=== mramm2 is now known as mramm === kadams54-away is now known as kadams54 [01:29] marcoceppi: rename to ouija ;-) === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away === beisner- is now known as beisner === jaywink_ is now known as jaywink === rcj is now known as Guest41677 [08:46] good morning, jujuers [08:49] morning [09:02] guys [09:02] can anybody tell me exactly when the hooks are called? I mean departed, broken, etc. [09:05] Muntaner : u mean when is it the hook are shown? [09:46] frankban, if you are interested, I discovered another very-little bug in juju-gui [09:47] Muntaner: I am [09:47] frankban, it's a very little problem. I am reproducing it right now, in google chrome, xubuntu 14.04 [09:49] well, I am in the "services" canvas, and zooming out and in with the mouse wheel. Then, I go on the "machines" tab, and back again to "services" tab. Into the canvas, the icons of the single charms are very very little and don't care about mouse wheel scrolling. They come back to normal by clicking on the charm and then clicking on the canvas again [09:50] Muntaner: yes, that's a known bug in chrome, we are tracking it [09:51] Muntaner: it shoud not affect other browsers, for instance firefox [10:24] frankban, I need to expose my charm into the store on the left [10:24] can I do that offline? [10:25] Muntaner: offline? [10:25] frankban, I mean [10:25] mantaining my charm on the laptop [10:28] Muntaner: without pushing the charm on lp it's not possible to have that on the left pane [10:29] ok frankban, actually it is a very experimental and rude charm, can I push it anyway? [10:29] Muntaner: sure, it will be under your user namespace [10:35] frankban, how should I do that? [10:36] Muntaner: push your charm as lp:~{yourself}/charms/trusty/{yourcharm}/trunk === Murali_ is now known as Murali [11:51] frankban, or others: can I execute an hook inside another hook? [11:54] Hi Jamespage [11:55] Hi lazyPower [11:55] I tried the https://github.com/whitmo/bundle-kubernetes/blob/master/specs/local.yaml example [11:56] I am seeing some issue with it [11:56] http://paste.ubuntu.com/10514375 [11:56] http://paste.ubuntu.com/10514379 [11:57] could you please check what is missing in my .yaml code [11:58] where to mention the local repository path for the charm to deploy via local [12:26] Murali: you can't target arbitrary machines [12:26] your picking 23 there for example [12:26] try 0 [12:27] Murali: http://pythonhosted.org/juju-deployer/config.html#placement [13:02] Is there a way I can use bundles with services from both charm store and my local machine ? [13:04] Anz: yes, using the juju-deployer directly you can do that. [13:04] @Thanks rick. Is there an example bundles.yaml for such a deployment which I refer to ? [13:05] Anz: I was just poking at http://pythonhosted.org/juju-deployer/config.html#services but it doesn't have an example there. looking [13:08] Anz: this bug references it https://bugs.launchpad.net/juju-deployer/+bug/1274067 [13:08] Bug #1274067: juju-deployer local-mods command is counter intuitive and thus confusing

[13:08] Anz: just make sure your ones from the charmstore are cs: urls [13:09] Ok [13:38] jamespage, have you got a sec for https://code.launchpad.net/~gnuoy/charms/trusty/nova-compute/1427660/+merge/251596 ? === kadams54-away is now known as kadams54 [13:52] whit, marcoceppi, asanjar, jcastro, arosales: any news about the merge of the tests? https://code.launchpad.net/~nicopace/+activereviews There are still 9 pending branches [13:52] whit, marcoceppi, asanjar, jcastro, arosales: good morning BTW :D [13:53] whit, marcoceppi, asanjar, jcastro, arosales: i know that you have other stuff to do. i just wanted to know because i have 2 hours left in this project :D [13:53] (at least for now) [14:09] rbasak: FYI we just released quickstart 2.0: https://jujugui.wordpress.com/2015/03/03/juju-quickstart-2-0/ [14:09] rbasak: note that it now includes a functional test suite: http://bazaar.launchpad.net/~juju-gui/juju-quickstart/trunk/view/head:/HACKING.rst#L70 [14:14] nicopace, I think the other guys have a plan to get that stuff in asap [14:14] jcastro: great! :D [14:15] mthaddon, if any of you guys are available to do some charm reviews, we could really use a hand === Guest41677 is now known as rcj === rcj is now known as Guest72440 [14:31] Murali: o/ [14:32] Murali: do you have the ENV variable JUJU_REPOSITORY set? [14:32] thats required to gain support for the local:foo/bar directive ina yaml [15:13] Hello - while trying to deploy my charm under "local" env, when it tries to downlload pip package for charmhelpers purpose I get messages like: unit-kimchi-0[871]: 2015-03-03 15:09:55 INFO unit.kimchi/0.install logger.go:40 E: Failed to fetch http://ports.ubuntu.com/ubuntu-ports/pool/universe/p/python-pip/python-pip_1.5.4-1_all.deb Cannot initiate the connection to ports.ubuntu.com:80 (2001:67c:1360:8001::18). - connect (101: Network is unreac [15:13] like if it is willing to use IPv6 when network is configured as : ubuntu@u0016001-local-machine-3:~$ ifconfig -a [15:13] eth0 Link encap:Ethernet HWaddr 00:16:3e:d3:41:fb [15:13] inet addr:10.0.3.119 Bcast:10.0.3.255 Mask:255.255.255.0 [15:13] inet6 addr: fe80::216:3eff:fed3:41fb/64 Scope:Link [15:13] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [15:13] RX packets:2996 errors:0 dropped:0 overruns:0 frame:0 [15:13] TX packets:1773 errors:0 dropped:0 overruns:0 carrier:0 [15:13] collisions:0 txqueuelen:1000 [15:13] RX bytes:12274149 (12.2 MB) TX bytes:219103 (219.1 KB) [15:13] My lxc config file is: k$ cat /etc/lxc/default.conf [15:13] lxc.network.type = veth [15:13] lxc.network.ipv4 = auto [15:13] lxc.network.link = lxcbr0 [15:13] lxc.network.flags = up [15:13] lxc.network.hwaddr = 00:16:3e:xx:xx:xx [15:13] thierry-ibm: o/ [15:14] thierry-ibm: can you please put multi-line output in a paste and link to it instead of flooding the channel? [15:14] thierry-ibm: http://paste.ubuntu.com is a great resource for this [15:15] sure - sorry about that - thanks for the advise: - doing it immediately [15:15] Thanks thierry-ibm [15:17] http://paste.ubuntu.com/10516073/ [15:19] http://askubuntu.com/questions/592177/is-it-possible-to-deploy-to-specific-machine-with-bundle [15:19] can someone peer review my answer please? [15:27] jcastro: looks sane, there's extant feature requests for the machine non zero thing, but its not implemented [15:29] thierry-ibm: are you on ipv6 net on the host? [15:31] hazmat: I connected using ipv4 but ifconfig gives both ipv4 and ipv6 addresses which is may be the issue [15:40] hazmat, I don't suppose there's a bug # for it? [15:40] jcastro: there is [15:41] nicopace: ack, thanks for checking on those reviews this week. We'll see if we can get those reviews turned around soon. [15:59] arosales: great! let me know! [16:04] hey guys [16:04] frankban, I need to call the "install" hook into another hook. Can I do that? [16:04] Muntaner: 1min on call [16:05] Muntaner: you just put "hooks/install" as an invokation [16:05] marcoceppi, thanks [16:05] Muntaner: that's for a bash charm [16:05] marcoceppi, it's my case, I'm writing in bash [16:06] Muntaner: then that's what you'd do === anthonyf is now known as Guest62520 === kadams54 is now known as kadams54-away === Guest72440 is now known as rcj [17:24] lazyPower: ping [17:24] sebas5384: pong [17:25] http://juju-docs.readthedocs.org/en/latest/charm-tests.html - theres no mention of running tests via 'make test' [17:25] is that still supported? [17:26] stokachu: we're moving more to the bundletester side of testing, and that document warrants refactoring due to changes in our testing infra and tooling [17:26] yesterday did the first version of the mount feature of vju https://github.com/sebas5384/vju/issues/7 [17:26] now you can make: vju mount drupal/0:/var/www ./www [17:26] stokachu: that document is so far out of date it's worth purging it forever [17:26] stokachu: https://jujucharms.com/docs/ [17:27] or that ^ [17:27] and would it be mounted using sshfs :P [17:27] sebas5384: nice bro :) [17:27] I wonder why not just use the vagrant bind mount directive vs using sshfs though? === kadams54-away is now known as kadams54 [17:27] lazyPower: someday of these i'll make the .deb package hehe [17:28] because nfs haves some problems with permissions [17:28] and it stucks [17:28] sebas5384: are you guys doing a lot of NFS work? [17:28] we where trying to use here directly to the /var/lxc containers rootfs [17:29] we where using [17:29] till some people of the team stop using [17:29] marcoceppi: lazyPower: is tests/* a hard requirement on where tests are located? [17:29] stokachu: yes [17:29] stokachu: bundletester sniffs for tests in tests/* that are +x [17:30] stokachu: unit tests should be stored in unit_tests [17:30] if that's what you're getting at [17:30] im reading https://github.com/juju-solutions/bundletester [17:31] so theoretically if tests are written in ruby, executed with a Rakefile, i'd need to set virtualenv: false, and add a Makefile with specific targets? [17:31] lazyPower: I would prefer nfs actually, but we had some bad experiences hehe [17:32] and its a little tricky since you have to do an inverse nfs [17:32] sebas5384: if sshfs works for you, im' all for it :) I was more curious than anything [17:32] lazyPower: :) === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 [18:10] jamespage, if you get a sec https://code.launchpad.net/~gnuoy/charms/trusty/rabbitmq-server/vhost-fix/+merge/251627 . It's very slightly different to your solution === rektide__ is now known as rektide === 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 [22:44] Hi guys. Can anyone explain why this merge is telling me that it has failed automated testing, but the test results page is blank? https://code.launchpad.net/~paulgear/charms/trusty/memcached/correct-nagios-service/+merge/250716 [22:46] blahdeblah_: looking now [22:47] thanks lazyPower [22:48] The same thing happened for several different charm merges. [22:48] looking at CI there is a heap of things in the build queue [22:50] blahdeblah_: CI has changed recently and the bot hasn't caught up: http://reports.vapour.ws/charm-test-details/charm-bundle-test-11063-results [22:50] the URL format changed [22:51] aahhh, thats right, i forgot the routes were changed [22:51] there should be a redirect put in place while the bot gets updated [22:51] so historical links will also work [22:53] marcoceppi, lazyPower: Is there an easy way to see the test results of the revision from which my merge was branched? That was a fix for showstopping typos (the variables provided didn't match what is used in the template), and I find it hard to believe my merge is the cause of those fails. [22:54] blahdeblah_: http://reports.vapour.ws/charm-tests/memcached [23:00] Is there anyone who can help me explain why these changes are failing? I really don't think the trivial nature of the changes in https://code.launchpad.net/~paulgear/charms/trusty/memcached/allow-rsync-firewall/+merge/250717 and https://code.launchpad.net/~paulgear/charms/trusty/memcached/correct-nagios-service/+merge/250716 could cause the sort of failures the test suite is seeing. Plus I've confirmed that they're working [23:00] on a real stack where I can see the charm's full logs. [23:01] (Did that whole line come out? I just realised it might be a bit long) [23:01] blahdeblah_: the logs are being collected here, just not linked inthe interface, not sure where they are atm === kadams54 is now known as kadams54-away === kadams54-away is now known as kadams54 === kadams54 is now known as kadams54-away