/srv/irclogs.ubuntu.com/2014/06/11/#juju.txt

=== vladk|offline is now known as vladk
=== vladk is now known as vladk|offline
=== vladk|offline is now known as vladk
=== vladk is now known as vladk|offline
=== CyberJacob|Away is now known as CyberJacob
=== CyberJacob is now known as CyberJacob|Away
=== alexlist` is now known as alexlist
=== isviridov|away is now known as isviridov
mhshamshi all, appreciate it if you have a look at this issue and help me to fix it: http://ubuntuforums.org/showthread.php?t=2228930&p=13046316#post1304631608:47
sparkiegeekmhshams: that sounds like https://bugs.launchpad.net/juju-core/+bug/130876708:56
_mup_Bug #1308767: juju client is not using the floating ip to connect to the state server <addressability> <landscape> <juju-core:Fix Released by axwalk> <https://launchpad.net/bugs/1308767>08:56
sparkiegeekmhshams: if you're feeling brave you can try using Juju 1.19.3 which contains the fix08:56
=== vladk|offline is now known as vladk
mhshams@sparkiegeek: thanks for the information. i'll try 1.19.3 for sure.09:38
=== rogpeppe1 is now known as rogpeppe
=== Ursinha is now known as Ursinha-afk
=== Ursinha-afk is now known as Ursinha
=== rbasak_ is now known as rbasak
=== jelmer_ is now known as jelmer
=== vladk is now known as vladk|offline
=== BradCrittenden is now known as bac
=== med_ is now known as Guest22673
=== vladk|offline is now known as vladk
avoinenoodles775: I've push my latest code for juju-ansible-role if you want to check: https://github.com/avoine/charm-ansible-roles13:29
=== tvansteenburgh1 is now known as tvansteenburgh
avoineI still need to migrate it to use your wsgi-app13:30
noodles775avoine: Excellent, thanks. I'll probably not get a chance to look through the diff today, but will do after the session (hopefully tomorrow).13:31
avoinegreat13:31
noodles775avoine: slides for the session here if you want to browse through - http://goo.gl/iewo1m13:32
avoinenoodles775: ok, I'll check that out. I've got tons of questions and comments for the session :-)13:32
noodles775avoine: great, as someone who's been using ansible for charms, it'd be great if you can join the hangout too.13:33
avoinenoodles775: sure13:33
jamespagegnuoy, a +1 on https://code.launchpad.net/~james-page/charms/trusty/openstack-dashboard/login-url-stable/+merge/221700 would be good if you have 5 mins13:42
gnuoylooking13:43
gnuoyjamespage, +1'd13:44
noodles775jcastro, jose or marcoceppi: do one of you guys setup the hangout for me, or is it something I do myself? http://summit.ubuntu.com/uos-1406/meeting/22244/simpler-charms-with-ansible/13:44
jcastroif you set it up all I need is the hangout URL13:45
jcastroand the youtube link it generates for you13:45
jcastroor I can do it for you, up to you13:45
noodles775jcastro: if it doesn't need to be linked to any special account, I can do it. If it requires me knowing a special url from which to start so it's linked to canonical, then I don't know that url.13:48
jcastronope, just as normal hangout on air13:49
jcastronote, on air, not a normal hangout13:49
jcastrowant me to just do it? :)13:52
noodles775jcastro: https://plus.google.com/events/cof2tovla6vsv84cq9d34u47b6c13:53
jcastronoodles775, ok, I just the need the URL to the hangout, right from your browser location and I'll take care of the rest13:54
=== avoine1 is now known as avoine2
avoinenoodles775: should I use this link too?13:57
noodles775avoine: Join #ubuntu-uds-dev13:59
noodles775avoine: hangout url is https://plus.google.com/hangouts/_/hoaevent/AP36tYeQwxrxdZGveiZVKpetKkySfiZswgNHf5t0qJUtsowh4qjIIg?authuser=0&hl=en-GB13:59
=== wwitzel3_ is now known as wwitzel3
=== mbarnett` is now known as mbarnett
=== vladk is now known as vladk|offline
=== avoine is now known as patrick
=== patrick is now known as patrick-uds
tech2How can I tell what juju is doing? Say I am working using the local environment and I have a slow connection to the internet, if I do a juju deploy whatever it needs to fetch an image for the instance (I can see the wget in htop), but all status tells me is the machine id is pending.16:09
marcoceppitech2: right, that fetch is only done once16:22
marcoceppithen it creates a template16:22
marcoceppiand subsequent deploys will clone taht templates16:23
tech2marcoceppi: okay, so that's a one-off time cost, that's cool, but is there a way of telling what it's up to in general?16:23
marcoceppitech2: not really, you can tail logs, and agent-state/status should show you what has just happened, but theres no real way to expose the event queue for a machine16:23
tech2marcoceppi: thanks again.16:24
bloodearnestrick_h_: just caught up with the roadmap uds session - machine view looks really nice!16:49
rick_h_bloodearnest: :) woot16:49
rbasakjamespage: so this juju-core SRU. If not going through every bug for SRU information, what other approach do you suggest?17:19
rbasakAn MRE applicatioN?17:19
pindongalazyPower, jcastro MAGIC! I just got juju-gui deployed locally with quickstart! finally :)17:19
pindongathanks a lot for the help17:19
pindongaone thing worth mentioning is.. I see a bunch of WARNING messages when running juju status17:19
pindongalike WARNING unknown config field "container"17:20
pindonganot sure what that is, but at least I got the local provider running17:20
* pindonga is happ(ier) :)17:20
lazyPowerpindonga: awesome!17:25
rick_h_pindonga: what issue were you having? /me missed what was up.17:38
cjohnstonmarcoceppi: any chance you could review https://code.launchpad.net/~ubuntu-ci-engineering/charms/precise/apache2/apache-apt-update/+merge/222846 for me please17:39
pindongarick_h_, hi.. the issue was that I couldn't get the local provider to bootstrap and deploy things properly17:39
rick_h_pindonga: ah ok, so the GUI is behaving itself then?17:39
pindongayep, that was never an issue, it was juju breaking and not cleaning up itself properly17:39
jcastromarcoceppi, omg juju kill17:40
pindongabut with the cleanup plugin and some persistence I got around it :)17:40
rick_h_pindonga: awesome, glad you got through it17:40
* pindonga too17:40
* rick_h_ ducks back into the shadows until the next time someone mentions quickstart/gui17:40
jcastrorick_h_, you're all set for the rest of the week17:40
rick_h_jcastro: now to get those releases out17:41
rick_h_jcastro: thanks for leashing me through the process there. :)17:41
jcastrocory_fu, lazyPower: nice work churning through the queue fellas17:41
lazyPowerjcastro: what are you talking about? we're not working the rev queue17:42
lazyPowerwe're playing broforce on steam right now17:42
jcastroheh17:42
lazyPoweroh i mean, yes, we're working feverishly on the rev q17:42
dpb1lazyPower: any change you could pause broforce and look at https://code.launchpad.net/~davidpbritton/charms/precise/apache2/vhost-config-relation  :)17:46
lazyPowerdpb1: you must have ESP because i'm already looking at that one now17:47
lazyPoweri'm kind of spooked about this really17:47
dpb1lazyPower: look behind you17:47
lazyPowerahhh!17:47
lazyPoweri see my cat17:47
dpb1haha17:47
jcastrohttps://plus.google.com/hangouts/_/hoaevent/AP36tYdCySsBAuEzAEONYU_JQO2lE6GNgDJyfQ2__c1KzGUX4Jc6jA?authuser=2&hl=en17:53
jcastromarcoceppi, charm school will be there ^^^17:54
jcastroand anyone else who wants to be in the actual hangout17:54
jcastromarcoceppi, 5 min warning17:54
mark06why did juju move to github?17:55
marcoceppimark06: increased exposure to the go-lang community? I imagine there's a number of reasons17:56
lazyPowerdpb1: so we are still base64 encoding the templates with this vhost-relation extension?18:00
dpb1yes, right.  I think that is a requirement, unless there is something else preferred?18:00
lazyPowerdpb1: is there a reason we are base64 encoding them vs just passing a string-blob on the wire for apache2 to parse/load/populate/deploy?18:00
khussi am trying to configure the maas server so that two interfaces are configured on the server. I setup two interfaces on the cluster but the server comes up with only one. any ideas?18:01
dpb1lazyPower: ^18:01
lazyPoweri dont think its necessarily required to be base64 encoded... i may be wrong. let me follow up with the ~charmers on this18:01
dpb1hm18:01
lazyPoweri think its added complexity that may not need to be there. becaus ehow cool would it be to just pass a heredoc or a file handle and let juju do its thing?18:01
dpb1ok sure.  I guess it would need to be encoded somehow or else the shell command (relation-set) would barf on it18:01
lazyPowerthats probably why the base64 encoding came up as a requirement now that i think about it18:02
dpb1but I haven't tried explicitly with a real template18:02
lazyPoweryeah, it'll need to be escaped and all kinds of voodoo18:02
lazyPowerdo you ahve an example charm to use for this or do i need to codify one real quick to validate?18:03
lazyPowerdpb1: ^18:06
dpb1lazyPower: the only one I have is in a landscape branch that is a bit hard to setup (since the code is proprietary -- as you know).  It wouldn't be a good test, but you are free to look at how it implements.18:08
dpb1lazyPower: we are waiting to submit until this one is up, since it will break without it.18:08
dpb1lazyPower: the landscape-charm branch: lp:~davidpbritton/landscape-charm/vhost-config-relation/18:09
lazyPowerstands to reason... i'm not clear on what i'm giving the relationshiop. you call a method that makes me think its using  jinja2 - template = get_template()18:09
lazyPowera great thing to have in here would be teh actual contents of an example template. even if its just plane ol yaml, it'll make it clearer and consicer than calling a ficticious method.18:09
lazyPowerah yeah, even in this example you're pulling it from unit data18:11
dpb1ah, got it.  This is basically a way to to specify 'vhost_http_template' and 'vhost_https_template' (current apache2 config settings) in a relation.  But I agree, those are not doced well to begin with I guess (because they are very application specific).18:11
lazyPowerso, it being in YAML format, whats the spec? vhost: {name} \n  port: {port} \n access_log: {logpath} \n directory_root: {dir} ?18:13
dpb1I think I explicitly documented that, let me check18:14
dpb1    relation-set vhosts="- {port: '443', template: dGVtcGxhdGU=}\n- {port: '80', template: dGVtcGxhdGU=}\n"18:15
lazyPowerah, i get that i was under the impression teh template itself was yaml spec18:15
dpb1oh, no, the template is a jinja2 apache config file18:16
dpb1so you can do {{ blah }} in the middle of it18:16
lazyPoweryeah, it supports all the niceties of liquid syntax18:16
lazyPowerdpb1: when you get some time, lets sync on this and clean up the apache readme a bit for new users so its more approachable.18:16
lazyPoweryou seem to have a lot of answers when it comes to the charm :)18:17
dpb1lazyPower: ok.  sure18:17
dpb1lazyPower: I'm going to eat lunch, want to set up a time and we can chat?18:17
lazyPowerI'm pretty booked today. Lets earmark this for post UOS?18:17
dpb1lazyPower: ok, sounds reasonable18:19
lazyPowerthanks for teh preemptive agreement. I'll try to get it on teh calendar for next week18:21
dpb1thx18:24
=== Ursinha is now known as Ursinha-afk
cjohnstonthanks marcoceppi18:43
=== patrick-uds is now known as avoine
=== Ursinha-afk is now known as Ursinha
sparkiegeekcory_fu: hi, I've addressed your comments on my reviewboard charm, ready for another round of review!18:54
sparkiegeekcory_fu: https://bugs.launchpad.net/charms/+bug/94203218:54
_mup_Bug #942032: Charm Needed: Review Board <Juju Charms Collection:Fix Committed by adam-collard> <https://launchpad.net/bugs/942032>18:54
=== roadmr is now known as roadmr_afk
cory_fuOk, I'll take a look.  :)18:54
sparkiegeekcory_fu: cheers!18:55
cory_fusparkiegeek: That was quick.  :)  I still need to test it, but it looks good.  You do need to update the config option descriptions, though.18:58
sparkiegeekcory_fu: ah yes, the dreaded documentation ;)18:58
cory_fuI've got a meeting right now, but I'll test it shortly and add my +118:58
cory_fuha18:58
=== CyberJacob|Away is now known as CyberJacob
sparkiegeekcory_fu: ok. Doc changes are made and pushed19:04
cory_fuThanks19:04
sparkiegeekcory_fu: as for speed, I didn't want to end up on the wrong side of that big long review queue. Got to make the most of your attention ;)19:04
cory_fusparkiegeek: I'm getting "no revisions to pull"19:12
sparkiegeekcory_fu: latest is r17...19:13
dpb1lazyPower: hey there -- so what do I do to get that same thing merged into trusty?  separate MP?  how are you guys handling that?19:13
sparkiegeekcory_fu: my bad, r18 just pushed19:14
cory_fuOk.  Do you think it's worth documenting the semi-magic "localhost" -> public-address behavior?19:16
lazyPowerdpb1: target trusty, make sure its got integration tests re: amulet or integration test method of your choosing.19:16
sparkiegeekcory_fu: it's in the README19:16
cory_fuSo ti is19:17
cory_fu*So it is19:17
cory_fuGreat.  It looks perfect.  I'll add my +1 and ping an official charmer (*ahem* lazyPower *ahem*) to get it promulgated shortly19:18
sparkiegeekpromulgate: a word I only ever here from Juju people ;)19:19
sparkiegeek*hear19:19
* lazyPower blinks19:19
lazyPowercory_fu: is it *good*19:19
lazyPowerlike *real good* ?19:19
sparkiegeeklazyPower: did you miss the "It looks perfect"? ;)19:19
lazyPowersparkiegeek: well... yes. i did.19:20
sparkiegeekI want stars, banners.... maybe even a pony? Can we juju deploy pony yet?19:20
sparkiegeek:)19:21
dpb1lazyPower: ok, unit tests don't suffice?  there are 0 integration tests (juju test) for apache2 right now.19:21
dpb1afaict19:21
dpb1unless I'm missing something (which is likely)19:21
lazyPowerdpb1: wait is apaceh2 in trusty? *looks*19:23
lazyPowerdpb1: ah it is!19:23
lazyPoweryep, same process. Let me run a quick deploy test, if its g2g i'll ack it and merge19:23
lazyPowerno need to fill out another MP. i may change my mind on the rquired MP in teh future, but today - nahhhh thats just extra busy work.19:23
dpb1lazyPower: ok, let me know.  I'll facilitate however need be19:24
lazyPowerdpb1: always a pleasure working with you my man. +1 on a personal level19:26
dpb1lazyPower: thanks!  you have been great to work with as well. :)19:27
lazyPowerdpb1: ack'd and pushed19:36
dpb1thx lazyPower, much appreciated19:37
sparkiegeeklazyPower: does reviewboard get official blessing? Want to get in whilst the going is good :)19:37
lazyPowersparkiegeek: its up on the list, i need a few to do a review as well19:37
sparkiegeeklazyPower: ok. third time's the charm... or so they say :)19:38
lazyPowersparkiegeek: unless you include a pizza with attempt #2, then attempt #2 + bribery is the charm19:39
sparkiegeeklazyPower: fair! :)19:39
lazyPowersparkiegeek: just to be sure, i am reviewing lp:~adam-collard/charms/precise/reviewboard/trunk  correct?19:40
sparkiegeeklazyPower: correct19:40
lazyPowerack, starting review now19:40
lazyPowersparkiegeek: nice charm!19:45
sparkiegeeklazyPower: why thank you :) I'd like to thank my family...19:46
lazyPowersparkiegeek: ackd, promulgated. it will be in the store shortly19:53
lazyPowerWelcome to the charm store :)19:54
sparkiegeeklazyPower: great! thanks a lot. Between you and cory_fb (not forgetting jose) the quality of the charm greatly increased through the review process.19:55
josewoohoo, congratulations sparkiegeek!19:55
lazyPowerThats our goal, with enough eyes we can squash papercuts19:55
sparkiegeekwas a little scary to see the review queue so long, but thanks to your efforts we turned it around in only 2 days19:55
lazyPoweri look forward to seeing your unit tests :) With those you'll be one step closer to a high quality mark on your charm. I've got a follow up todo after its ingested to fill out teh quality data.19:56
sparkiegeeklazyPower: right. Amulet tests are unfortunately, a non-starter. But unit tests are plausible (and desirable)19:56
lazyPowersparkiegeek: the review queue is kind of a fibber. it doesn't calculate based on last interaction, it reads from beginning of item - so, its quick to scatter the result.19:56
sparkiegeeklazyPower: top of my todo list :)19:57
sparkiegeeklazyPower: is there a way of subscribing to a particular charm's bugs in LP?19:59
sparkiegeekah n/m found i19:59
sparkiegeek*it19:59
=== vladk|offline is now known as vladk
=== roadmr_afk is now known as roadmr
lazyPowersparkiegeek: ty for updating the bug to fix released, i just noticed i missed that20:27
sparkiegeeklazyPower: np20:27
=== vladk is now known as vladk|offline
=== jcsackett_ is now known as jcsackett
=== ming is now known as Guest58318
=== Beret- is now known as Beret
=== CyberJacob is now known as CyberJacob|Away
AskUbuntuHow to Access HBase after JUJU | http://askubuntu.com/q/48203623:41

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!