[01:29] okey guys i have juju runnung on my AWS :) [01:30] i just had an issue though which got resolved after i restarted my client machine [01:34] koolhead17, nice [01:34] hspencer: hey there. how have you been man? [01:36] been workin ahrd [01:36] hard [01:39] gettin ready for this Euca 3.0 drop [01:39] next week dog [01:39] its coming out [01:39] its lookin good [01:41] cool [01:42] i hanve' t had chance to take care of that walrus issue [01:42] i was lookin through boto code [01:42] cause it works with boto [01:42] wanted to get an idea of how it was fixed there [01:44] k === grapz is now known as grapz_afk === grapz_afk is now known as grapz === grapz is now known as grapz_afk === grapz_afk is now known as grapz [11:47] hey guys, I am working on a charm: is there a way to rerun the relation-name-relation-changed hook when a config is updated? e.g. from the config-changed hook? I am in a situation where a subscriber configuration value must be sent to the provider charm [12:44] <_mup_> juju/trunk-merge r278 committed by kapil.thangavelu@canonical.com [12:44] <_mup_> merge trunk [13:25] Good morning everybody [14:54] frankban, not at the moment [14:55] hazmat: thanks [14:58] frankban, you can run other hooks from any given hook, but particular calls to the juju cli api for relations want certain parameters passed either on the cli or environment, normally juju takes care of that when executing them, but its not possible at the moment for other hooks to pass in the correct parameters to drive that usage of the hook cli api (stuff like relation-get, relation-set) [15:05] hi all, [15:05] running 'juju status' seems to spawn tens of ssh connections leading to trashing, [15:06] additionally I can't ctrl-C it and have to kill it [15:06] any idea on how I could debug ? [15:08] <_mup_> txzookeeper/errors-with-path r45 committed by kapil.foss@gmail.com [15:08] <_mup_> merge trunk [15:09] ha, 'juju -v status' is a bot more verbose, but it mentions timimg out on 3 connections while a 'top' lists tens of processes [15:09] s/bot/bit/ [15:11] the machine have trashed for a few and finally the ctrl-c has been taken into account [15:18] ok, fixed, my ssh configuration was broken recently [15:20] vila: you can run 'juju open-tunnel' to keep an ssh tunnel open [15:22] SpamapS: ironically I broke my config when I grew bored of warnings about not being able to install a tunnel already in place and commented out too much ;) [15:25] Attention juju LXC users on 11.10, this stable update needs verification: bug #922645 [15:25] <_mup_> Bug #922645: ubuntu template should fail on error < https://launchpad.net/bugs/922645 > [15:51] <_mup_> Bug #926146 was filed: destroy-environment says: 'NoneType' object is not iterable < https://launchpad.net/bugs/926146 > === fenris is now known as Guest21494 === Guest21494 is now known as ejat [17:17] hi all [17:26] SpamapS: i was able to execute wordpress charm but add service failed for me :( [17:27] i am trying again now. :P [17:45] jcastro: are you working on some documentation with the recent juju commands in it [17:51] no [17:52] I've submitted a branch with some doc fixes for mentioning READMEs for charms though [17:52] does image initialization on AWS takes lots of time? [17:52] :( [17:53] jcastro: was thinking to use charm get command [17:53] oh we don't need that [17:53] usage: charm get name_of_charm [ local_charm_repo ] [17:53] soon the charm store will land [17:53] and we won't need any of that [17:54] ooh. so for the 9th demo i should not use that command? i should simply show them wordpress charm which comes with the example? [17:54] hi hspencer [17:54] don't use the example charm, it's out of date [17:54] what I do is do the charm get [17:55] ok [17:55] and then tell them "we've got some UI things here that will make it easier, so ignore this ugly --repository, etc. part." === fenris is now known as Guest21540 [17:55] and then just tell them it'll be just like apt is [17:55] ok === Guest21540 is now known as ejat [17:56] hmm. so when i do charm get wordpress, it will bring mysql charm too :) [17:58] now am getting error from AWS dashboard wired [17:59] i was using asia pacific since its nearer [18:00] instance reachablity check failed [18:34] hey koolhead17 [18:34] how you doin? [18:35] hspencer: am good thanks [19:40] SpamapS: thx for adding me :) [19:46] negronjl: np, I'll explain what it is shortly. ;) [19:46] SpamapS: perfect ... already poking at it too :) [20:34] niemeyer, hi, you mentioned that lpad needs to sign all requests so a Noop Auth is not enough as an anonymous login method. But from LP docs I understand that one should be able to get public data without needing to authenticate. [20:34] The latter seems useful to me and simple to add [20:48] janimo: What I said is in agreement, I believe [20:48] janimo: I said we want to send the consumer of the data [20:48] janimo: That's not authentication [20:49] janimo: It requires us to sign the data [20:49] niemeyer, ah I misunderstood. But would LP not answer if we did not set consumer? [20:49] janimo: It would.. [20:50] for public data it is the simplest, just as if it was browsed anonymously via a browser [20:50] I used lpad for a test from GAE and needed the noop auth hack again [20:51] janimo: I don't get what's the problem [20:51] as it does not use files and making an auth for it's datastore is overkill imo if one only wants public data anony7mously [20:51] janimo: Just use OAuth{Anonymous: true}? [20:51] ah, I though only ConsoleAuth is to be used not auth directly [20:52] thanks, I'll check that out [20:52] I did not read the file thoroughly and assumed OAuth is an interface name only not an implementation [21:12] janimo: Ah, no worries [21:12] Soooo.. [21:12] Symbolic links.. [21:12] Hmm [21:32] <_mup_> juju/unit-stop r422 committed by kapil.thangavelu@canonical.com [21:32] <_mup_> draft spec on stopping service units cleanly [21:57] hi, I'm having some problems trying to deploy on openstack [21:58] seems to be stuck in: provision:ec2: juju.agents.provision INFO: Starting machine id:1 [21:58] the bootstrap worked ok, then I did a deploy [21:59] and looking at the debug-log I see this error: EnvironmentNotFound: juju environment not found: is the environment bootstrapped? [21:59] after that, it keeps logging "Starting machine id:1: [21:59] any idea what's going on? [22:06] verterok: juju -v status [22:06] lets see what do we get from that [22:09] koolhead17: http://paste.ubuntu.com/828143/ [22:10] verterok: hal-bot ? [22:10] koolhead17: an irc bot [22:10] verterok: i think it will take sometime for the instance to come up. [22:10] because juju status looks perfect, error free :) [22:11] koolhead17: this is the debug-log: http://paste.ubuntu.com/828146/ [22:11] which has an error, but I have no idea what it means [22:13] verterok: is this custom charm you have written testing? [22:14] koolhead17: yes, this is a custom charm [22:14] and a very simple one btw [22:14] verterok: charm is charm!! :) [22:15] verterok: can you try running allready submitted charm from the repository to test if local env works well with it [22:15] sure [22:16] verterok: juju destroy-enviornment [22:16] and try bootstrap again [22:16] koolhead17: done and done [22:17] verterok: juju -v status? [22:17] koolhead17: not ready yet, "Retrying connection: No machines have assigned addresses" [22:19] hmm. verterok unfortunately i have not been lucky yet to get charms running on my inhouse openstack enviornment [22:19] what i did was i used LXC and did all charming on my local system :P [22:20] verterok: pastebin the juju -v please [22:20] koolhead17: ok, now it's up [22:21] verterok: https://juju.ubuntu.com/ [22:21] juju deploy --repository /usr/share/doc/juju/examples local:mysql [22:21] koolhead17: status output: http://paste.ubuntu.com/828154/ [22:22] koolhead17: ok, running that charm [22:22] COOL [22:24] koolhead17: same error in the provision-agent :/ [22:24] EnvironmentNotFound: juju environment not found: is the environment bootstrapped? [22:24] koolhead17: this is from the debug-log output [22:24] verterok: but the environment is already bootstraped IMO [22:24] koolhead17: same status: http://paste.ubuntu.com/828158/ [22:25] koolhead17: yes [22:25] verterok: are you using PPA [22:26] koolhead17: yes [22:27] verterok: would you mind pasting your enviornment.yaml [22:27] also the debug-log your running from same user account!! :) [22:28] as in same envirnment [22:30] koolhead17: the yaml is: http://paste.ubuntu.com/828167/ [22:31] looks perect [22:31] perfect [22:33] verterok: am sorry but my limited knowledge is unable to solve your issue :( === arosales1 is now known as arosales [22:33] koolhead17: the debug-log is the same as with my charm... :( [22:34] koolhead17: np, thank. I'll ping one of the core devs [22:34] thanks for the help [23:18] verterok: hi, did you get your services started on openstack?