[02:27] <davecheney> marcoceppi: can I ask what the plan for testing trusty charms is ?
[02:27] <davecheney> i have a machine that can only use trusty (don't ask)
[02:28] <davecheney> so to test the local provider I need a good selection of trusty charms
[02:28] <davecheney> currently i'm doing this with local forks of the charms
[02:43] <davecheney> lazyPower: charm-helper-sh
[02:43] <davecheney> which repo/ppa defines that package ?
[07:53] <vila> it seems I encounter issues to bootstrap with 1.17.5 https://pastebin.canonical.com/106545/
[07:54] <vila> 1.17.5 on saucy (my desktop) that is, for the exact same cloud/env I can boostrap from my laptop (trusty) with 1.17.4
[08:02] <vila> jam: around ? The above is critical for us we have a demo coming like... tomorrow
[08:04] <jam> vila: investigating
[08:05] <vila> jam: thanks !
[08:06] <vila> jam: it may just be a matter of uploading 1.17.5 to whatever is behind tools-metadata-url (shared for the team) but I've never did that before (and hope I was write access there, failing that I'll setup mine)
[08:06] <vila> s/was/have/
[08:07] <vila> jam: doing 'juju sync-tools --public --local-dir=. --all --debug -e hp' as a first step
[08:07] <vila> done
[08:08] <vila> jam:   concretely  tools-metadata-url: https://region-a.geo-1.objects.hpcloudsvc.com/v1/11289530460295/tools
[08:08] <jam> vila: http://streams.canonical.com/juju/tools/releases/ doesn't have 1.17.5
[08:08] <jam> I think we need to poke Curtis (sinzui)
[08:08] <vila> jam: he's sleeping right now right ?
[08:08] <jam> yes
[08:08] <jam> he's UTC-6 IIRC
[08:08] <jam> Curtis and Ben Howard manage streams.canonical.com
[08:08] <vila> jam: won't do then :-/ I need something that worked yesterday :-(
[08:09] <vila> jam: is there a workaround ?
[08:09] <vila> jam: it's for hpcloud, my understanding is that we already had to setup something ourselves
[08:09] <jam> vila: you can try "juju bootstrap --upload-tools" which generally requires you to only deploy to a matching series, but should work
[08:09] <jam> vila: there *are* no official juju tools for 1.17.5 because they haven't been built yet
[08:10] <jam> there is only the client
[08:10] <jam> which, we shouldn't be publishing to Trusty until we have tools for it to use
[08:10] <jam> but...
[08:10] <vila> jam: ouch, so saucy is busted for now is what you're saying ?
[08:10] <vila> jam: first things first. Hello there and thanks for helping ;)
[08:10] <jam> vila: it shouldn't be in Saucy. Are you using a PPA there?
[08:11] <jam> vila: hi
[08:11] <vila>  *** 1.17.5-0ubuntu1~ubuntu13.10.1~juju1 0
[08:11] <vila>         500 http://ppa.launchpad.net/juju/devel/ubuntu/ saucy/main amd64 Package
[08:11] <jam> vila: "juju/devel" is in there
[08:11] <vila> jam: is there a way to use a different ppa ?
[08:11] <vila> meh, is there a different ppa I can use :)
[08:12] <jam> vila: unfortunately, we probably don't publish 1.17.4 in another place, and ppa's don't keep a backlog :(
[08:13] <vila> jam: ok, back to --upload-tools, what will that do ? Override tools-metadata-url for the current env ?
[08:13] <vila> jam: any risk I break something else by trying ?
[08:14] <jam> vila: that will look on your local system for a 'jujud' binary, package it up into a tarball, and upload it to the private bucket for your environment
[08:14] <vila> jam: cool, sounds like it, trying
[08:15] <vila> jam: we deploy against precise though... but worth a try
[08:15] <jam> vila: so we fake the tools against the LTS
[08:15] <jam> and for most purposes it works
[08:15] <vila> 2014-03-17 08:15:13 INFO juju.environs.sync sync.go:176 downloaded tools/releases/juju-1.17.5.1-precise-amd64.tgz (5051kB), uploading
[08:15] <vila> sounds encouraging
[08:16] <jam> the binaries are statically linked, so usually the Series would work
[08:16] <jam> (is unimportant?)
[08:16] <jam> vila: you won't be able to do stuff like deploy i386 machines, etc.
[08:16] <vila> jam: ack, only amd64 I'm pretty sure
[08:17] <vila> jam: I'm not taking you out of bed I hope (lost track of your TZ ;)
[08:17] <jam> vila: I'm UTC+4 now. Right at noon time
[08:17] <jam> I need to go make lunch, but otherwise no problem
[08:18] <vila> jam: ack, thanks, I'll let you know how it goes
[08:19] <vila> jam: just to summarize, 1.17.5 shouldn't have entered the ppa ?
[08:19] <vila> jam: or did it enter as part of a bigger work that got interrupted ?
[08:20] <jam> vila: well, we need to release, so 1.17.5 should be added to the ppa, but IMO it shouldn't be added until we've published the corresponding tools to streams.canonical.com/
[08:21] <vila> jam: ack, so wip and I was unfortunate to upgrade at the wrong time ?
[08:21] <jam> vila: well, I think our process is wrong that we exposed you to it, but yes
[08:22] <jam> hmm.... looks like we did publish tools to s3 http://juju-dist.s3.amazonaws.com/
[08:22] <vila> jam: ack, we need the devel version, it's unfortunate that we run into this at this time but... if --upload-tools works I'm good and will warn others to *avoid* upgrading
[08:23] <vila> at least until we get confirmation it's safe to do
[08:23] <jam> ah, nm, those are "testing" one
[08:23] <jam> ones
[08:26] <jam> vila: is it working now?
[08:26] <vila> jam: oncee the node 0 is bootstrapped I should be fine right ? But I'll need to use --upload-tools each time I destroy the env correct ?
[08:27] <vila> jam: still uploading, network between me and hp is... slow
[08:27] <jam> vila: it should use the same tools for everything you deploy, and yes if you destroy the environment, the tools should get removed at the same tiem.
[08:27] <vila> ack
[08:27] <jam> vila: I thought you had good upload in Strausberg
[08:28] <vila> jam: thought so too until I start using hp ;)
[08:31] <vila> urgh timeout :-/
[08:31] <vila> https://pastebin.canonical.com/106555/
[08:32] <vila> re-trying
[08:44] <vila> jam: re-trying went further, bootstrapped, pfew trying to deploy now
[09:04] <vila> jam: deployment completed, looks like I'm back in business, thanks for the tip and keep me posted about 1.17.5 progress !
[09:05] <jam> vila: thanks for being a guinea pig and noticing we broke stuff
[09:05] <jam> looks like we already uploaded tools for ec2 and canonistack, I don't quite understand why we don't have it for HP and streams.canonical.com
[10:12] <zchander> ping marcoceppi
[12:20] <marcoceppi> zchander: pong
[12:21] <bodie_> boink
[12:55] <rick_h_> jcastro: blog posts up at http://blog.mitechie.com/2014/03/17/juju-quickstart-and-the-power-of-bundles/ and http://jujugui.wordpress.com/2014/03/17/video-walkthrough-of-juju-quickstart/
[12:55] <rick_h_> jcastro: want to give a final look before I put them up somewhere?
[13:22] <jamespage> any charm-helpers around? need an ack on https://code.launchpad.net/~james-page/charm-helpers/ceph-fix-zap/+merge/209539
[13:29] <mbruzek> Hi james-page  I can take a look
[13:32] <jcastro> rick_h_, got it
[13:33] <rick_h_> jcastro: cool, I put up on my twitter/G+ for now. Will try to put on reddit/HN later if you think it's a good idea.
[13:40] <marcoceppi> jamespage: I'll merge in a few
[13:53] <jcastro> rick_h_, yeah I'll try to coordinate that
[13:54] <rick_h_> jcastro: rgr
[13:55] <bodie_> is there a virtualbox or vagrant image specifically for dev?
[13:56] <marcoceppi> bodie_: kind of
[13:57] <marcoceppi> bodie_: https://juju.ubuntu.com/docs/howto-vagrant-workflow.html
[13:58] <bodie_> ah yeah, I mean for core dev, with go and such
[13:58] <bodie_> I'm just gonna set one up I think
[13:58] <bodie_> been having a hell of a time getting 14.04 to work with this
[13:58] <marcoceppi> bodie_: ah, no, bit one would be nice
[14:00] <marcoceppi> bodie_: i have experience getting go set up, etc, feel free to ping me if you run in to issues
[14:00] <bodie_> ok, much appreciated
[14:19] <hazmat> bodie_, i thought the pinned version on gwacl resolved for you?
[14:19] <bodie_> yeah, I got that taken care of
[14:19] <bodie_> it's giving me problems with mongo now I think
[14:19] <bodie_> setting up a 12.04 box just to make sure
[14:26] <ghartmann> is there any way to control how a service is exposed in juju ?
[14:27] <ghartmann> mainly I have my lan on 192.168.0.1 network .. and the exposed juju service is on 10.0.3.x
[14:28] <marcoceppi> ghartmann: that's the local provider, not exactly ajuju problem
[14:28] <ghartmann> I wanted to use lxc
[14:28] <marcoceppi> ghartmann: you'll have to bridge your network
[14:28] <marcoceppi> with the lxc network
[14:28] <ghartmann> I see
[14:29] <ghartmann> I thought that you could expose and he would take care of iptables, etc
[14:30] <marcoceppi> ghartmann: exposing is something that works against the providers firewall, not at the network level
[14:30] <ghartmann> ok, makes sense now ! Thanks !
[14:42] <mbruzek> Hello #juju I am working on the review queue today and can not review my own request.  Can I  get a charmer to review https://code.launchpad.net/~mbruzek/charms/precise/rabbitmq-server/tests/+merge/202573
[15:15] <lazyPower> mbruzek: i'll take a look at that MP shortly
[15:15] <mbruzek> ack
[15:39] <mbruzek> Hello #juju.  The merge proposal https://code.launchpad.net/~hopem/charms/precise/rabbitmq-server/lp1281753/+merge/209320 has passed my +1 review, and needs a review by an official charmer
[15:40] <lazyPower> mbruzek: can you make cards for promulgation on the board?
[15:41] <mbruzek> yes will do
[16:14] <hazmat> rbasak, ping
[16:15] <rbasak> hazmat: pong. otp, but can chat mostly.
[16:18] <hazmat> rbasak, was looking at the jujuclient / test_jujuclient.. wrt to mir.. afaicr the underlying ws lib is py2... is that okay re main.. also the test_jujuclient wants an existing juju environment available.. i can write some additional tests that are mock based though
[16:18] <hazmat> rbasak, basically it expects the user to provide env vars to point to a test env and creds, and then runs against it
[16:19] <rbasak> hazmat: ah yes. I noticed that yesterday.
[16:19] <rbasak> hazmat: py2 is fine in main.
[16:19] <rbasak> (AFAIK)
[16:20] <rbasak> hazmat: mock tests would be nice. I was going to proceed with the MIR on the basis of what's there now, and see what the MIR approval team has to say about it.
[16:20] <hazmat> rbasak, okay.. i'll add in some mock tests then (will result in version incr)
[16:22] <rbasak> hazmat: I'm almost done with preparing MIR reports and fixing up this dependency stack to main quality now. I just have to disable some tests in websocket-client, since some expect echo.websocket.org or something to be reachable.
[16:22] <rbasak> hazmat: thanks!
[16:30] <whit> hi all!
[16:31] <marcoceppi> Hi whit o/
[16:31]  * whit is looking for antonio @ canonical but doesn't know his nick
[16:31] <bodie_> howdy
[16:31] <whit> Hi marcoceppi !
[16:31] <marcoceppi> whit: arosales is Antonio
[16:31] <whit> ah… duh.
[16:31] <lazyPower> Allo whit o/
[16:32]  * whit realizes he knew that… 
[16:32]  * whit is a little punch drunk after babysitting a friend's kids late in the night while they went and had a baby
[16:32] <whit> marcoceppi: danke!
[16:32] <arosales> whit: hello
[17:54] <arosales> lazyPower: congrats one your sucessfull application into ~charmers
[17:57] <lazyPower> Thanks arosales! Very proud to sport the title in my lower third.
[17:59] <arosales> lazyPower: :-)
[18:03] <sarnold> lazyPower: woo :) congratulations :)
[18:03] <lazyPower> sarnold: I owe it all to tiny chocolate donuts
[18:03] <lazyPower> \o/
[18:04] <sarnold> lazyPower: hooray for little chocolate donuts!
[18:04] <lazyPower> I've been waiting forever to use that quote...
[18:13] <lazyPower> hey marcoceppi, question for you. I've got an open pending MP that extends the config section of nagios. Should I write tests for teh current charm - as-is, or should i point to my branch, write tests for that pending the merge?
[18:13] <marcoceppi> lazyPower: you can make your branch a pre-requisite for your testing branch
[18:13] <lazyPower> ack
[18:13] <lazyPower> talking about in the MP itself right?
[18:13] <marcoceppi> yes
[18:13] <lazyPower> kk, i'll do that.
[18:13] <mbruzek> marcoceppi, How would one make a pre-req
[18:14] <marcoceppi> mbruzek: when you create the merge proposal you can specify a pre-requisite branch
[18:14] <mbruzek> Oh cool.
[18:38] <zchander> ping marcoceppi
[18:38] <marcoceppi> zchander: pong
[18:42] <zchander> I four that you have developed a gluster charm? Do you have any info how to use it properly? I keep getting a failed hook
[18:43] <zchander> flu == found
[18:49] <marcoceppi> zchander: it hasn't been updated in a long time
[18:50] <zchander> I noticed.... ;) I am looking for a nice/simple (read: easy) way to create a storage environment to use with owncloud
[18:52] <zchander> Mmmm, noticed I have a mysql charm which is dying for 6 hours now.....
[18:52] <zchander> Is there a way to force the charm to stop
[18:57] <marcoceppi> zchander: yes, juju terminate-machine --force <machine mysql is on>
[18:57] <marcoceppi> zchander: I would recommend ceph
[18:58] <marcoceppi> it's way more tested and in the charm store
[19:20] <zchander_> marcoceppi: I'm back again....
[19:21] <zchander_> marcoceppi: Just read back the irc log. I am going to try to terminate the charm
[19:22] <zchander_> marcoceppi: Seems there is no machine assigned to mysql (??)
[19:23] <marcoceppi> zchander: just do juju destroy-service mysql --force
[19:35] <zchander_> marcoceppi: error: flag provided but not defined: --force
[19:43] <themonk> is subordinate charm share same filesystem with master charm?
[19:52] <themonk> marcoceppi: is subordinate charm shares same filesystem with master charm?
[19:52] <marcoceppi> themonk: yes
[19:52] <themonk> marcoceppi: hi thanks :)
[20:24] <jose> marcoceppi, lazyPower: hey, any of you guys around? I'm having a problem with a nested if statement on the mailman charm
[20:29] <lazyPower> jose: surely
[20:29] <lazyPower> jose: link to branch or gist or pastebin?
[20:30] <jose> lazyPower: https://code.launchpad.net/~jose/charms/precise/mailman/trunk
[20:31] <jose> when I deploy it, I don't know what happens but says it's not stated, and exits with error code
[20:31] <lazyPower> jose: josewell, on line 70 your if syntax is incorrect..
[20:31] <lazyPower> and i suggest encasing your variables with quotes as well.
[20:32] <lazyPower> if [ "$currentpassword" != "$PASSWORD" ]; then #do stuff fi
[20:32] <jose> oh
[20:32] <lazyPower> you have the same thing ocurring up on line 64
[20:32] <jose> yeah
[20:33] <jose> after the then, should a 'do' go there?
[20:33] <lazyPower> are you going to be looping?
[20:34] <jose> nope
[20:34] <lazyPower> then i would abstain from using the do operator. Do's are typically used in conjunction with a while statement.
[20:34] <jose> got it :)
[20:34]  * lazyPower thumbs up
[20:34] <lazyPower> go forth and conquer my friend
[22:13] <jose> hey lazyPower, still having troubles with the password variable on the latest revision in config-changed
[22:13] <jose> any clues?
[22:14] <lazyPower> jose: whats the error output?
[22:14] <jose> lazyPower: something like parameter not set
[22:15] <jose> but the if is testing if the parameter is null
[22:20] <lazyPower> jose: ok give me a few i'm not infront of my PC to look over the code
[22:20] <jose> np
[22:27] <lazyPower> jose: missing a ; on line 69
[22:27] <lazyPower> the then statement should be inline with the if as well, but thats coding convention
[22:27] <jose> urgh
[22:28] <jose> it's always because of a sign
[22:28] <jose> last time me code was wrong because of a - was placed instead of a _
[22:28] <lazyPower> jose: happens to the best of us, best thing you can do when encountering issues like this is take a step back and start unwinding complexity one layer at a time
[22:28] <jose> will do next time :)
[22:29] <jose> now let's hope it's good for deployment, trying
[22:30] <jose> marcoceppi: hey, wanna host that livestream at ubuntuonair?
[22:31] <jose> I just thought of creating a series of challenges to be done live :P
[22:31] <marcoceppi> jose: well, I'm doing it through twitch, since I can impose my camera on my feed
[22:31] <jose> that's right
[22:31] <jose> good enough
[22:31] <marcoceppi> jose: this is more like "I think I can do this in under and hour"
[22:31] <marcoceppi> but I'd love to do some charming based challenges in the future
[22:32] <jose> link me when it starts, I'd like to see it
[22:48] <jose> lazyPower: well, it was all when defining the variables: I wrote PASSSWORD instead of PASSWORD
[22:48] <lazyPower> Right on, i didn't notice that looking over it on my phone. Good catch.
[23:06] <jose> lazyPower, marcoceppi: when you're on the mood for some charm reviewing, the mailman one is waiting for you
[23:06] <marcoceppi> jose: mbruzek is the review on duty this week
[23:06] <lazyPower> jose: outlook is for tomorrow - its getting late on the east coast.
[23:06] <jose> lazyPower: enjoy the rest of your day!
[23:06] <lazyPower> I'm only goign to be online for another 30 minutes or so before i head off to go rabble rouse on steam :)
[23:13] <mbruzek> jose, What can I do for you?
[23:13] <jose> mbruzek: hey! I was wondering if you could please review the mailman charm
[23:13] <jose> just did a successful deployment
[23:13] <jose> https://bugs.launchpad.net/charms/+bug/1199052 should be the bug link
[23:13] <_mup_> Bug #1199052: New charm: mailman <Juju Charms Collection:New> <https://launchpad.net/bugs/1199052>
[23:14] <mbruzek> Sure, but it will have to be tomorrow I was also heading out for the day.
[23:14] <jose> that's fine :)
[23:14] <jose> enjoy your evening!
[23:14] <mbruzek> You too.  Thanks for being responsive
[23:14] <mbruzek> I saw that you updated that today
[23:25] <ev> is there any way to tell juju not to care about self-signed certs?
[23:26] <ev> it doesn't like that swift's certificate cannot be verified
[23:46] <lazyPower> sarnold: why didn't you tell me about ipdb sooner?
[23:46] <sarnold> lazyPower: because I've never heard of it? :)
[23:46] <lazyPower> OH MY
[23:46] <lazyPower> http://blog.dasroot.net/ipdb-the-repl-you-didnt-know-about/
[23:46] <lazyPower> i find it so awesome, i wrote about it
[23:50] <sarnold> lazyPower: that's cool! :) thanks
[23:50] <lazyPower> right?!
[23:50] <lazyPower> i'm jazzed about this find
[23:51] <sarnold> okay, scribbled into my ~/.notes file, hopefully next time I need it it'll be at hand :) thanks