| === natefinch-afk is now known as natefinch | ||
| === menn0 is now known as menn0-afk | ||
| === menn0-afk is now known as menn0 | ||
| kjackal_ | Hello Juju World! | 07:15 |
|---|---|---|
| === frankban|afk is now known as frankban | ||
| === jamespag` is now known as jamespage | ||
| === gnuoy` is now known as gnuoy | ||
| === ant_ is now known as Guest45561 | ||
| === arturt__ is now known as arturt | ||
| KpuCko | hello is it that any way to change the juju controller ip address? I've installed juju with lxdbr0 random generated network as a bridge, now i prefere to use manually created bridge to access my vm's in easy ways, so now when i type juju status, juju cannot connect to juju controller | 08:43 |
| === dpm_ is now known as dpm | ||
| magicaltrout | ah one of those quality days where you try and start ec2 nodes and they just fail in error :) | 09:39 |
| magicaltrout | anybody around who know what on earth you do in beta16 now upload-tools seems to have vanished? | 09:47 |
| admcleod_ | magicaltrout: sync-tools instead? | 09:48 |
| magicaltrout | what the f**k | 09:49 |
| magicaltrout | you need to tell your overlords to stop being so pedantic | 09:49 |
| admcleod_ | im just guessing. i remembered upload-something was obsolete but that appears to be upload-series | 09:49 |
| admcleod_ | also, i cant really help since im mainly using 1.25 | 09:52 |
| magicaltrout | you make me sad admcleod_ | 09:52 |
| magicaltrout | get with the times | 09:52 |
| admcleod_ | thats why im here ;) are you going to strata ny? | 09:52 |
| magicaltrout | i'm not. I'm doing spain and pasadena twice before the end of the year | 09:53 |
| magicaltrout | any more and the mrs might kill me | 09:53 |
| admcleod_ | also are you saying --upload-tools is missing from bootstrap? | 09:53 |
| magicaltrout | that is correct | 09:53 |
| magicaltrout | it was there in 15 | 09:53 |
| magicaltrout | sync-tools does look suspiciously like its what its moved to | 09:54 |
| magicaltrout | but it did used to live in the bootstrap | 09:54 |
| babbageclunk | magicaltrout: There's a thread about it here: https://lists.ubuntu.com/archives/juju-dev/2016-August/005893.html | 10:05 |
| babbageclunk | I guess it was only discussed on the juju-dev list because people assumed that it was only used by devs. Oops! | 10:06 |
| magicaltrout | thanks babbageclunk it probably ended up in a release email somewhere I failed to read as well | 10:07 |
| magicaltrout | no problem | 10:08 |
| rock_ | Hi. I deployed OpenStack on LXD using https://github.com/openstack-charmers/openstack-on-lxd. I have created "cinder-storagedriver" charm . I pushed the created charm to public charm market place(charm store). So Using JUJU GUI when i was trying to deploy "cinder-storagedriver" charm by adding relation to "cinder" it was throwing an error. | 11:20 |
| rock_ | ERROR: Relation biarca-openstack:juju-info to cinder:juju-info: cannot add relation "biarca-openstack:juju-info cinder:juju-info" : principal and subordinate applications' series must match. | 11:20 |
| rock_ | Manually , When I was deployed our charm by taking from charm store then it is able to deploy and able to add relation to cinder. | 11:20 |
| rock_ | As part of debug, I deployed juju-gui as $juju deploy cs:juju-gui-134 on our setup. But it was showing series as trusty even we have choosen xenial. | 11:21 |
| rock_ | I am thinking this is JUJU-GUI issue. | 11:22 |
| rock_ | (or) Is it some other issue? Please anyone respond to this issue. | 11:23 |
| kjackal | Hi rock_ the series of the deployed charms should be shown on your juju status | 11:39 |
| kjackal | can you make sure you have biarca-openstack and cinder charms on the same series? | 11:39 |
| kjackal | rock_: you are on juju 2.0? | 11:40 |
| kjackal | you could choose the series with the --series flag at deploy time | 11:41 |
| rock_ | kjackal: Hi. biarca-openstack and cinder charms are in the same series(Xenial). Yes juju 2.0 | 11:42 |
| rock_ | But I can able to deploy my charm through juju cli by taking from charm store. | 11:43 |
| rock_ | And I can able to add relation to cinder successfully. | 11:43 |
| rock_ | When I did from Juju CLI everything working fine. | 11:44 |
| D4RKS1D3 | Hi, I want to change the mtu of juju-br0, I changed the value of the associated interface but when i restart the machine change other time to 1500. | 11:45 |
| D4RKS1D3 | Someone knows how to change always? the interface in /etc/networks/interfaces is in manual and i can not change in this config file. | 11:46 |
| rock_ | kajackal: But from JUJU gui, I am not able to add relation to cinder . It was giving ERROR: Relation biarca-openstack:juju-info to cinder:juju-info: cannot add relation "biarca-openstack:juju-info cinder:juju-info" : principal and subordinate applications' series must match. | 11:47 |
| kjackal | rock_: where is your cinder-storagedriver charm? | 11:47 |
| kjackal | rock_: can I see it? | 11:47 |
| rock_ | Yes. Type "biarca" in charm store. | 11:48 |
| kjackal | rock_: this one: cinder-storagedriver | 11:49 |
| kjackal | rock_: this one: https://jujucharms.com/u/siva9296/biarca/0 | 11:49 |
| kjackal | rock_: Looks ok to me. You could ask at #juju-gui and #openstack-charm . It might be a bug | 11:51 |
| rock_ | kjackal: ok. Thank you. | 11:54 |
| magicaltrout | T-1hour | 12:00 |
| magicaltrout | still working on my demo \o/ | 12:00 |
| PCdude | hé magicaltrout , hru? | 12:09 |
| magicaltrout | lol | 12:09 |
| magicaltrout | tired... still hacking | 12:09 |
| magicaltrout | you know | 12:09 |
| magicaltrout | the usual before a 50 minute presentation ;) | 12:09 |
| PCdude | haha, 50 minutes and u are wasting time here ;) , good time managment | 12:10 |
| PCdude | but anyway, good luck! | 12:10 |
| magicaltrout | thanks :P | 12:10 |
| marcoceppi | good luck magicaltrout | 12:10 |
| magicaltrout | aye... then i need to come back to the speakers room and start tomorrows :P | 12:11 |
| magicaltrout | which you guys claim to want to film and slap on youtube... never good | 12:11 |
| magicaltrout | woop thanks for that dump cmars | 12:31 |
| magicaltrout | it started working | 12:31 |
| cmars | magicaltrout, awesome! | 12:31 |
| magicaltrout | don't think i got my original config.yaml stanza correct | 12:31 |
| admcleod_ | magicaltrout: good luck | 12:48 |
| lazyPower | magicaltrout knock em downnnn | 12:53 |
| admcleod_ | break someones leg | 12:54 |
| marcoceppi | smash 'em mash 'em stick 'em in a stew! | 12:56 |
| marcoceppi | or, whatever | 12:56 |
| admcleod_ | ill take 2 whatevers | 12:58 |
| lazyPower | petevg Thanks for https://github.com/juju/docs/pull/1317 | 12:59 |
| lazyPower | petevg looks good from my house, and it gets us some proper representation of what charmers should be thinking about as they organize and structure their modules. | 13:00 |
| kjackal | hey petevg, are you there? | 13:02 |
| magicaltrout | all done | 13:45 |
| magicaltrout | not a bad turn out, must have been 30 odd | 13:45 |
| admcleod_ | did you tell any jokes? | 13:49 |
| magicaltrout | I don't believe i did | 13:50 |
| magicaltrout | i did ask who'd heard of Juju | 13:51 |
| magicaltrout | and a whole 2 hands went up | 13:51 |
| Randleman | Nice | 13:52 |
| magicaltrout | now i need to write my talk and demo for tomorrow :P | 13:53 |
| jrwren | magicaltrout: were they your two best friends, or were they complete strangers? | 13:57 |
| magicaltrout | pfft | 13:57 |
| magicaltrout | actually the only guy i knew in the room was a guy that hacks on Karaf in his spare time | 13:58 |
| magicaltrout | not got many friends at this one =/ | 13:58 |
| magicaltrout | anyway! in reality whilst we all wish everyone knew of and used juju at talks its probably better to talk to a room of people who don't know what it is | 13:59 |
| magicaltrout | hell, it was in my talk title and still 30 people showed up ;) | 13:59 |
| jrwren | magicaltrout: well, that is pretty good then! | 13:59 |
| magicaltrout | aye | 14:00 |
| magicaltrout | and appears 30odd are likely to turn up tomorrow to my Business Intelligence juju talk | 14:01 |
| magicaltrout | so at least its getting faces in front of the software | 14:01 |
| magicaltrout | appears to be standing room only at Bluefin tomorrow night | 14:05 |
| magicaltrout | better come up with something half decent then | 14:05 |
| petevg | kjackal: I am here, though apparently I'm not paying attention to my IRC client ... | 14:13 |
| kjackal | petevg: :) | 14:14 |
| petevg | kjackal: I just pushed a new revision of the Zookeeper charm. If those timeouts are legit, it should fix them. | 14:14 |
| kjackal | I wanted to ask you something on the zookeeper charm | 14:14 |
| petevg | kjackal: ask away. | 14:14 |
| petevg | lazyPower: thank you for the positive feedback on that testing PR :-) | 14:15 |
| kjackal | I see that now the zookeeper waits for a service restart when new units are added | 14:15 |
| petevg | kjackal: It does. We did that because the Zookeeper docs advised that it's not always stable after removing or adding units. | 14:16 |
| kjackal | petevg: have you tested this behaviour with the HA bundles that expect ZK to have 3 units? | 14:16 |
| petevg | kjackal: I haven't. That is a very good test to add to the bundle tests. | 14:16 |
| kjackal | petevg: Yes I remember that, I am not sure how this behavior will play along with the hadoop processing and the rest of the bundles | 14:16 |
| petevg | kjackal: it probably will function well. The Zookeepers don't stop working as they're waiting for the restart. They just won't pick up new peers until you do the restart. | 14:17 |
| kjackal | Hm... so we need the restart to have Zookeeper HA | 14:18 |
| petevg | kjackal: if you specify three zookeepers in your bundle, I believe that this works around the issue; Zookeeper will only wait for a restart if it has already setup a quorum that then changes. | 14:19 |
| petevg | Let me verify that ... | 14:19 |
| kjackal | I have tested "juju deploy zookeeper -n 3" and the restart is needed there | 14:20 |
| valeech | Sorry guys. I just registered for the charmers summit. I hope I don’t hold back too many people :) | 14:20 |
| kjackal | petevg: And what happens upon restart? Do the ZK records of all peers get merged? | 14:21 |
| petevg | kjackal: I'm not sure what Zookeeper does internally. I just know that the docs advise that you do a rolling restart after you've changed the number of peers. | 14:21 |
| petevg | ... if you don't, in the Zookeeper version that we're using, a job can get left without a peer to attach to. | 14:22 |
| petevg | kjackal: darn it. You're right about the peers needing a restart even on setup. I bet that the timing of things changed when we ripped out the openjdk relation :-/ | 14:27 |
| jamespage | marcoceppi, hey - I need to add uosci-testing-bot to the charmers team to support our automated push/publish process for promulgated charms - is that OK with you? | 14:45 |
| marcoceppi | jamespage: you don't actually need to do that | 14:47 |
| marcoceppi | jamespage: we can just give uosci-testing-bot write access to the charms that it needs to publish | 14:47 |
| jamespage | beisner, ^^ | 14:47 |
| jamespage | marcoceppi, oh yeah that's a good idea | 14:48 |
| jamespage | beisner, ^^ | 14:48 |
| * jamespage faceplatns | 14:48 | |
| marcoceppi | beisner jamespage just give me a list of entities and I'll give you guys the commands, I'm sure a few of your are still ~charmers ;) | 14:48 |
| beisner | jamespage, marcoceppi - hrm yah, looks like perms are good for some of them, but not all, for the bot user. | 14:48 |
| marcoceppi | beisner jamespage tl;dr: charm grant <charm> --channel stable --acl write uosci-testing-bot | 14:49 |
| beisner | ack marcoceppi thank you | 14:50 |
| valeech | running juju2.0beta15 when trying to deploy to lxd container, I get the following error: 'failed to ensure LXD image: unable to get LXD image for ubuntu-xenial: The requested image couldn''t be found.' If I ssh into the machine and run “sudo -E lxc launch ubuntu:16.04” it starts just fine. Here is the yaml file I am using to deploy: http://pastebin.com/hzx6icGk | 14:54 |
| magicaltrout | itsn't it some naming thing | 14:55 |
| magicaltrout | isn't | 14:55 |
| magicaltrout | like | 14:55 |
| magicaltrout | you download the image and tag it with a custom name, in your case ubuntu-xenial | 14:55 |
| magicaltrout | its been a while, my memory might be failing me | 14:55 |
| petevg | kjackal: on that failing zookeeper test run, are there tracebacks in the juju logs? | 14:56 |
| valeech | magicaltrout: That makes sense. I am learning so I am cutting and pasting from different sources and I bet that is the case. | 14:57 |
| kjackal | petevg: I am afraid not | 14:57 |
| magicaltrout | valeech: | 14:57 |
| magicaltrout | if you run | 14:57 |
| magicaltrout | lxc image list | 14:57 |
| magicaltrout | you'll see the alias column on the left | 14:58 |
| magicaltrout | thats what it uses | 14:58 |
| valeech | magicaltrout: I see that. How do you set that on newly created machines from maas? | 14:59 |
| magicaltrout | erm | 14:59 |
| magicaltrout | well you do it when you do an image pull | 14:59 |
| magicaltrout | does that happen in MAAS or is it magic? | 14:59 |
| valeech | magicaltrout: It is very possible there is a gap in my understanding, but doesn’t maas deploy a vanilla image on the machine then hand it over to juju? At that point, juju would need to pull down whatever images it needs for lxd? | 15:01 |
| magicaltrout | ah yeah, now i get you | 15:01 |
| magicaltrout | so this is during juju bootstrap? | 15:02 |
| magicaltrout | or post bootstrap? | 15:02 |
| magicaltrout | like, if you do a juju bootstrap it should just dump it on the metal | 15:02 |
| magicaltrout | so no lxd involved to my knowledge | 15:02 |
| valeech | Correct. I have a new juju model setup and I am trying to deploy my first charm to a container. So juju has to get a machine from maas to work with | 15:02 |
| magicaltrout | k | 15:03 |
| valeech | that part works. when juju tries to setup the container to push the app to the container fails to start. | 15:03 |
| magicaltrout | well, i asked around the other day and my understanding was that whilst it needs an image, thats not a lxd image | 15:03 |
| magicaltrout | okay, so you have MAAS spin up a node, then you try and push a LXD charm to it? | 15:04 |
| valeech | Kinda. juju asks maas to spin up a node which will become machine 0 in juju. Then juju attempts to initialize a container on machine 0 which would become machine 0/lxd/0 in juju. Then juju would push the charm to that container. | 15:05 |
| magicaltrout | i'm sufficently confused, I tried to get MAAS running on virtualbox the other day and it failed, but i did ask around and was told that if you "juju deploy mycharm" it would go onto the bare metal not a lxd container | 15:07 |
| skay | does mojo with with juju2 yet? | 15:07 |
| lazyPower | valeech - there has been a ton of work around that. this is juju 2.0-beta16 right? | 15:07 |
| lazyPower | skay - last i heard it was still 1.25 only | 15:07 |
| valeech | That is correct. It will deploy to bare metal. I am trying to get the next step which is deploy inside a container on the bare metal. | 15:08 |
| skay | aw | 15:08 |
| valeech | lazyPower: it is beta 15. I can load beta 16. | 15:08 |
| lazyPower | shouldn't make too big of a difference. | 15:08 |
| lazyPower | do you have logs from the container thats failing to start? | 15:09 |
| magicaltrout | valeech: i'd update anyway | 15:09 |
| magicaltrout | because the charm store will make your instance explode as soon as you get going :) | 15:09 |
| valeech | ok, I will update. as far as container logs, how would I go about getting those? | 15:09 |
| valeech | haha | 15:10 |
| babbageclunk | valeech - I've seen that as well - I think it's resolved by beta16. | 15:21 |
| valeech | babbageclunk: great. I am upgrading to beta16 now but having some trouble... | 15:22 |
| babbageclunk | valeech: trouble? | 15:35 |
| valeech | babbageclunk: I don’t think I upgraded properly. I did an apt-get upgrade juju which upgrade the juju client host to beta16. Then in the controller model I did juju upgrade-juju. It said is was upgrading but when I would issue a juju status I would get this error: “ERROR invalid entity name or password (unauthorized access)” Maybe I didn’t wait long enough? | 15:39 |
| babbageclunk | valeech: hmm - I'm not really familiar with using upgrade-juju. Are you in a position where you can rebootstrap instead? | 15:40 |
| valeech | babbageclunk: That is what I did :) It is adding my redundant controllers right now from juju enable-ha. | 15:41 |
| babbageclunk | valeech: ok cool - so things are looking alright at the moment? | 15:42 |
| valeech | babageclunk: everything looks good. About to deploy this charm to the container and see what happens. | 15:44 |
| babbageclunk | valeech: fingers crossed. | 15:46 |
| valeech | babbageclunk: it spun up the container!! Now it’s installing the charm :) | 15:53 |
| * babbageclunk dances | 15:54 | |
| valeech | babbageclunk: Woohoo! mysql/0 active idle 0/lxd/0 10.0.129.59 Unit is ready | 15:56 |
| valeech | many thanks to babbageclunk, magicaltrout and lazyPower | 15:56 |
| jacekn | is threre a known problem with "charm build" not working any more? I'm getting "ERROR unrecognized command: charm build" when I try to compose charm using charm 2.1.1-0ubuntu1 | 15:58 |
| lazyPower | nice valeech | 15:59 |
| lazyPower | jacekn - try charm-tools from the snap channels | 16:00 |
| jacekn | lazyPower: ah that could be it, thanks | 16:00 |
| magicaltrout | jacekn: also it got broken into a bunch of binaries | 16:24 |
| magicaltrout | charm-build | 16:24 |
| magicaltrout | for example | 16:24 |
| jacekn | magicaltrout: tha's fine as long as dependencies and/or sensible error messages are there | 16:27 |
| magicaltrout | sensible error messages? | 16:27 |
| magicaltrout | this is juju | 16:27 |
| valeech | ok, next challenge, how do I set the lxd container to use a specific bridge? when juju created it inside of machine 0, lxdbr0 was added to machine 0 with what looks like a random IP segment. Is there a mechanism that I can tell juju to use a specific interface for containers so I can make sure the containers end up inthe correct IP segment? | 16:31 |
| PCdude | magicaltrout: how did it go? | 16:37 |
| magicaltrout | not bad PCdude | 16:37 |
| magicaltrout | 30 odd people and i streched it out to 40 mins | 16:37 |
| magicaltrout | so i guess it was a success | 16:37 |
| PCdude | sounds like a succes | 16:37 |
| magicaltrout | just tomorrow to do now | 16:37 |
| PCdude | whats tomorrow? | 16:38 |
| magicaltrout | business intelligence and juju talk for a user group in london | 16:38 |
| PCdude | ah ok, u keep urself busy haha | 16:38 |
| magicaltrout | initially both talks were on the same day | 16:39 |
| magicaltrout | one in ams one in london | 16:39 |
| PCdude | well it is possible, but convenient is something else | 16:39 |
| PCdude | magicaltrout: let me ask u, sometimes questions on stackexchange takes years before they get answered and on IRC is sometimes hard too to find answers, were do u go when u have questions? | 16:42 |
| magicaltrout | IRC | 16:42 |
| magicaltrout | mailing lists | 16:42 |
| magicaltrout | i do a lot of apache stuff | 16:43 |
| magicaltrout | thats all mailing lists | 16:43 |
| PCdude | I never used mailings lists, how does that work exactly? | 16:43 |
| magicaltrout | you sign up | 16:44 |
| magicaltrout | and send emails | 16:44 |
| magicaltrout | ... mailing.... lists ;) | 16:44 |
| babbageclunk | valeech: I'm not sure - I think the best people to help you have dropped off for the day, sorry. | 16:44 |
| magicaltrout | PCdude: juju relies heavily on IRC and mailing lists | 16:47 |
| magicaltrout | if people are offline on IRC | 16:47 |
| magicaltrout | you generally post to the list and get an answer there | 16:47 |
| magicaltrout | depends on the project though, not all use mailing lists but many do | 16:47 |
| PCdude | magicaltrout: ah ok, so if I am correct, I add my e-mail address there and when I have a question that e-mail is send to anyone in that list? | 16:48 |
| PCdude | and anyone can repond to that mail | 16:48 |
| magicaltrout | https://lists.ubuntu.com/mailman/listinfo/juju | 16:48 |
| magicaltrout | you subscribe there | 16:48 |
| magicaltrout | then you will recieve mails sent to the list | 16:48 |
| magicaltrout | and you can send to the list | 16:48 |
| PCdude | but on what speed are those e-mails send? I mean a regular forum goes so fast that means 5 mails a second | 16:49 |
| magicaltrout | few a day | 16:50 |
| jose | it's a discussion list. not extremely heavy volume, but not as low as an announcement only list. | 16:50 |
| magicaltrout | PCdude: i get about 800 a day from apache projects, you'll be fine ;) | 16:51 |
| PCdude | ah ok, good to know, lets not use my primary mail address haha, I just wanna get closer to the heat :) | 16:51 |
| holocron | hi all, quick question i hope.. I already have some machines that are added to my model and I'd like to specify them sa the targets in a juju bundle | 16:51 |
| magicaltrout | i just have a bunch of filters | 16:52 |
| holocron | the machines were manually added: | 16:52 |
| holocron | whenever i attempt to deploy a bundle, it goes and tries to make 4 more machines, but i'd rather it just used the 4 i already have | 16:53 |
| magicaltrout | holocron: you probably need to use --to:... in the deploy stanza | 16:53 |
| magicaltrout | to tell it where to go | 16:53 |
| holocron | magicaltrout: sure I can direct charm deployment to the machines | 16:54 |
| holocron | but i want to do it from a bundle | 16:54 |
| magicaltrout | i suspect you're out of luck | 16:54 |
| magicaltrout | but lazyPower might know better | 16:54 |
| * lazyPower reads scrollback | 16:55 | |
| lazyPower | bundle machine #'s dont necessarily correspond to machine #'s already in the model | 16:55 |
| lazyPower | i'm not certain how to do that other than to use maas tagging holocron, but that assumes you're using the maas juju provider | 16:55 |
| holocron | unfortunately i'm not lazyPower, I'm using the lxc provider | 16:56 |
| === frankban is now known as frankban|afk | ||
| holocron | lazyPower, can you tell me where some of this might be documented better? https://jujucharms.com/docs/devel/charms-bundles is not very telling | 17:17 |
| holocron | in particular, i'm looking for the specification on placement directives | 17:21 |
| lazyPower | holocron - that seems bug worthy holocron. If you dont mind opening a bug at http://github.com/juju/docs/issues with the specifics you're looking for I can make sure we get it addressed | 17:31 |
| holocron | lazyPower okay.. i think i answered my own question after reading the page i linked for a 2nd time | 17:32 |
| holocron | there's simply no way to use a pre-defined machine directly from a bundle unless it's MAAS controlled? | 17:32 |
| lazyPower | holocron - can you pastebin me your bundle? | 17:33 |
| kwmonroe | petevg: i notice you have a convention with branch names at https://github.com/juju-solutions/layer-apache-bigtop-base. i'd like to adopt it.. what are your keys? i see bug/foo and feature/foo. anything else? | 17:35 |
| petevg | kwmonroe: I've just been using bug and feature. If you have something that doesn't fit, feel free to add a word, and let me know about it, so that I can use it, too :-) | 17:36 |
| holocron | lazyPower http://pastebin.com/getn0Vzm | 17:36 |
| holocron | that isn't importable due to "The following errors occurred while retrieving bundle changes: placement "3" refers to a machine not defined in this bundle.." etc etc | 17:36 |
| holocron | i have machines 1-4 defined to juju already | 17:37 |
| holocron | but there's no way to indicate them from the bundle it seems | 17:37 |
| cory_fu | kwmonroe, petevg: If there is a corresponding bug / issue number, what do you think about including it in the branch name so that it's easy to associate them? Something like "feature/123-foo"? | 17:37 |
| petevg | cory_fu: that's officially part of the convention that I've borrowed, though I'm not always great at remembering. I'll try to be better about it :-) | 17:39 |
| cory_fu | Oh, I didn't realize. Carry on, then | 17:39 |
| kwmonroe | cory_fu: i think the 123 won't be useful in practice. how would you know if it was a gh issue, lp bug, or jira? | 17:40 |
| cory_fu | kwmonroe: Hrm. What about feature/gh-123/foo? | 17:41 |
| kwmonroe | i suppose you could say bug/gh123-foo, feature/lp-bug... heh.. yeah cory_fu | 17:41 |
| lazyPower | holocron - i think this is a regression. Bugs would def. be welcome around this so we can get eyes on it. | 17:41 |
| holocron | okay lazyPower, i'll open a bug .. LP or github? | 17:41 |
| cory_fu | kwmonroe, petevg: Actually, is knowing that something is a feature vs bug useful info? | 17:41 |
| lazyPower | https://launchpad.net/juju/ would be preferrable, if its closed fairly quickly with instructions we can port those into the docs. | 17:42 |
| kwmonroe | i think so cory_fu.. i (will) tend to look at bug branches before features.. i think. | 17:42 |
| cory_fu | Also, should we worry about cleaning up our branches after they're merged? | 17:42 |
| cory_fu | kwmonroe: Fair enough | 17:43 |
| lazyPower | I'm a +1 for cleaning up branches after merging | 17:43 |
| lazyPower | but nobody asked me :) | 17:43 |
| cory_fu | lazyPower: We appreciate your input all the same, you special snowflake, you. ;) | 17:43 |
| kwmonroe | +1 for lazyPower being a good person. and yes cory_fu, i'm all about deleting branches, whether they're merged or not. | 17:44 |
| cory_fu | lol | 17:44 |
| kwmonroe | ;) | 17:44 |
| cory_fu | I should create a bot that deletes kwmonroe's branches as soon as he pushes them | 17:44 |
| petevg | I'm merciless about cleaning up branches locally. | 17:44 |
| petevg | I would be down with cleaning them up remotely, too. | 17:44 |
| lazyPower | @cory_fu have you seen: https://github.com/jfrazelle/ghb0t | 17:44 |
| cory_fu | I wish GH had a "Merge and delete this branch" button | 17:45 |
| petevg | That actually fixes the issue where you squash a branch to get it ready to merge, and then you confuse someone who has the branch checked out locally. | 17:45 |
| petevg | If the branch is just gone, the error is much less confusing. | 17:45 |
| petevg | cory_fu: the delete button does appear right after you click the merge button ... but yeah, it would be nice to do it in one go. | 17:45 |
| cory_fu | lazyPower: Nice | 17:45 |
| kwmonroe | ok cory_fu petevg kja-tab-tab-tab, so here's our new thing: branches are bug/lp-123/<short-desc>, feature/gh-123/<short-desc>, bug/jira-123/<short-desc> and the merger deletes the branch. sound good? | 17:46 |
| cory_fu | +1 | 17:46 |
| lazyPower | LOL @ tab-tab-tab | 17:46 |
| petevg | Ooh. Extra forward slashes. Makes it look all neat and organized. | 17:46 |
| lazyPower | RIP | 17:46 |
| petevg | +1 :-) | 17:46 |
| kwmonroe | aight cory_fu petevg, wiki step 2 updated: https://github.com/juju-solutions/bigdata-community/wiki/Bigtop-Patch-Process | 17:53 |
| petevg | Nice! | 17:53 |
| holocron | https://bugs.launchpad.net/juju/+bug/1618996 | 17:56 |
| mup | Bug #1618996: unable to specify manually added machines from bundle.yaml <juju:New> <https://launchpad.net/bugs/1618996> | 17:56 |
| lazyPower | jcastro ping | 18:00 |
| Prabakaran | hello Team, I am not getting any proper info on the error when i am using make lint command on the build charm. I meant, make lint command is not showing from which file i am getting these error. Could someone please explain me on this? | 18:03 |
| === rmcall_ is now known as rmcall | ||
| kwmonroe | Prabakaran: will you pastebin your charm's Makefile and 'make lint' output? | 18:06 |
| Prabakaran | ya sure kwmonroe | 18:06 |
| lazyPower | holocron thanks, i'll bring this up | 18:06 |
| Prabakaran | kwmonroe: i have pasted here http://pastebin.ubuntu.com/23117014/ | 18:07 |
| kwmonroe | ah, that's an easy one Prabakaran :) your machine is out of space.. what does "df -h" show? | 18:08 |
| Prabakaran | kwmonroe: link is here http://pastebin.ubuntu.com/23117020/ | 18:09 |
| kwmonroe | yup Prabakaran.. your / filesystem at /dev/sda3 is 100% full | 18:09 |
| Prabakaran | kwmonroe: you meant all these error are because of memory of the machine | 18:10 |
| kwmonroe | yup Prabakaran. the "apt-get install $apt_prereqs" failed because it didn't have enough space to write data to /var/lib/apt. | 18:11 |
| magicaltrout | sad times | 18:12 |
| Prabakaran | thanks kwmonroe | 18:12 |
| Prabakaran | i wil clear my machine space and try this out | 18:12 |
| kwmonroe | np Prabakaran. if you've downloaded any charms from magicaltrout, i'd suggest deleting those first. | 18:13 |
| Prabakaran | kwmonroe: magicaltrout means? | 18:13 |
| kwmonroe | heh, sorry Prabakaran.. it was a joke. magicaltrout is the guy in this channel that causes so many problems. i like to blame all things on him first ;) | 18:14 |
| magicaltrout | lol | 18:14 |
| Prabakaran | thats no problem :) | 18:14 |
| magicaltrout | kwmonroe: you're so mean | 18:58 |
| magicaltrout | you make me sad | 18:58 |
| magicaltrout | in other todos... graylog, we should get graylog onto juju | 18:58 |
| kwmonroe | pretty sure it's the weather that's making you sad magicaltrout. you need southern california stat. | 18:59 |
| magicaltrout | hehe | 18:59 |
| magicaltrout | our summer has been very nice | 18:59 |
| magicaltrout | compared to the last few years | 18:59 |
| magicaltrout | but i do need southern california | 19:00 |
| magicaltrout | i even have a car god help you all | 19:00 |
| x58 | stub: Are you stub42 on Github? | 19:27 |
| PCdude | magicaltrout: for some weird reason is the MAAS error gone now :D | 19:32 |
| magicaltrout | magic | 19:34 |
| magicaltrout | told you PCdude we all have those | 19:34 |
| PCdude | magicaltrout: best guess is something with the networkcard, when installing with only one card it fails, but thats it and frankly I dont care anymore it works now | 19:35 |
| x58 | Is there a helper in charmhelpers for restarting a systemd service? | 19:51 |
| lazyPower | x58 - i think thats a missing feature, however there is a juju-reboot hook tool that will safely reboot a unit in the context of juju + containers that may be running on the host (lxd based) | 20:17 |
| lazyPower | x58 https://jujucharms.com/docs/stable/reference-hook-tools#juju-reboot | 20:23 |
| x58 | lazyPower: I found charmhelpers.core.services.base which has service_restart which is a wrapper around charmhelpers.core.host.service_restart | 20:38 |
| x58 | or something like that. | 20:38 |
| x58 | Anyway, my @when_file_changed hook correctly restarts my service. | 20:38 |
| x58 | As for other accomplishments... published my second charm today =) | 20:39 |
| x58 | https://jujucharms.com/u/bertjwregeer/snmpd | 20:39 |
| lazyPower | nice! ^5 | 20:39 |
| x58 | ^5 | 20:39 |
| lazyPower | oo your readme is from teh apt layer | 20:40 |
| x58 | marcoceppi: https://jujucharms.com/u/bertjwregeer/snmpd (another charm built in anger ;-)) | 20:40 |
| x58 | Refresh a couple of times.. | 20:40 |
| x58 | It should show my .rst | 20:40 |
| x58 | which isn't rendered ;-) | 20:41 |
| x58 | Unless I need to make cs:~bertjwregeer/snmpd-2 public? Although I am not sure that should be required since I can see it... | 20:41 |
| marcoceppi | x58: weird the RST isn't rendering, this might be a bug in our GUI | 20:44 |
| x58 | marcoceppi: It is. I have an open bug for it on Github | 20:44 |
| x58 | marcoceppi: https://github.com/CanonicalLtd/jujucharms.com/issues/313 | 20:46 |
| marcoceppi | x58: cool thanks! | 20:46 |
| marcoceppi | charm looks awesome though | 20:46 |
| x58 | Thanks :-) | 20:47 |
| x58 | Ultimately ended up being fairly simple, but the docs aren't all that nice to get to that point ;-) | 20:47 |
| x58 | Also, charm proof is angry that I don't have any "provides", but I am not sure what my charm would provide exactly. | 20:48 |
| kwmonroe | cory_fu: halp!! | 20:55 |
| cory_fu | kwmonroe: What up? | 20:55 |
| kwmonroe | cory_fu: http://paste.ubuntu.com/23117537/ | 20:55 |
| kwmonroe | that's pip 8.1.2 in charmbox.. is it me pip or me python-distutils-extra? | 20:55 |
| cory_fu | kwmonroe: Do you have an old version of the layer:apt? | 20:57 |
| x58 | marcoceppi: BTW, even with layer.yaml in layer:apt having ignore: ['README.md'] in my layer.yaml won't let charm build succeed. | 20:58 |
| cory_fu | kwmonroe: Or a modified version? There's no wheelhouse.txt file in the apt layer | 20:59 |
| === natefinch is now known as natefinch-afk | ||
| kwmonroe | yeah cory_fu.. i think you're right | 21:00 |
| kwmonroe | (rebuilding) | 21:00 |
| kwmonroe | all good kwmonroe | 21:01 |
| kwmonroe | heh | 21:01 |
| kwmonroe | derp | 21:01 |
| kwmonroe | all good cory_fu | 21:01 |
| cory_fu | :) | 21:01 |
| kwmonroe | you were right.. this time ;) | 21:01 |
| lazyPower | kwmonroe - you can embed github repos in a wheelhouse like the openstack-layer has done - https://github.com/openstack/charm-layer-openstack/blob/master/wheelhouse.txt | 21:04 |
| lazyPower | just fyi | 21:04 |
| kwmonroe | gracias lazyPower! | 21:06 |
| lazyPower | anytime homie | 21:06 |
| kwmonroe | two Ms in hommie, hommie. | 21:06 |
| x58 | Cue "you are not my hommie, pal" :P | 21:09 |
| kwmonroe | lol | 21:14 |
| lazyPower | dont call me buddy, guy! | 21:14 |
| x58 | Abbott and Costello. My favourite still has to be whose on first. | 21:15 |
| lazyPower | thats better than my weak reference to an old adam sandler routine before he lost any sense of credibility :) | 21:18 |
| x58 | lol | 21:19 |
| x58 | For those nostalgic: https://www.youtube.com/watch?v=kTcRRaXV-fg | 21:20 |
| catbus1 | Hi, is agenda available for the upcoming juju charmers summit? | 22:31 |
| marcoceppi | catbus1: yes, a tentative one is | 22:33 |
| catbus1 | marcoceppi: Can I see it now and share with an attendee from a technology partner company? Or is it too early to share now? | 22:36 |
| marcoceppi | catbus1: nope, we mailed the Juju mailing lits on it | 22:42 |
| marcoceppi | catbus1: https://lists.ubuntu.com/archives/juju/2016-August/007743.html | 22:43 |
| catbus1 | marcoceppi: thank you! | 22:43 |
| bdx | marcoceppi: I seem to be hitting some kind of limit on my aws charm-dev account -> http://paste.ubuntu.com/23118016/ | 22:48 |
| bdx | marcoceppi: can you clear out my instances ... or something .. | 22:49 |
| bdx | pls | 22:49 |
| rajith | hi while bootstraping on juju2 beta 16 getting error : ERROR cmd supercommand.go:458 creating LXD client: Get https://10.35.77.1:8443/1.0: Unable to connect to: 10.35.77.1:8443 ERROR failed to bootstrap model: subprocess encountered error code 1 | 23:21 |
| bdx | rajith: you need to run `sudo lxd init` again, and configure lxd for api access | 23:37 |
| bdx | rajith: to my knowledge, this means you need to first wipe your existing lxd containers | 23:38 |
| bdx | rajith: and images | 23:39 |
| bdx | mbruzek: http://paste.ubuntu.com/23118148/ | 23:49 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!