[00:19] hi!, still remains this bug http://askubuntu.com/questions/602527/provide-data-doesnt-send-data-when-required-keys-not-satisfied-juju-charm-usin in services framework [00:19] ?? === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === natefinch-afk is now known as natefinch === axw_ is now known as axw === natefinch is now known as natefinch-afk === natefinch-afk is now known as natefinch === axw_ is now known as axw === urulama is now known as urulama|lunch === urulama|lunch is now known as urulama [11:38] evilnickveitch, ping? [11:39] mattyw, pong [11:39] what's up? [11:39] evilnickveitch, still thinking about the documentation thing.... It occurred to me that unit numbering is going to go through a slight change in juju 1.25 [11:40] evilnickveitch, and I wonder where I should put the docs [11:40] mattyw, oh goodie [11:41] evilnickveitch, basically unit numbering will be like machine numbering - it will always go up, and be unique. so if you have mongo/0 and mongo/1 then you destroy a unit and add a new one instead of still having mongo/0 and mongo/1 you'll have mongo/0 and mongo/2 [11:41] mattyw - hmmm, okay. That doesn't sound too complex [11:42] however, the best thing to do is to create a new page [11:42] and put all the stuff in that [11:42] PR it to the docs [11:42] evilnickveitch, will do [11:42] and then it will be in the dev docs and we can work out which bits need to go where when it lands [11:43] when filing bug report at launchpad i got: "sstmp" does not exist in Juju Charms Collection; there is no doubt that charm exits https://code.launchpad.net/charms/trusty/+source/sstmp [11:44] what shall i do? play that im dummy and say that i don't know package name? [11:45] mattyw - I forwarded you a mail which may help [11:46] uf [11:46] Sorry, something just went wrong in Launchpad. We’ve recorded what happened, and we’ll fix it as soon as possible. Apologies for the inconvenience. (Error ID: OOPS-6b0d7c3b791acfb2e06289ac42f97c68) [11:49] evilnickveitch, much obliged [11:50] and to be even more convenient ssmtp charm does not exist in github mirror :) [11:50] schkovich, I'd probably just report a bug here https://bugs.launchpad.net/charms [11:51] mattyw: i already tried that [11:53] mattyw: if i answer to In what package did you find this bug? ssmtp in response i am getting error that such package does not exists; if i opt to play dummy and say that i don't know i am getting Error ID: OOPS-6b0d7c3b791acfb2e06289ac42f97c68 [11:56] schkovich, sounds like you're doing everything right. can you pastebin the details you want to enter and I'll see if I can get it in. Otherwise I'll ping a charmer when some are around [12:00] mattyw: thanks! http://pastebin.com/5XmrqX7N [12:03] schkovich, https://bugs.launchpad.net/charms/+bug/1457860 [12:03] Bug #1457860: sstmp charm: Configuration cannot be changed/updated neither on installing nor later on using GUI/CLI. [12:03] schkovich, if you can post a comment on the bug just so we can remember it's you we need to talk to :) [12:07] mattyw: done. thanks. coffee break. :) === rogpeppe1 is now known as rogpeppe [12:41] I've just tried the demo web UI of Juju, but I can't seem to find Actions in the UI (for example, mysql backup). Are they exposed in the UI? [13:04] hi guys [13:06] I'm not able to deploy local bundle on juju... I got this error: "No charm metadata @ %s', 'trusty/e-commerce/metadata.yaml". can someone help me? [13:08] obviously the metadata is in the e-commerce's directory! [13:09] turicasto: most likely either you don't have metadata.yaml, it is misconfigured or you submitted the wrong path to the local charmstore [13:10] turicasto: command should look like juju deploy --repository=opt local:trusty/puppet --to 4 where opt is the root of the local charmstore [13:11] schkovich, for the bundle too? [13:12] turicasto: sorry, i don't know, i guess that the command should be similar [13:13] schkovich, ok thank you! [13:14] turicasto: no mention. i just didn't evolve to using bundles yet ;) [13:15] turicasto: pastebin the cmd and output you're seeing [13:20] schkovich, make sure you're subscribed to that bug so you get updates about it [13:21] mattyw: i already did it :) [13:22] schkovich, awesome, thanks [13:25] mattyw: thank you Matthew. :) i truly don't get why i was not able to file the bug report. never happened to me in almost 10 years since im on launchpad :( [13:27] schkovich, np - it's happened to me once or twice before [13:44] skay: did you notice the python-django charm is geting some additional love? :) [13:44] skay: https://code.launchpad.net/~thumper/charms/trusty/python-django/support-1.7/+merge/259889 [13:45] lazyPower: I did not! I'm subscribed to teh repo but my emails got lost in all my other launchpad emails [13:45] i know those feels skay [13:45] lazyPower: I am happy! [13:45] Thumper and I had a quick hangout lastnight over the future of python-django [13:45] it looks like he's goign to start paving a path forward for co-maintainership [13:45] cool === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 [13:46] hopefully I will have time freed up so I can start making merge requests with some suggestions [14:38] guys juju-quickstart is the only way to deploy bundle? [14:42] turicasto - not the only way, you can use the juju-gui [14:42] or juju deployer [14:43] evilnickveitch, there is a way to specify the local repository of my bundle? [14:43] if you are using the gui, you can just drag and drop it [14:45] evilnickveitch, via command line? [14:46] turicasto, if you don't want to use the gui OR quickstart then you will need to use juju-deployer [14:46] and yes, you can specify a local bundle [14:47] evilnickveitch, how? [14:47] http://pythonhosted.org/juju-deployer/config.html [14:47] ^ docs for deployer - i can't say if they are up to date, we are phasing out the use of deployer [14:48] evilnickveitch, because I have problem with the deploy of my bundle, [14:48] 'juju help deployer' would probably give better usage info [14:49] evilnickveitch, from gui i got this error: "error occurred while deploying the bundle: ('No charm metadata @ %s', 'trusty/e-commerce/metadata.yaml')" [14:50] turicasto, is this a charm/bundle you have written yourself [14:50] ? [14:50] evilnickveitch, yep [14:50] do you know if the *charm* itself works? [14:50] yeah it works [14:51] how did you make the bundle? [14:57] Hey lazyPower http://juju-ci.vapour.ws:8080/job/charm-bundle-test-lxc/67/console [14:57] This flannel-docker test passed on LXC? [14:57] wat [14:57] welp, nothing to do here [14:57] I don't know what is going on [14:58] * lazyPower whistles and wanders off [15:15] howdy... I am running into issues bootstrapping vivid nodes via maas and manual using juju 1.23.2-trusty-amd64. Maas without juju is able to install nodes that I can log in to, but after bootstrapping the same node with juju I can not ping the instance. [15:17] If I deploy the node via maas, setup a manual environment specifying the node, and then attempt to bootstrap the node it appears to work, but jrunning uju status on the environment hangs [15:18] nodtkn: juju status is reaching out to the unit on port 17070 - a couple things may have happened that would cause status to hang... 1 being that port is blocked, or the jujud service came online and tanked shortly thereafter [15:18] Is anyone having good experiences with a stable release of juju and vivid on a node. [15:19] nodtkn: admittely we typically stick with LTS releases on juju - but i know there is some work in teh beta repository wrt vivid and the new systemd init components. [15:19] i'm not positive those made it into 1.23.3 [15:19] the fact its having an issue standing up w/ the 'maas provider' on juju, it leads me to beleive thats not the case. [15:20] natefinch: i know you had run into some differences w/ juju and vivid - can you confirm that the 1.24 branch has the fixes in place w/ teh systemd init stuff? [15:20] nodtkn: if you can get into one of the borked nodes, look at /var/log/cloud-init-output.log and /var/log/juju/* [15:20] it may be an init system issue that you need a newer juju for the fix [15:21] lazyPower: The last time I attempted that it did not end well... Do you have a link for how to do this? [15:21] nodtkn: how to upgrade juju to the currently released -devel branch? [15:21] nodtkn: for instance bug 1452511 [15:22] Bug #1452511: jujud does not restart after upgrade-juju on systemd hosts

[15:22] and various other systemd related issues [15:23] lazyPower: How to set a password via preseed or how to connect to the node that failed to bootstrap. [15:23] mgz: thanks [15:23] ah im not that fluent with the specifics of the preseed process, i can however point you to #maas and ping for assistance in doing so [15:24] let me see if anyone from teh maas team is around atm to help troubleshoot nodtkn [15:26] nodtkn: as soon as i get a response i'll ping again - i have an open request w/ the maas team that we need help. [15:28] lazyPower: thanks. I have node that I first deployed via maas, then attempted to bootstrap directly. The cloud-init.log is not very interesting [15:28] nodtkn: i'm fairly certain the issues you're running into are due to running on stable, there's a ton of fixes inc. w/ the 1.24 release and beyond that are specific to vivid/systemd bugs [15:31] lazyPower: ok... I could have sworn I got further about a month ago on older cloud init images. I must of not actually been running on vivid. [15:32] really sorry you ran into this snag, but the mailing list has been fairly active the last month with discussion on this very subject [15:35] lazyPower: I am here [15:36] nodtkn: can you relay to blake_r the situation when attempting to connect to the maas unit post juju bootstrap? he can lend a hand in getting you into the unit so we can collect the logs [15:41] blake: thanks [15:44] blake: If I deploy the node via maas I can ping and ssh into the node. If release and bootstrap the node via juju I can not ping the node or ssh into it. What can I do via preseeds or modifying the cloud image to set the root password so that I can log into the node and look at the juju logs? [15:53] blake_r: ^ === urulama is now known as urulama__ [15:57] blake_r: I sorry I referred to use as blake please read my comments above [15:58] sorry [15:58] taking a look now [15:59] nodtkn: hm that is wierd, so deploying with maas works fine [15:59] nodtkn: can you view the console of the machine? does cloudinit show an error when running userdata [15:59] nodtkn: because the only difference is the userdata from MAAS point of view [16:00] nodtkn: if you want to set the root passwd you can do that with some preseed modification [16:01] blake_r it hangs* during sending of the ssh keys. If I hit enter the console continues to scroll to a log in prompt. So hangs is not the best description...lingers? [16:03] blake_r: run_parts has claimed to finish restaring the juju bridge interface. and the sending of ssh keys has started. I hit enter a few times and it will show the login prompt [16:04] nodtkn: okay so the only real way to get in is using password as it might be a bridge issue [16:04] nodtkn: give me a moment and I will try to come up with a preseed modification to do what you want [16:08] blake_r: thank you === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away [19:38] In the new bundle format are all container technologies treated with the same synax? ie) lxc:new kvm:new [21:17] blake_r: ping === JoshStrobl is now known as JoshStrobl-Nukun [23:21] lazyPower: ping [23:23] hmm later for lazyPower than I thought