it runs to completion [14:28] If I uninstall it while it is setting up the machine, I think this will break badly :) [14:29] nah, there's probably an easy solution here [14:29] using ds= on the kernel command line doesn't make a change [14:31] I read that was broken, not sure in which version though [14:32] how do I know what is run during "cloud-init final"? [14:32] I could disable cloud-init init scripts from here if I could run a command [14:33] you could ofcourse just pass in some ordinary cmd in your user-data, like systemd disable cloud-init [14:33] but I'm not confortable with that, so smoser or harlowja_away should just wake up - they'll know :) [14:34] yes, that would be a solution but then I may skip some tests [14:34] but if I understand cloud.cfg correctly, runcmd is run during "config" [14:34] but I could disable cloud-init in the user scripts since they are run in final module [14:34] I'll try that [14:42] harmw: well, that seems to work reliably [14:45] :) === zz_gondoi is now known as gondoi === harlowja_away is now known as harlowja_ === harlowja_ is now known as harlowja_away === harlowja_away is now known as harlowja_ [17:09] i wake up [17:10] harmw ya fiddled with distro code, it was chopping off some of the values, so fixed that :) [17:11] i guess thats what tests are for (to find that, ha) [17:20] harmw ie, re.compile(r'^(\w+)="?(\w+)"?') was chopping things in quotes that were like spaces and such [17:20] like "a b" [17:21] anyways [17:37] ah lol [17:37] hehe [17:37] stupid [17:37] *me [17:37] all good [17:37] ill go ahead and apply it in a whil [17:38] kk [17:47] AttributeError: 'TestNetCfgDistro' object has no attribute 'patchUtils' [17:47] would that be my pthon env missing some packages? [17:48] hmmm [17:48] patchUtils is like a cloud-init thing i made [17:48] so probably not env [17:48] oh ok [17:48] are u deriving from the right class [17:49] patchUtils is in http://bazaar.launchpad.net/~cloud-init-dev/cloud-init/trunk/view/head:/tests/unittests/helpers.py#L161 [17:49] so u have to derive tests from that sucker if u want to use that function [17:49] oopsiedoopsie [17:49] I had some old lines of code in thre [17:49] it works now :) [17:49] kk [17:57] ok, done [17:57] pushed [17:57] cool [17:57] smoser: ping === gondoi is now known as zz_gondoi === zz_gondoi is now known as gondoi [18:14] harmw '10 files modified (has conflicts)' :-/ [18:14] the hell [18:14] ya, must be a conflict somewhere [18:15] and I did a merge from trunk like 5 days ago [18:15] Conflicts could easily be caused by my pep8 fixing patch [18:15] that went in two days ago [18:15] that touched a lot of files [18:17] aha! [18:17] well, since I'm way to lazy Ill just let harlowja_ sort it :> [18:17] :-/ [18:17] lol [18:17] i'm just gonna let smoser handle it [18:17] lol [18:17] :P [18:17] responsibilty deferred :-P [18:17] +1 [18:18] pass the buck ftw [18:18] lol [18:18] harmw btw, a guy from yahoo (not sean as u know) i think gonna get involved with the freebsd [18:18] guy pinged me yesterday [18:19] going by the name of hiren? [18:19] https://www.linkedin.com/in/hirenpanchasara [18:19] damn [18:19] how u know all this stuff [18:19] lol [18:19] yea well [18:19] :p [18:19] is the freebsd people like a group of like 10 people that all know each other [18:19] and live in a communue? [18:19] freebsd commune [18:19] lol [18:20] lol [18:25] harmw added a few more review comments to that freebsd one [18:25] beasite? [18:25] wtf [18:25] lol [18:25] ah lets see [18:26] *beastie [18:26] lol [18:26] :-) [18:26] haha yea, given the linux world is using fedora/ubuntu [18:26] hiren_ yo yo [18:26] howdy [18:26] ah look who's there [18:26] ah i get it, reference to the freebsd demon picture thingy [18:26] magic! [18:26] no shit sherlock [18:26] so hiren_ i think https://code.launchpad.net/~harmw/cloud-init/freebsd/+merge/231024 is just going through final adjustments [18:27] if u want to look it over, please do (i'm no freebsd expert that might miss something there) [18:27] k, let me [18:27] hiren_ but after that goes in, we can start building freebsd images and seeing what to do next :-P [18:27] harmw did u need anything else put in freebsd ports (maybe hiren_ knows people or something that can make this happen faster?) [18:28] nope, the dependencies are taken care of already [18:28] if this is merged and version bumped Ill submit the c-i port [18:28] cool [18:29] yep! Super excited to see this finally happening :-) [18:29] harmw: yes do let me know if you need any help from ports people. [18:29] nah, excitement will kick in once smoser version bumps :> [18:29] heh [18:29] ok, though last time that was no problem [18:30] cool. [18:30] marino talked me through some bits, got everything in just a few days [18:30] mario? [18:30] the mario? [18:30] read [18:30] damn [18:31] ;) [18:31] lol [18:31] :P [18:31] ha. [18:31] harmw: cool. [18:32] harlowja_: diff comments, wtf is up with that :p [18:32] I can see a conflict, with the <<< and such [18:32] ya, thats bzr/launchpad for u [18:32] i've never understood why it shows commits like that [18:32] lol [18:32] *commits that have conflicts [18:32] hmk [18:33] +<<<<<<< TREE [18:33] 9 def _resize_ufs(mount_point, devpth): [18:33] 10 return ('growfs', devpth) [18:33] 11 +======= [18:33] 12 +def _resize_ufs(mount_point, devpth): # pylint: disable=W0613 [18:33] 13 + return ('growfs', '-y', devpth) [18:33] 14 +>>>>>>> MERGE-SOURCE [18:33] that one right [18:33] pylint stuff ;) [18:33] yea [18:33] but thats all? [18:33] seems like it [18:33] since it puts the rest of the patch there as well [18:33] which kinda draws away my attention :> [18:33] (though I get it, since its still part of the full merger [18:33] ) [18:34] i'm gonna say its all smoser fault [18:34] lol [18:34] ofcourse [18:34] like you always do :P [18:34] :) === gondoi is now known as zz_gondoi === zz_gondoi is now known as gondoi [19:27] I’m using the “set RANDOM password” feature of cloud-init and I just noticed that its being logged to the console when its performed which means its getting logged to “/var/log/boot.log”. Is there a way to make cloud-init not do that? === gondoi is now known as zz_gondoi