=== otubo[AFK] is now known as otubo === otubo is now known as otubo[AFK] [01:42] <_mup_> ensemble/expose-docs r311 committed by jim.baker@canonical.com [01:42] <_mup_> Corrected example ensemble status output for exposed service in doc === otubo[AFK] is now known as otubo === otubo is now known as otubo[AFK] === otubo[AFK] is now known as otubo [02:56] <_mup_> Bug #824279 was filed: Security group functions for EC2 provider should retry < https://launchpad.net/bugs/824279 > === otubo is now known as otubo[AFK] [09:35] <_mup_> Bug #824403 was filed: ensemble should support openstack as a provider < https://launchpad.net/bugs/824403 > === otubo[AFK] is now known as otubo [13:30] m_3: I updated the hadoop post to point to the updated formula locations [13:56] kim0: cool... I [13:56] 'll update [14:13] <_mup_> Bug #824557 was filed: Update formula writer tutorial for port exposing functionality < https://launchpad.net/bugs/824557 > [14:16] hmm, I see the wordpress example uses "open-port" in the db-relation-changed hook [14:16] Wouldn't the install hook be a better place for this? [14:16] I'm updating the tutorial to add this, hence the question [14:25] kim0: install would be too early [14:25] kim0: the idea is to open the port when the service is ready to serve users. [14:25] s/users/clients/ [14:25] Yeah .. ok [14:58] fwereade: http://paste.ubuntu.com/663457/ [14:58] RoAkSoAx: cheers [15:05] fwereade: so yeah, given that thte provisioning agent gets started late, and appears to be after a machine has been obtained for the deploy [15:06] fwereade: it tries to shutdown that machine first [15:06] fwereade: so basically, if the agent gets started when it should, we should not see that issue [15:06] RoAkSoAx: yep, I think so [15:12] kim0, the tutorial looks good to me except for the following: we should have one more ensemble status after the ensemble expose === otubo is now known as otubo[AFK] [15:24] jimbaker: mm ok adding [15:24] kim0, cool [15:25] <_mup_> ensemble/expose-docs r312 committed by jim.baker@canonical.com [15:25] <_mup_> YAML serializes as true, not True [15:33] <_mup_> ensemble/trunk r310 committed by jim.baker@canonical.com [15:33] <_mup_> merge expose-docs [r=jorge][f=824218] [15:33] <_mup_> [trivial] Move expose docs from drafts to main, along with minor [15:33] <_mup_> updates, given comprehensive expose functionality now in trunk. [15:38] jimbaker: ok pushed the one with status at the end [15:42] kim0, just taking a look - the drupal service in that status output doesn't have an exposed: true [15:44] jimbaker: yeah .. I didn't want to change machine names, so I'm manually adding what I think should go there [15:45] kim0, no worries, that's the only danger in doing that :) [15:45] yeah hehe :) [15:46] jamespage: What are "labels" in the jenkins-slave interface? [15:47] RoAkSoAx: # TODO: Obtain only systems in the management classes [15:47] kim0, anyway, good to have one more lookover, but with that change, it's ready to merge. looks like you're empowered, so please commit when ready [15:47] m_3: labels are used by jenkins to control which builds run where; you automagically get some if you don't specify OR you can set them manually for a service [15:47] RoAkSoAx: I think it's trying to kill *everything* [15:48] jimbaker: cool, thanks [15:48] RoAkSoAx: I have a fix for that, but it hasn't made it into shadow-trunk yet [15:48] RoAkSoAx: consider me on it [15:48] kim0, just add a [trivial] tag, since it's that [15:48] sure [15:48] jamespage: cool thanks... (I'm documenting interfaces, authors can update them later) [15:48] kim0, see our bzr log for the formatting on that... rather trivial [16:23] adam_g: is rabbitmq really two different interfaces? [16:24] maybe rabbit and rabbit-peer? [16:24] or rabbit-replication [16:24] RoAkSoAx: lp:~fwereade/ensemble/spike-catchup [16:46] m_3: AFAIK, yes. one for clients and another for clustered rabbitmq servers [16:52] adam_g: cool... you good with 'rabbitmq' and 'rabbitmq-node'? [16:54] <_mup_> Bug #824654 was filed: no MachineProvider base class < https://launchpad.net/bugs/824654 > === otubo[AFK] is now known as otubo === otubo` is now known as otubo [19:08] speaking of phones and laptops: http://dilbert.com/2011-08-03/ [19:09] niemeyer: http://golang.org/doc/go_tutorial.html uh, what? 'fmt.Printf'? [19:09] i can't do that [19:09] LOL [19:10] hallyn: There's println for debugging [19:10] hallyn: Just to relieve you a bit ;) [19:10] ok then :) [19:14] hallyn: The uppercasing has some good background, even though it takes a bit to get used to. Every const/var/method/field/function/etc is public if it's uppercased, and package-private if it's lowercased [19:15] hallyn: Besides the ability to tersely define scope, this has the non-immediately obvious benefit of informing scope not only on declaration, but on every use [19:20] to deploy into oneiric machines, do I need to give a specific ami in my environments.yaml, or can I just use "oneiric" as a keyword? [19:20] SpamapS said this was under development ... === otubo is now known as otubo[AFK] [19:21] jcastro: http://www.reddit.com/r/devops/comments/j8qc1/please_talk_me_out_of_writing_my_own_puppettype/ [19:26] <_mup_> ensemble/expose-cleanup r311 committed by jim.baker@canonical.com [19:26] <_mup_> Spike on wait on terminated instances, then delete security groups === otubo[AFK] is now known as otubo [20:15] <_mup_> ensemble/formula-state-with-url r310 committed by kapil.thangavelu@canonical.com [20:15] <_mup_> add formula bundle url to formula state [20:17] kirkland, it looks like it needs an ami, adding support for release name would be easy though [20:18] kirkland, and just grab a daily [20:18] or release [20:21] kirkland, https://ensemble.ubuntu.com/docs/provider-configuration-ec2.html [20:24] <_mup_> Bug #824787 was filed: default-ami appears in schema, but default-image-id is used in code < https://launchpad.net/bugs/824787 > [20:37] <_mup_> Bug #824792 was filed: cli api to store unit local data < https://launchpad.net/bugs/824792 > [21:24] kirkland: default-image-id: ami-1335f37a [21:25] m_3: thanks, trying ....... [21:31] kirkland: lemme know [21:31] m_3: oneiric!!!!!!! [21:33] niemeyer: lp:~kirkland/ensemble/byobu-tmux is working nicely :-) [21:33] niemeyer: shall I send over a merge proposal? [23:39] <_mup_> ensemble/formula-state-with-url r311 committed by kapil.thangavelu@canonical.com [23:39] <_mup_> formula publisher now creates formula states with bundle url