[00:00] catbus1: you can find it by going to jujucharms.com/charm and it'll be listed on the page, https://jujucharms.com/wordpress/ says it's revision 10 [00:00] catbus1: you can use the cs:charm-rev, not sure if that actually works, but if it did work before it should still going forward [00:01] catbus1: I lied, the revision is in the URL on the right [00:01] so that version of WordPress is version 1 [00:01] ok. thanks, marcoceppi [00:11] marcoceppi: do you know if there is any documentation about how to mirror the juju charm store? I have a friend in Asia who is in need of a local mirror as he sometimes has problems getting charms. [00:12] catbus1: you can run an old script, "charm getall", which will download all of the precise charms. It'll be updated in the next few weeks so that you can use it with other series [00:12] once you have it downloaded, you just need to put it in a local JUJU_REPOSITORY [00:12] and then use it as a local deploy [00:15] That would require building a local repo for each juju. [00:16] my friend was thinking about hosting a mirror. [00:16] so anyone can access it, not just for his juju instance. [00:34] am I right that for every deployed charms, Juju caches them in the ~/.juju/charmcache folder? [01:11] lazyPower: hye lazyPower are u available at this time... === rcj is now known as Guest82790 === ericsnow_ is now known as ericsnow === Guest82790 is now known as rcj === rcj is now known as Guest563 [04:56] halcyon: o/ [04:56] halcyon: i'm a bit late, had some things to wrap up downstairs. Hopefully you're still around though === kadams54 is now known as kadams54-away === Murali_ is now known as Murali [05:59] Hi gnuoy: [06:00] I would like to do some changes to nova-compute charm and would like to test locally [06:01] And I want deploy using juju-deployer. is it possible ? [06:04] lazyPower: hye [06:04] lazyPower: I'm here [06:05] lazyPower: currently I destroy back my environment [06:09] halcyon: ey [06:10] so you're having problems with the local provider since you edited your bridge settings, as i understand it? [06:11] lazyPower: yes, as i change according ur tutorial but now I'm doing it once again and let u see the log ..is it okay?? [06:12] lazyPower: i hope u can guide me from there [06:13] halcyon: There's a bit i'll need to know - like whats your current CIDR for your network [06:13] halcyon: and can you pastebin the edits you made? [06:14] lazyPower: sure can paste it here but [06:14] lazyPower: sorry what is CIDR [06:14] halcyon: please pastebin it, then handoff a link. [06:15] halcyon: the ip range of your LAN in this case - 10.0.10.0/24 [06:15] ^ as an example [06:15] lazyPower: how should I handoff a link? [06:15] halcyon: paste.ubuntu.com - paste the contents of the file there and give me the link to the paste. [06:16] lazyPower: ok [06:18] lazyPower: current CIDR [06:20] lazyPower: 10.11.66/24 [06:21] ok, in /etc/default/lxc-net - is your LXC_NETWORK set to 10.11.66.0/24? [06:22] halcyon: better yet, just pastebin the contents of /etc/default/lxc-net for me :) [06:22] OK [06:22] yes set to that CIDR [06:28] lazyPower: this is my file http://paste.ubuntu.com/10535681/ [06:31] halcyon: ok, this looks good so far. and if you run `ifconfig br0` you should see the ip address that you defined 10.11.66.131 [06:31] lazyPower: are u still here right? [06:31] lazyPower: if I run ifconfig it give me another ip for br0 [06:32] halcyon: ok, this probably means your parent interface is set to DHCP - the post assumes the host IP will be static. there's issue #! [06:33] *#1 [06:34] lazyPower: ok wait I set the file in etc/network/interfaces like this...I pastebin to show u... [06:37] lazyPower: http://paste.ubuntu.com/10535757/ [06:39] hmm... ok, can you pastebin the local provider snippet from ~/.juju/environments.yaml and the contents of /etc/lxc/default.conf? [06:41] lazyPower: ok...but the default.conf I move to file name lxc.conf [06:43] halcyon: there are 2 files in /etc/lxc - it shoudl be default.conf and lxc-usernet [06:44] lazyPower: yes 2 file at first but still now the default.conf is there [06:45] lazyPower: now this is the file ...environments.yaml [06:45] http://paste.ubuntu.com/10535822/ [06:45] lazyPower: that is the local part [06:48] lazyPower: this is the default.conf file http://paste.ubuntu.com/10535839/ [06:49] that lxc-clone line looks like its missing an indent - possibly a paste artifact? [06:50] lazyPower: paste artifact?? [06:50] what is that? [06:50] side-effect when pasting? [06:50] is the lxc-clone line in ~/.juju/environments.yaml under the local provider indented to match the rest of the provider declaration? [06:50] yes yes it should be same like the rest [06:51] yes side effect from pasting [06:51] ok. since you moved the file default.conf back into place - reboot and try bootstrapping and deploying cs:trusty/ubuntu [06:53] lazyPower: ok [06:53] lazyPower: u mean juju deploy cs:trusty/ubuntu... [06:53] is the command like that [06:54] yep [06:57] lazyPower: u hv any idea when i run command juju status but it take some time to execute the output [06:58] halcyon: does it eventually come back or is it hung? [06:58] lazyPower: is something like hung [06:59] not sure what would cause that to be slow :( [06:59] should I reboot once again [07:00] if you've rebooted once - you shouldn't have to reboot again [07:01] you've bootstrapped and run juju deploy cs:trusty/ubuntu right? [07:02] I did bootstrap but not yet run juju deploy cs:trusty/ubuntu [07:03] halcyon: in ~/.juju/environments/local.jenv - you should have a couple lines denoting your state server. for example, mine looks like: http://paste.ubuntu.com/10535978/ [07:04] do you see your br0 ip address listed as one of the state server entries? [07:06] lazyPower: now my status come out and again lazyPower i see the machine id 1 the status is container failed to start [07:06] halcyon: ok - run sudo lxc-ls --fancy and pastebin the output for me [07:07] ok [07:08] lazyPower: suprisingly when i run lxc-ls --fancy no output [07:08] means i hv no container [07:08] hmmmm [07:08] halcyon: did you juju deploy trusty/ubuntu? [07:08] halcyon: the state server doesn't create a container - its actually running on your desktop. [07:09] i deploy [07:09] juju deploy cs:trusty/ubuntu [07:09] ok, so it failed before it even greated the template container... weird [07:09] *created [07:10] oh u mean when i deploy trusty/ubuntu then it should create that service as if in my desktop itself?? [07:10] is it right?? [07:11] halcyon: ps aux | grep wget [07:11] do you see wget fetching a cloud archive image? [07:12] not see any image [07:12] let me pastebin u my local.jenv file [07:12] ok, i'm thinking there's something else going wrong here that juju is kind of hiding due to the fact it didnt even attempt to create teh container. [07:13] halcyon: lets try this, we're going to create a lxc container manually [07:13] ok [07:13] sudo lxc-create -t download -n test -d ubuntu -r trusty -a amd64 [07:14] lazyPower: done create lxc container [07:14] halcyon: ok now run sudo lxc-start -n test -d [07:15] halcyon: you should see the container running and with an ip address when you sudo lxc-ls --fancy now [07:16] I got this message lxc_container: lxc-start.c main: 296 Executing '/sbin/init' with no configuration file may crash the host [07:16] wat [07:16] hang on [07:17] it think the container doesnt start [07:18] halcyon: can you try starting the container again with -o lxc.log -l TRACE [07:18] ok [07:18] -o lxc.log -l TRACE [07:19] the command like above one lazyPower i write ya [07:19] halcyon: sudo lxc-start -n test -d -o lxc.log -l TRACE [07:19] OK [07:20] lazyPower: I got the same error message :( [07:20] halcyon: anything in lxc.log? [07:20] or did it bail before it even attempted to log [07:21] lazyPower: http://paste.ubuntu.com/10536067/ [07:22] lazyPower: the log file for lxc [07:23] halcyon: run "dpkg -l lxc" - are you running 1.0.7-0ubuntu0 ? [07:24] u ask about juju version?? [07:25] halcyon: no, just wanting to confirm you're on the latest stable release of lxc [07:25] yes 1.0.7-0ubuntu0.1 [07:26] halcyon: is it possible to set your ethernet to a static ip? even if just temporarily? [07:26] yes [07:26] before I edit br0 the ip is static one [07:27] ok, can you set it to the 10.11.66.131 address [07:27] u mean I change eth0 to 10.11.66.131 to static?? [07:29] halcyon: no, we want to keep teh bridge around [07:29] halcyon: so we need to leave teh physical interface to manual, let the br0 be auto, and make it static [07:30] halcyon: like so: http://paste.ubuntu.com/10536134/ [07:30] ok u mean I change this file /etc/network/interfaces [07:30] halcyon: correct [07:31] lazyPower: ok understood but before i do that i destroy environment right? [07:32] halcyon: yeah, we'll use the lxc-container we created as our litmus test [07:39] lazyPower: ok done edit that file [07:40] lazyPower: [07:40] halcyon: did you bring the interface up/down so the static ip is in effect? [07:40] so now i bootstrap again ya [07:40] not yet [07:41] lazyPower: now when i run ifconfig [07:42] my ip for br0 is 10.11.66.131 [07:42] i just reboot machine after i edit that file just now [07:42] ok, with the static ip in place, i think that will fix the problem [07:42] if not, i'm out of ideas :( [07:44] lazyPower: huhu.. [07:45] lazyPower: wait ya.. [07:50] lazyPower: ok i cant ping top google.com [07:50] halcyon: can you reach your gateway? [07:50] yes... [07:51] and i can ping 8.8.8.8 [07:51] as well [07:53] hmmm kind of weird..I am sorry lazyPower coz troublesome for u...;( [07:54] halcyon: with all the issues that are cropping up and me not having a good answer it sounds like its time to start reverting back to the shipped config [07:54] halcyon: do you need me to pastebin any of those configs for you? [07:56] ok ...with all those issue I would like to again revert back to the original state where the host itself is clean [07:56] ok sure u can pastebin those configs for me [07:57] i myself do not hv much knowledge about lxc [07:57] - /etc/lxc/default.conf - http://paste.ubuntu.com/10536369/ [07:57] lazyPower: do u think those problems regarding the container is it ?? [07:58] - /etc/default/lxc-net - http://paste.ubuntu.com/10536382/ [07:58] YOu will need to change the bridge to lxcbr0 in ~/.juju/environments.yaml [07:59] by removing br0 from /etc/network/interfaces, and putting configuration back under your ethernet device, you should be g2g after a reboot. [07:59] (dont forget to set your ethernet device back to auto as well) [08:00] ok understood [08:00] thank you lazyPower [08:00] halcyon: np, sorry i couldn't get you up and running :( [08:01] bloodearnest: i took a look - are you referring to this MP? https://code.launchpad.net/~bloodearnest/charms/trusty/rabbitmq-server/add-nagios-service-groups/+merge/251790 [08:01] :( is okay [08:08] lazyPower: hope can communicate again! [08:08] halcyon: good luck, i'm about to head out and get some sleep. its 3am here :) [08:11] ok :) [09:46] lazyPower, I found the issue. The linting uses the venv, which uses --system-site-packages, and we don't have pinned requirements, which means I was using an older version of flake8 that was in the system [09:46] I will pin the reqs, should fix it [09:47] although the amulet failures are a bit frustrating, as they are failing in tip, not because of my changes === dooferlad_ is now known as dooferlad === frankban__ is now known as frankban [12:03] Is there a limit to how large a file that can be included in a charm? I've got a 95MB tarball in the directory I'm deploying from that isn't ending up on the deployed units [12:12] Hmm.... juju is fine. Looks like juju-deployer is what is griefing me. [12:37] marcoceppi: Looking at http://marcoceppi.com/2014/06/deploying-openstack-with-just-two-machines/ I have a question - working on Kimchi charm - kimchi being a monitor for all VMs on system, I believe the idea is to deploy it onto 0 - that sounds possible for Manual or OpenStack env , but what would I configure for a Local env ? [12:44] thierry-ibm: you don't [12:44] stub: details? [12:44] thierry-ibm: local env machine 0 is not a container, its the host, its 'special' [12:44] no units get deployed to it [13:16] hazmat: so is there a purpose to deploy this charm in a local env ? may be to verify it starts properly but then I think I will have to redirect nginx associated port to a port on main host - what do you think ? [13:16] thierry-ibm: if your on the host you can directly navigate to the container ip address .. if your using a vm, you'll need to port forward [13:17] s/on vm/off the host [13:17] hazmat: understand that - thanks === Guest563 is now known as rcj === xerxas_ is now known as xerxas [15:29] I like having a local directory mounted in a container so that I can do work with my repo locally. vaguely, is it easy to make a post-deploy(?) script in a mojo spec that would mount the directory for me so that I could do this? [15:29] currently I use a script that does not involve juju that sets up a container and does hte mounting work for me [15:29] what I'd love is to have a stage that would do everything for me [15:30] another thing that would be nice is to be able to deploy some units with lxc and some to a remote openstack environment [15:30] is that possible? [15:31] roadmr: it would be like if I could have juju deploy a local development environment for me that behaves like your develop-in-lxc.py script [15:33] skay: I don't think you can merge local and lxc :/ at least not local lxc :( [15:34] roadmr: it would be cool if we could have a situation where we used swift externally but everything else locally [15:34] skay: that'd be so cool :) Usually the lxc mount-a-local-dir config has to be done *before* booting the container, it's a modification of the fstab so it remains to be seen if it can be done while the lxc is already booted [15:34] roadmr: so maybe one of the predeploy scripts in a mojo spec [15:34] I need to beef up my lxc-fu [15:36] skay: I don't see stgraber to ask him for confirmation, but given how the host resource looks like in the mtab (it's not even in fstab) I'd say it can't be done in a running container :/ [15:37] skay: plus it wouldn't make sense security-wise; if a container were able to arbitrarily mount resources in the host system it'd be a huge hole [15:37] skay: so prior to boot we tell lxc which host resources to expose to the container, and once it's booted they can't be altered === urulama is now known as urulama__ === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 [20:12] hey wwitzel3 === urulama__ is now known as urulama === 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 === kadams54 is now known as kadams54-away [21:39] anyone here thinking about going to dockercon? doing the hackathon? [22:00] kwmonroe: hey the documentation for the charm you're looking at is actually here: http://chuckbutler.github.io/docker-charm/user/getting-started.html [22:01] kwmonroe, we'll fix the readme, thanks for spotting that [22:08] np whit. thanks for being nice to me. lazyPower could learn a thing or two from you. [22:12] can anyone tell me a story about logstash-indexer on trusty? [22:28] lazyPower: What does this mean? ERROR: Unable to find an inventory file, specify one with -i ? [22:29] cory_fu: that sounds like an ansible error - can you pastebin me the output of what threw that/ [22:29] jrwren: once upon a time, in a log entry far far away ;-) [22:29] make lint on the docker charm [22:29] I can't get it to lint properly [22:29] lazyPower: NB: Using charmbox to review [22:29] cory_fu: its barfing on not having an inventory [22:29] ok, charmbox wasn't a thing when that portion was written, its isolated to a venv [22:29] cory_fu: typically that means its not operating in local mode [22:30] lazyPower: The venv construction in the charm is broken [22:36] well.. thats fun :( [22:37] lazyPower: Ok, fixed the inventory issue, since it was my fault, but getting "provided hosts list is empty" now [22:37] Any idea? [22:38] cory_fu: 1 sec let me double check the makefile [22:38] lazyPower: I've got a list of comments regarding that Makefile. ;) [22:38] cory_fu: i think, i'm not positive - but i think what the issue is with the line about --syntax-check @ansible-playbook --syntax-check playbooks/site.yaml [22:39] my in dev copy actually has that removed, and lint passes fine with just ansible-lint plugin [22:39] lazyPower: Strange thing is that lint worked fine for asanjar, so I'm not sure why it's failing for me [22:40] cory_fu: it might be due to a missing dependency, mismatch in ansible versions [22:40] I didn't have ansible installed already, though, so I did an apt-get install ansible [22:40] those are teh 2 things that come to mind [22:40] Ok, I removed that line, and now it says that .venv/bin/ansible-lint not found [22:41] I'll just note these things on the review and move on [22:42] cory_fu: weird, the make virtualenv target should have installed all those deps [22:42] It doesn't. The venv target doesn't install anything ansible related [22:42] cory_fu: but - http://bazaar.launchpad.net/~lazypower/charms/trusty/docker/trunk/view/head:/Makefile [22:43] it says it does [22:43] * lazyPower is cornfused [22:43] Oh [22:43] I was using the github version [22:43] I thought that was the more up-to-date one [22:43] And for some reason I already had it checked out [22:46] cory_fu: looks like i vendored something without getting it merged into master [22:46] #fwp with my vendoring process [22:46] Well, the LP branch is what the review points to, so I should be using that === kadams54-away is now known as kadams54 === ericsnow is now known as ericsnow_on_an_e === ericsnow_on_an_e is now known as ericsnow_afk === kadams54 is now known as kadams54-away === ericsnow_afk is now known as ericsnow