2012/04/05/#juju.txt

[00:33] <hazmat> SpamapS, i thought it was going to have the default config written out with ssl check enabled
[00:34] <SpamapS> hazmat: the branch I had proposed does that
[00:35] <hazmat> SpamapS, not the version in reitveld afaics
[00:35] <hazmat> https://codereview.appspot.com/5934054/
[00:36] <hazmat> SpamapS, besides that LGTM
[00:40] <SpamapS> hazmat: https://codereview.appspot.com/5934054/patch/11001/12003
[00:42] <_mup_> juju/relation-id-option r521 committed by jim.baker@canonical.com
[00:42] <_mup_> Fix setup of local api factory used by testing
[00:49] <_mup_> juju/relation-id-option r522 committed by jim.baker@canonical.com
[00:49] <_mup_> Remove usage of invoker attr in favor of using the factory to get it
[00:51] <_mup_> juju/relation-id-option r523 committed by jim.baker@canonical.com
[00:51] <_mup_> Code rearrangement
[00:56] <_mup_> juju/relation-ids-command r511 committed by jim.baker@canonical.com
[00:56] <_mup_> relation-ids command without a relation name to qualify returns all relation ids
[00:58] <_mup_> juju/relation-id-option r524 committed by jim.baker@canonical.com
[00:58] <_mup_> Merged upstream & resolved conflict due to moving relation-ids command tests to upstream
[00:59] <_mup_> juju/relation-hook-context r526 committed by jim.baker@canonical.com
[00:59] <_mup_> Merged trunk for this already merged branch to workaround lbox prereq issue
[01:11] <_mup_> juju/relation-ids-command r512 committed by jim.baker@canonical.com
[01:11] <_mup_> Change test setup to verify that relation-ids on a relation name does in fact only return those relation ids
[01:11] <_mup_> juju/relation-id-option r525 committed by jim.baker@canonical.com
[01:11] <_mup_> Merged upstream
[03:20] <_mup_> juju/relation-ids-command r513 committed by jim.baker@canonical.com
[03:20] <_mup_> Test relation ident enumeration from nonrelational hook context and departed relation hook context
[03:35] <_mup_> juju/relation-id-option r526 committed by jim.baker@canonical.com
[03:35] <_mup_> Merged upstream
[06:45] <lynxman> if someone can promote charm attached to bug #956259 looks alright :)
[06:45] <_mup_> Bug #956259: Charm needed: znc  <new-charm> <Juju Charms Collection:Fix Committed by lynxman> < https://launchpad.net/bugs/956259 >
[06:47] <SpamapS> lynxman: you're not in ~charmers ?
[06:47] <lynxman> SpamapS: yeah, so how can I promote it myself? :)
[06:48] <SpamapS> lynxman: bzr push lp:~charmers/charms/oneiric/znc/trunk && charm promulgate
[06:48] <lynxman> SpamapS: cool, thanks
=== almaisan-away is now known as al-maisan
[07:27] <res0nat0r> /loa/clear
[07:27] <res0nat0r> oops
[07:47] <shazzner> ok
[07:47] <shazzner> I (poorly) rewrote the parse_upstream php file into python
[07:48] <SpamapS> shazzner: eh?
[07:48] <imbrandon> heh why ? j/k
[07:48] <shazzner> http://paste.ubuntu.com/915662/
[07:48] <shazzner> well
[07:48] <shazzner> mostly to do with how it's broken in precise
[07:48] <imbrandon> :)
[07:48] <SpamapS> oh the https
[07:48] <shazzner> ya
[07:48] <SpamapS> wget's not broken.. just wget the damn thing ;)
[07:49] <imbrandon> i'm actully converting SpamapS's python for the multi nginx stuff to php now
[07:49] <shazzner> I wish!
[07:49] <shazzner> haha
[07:49] <SpamapS> imbrandon: hater
[07:49] <imbrandon> started it like 15 min ago
[07:49] <imbrandon> lol
[07:49] <imbrandon> SpamapS: hehe , well i wanted it all to match
[07:49] <imbrandon> e.g. all the othersgonna be php
[07:50] <SpamapS> imbrandon: I'd like for that to all be generic as soon as subordinates land
[07:50] <imbrandon> nothing against your implmentation though
[07:50] <SpamapS> imbrandon: shouldn't need to all be one language
[07:50] <imbrandon> no, it dont, its a prefrence thing
[07:50] <SpamapS> imbrandon: but I'm actually glad to see non-python/bash charm hooks :)
[07:50] <imbrandon> just being anal
[07:50] <imbrandon> :)
[07:51] <SpamapS> I always figured ruby would be more prevalent.. not .. php
[07:51] <imbrandon> i'm trying to figure out the best way to do the most in php thought, since its not installed by default i have to bash bootstrap it minimum
[07:51] <imbrandon> i'm sure rubys and gems and diimonds and poneies will be next :)
[07:51] <SpamapS> imbrandon: I want to create a more declarative way to pull in packages at some point to handle this case
[07:52] <imbrandon> yea thats would be cool
[07:52] <shazzner> I just want to say xml handling is so much easier than python
[07:52] <shazzner> *in php I mean
[07:52] <SpamapS> imbrandon: I'd like for there to be a way to write a charm that is 99% just lists of packages/files to install.
[07:52] <imbrandon> i duno about 99% then its just a deb :)
[07:53] <imbrandon> deb + rules file thats not debuild'd
[07:53] <SpamapS> shazzner: seems like you're iterating when you could have done xpath
[07:53] <imbrandon> SpamapS: but i agree, more than now would be nice
[07:53] <shazzner> SpamapS: I have no doubt I did it poorly :p
[07:53] <SpamapS> imbrandon: the 1% would be the bits that run on hook changes.
[07:54] <imbrandon> ahh ok
[07:54] <imbrandon> then yea i;'m right with ya
[07:54] <SpamapS> so maybe 90%
[07:54] <shazzner> if you have suggestions or changes, please go for it
[07:54] <SpamapS> Like, let me write the code only when code is warranted.
[07:54] <imbrandon> right
[07:54] <imbrandon> yea i just equated that in my head as more, but yea, i fully agree then
[07:55] <imbrandon> but yea , its nothing against your implmentation, infact its quite simple and easy to understand even for a python newb but i'm just being picky
[07:55] <imbrandon> is all :)
[07:56] <SpamapS> The reason I've gravitated to python away from other languages has been that the code is usually easy to understand (except when its twisted based ;)
[07:57] <imbrandon> btw i got importing from github going i think, its not imported yet but its in the queue for RSN
[07:57] <imbrandon> yea i've just stayed away from it since the rules seem like they were written by rms
[07:58] <imbrandon> heh and really i havent stayed away just try not to use it when not nessesary
[07:58] <imbrandon> lol
[07:58]  * SpamapS will like github when celery tastes like candy
[07:58] <imbrandon> hahaah
[07:58] <shazzner> hehe
[07:58] <imbrandon> btw they did a huge upgrade to the issue tracker this week
[07:58] <SpamapS> urllib2.URLError: <urlopen error [Errno 104] Connection reset by peer>
[07:58] <imbrandon> might peek at it again
[07:58] <SpamapS> shazzner: thats what I get
[07:59] <shazzner> huh
[07:59] <imbrandon> milestones and proraties and stuff
[08:00] <shazzner> haha it seems like the same issue php had
[08:00]  * imbrandon slides off to find some nicoteine and finish bootstrapping drupal 8 into a working state
[08:00] <imbrandon> semi afk
[08:01] <imbrandon> btw morning fellas 0/ :)
[08:01] <SpamapS> imbrandon: 'bout to be sleep for me
[08:01] <imbrandon> ahh
[08:01] <shazzner> SpamapS: are you behind a proxy perhaps?
[08:02] <SpamapS> shazzner: definitely not
[08:02] <shazzner> same here, it's like 3am
[08:02] <shazzner> err huh
[08:02]  * imbrandon just woke up, i keep odd hours for most, its 3am here too
[08:02] <SpamapS> shazzner: have you tried with openssl s_client
[08:02] <SpamapS> I think their servers may just suck
[08:03] <shazzner> probably
[08:03] <SpamapS> probably out of entropy
[08:03] <imbrandon> where ya grabbin it from ?
[08:03] <SpamapS> openssl s_client -host sourceforge.net -port 443
[08:03] <SpamapS> stalls out
[08:03] <shazzner> I ran that script several times testing it though, all consistant
[08:03] <imbrandon> nice
[08:03] <shazzner> hrm
[08:03] <SpamapS> telnet to that host/port also stalls
[08:04] <SpamapS> shazzner: sure, you used up all their entropy ;)
[08:04] <shazzner> damnit! sorry guys
[08:04] <imbrandon> use lp to import  the sf repo and then grab from there ;)
[08:04] <SpamapS> actually
[08:04] <SpamapS> this is likely an openssl issue
[08:04] <SpamapS> on a centos 6 box, no issue, connects right up
[08:04] <imbrandon> i'll try from osx one sec
[08:05] <imbrandon> yea right up here
[08:05] <shazzner> yeah that openssl line stalls for me too
[08:05] <imbrandon> http://paste.ubuntu.com/915676/
[08:05] <SpamapS>     Protocol  : TLSv1
[08:05] <shazzner> imbrandon: same result
[08:06] <shazzner> actually wait, mines a little different
[08:07] <imbrandon> mines from osx so it might be a bit diffrent
[08:07] <shazzner> ah
[08:07] <SpamapS> wtf
[08:07] <imbrandon> http://paste.ubuntu.com/915676/
[08:07] <SpamapS> I think I accidentally /quit there or something
[08:07] <bkerensa> SpamapS: you lost terminal
[08:07] <bkerensa> :D
[08:07] <imbrandon> heh says lost term
[08:07] <imbrandon> 03:06 -!- SpamapS [~clint@ubuntu/member/spamaps] has quit [Quit: Lost terminal]
[08:07] <bkerensa> SpamapS: How is contest coming along ? :P
[08:08] <shazzner> http://paste.ubuntu.com/915679/
[08:08] <SpamapS> Yeah must have torched byobu accidentally
[08:09] <SpamapS> bkerensa: wrapping up judging this week :)
[08:09] <bkerensa> :P
[08:09] <shazzner> I use guake and have move between tabs be ctrl-left & right and close tabs ctrl-down
[08:09] <SpamapS> RC4-SHA
[08:09] <SpamapS> I wonder if Ubuntu's openssl has that disabled
[08:09] <shazzner> I've closed my irc session many times :/
[08:10] <shazzner> the channel topic should probably change too :p
[08:12]  * imbrandon gave up OPer when lilo passed, no haxoring the topic from me
[08:12] <SpamapS> aha
[08:12] <SpamapS> RC4 is in fact disabled
[08:12] <imbrandon> nice
[08:12] <SpamapS> if you explicitly request it
[08:12] <SpamapS> it works
[08:12] <SpamapS> *thank you ubuntu security team for making sure we are safe*
[08:12] <SpamapS> and sourceforge.. I spit on your RC4
[08:12] <imbrandon> heh
[08:13] <SpamapS> shazzner: so your PHP implementation will probably work fine, if you add some options
[08:14] <shazzner> oh yeah?
[08:14] <SpamapS> http://www.php.net/manual/en/context.ssl.php
[08:15] <SpamapS> specifically, 'ciphers'
[08:15] <SpamapS> I would set it to AES256-SHA,RC4-SHA
[08:16] <imbrandon> if google glass is not a april fools joke i'm so getting some
[08:17] <shazzner> yeah those look pretty rad
[08:17] <shazzner> SpamapS: I'll take a look at that, thanks!
[08:18] <shazzner> I'm off to bed though, have do to taxes tomorrow :(
[08:18] <shazzner> night ya'll
=== shazzner is now known as shazzner-away
[08:19] <SpamapS> ugh yeah I have to wrap my taxes up this week too :-P
[08:19] <SpamapS> what a lovely thought to go to sleep to
[08:19]  * SpamapS passes out
[08:27] <res0nat0r> Any guess why I can't test out and deploy the cassandra charm? I see it in the repo https://code.launchpad.net/~charmers/charms/oneiric/cassandra/trunk-1
[08:27] <res0nat0r> I'm getting a not found
[08:53] <yolanda> hi, good morning
[08:53] <yolanda> i'm having a problem running juju in a local container in precise
[08:54] <yolanda> error: internal error Network is already in use by interface virbr0
=== al-maisan is now known as almaisan-away
[09:54] <arashbm> yolanda: just re login and it will be solved!
[10:13] <yolanda> hi, yes, i rebooted and works
[11:44] <yolanda> hi, another question. In what directory should i put my local.yaml file with config settings? i tried to put it under the precise/ directory of the repo and gives an error
[13:25] <niemeyer> Good morning jujuers!
[13:36] <TheMue> niemeyer: morning
[13:36] <niemeyer> TheMue: Good morning!
[13:38] <yolanda> hi
[13:38] <hazmat> g'morning folks
[13:38] <yolanda> i have this problem when trying to get content from a private repo:
[13:38] <yolanda> W: Failed to fetch https://private-ppa.launchpad.net/canonical-isd-hackers/ppa/ubuntu/dists/precise/main/binary-i386/Packages  Proxy CONNECT aborted
[13:38] <yolanda> i'm using local containers
[13:38] <yolanda> in my local machine works, but juju local container gives that error
[13:38] <yolanda> any clue?
[13:39] <koolhead11> yolanda, your running your machine behind a proxy
[13:40] <yolanda> no, i didn't add any config for that. Locally i can get that repo without problems
[13:40] <yolanda> but when i create a juju local machine, it gives that error
[13:40] <hazmat> yolanda, juju setups the local containers for an apt cacher ng proxy on the host
[13:40] <yolanda> hazmat, how can i solve the problem?
[13:40] <hazmat> yolanda, you may have to adjust the cacher-ng proxy settings on the host
[13:41] <yolanda> hazmat, where can i check it?
[13:43] <hazmat> yolanda, its in /etc/apt-cacher-ng   its got a web page for its stats at http://localhost:3142/acng-report.html
[13:44] <hazmat> yolanda, do you have that ppa enabled on the host?
[13:44] <yolanda> hazmat, but in my local machine i can access the repo, the problem is on the juju instance
[13:45] <yolanda> and in the juju instance i cannot see any apt-cacher-ng directory when i debug
[13:45] <hazmat> yolanda, the apt-cacher-ng stuff is on the host not the container
[13:45] <hazmat> the containers are all configured to point to it
[13:47] <hazmat> yolanda, so you trying to add a ppa and install a package from it in the container in a charm hook?
[13:48] <hazmat> yolanda, the private ppa has some form of auth.. it sounds like that it causing a problem with proxy.. but if thats not adding ppa auth to the container, or config to the proxy i don't know
[13:48] <yolanda> hazmat, i added the right user and pass to the sources.list
[13:49] <yolanda> perhaps is some problem with https?
[13:52] <hazmat> yolanda, could you pastebin anything relevant from the cacher-ng logs at /var/log/apt-cacher-ng on the host
[13:53] <yolanda> let me check
[14:11] <yolanda> hazmat, i cannot see any content about this ppa in logs
[14:25] <koolhead11> jcastro, hi there
[14:26] <jcastro> hi
[14:26] <koolhead11> let me know how can i help with juju doc :)
[14:27] <koolhead11> i think i did do some changes with the doc and its waiting 4 approaval
[14:29] <yolanda> hazmat, i'm totally lost with that problem, what else can i check?
[14:31] <jcastro> koolhead11: that was your one before the docs were split out
[14:31] <jcastro> everything is integrated now
[14:32] <hazmat> yolanda, you could try turning up the logging in apt-cacher-ng (VeboseLogging=1), the errors should be here. /var/log/apt-cacher-ng/apt-cacher.err
[14:33] <koolhead11> jcastro, as in all documenation related stuff is handled?
[14:33] <hazmat> yolanda, er.. VerboseLog=1
[14:34] <hazmat> yolanda, or email the list, i'm not that familiar with apt-cacher-ng.. or if it supports https
[14:35] <hazmat> koolhead11, all the docs are in a separate repo so they don't have to wait in the code merge queue, and there's a larger group that has direct access.
[14:35] <hazmat> lp:juju/docs
[14:36] <koolhead11> hazmat, yes saw the ail in mailinglist :)
[14:36] <hazmat> cool
[14:37] <yolanda> let me check
[14:41] <yolanda> hazmat, i enable log, but still any clue about canonical, i see some logs there, but nothing related to that entry
=== hazmat is now known as kapilt
[16:01] <SpamapS> imbrandon: OpenSSL problem identified btw.
[16:02] <marcoceppi> SpamapS: what was it?
[16:02] <SpamapS> bug #965371
[16:02] <_mup_> Bug #965371: HTTPS requests fail on some sites on Ubuntu 12.04 <rls-p-tracking> <OpenSSL:Confirmed> <openssl (Ubuntu):Triaged by cjwatson> <openssl (Ubuntu Precise):Triaged by cjwatson> <openssl (Debian):New> < https://launchpad.net/bugs/965371 >
[16:02] <SpamapS> fixed in the latest openssl
[16:02] <SpamapS> has to do with the size of the initial handshake packet
[16:03] <marcoceppi> awesome
[16:03] <SpamapS> so, dist-upgrade people!
[16:03] <SpamapS> was fixed on Friday
[16:16] <imbrandon> :)
[16:21] <jcastro> SpamapS: marcoceppi: m_3_ charm contest, let's finish it </kombat>
[16:21] <marcoceppi> jcastro: might as well, still waiting on s3cmd
[16:22] <jcastro> http://pad.ubuntu.com/charmcontest
[16:26] <jcastro> SpamapS: ^ those are the entrants afaict who are eligible. Confirm pls.
[16:28] <jcastro> oh, gitolite
[16:28] <jcastro> https://bugs.launchpad.net/charms/+bug/906176
[16:28] <_mup_> Bug #906176: Gitolite charm <new-charm> <Juju Charms Collection:Fix Committed by shazzner> < https://launchpad.net/bugs/906176 >
[16:28] <jcastro> we need to finish this review
[16:29] <jcastro> ditto for drupal
[16:29] <jcastro> https://bugs.launchpad.net/charms/+bug/964936
[16:29] <_mup_> Bug #964936: Drupal Charm: superchared Drupal charm with nginx,, apc, php-fpm, all setup to scale to the moon and be Best Practices. <new-charm> <Juju Charms Collection:Fix Committed by imbrandon> < https://launchpad.net/bugs/964936 >
[16:32] <SpamapS> jcastro: dunno if you saw, but lynxman promulgated znc last night
[16:32] <jcastro> yep, on the list
[16:33] <jcastro> SpamapS: so I'm thinking, check these two, promulgate or kickback, and then decide?
[16:33] <marcoceppi> Drupal hasn't been promulgated yet, still needs a finaly review
[16:35] <jcastro> http://pad.ubuntu.com/charmcontest <-- am I missing anyone?
[16:35] <jcastro> I left out myself, marco, and nick's charms that promulgated or where finished during the period
[16:37] <bkerensa> jcastro: thanks for the shipment
[16:37] <bkerensa> :D
[16:37] <bkerensa> I assume you had Cezz send the UPS thingy
[16:37] <jcastro> I didn't send you anything yet
[16:37] <bkerensa> -.-
[16:37] <bkerensa> jcastro: no?
[16:37] <jcastro> nope, I'm not mailing out the prizes until tomorrow
[16:37] <bkerensa> uhh well I didnt order anything and Merchandise Mania is shipping me something
[16:37] <bkerensa> so hmm
[16:37] <bkerensa> LOL
[16:37] <bkerensa> wth
[16:38] <jcastro> stealth swag
[16:38] <_mup_> juju/relation-id-option r527 committed by jim.baker@canonical.com
[16:38] <_mup_> Addressed minors
[16:38] <bkerensa> jcastro: indeed
[17:01] <bkerensa> jcastro: lol its from Mozilla :D they use Merchandise Mania too
[17:09] <marcoceppi> Are there any instructions or documentation on possibly adding a provider to juju?
[17:15] <zul> SpamapS: hold on http://paste.ubuntu.com/916240/
[17:19] <jcastro> dude: "authorized-keys: chuck"
[17:20] <jcastro> you can do that?
[17:20] <zul> jcastro: according to the docs you can
[17:20] <zul> but that might be the problem
[17:22] <aiurea> hi
[17:22] <aiurea> I have a quick question. Is support for Debian & Redhat in the Juju Roadmap?
[17:23] <bkerensa> aiurea: good question.... I think jcastro might be able to answer that
[17:24] <aiurea> bkerensa: I'm a Debian developer starting a new project and I would love to use juju. And I could contribute afterwards. But I would like to know that support for RedHat/Debian is incoming.
[17:26] <bkerensa> aiurea: If for some reason you have to go and dont get a answer before you leave feel free to ping the Cloud mailing list to get a answer
[17:26] <bkerensa> https://lists.ubuntu.com/mailman/listinfo/Ubuntu-cloud
[17:27] <aiurea> bkerensa: thank you.
[17:31] <marcoceppi> aiurea: bkerensa juju mailing list works well too: https://lists.ubuntu.com/mailman/listinfo/juju
[17:31] <aiurea> marcoceppi: got it.
[17:31] <bkerensa> marcoceppi: ok :)
[17:32]  * bkerensa is interest to know the answer to aiurea's question since I hope to talk about juju in the futurew
[17:33] <marcoceppi> From what I understand, and I might be wrong, juju client should be able to run on any linux system. The charms are all apt-get specific so outside of aptitude they won't deploy. Theoretically you could deploy to Debian images in say ec2 but I don't think anyone has tried. As for Redhat while the client might run I think yum is missing a few vital packages that would be required for redhat images
[17:33] <marcoceppi> to run
[17:34] <marcoceppi> So it's a giant maybe. I don't know if the dev team has any plans to support other systems and if they do when they would be doing that
[17:35] <zul> SpamapS: http://paste.ubuntu.com/916313/
[17:40] <aiurea> I see.
=== kapilt is now known as hazmat
[17:43] <SpamapS> zul: that means zookeeper failed to start
[17:43] <SpamapS> zul: check console log
[17:43] <zul> k
[17:44]  * SpamapS puts $5 on "Hash sum mismatch"
[17:44] <zul> SpamapS: last line is this: http://paste.ubuntu.com/916321/
[17:45] <SpamapS> zul: that means juju failed to install. Look up higher.
[17:46] <zul> SpamapS: wha....it tried to fetch from the ec2 archive: http://paste.ubuntu.com/916324/
[17:49] <SpamapS> zul: whats 'nov.ec2' ?
[17:49] <SpamapS> zul: is that some kind of weird thing that cloud-init is doing because the "region" is nov?
[17:49] <zul> SpamapS: im not exactly sure
[18:35] <marcoceppi> SpamapS: getting a wall of "[  183.541157] 1 multicall(s) failed: cpu 0" when bootstrapping a t1.micro, is that Amazon's gentle way of saying NO CPU FOR YOU?
[18:42] <SpamapS> marcoceppi: never seen that
[18:42] <SpamapS> marcoceppi: what region?
[18:43] <marcoceppi> eu-west-1
[18:43] <marcoceppi> The last several micros I've tried to boostrap in eu-west-1 and us-east-1 have failed
[18:43] <marcoceppi> this is just the first time I've actually checked the logs
[18:50] <SpamapS> java bug.. very sad. ;)
[18:52] <SpamapS> jimbaker: did I see relation id support land this morning?
[19:34] <jimbaker> SpamapS, yes you did
[19:35] <jimbaker> SpamapS, definitely a nice feature - charms can now work with relations from any hook
[19:35] <marcoceppi> jimbaker: wait, what?
[19:37] <jimbaker> marcoceppi, you can list available relation ids using the new relation-ids command
[19:37] <jimbaker> then use -r <relation id> with relation-get|relation-set|relation-list
[19:38] <marcoceppi> Ah, so you can get, set, and list relation data from any hook now?
[19:38] <jimbaker> marcoceppi, correct
[19:38] <marcoceppi> epic
[19:41] <SpamapS> truly epic
[19:42] <SpamapS> you can ripple an event through the system now
[19:42]  * SpamapS goes to get lunch
[19:43] <avoine> awesome
[19:44] <jimbaker> SpamapS, yes, this is a very nice aspect of what has landed - generalizing the orchestration
[19:45] <jimbaker> with ultimately just a very simple addition of functionality
[21:05] <bkerensa> SpamapS: I think the next charm I am going to write is for openphoto
[21:07] <jcastro> wow, I didn't even know that existed
[21:07] <jcastro> bkerensa: if you want you can file a wishlist bug on lp under "charms" and link the home page there
[21:07] <jcastro> so we have it
[21:08] <bkerensa> jcastro: I will do... I am joining the openphoto team as a Community Manager
[21:08] <bkerensa> :P
[21:08] <bkerensa> so it naturally makes sense
[21:08] <jcastro> <3
[21:12] <SpamapS> bkerensa: *sweeeeet*
[21:13] <SpamapS> bkerensa: btw, when you're done w/ that charm, I'll definitely +1 you for ~charmers (and I bet somebody else will too)
[21:14] <bkerensa> heh
[21:14] <bkerensa> :D
[21:15] <lifeless> SpamapS: oh hai
[21:15] <_mup_> Bug #974650 was filed: relation-ids should default to $JUJU_RELATION, or if not available, raise an error <juju:In Progress> < https://launchpad.net/bugs/974650 >
[21:15] <lifeless> SpamapS: maybe you know the answer to my question in #ubuntu-cloud ?
[22:42] <shazzner> yay taxes are done! :D
[23:05] <SpamapS> shazzner: did you see above, the openssl issue is solved in the latest package in precise?
[23:06] <shazzner> Spamaps: no I didn't, let me give it a shot
[23:07] <SpamapS> shazzner: something to do with broken BigIP's
[23:07] <shazzner> huh
[23:07] <shazzner> does that python script work for you now?
[23:08] <shazzner> I'm still getting the ssl timeout with the php parse_upstream script
[23:08] <shazzner> unless, this is something totally different
[23:11] <SpamapS> shazzner: I lost the python script.. (/tmp .. rebooted.. ;)
[23:12] <SpamapS> shazzner: but openssl s_client works now
[23:12] <SpamapS> shazzner: the error 104 would be different
[23:12] <shazzner> here it is in all it's glory!: http://paste.ubuntu.com/916729/
[23:19] <shazzner> so when I call juju-log
[23:19] <shazzner> where exactly does it log it to?
[23:20] <SpamapS> shazzner: the charm.log
[23:20] <SpamapS> shazzner: which is visible via 'debug-log'
[23:21] <SpamapS> shazzner: or on disk, as /var/lib/juju/units/xxx-#/charm.log
[23:21] <SpamapS> shazzner: or if you're using the local provider, data-dir/units/xxx-#/unit.log
[23:21] <shazzner> SpamapS: got it, thanks! :)
[23:23] <shazzner> hmm ok, so the root of the matter is I added a bin/parse_upstream.py file along with my charm
[23:23] <shazzner> when I try to call it, it can't seem to find it
[23:23] <shazzner> I'm basing this off the phpmyadmin charm, which seems to be correct but it keeps failing there
[23:24] <shazzner> what would be the absolute path to the charm on the machine?
[23:24] <shazzner> oh wait I think I found it
[23:24] <shazzner> same path as the charm.log
[23:26] <SpamapS> python /tmp/foo.py
[23:26] <SpamapS> http://sourceforge.net/projects/kusabax/files/Kusaba%20X%200.9.3/Kusaba%20X%200.9.3%20Full/kusabax_0.9.3_full.zip/download|fbd051a3b0c3ddd9d08e56309b4f2e23
[23:26] <SpamapS> shazzner: yep, works now
[23:26] <shazzner> yay! :)
[23:26] <SpamapS> shazzner: don't assume that path
[23:27] <SpamapS> shazzner: your charm will always start executing in the root of the charm
[23:27] <shazzner> yeah I was noticing that
[23:27] <SpamapS> shazzner: $CHARM_DIR will always be set to that dir
[23:27] <shazzner> hrm
[23:27] <SpamapS> those are two assumptions you can make
[23:28] <shazzner> it fails here: HASHFILE=`bin/parse_upstream.py` with /var/lib/juju/units/kusabax-0/charm/hooks/install: 20: local: not in a function
[23:29] <shazzner> it's probably something stupid on my end
[23:30] <SpamapS> thats a syntax error
[23:30] <SpamapS> local is only allowed in functions
[23:30] <shazzner> oh
[23:30] <shazzner> like I said, something stupid :p
[23:30] <SpamapS> because she'll scope for non-functions is always global :-P
[23:31] <shazzner> got it
=== Furao_ is now known as Furao