=== kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away === kadams54 is now known as kadams54-away === Guest33721 is now known as benonsoftware === Murali_ is now known as Murali === benonsoftware is now known as Guest48867 === Guest48867 is now known as benonsoftware [08:27] Hi Jamespage [08:38] Murali, good morning [08:38] Good Morning Jamespage [08:38] finally we are able to successfully launch the openstack [08:39] we are able to see the openstack-dashboard [08:39] Murali, \o/ [08:39] congrats! [08:39] Now we have issue in logging to instances we created in openstack [08:40] Murali, oh yes [08:40] * jamespage braces for mtu confusion [08:40] Murali, you've done the normal stuff like enabled SSH and ping access to the instances? [08:41] its says "console is currently unavailable. please try again later" from openstack dash-board [08:41] Murali, oh the console access is disabled by default [08:41] ok [08:41] Murali, it does not scale that well and is not fully security supported in Ubuntu, so we can't enable it by default [08:42] Murali, we try to stick to fully supported options by default [08:42] how to make to enable? [08:43] or how to access the instance we created in openstack [08:48] Murali, the best way IMHO is with SSH [08:48] Murali, the bundle on jujucharms.com has some details on how to configure the firewall to allow access [08:49] Murali, you can inject an SSH key as part of booting an instance as well [08:49] Murali, the dashboard is OK for a quick drive through, but for any serious use you'll want to get to know the command line tooling :-) [08:49] on cli also we tried Jamespage [08:50] Murali, what problem did you hit? [08:52] we are now able to login via CLI Jamespage [08:52] Murali, good [08:52] Thanks a lot for kind response [09:00] Murali, no problem [09:04] Murali, I was thinking about why you ended up with no relations to start with - I suspect it was due to the error on the neutron-gateway during deployment [09:04] Murali, I *think* that any errors like that will stop the second phase of adding relations from being executed [09:05] Murali, I also believe that the juju-gui is switching to not work like that - but rick_h_ would need to confirm on that - are we moving to a deploy and related, [09:05] ? [09:05] not waiting for services to deploy before relating [09:06] we first ran the jujuquickstart [09:07] it successfuly deployed all the services [09:07] we waited for longtime couldnt see the relations added [09:08] and juju-gui showed some error saying "error occured but information" [09:09] then we applied juju-deployer bundle.yaml as per you [09:10] But we saw that neutron-gateway deployment was success [09:44] jamespage: Murali yes, depending on the gui version it would stop the deployment if it hit an error. There's plans to update that to not be the case, but it's part of a bigger chunk of how bundles are handled and not ready yet. [10:05] ok [11:48] Hi Jamespage [11:48] sorry to trouble you with questions [11:48] we would like add the local jujucharm which we developed to bundle.yaml [11:49] is it possible? [11:56] Murali, it is - you can just add it to the deployed environment and the export the bundle from the GUI [12:03] is there any changes required in bundle.yaml Jamespage? [12:05] how to add the repository path in bundle.yaml file? [12:06] Murali, oh I see [12:06] tricky that one [12:15] Murali: the GUI doesn't support bundles with local charms because they're not reusable then. I don't recall if the deployer had a way to do that. I think it tends to use vcs locations in order to support a charm not in the main charm store [12:16] ohh ok rick_h_ [12:16] is there any other way to deploy the local charms with out juju-gui [12:17] and using jujudeployer command [12:18] is it possible via CLI juju deployer [12:36] is it possible to loosen the permissions on log files for the local provider? Currently 0500, syslog:syslog [12:37] 0600 rather [12:37] I can edit the /etc/rsyslog/conf.d/ juju file, but was wondering [12:38] if it would get rewritten on bootstrap, or if it was worth loosening by default? Given the local provider is explicitly for dev only? [14:03] rick_h_: you can use local charms with deployer if you have JUJU_REPOSITORY env set [14:03] Murali: ^ [14:03] we make extensive use of this in our testing of bundles that dont yet exist or have a place in the charm store - as ingest in personal namespace introduces a latency between changes. [14:05] Murali: if you need an example of a bundle leveraging local charms - see: https://github.com/whitmo/bundle-kubernetes/blob/master/specs/local.yaml === Guest81339 is now known as rcj === rcj is now known as Guest41495 [14:49] Thanks lazyPower [15:07] I'm looking for some help on the correct way to configure the charm(s) to deploy Nova correctly to use either Neutron or FlatDHCPManager (whatever is easier) when deployed by Juju on a MAAS node. The node only has 1 network interface that is brought up (eth0) and it is already bridged to juju-br0. I'm running trusty/icehouse. === whit_ is now known as Guest39130 [15:36] bloodearnest: I opened a bug on that, about changing ownership of the logs, but it is won't fix as the team are trying to make the local provider behave more like the rest of the providers rather than being full of special cases. [15:39] stub, ack, thanks, good to know [15:39] approve of that goal [15:40] yeah, agreed [16:35] Hi, I've recently installed a juju master (charm development) and when I run a "charm proof" command on my charm, I get this error message [16:35] I: Includes template icon.svg file. [16:35] W: README includes line 6 of boilerplate README.ex [16:35] I: relation website has no hooks [16:35] I: missing recommended hook config-changed [16:35] ERROR subprocess encountered error code 100 [16:36] I also ran this on a colleague's charm and got different warning messages, but also the same error code 100. [16:36] Is there a FAQ for error codes? I tried searching on google first, but no "100" hits in the first pages === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === scuttlemonkey is now known as scuttle|afk === scuttle|afk is now known as scuttlemonkey === scuttlemonkey is now known as scuttle|afk === scuttle|afk is now known as scuttlemonkey === kadams54 is now known as kadams54-away [21:06] guys, the HACKING.txt from charms-tools is outdated. Could I submit a merge request with updated instructions based on the latest Makefile? [21:19] ayr-ton: sounds good to me [21:21] thumper: okay :) === kadams54-away is now known as kadams54 [22:40] hello, I have armhf board with debian on it, and ubuntu trusty inside an lxc on it, and I run juju on that ubuntu. and when I try to bootstrap with local env (meaning that would create nested lxc containers) - that doesn't work (lxc starts but agent fails to start) [22:40] how do I fix this and make agent start up properly? [22:41] I'm trying to deploy juju-gui [22:42] http://code.re/7oB [22:42] http://code.re/7oC [22:42] http://code.re/7oD [22:44] http://code.re/7oE === kadams54 is now known as kadams54-away