| bdx | but it needs the full path | 00:00 |
|---|---|---|
| valeech | bdx: ok, so I just need to feed it a config.yaml, right? | 00:00 |
| bdx | valeech: juju deploy cs:~blake-rouse/maas-region-4 | 00:00 |
| bdx | valeech: correct | 00:00 |
| arosales | valeech: ubuntu deploys looks good | 00:01 |
| arosales | bdx: is the latest blake-rouse or maas-maintainers | 00:01 |
| valeech | arosales: so the status unknown is ok/expected? | 00:02 |
| bdx | valeech, arosales: the ha functionality isn't there yet as the docs portray, but if you just want 1 rack-controller + 1 region-controller it works great | 00:02 |
| valeech | bdx: curses! I was hoping for the HA pieces :) | 00:02 |
| bdx | arosales: as far as I know, yes - I have cs:~blake-rouse/maas-region-4 deployed | 00:03 |
| arosales | valeech: workload state looks to be "unknown" which is ok | 00:03 |
| valeech | arosales: ok | 00:03 |
| valeech | how do I go about removing those stranded mariadb and wordpress apps? | 00:03 |
| bdx | valeech: from what I hear the ha portion will be getting some cycles again soon | 00:03 |
| arosales | valeech: in juju 2.0 beta9 the new syntax is "juju remove-application mariadb" | 00:04 |
| arosales | and juju remove-application wordpress | 00:04 |
| valeech | bdx: cool. How does one go about contributing to something like that? I am new to this, but would love to give back | 00:05 |
| valeech | arosales: I have run both of those and the apps persist, they’re like extended family, they just won’t leave! | 00:06 |
| bdx | valeech: start here -> https://jujucharms.com/docs/devel/developer-getting-started | 00:06 |
| arosales | bdx, re ha no available, the docs @ https://maas.ubuntu.com/docs2.0/ha.html are very misleading then :-( | 00:06 |
| valeech | arosales: Yes, the docs are misleading :) | 00:06 |
| * arosales filing a bug on that | 00:07 | |
| bdx | valeech: then go here -> https://jujucharms.com/docs/devel/developer-layer-example | 00:07 |
| arosales | valeech: huh remove-application doesn't want to work for you . . . | 00:07 |
| arosales | let me see if there is a --force | 00:07 |
| arosales | valeech: were you interested in maas ha contributions or charm contributions? | 00:08 |
| arosales | or both :-) | 00:08 |
| valeech | arosales: both! I plan on using these products heavily | 00:08 |
| bdx | valeech: I'm with you there .... I've got the hardware already set aside for it :-) | 00:09 |
| arosales | valeech: great to hear, the docs that bdx pointed are are excellent. I would also add https://jujucharms.com/docs/devel/developer-getting-started and suggest to join the mail list I pointed at earlier | 00:10 |
| valeech | bdx: me too! trying to get this lab up and running so the my product managers can push it into production before it’s ready! | 00:11 |
| valeech | arosales: I joined the mailing list! I’ll spend some time on the docs as soon as I can get a functioning concept of all of this :) | 00:12 |
| bdx | valeech: the docs will aid you in getting the concept down too though:-) | 00:12 |
| valeech | bdx: good point | 00:13 |
| arosales | valeech: for maas contributions see http://maas.ubuntu.com/docs/about.html#contributing | 00:13 |
| arosales | valeech: but your manual setup is running | 00:14 |
| arosales | just not a lot of xenial charms yet | 00:14 |
| arosales | suggest to spin up some trusty machines | 00:14 |
| arosales | or if you have maas configured just point juju at that :) | 00:14 |
| valeech | arosales: true. I was trying to use juju to deploy maas ha :) Then I would point juju at maas. I guess I need to get maas ha up manually. | 00:15 |
| valeech | would it make sense to deploy 2 maas region controllers with juju and then do the work externally to ha them togeher? or is that more work than its worth? | 00:16 |
| arosales | valeech: try: juju deploy --to 3 cs:~maas-maintainers/trusty/maas-region-3 | 00:17 |
| valeech | Added charm "cs:~maas-maintainers/trusty/maas-region-3" to the model. | 00:17 |
| valeech | Deploying charm "cs:~maas-maintainers/trusty/maas-region-3" with the user specified series "trusty". | 00:17 |
| valeech | ERROR cannot add application "maas-region": cannot deploy to machine 3: series does not match | 00:17 |
| arosales | ugh | 00:18 |
| arosales | xenial | 00:18 |
| arosales | same issue as wordpress, no xenail charm for maas-region charm | 00:18 |
| arosales | valeech: you can try appending --force to see if deploy maas-region on xenail | 00:20 |
| arosales | but that is uncharted territory | 00:20 |
| valeech | arosales: sure | 00:20 |
| arosales | as the charm maintainer may have to make special accommodations for xenail especially given systemd | 00:21 |
| arosales | juju deploy --to 3 cs:~maas-maintainers/trusty/maas-region-3 --force | 00:21 |
| arosales | may get you a deploy, but results may vary if it works on a xenial machine | 00:21 |
| arosales | valeech: may be better to just spin up a trusty machine and go from there | 00:21 |
| valeech | arosales: yeah, perhaps. I’ll give that a shot and report back on results. | 00:22 |
| arosales | valeech: good luck | 00:23 |
| valeech | the —force won’t take because between the time you asked to try the mass-region-3 and it failed and asking me to try the —force I install maas-region-4 and it took. I didn’t give it a yaml file though, I just wanted to see what would hapen. | 00:23 |
| * arosales goes to file bug against maas docs | 00:23 | |
| valeech | but the remove-application function isn’t working so I can’t remove it and try the maas-region-3 —force install :) | 00:24 |
| valeech | arosales bdx lazyPower thanks for all the help! | 00:24 |
| lazyPower | valeech anytime :) | 00:25 |
| arosales | valeech: I will also file a bug on remove-application | 00:25 |
| arosales | that should remove the application from the model | 00:25 |
| arosales | valeech: could you pastebin me "juju remove-application wordpress --debug" | 00:25 |
| valeech | arosales: sounds good. Is there any logs or data I should gather? | 00:26 |
| valeech | sure | 00:26 |
| arosales | juju debug-log --replay may be interesting | 00:27 |
| arosales | but the --debug with remove-application I think will have the interesting bits | 00:27 |
| valeech | http://pastebin.com/xyjcMz9m | 00:27 |
| arosales | valeech: thanks | 00:30 |
| valeech | arosales: np! | 00:31 |
| arosales | valeech: the heavy handed approach would be to destroy the model and create a new one once you are done experimenting given remove-application doesn't seem to be happy atm. | 00:31 |
| valeech | arosales: I tried that too and it just runs forever waiting for the service to stop… | 00:32 |
| arosales | valeech: nice :-/ | 00:32 |
| valeech | :) | 00:32 |
| arosales | and no --force on destory-model | 00:33 |
| arosales | there is on the controller, but at that point its a new bootstrap | 00:34 |
| valeech | This is neat: http://pastebin.com/iSnUi1ez | 00:35 |
| arosales | valeech: ah you may have to remove the units first, given they are manually added | 00:36 |
| arosales | valeech: try "juju remove-unit 4" | 00:38 |
| arosales | note if you have ubuntu on that still you may want to at least try "juju remove-application ubuntu" | 00:38 |
| thumper | 4 what? | 00:39 |
| valeech | error: invalid unit name "4" | 00:39 |
| arosales | valeech: sorry try remove-machine | 00:40 |
| arosales | so "juju remove-machine 4" | 00:40 |
| arosales | thumper: you just trying to cause trouble :-) | 00:40 |
| valeech | that worked | 00:41 |
| * thumper lobs hand grenades from the safety of another country | 00:41 | |
| arosales | thumper: remove-application not being helpful tonight | 00:41 |
| thumper | well that sucks | 00:41 |
| arosales | thumper: http://pastebin.com/xyjcMz9m | 00:41 |
| arosales | valeech: try the same with 3 | 00:42 |
| valeech | that worked too for 3 | 00:42 |
| valeech | but the apps are still there :) | 00:42 |
| arosales | valeech: what does "juju destroy-model default" return now | 00:42 |
| valeech | ok, now destroy-model worked! | 00:43 |
| thumper | arosales: ooo... | 00:43 |
| thumper | 2016-06-17 00:26:42 DEBUG juju.api apiclient.go:520 health ping failed: connection is shut down | 00:43 |
| thumper | that line right there | 00:43 |
| arosales | thumper: timeout? | 00:43 |
| thumper | uh | 00:43 |
| thumper | maybe not | 00:43 |
| arosales | valeech: ok suggest to start a new model, and spin up some trusty images | 00:43 |
| thumper | debug on the client when things aren't working don't tell us much | 00:43 |
| thumper | the debug log of the controller is much more useful | 00:43 |
| arosales | thumper: well deploy ubuntu worked | 00:43 |
| arosales | thumper: but couldn't remove wordpress | 00:44 |
| thumper | if you have debug logging on | 00:44 |
| thumper | grab the output by doing this: | 00:44 |
| thumper | juju debug-log -m controller --replay --no-tail > foo.log | 00:45 |
| arosales | valeech: also if you want some run time on AWS and are interested in contributing to juju charms please feel free to enroll in https://developer.juju.solutions/ | 00:45 |
| thumper | that will get the apiserver calls and internal errors or issues with removal | 00:45 |
| arosales | valeech: that will give access to more "machines" if you want to focus on just the charm piece if you don't have a lot of spare hardware in your maas to test against. | 00:46 |
| arosales | thumper: already destroyed the controller :-/ | 00:46 |
| arosales | thumper: I'll try to recreate on AWS, but this issue was on manual provider so I may try Digital ocean to reproduce | 00:47 |
| thumper | ack | 00:47 |
| arosales | thumper: but " juju debug-log -m controller --replay --no-tail > foo.log" noted if I am able to reproduce | 00:47 |
| thumper | yeah, that is a handy command | 00:47 |
| valeech | arosales and thumper I can spin it back up again | 00:47 |
| * thumper is in the bowels of juju ripping at the bad code he left several years ago | 00:48 | |
| arosales | thumper: make sure you get _all_ the bad code then | 00:48 |
| arosales | valeech: if you have time, sure. If not I can also try to produce | 00:48 |
| valeech | arosales: sure. It doesn’t take long to setup | 00:49 |
| arosales | valeech: thanks | 00:50 |
| thumper | arosales: no chance... | 00:50 |
| thumper | but I'll get the cancer called fslock out | 00:50 |
| arosales | thumper: well at least there is that :-) | 00:51 |
| * arosales filed https://bugs.launchpad.net/maas/+bug/1593516 in reference to the incorrect syntax for deploys | 01:31 | |
| mup | Bug #1593516: docs2.0/ha juju deploy maas-region not correct syntax <MAAS:New> <https://launchpad.net/bugs/1593516> | 01:31 |
| Yash | nova-compute/10 error idle 2.0-beta7 2 10.100.100.200 hook failed: "install" | 07:25 |
| Yash | How to solve this? | 07:25 |
| Yash | Please help. | 07:25 |
| admcleod | Yash: juju debug-log -i nova-compute/10 -n 30 | 07:35 |
| Yash | admcleod: yash@ugocloud:~$ juju debug-log -i nova-compute/10 -n 30 ERROR invalid entity name or password yash@ugocloud:~$ juju debug-log -i nova-compute/9 -n 30 ERROR invalid entity name or password yash@ugocloud:~$ juju debug-log -i nova-compute/11 -n 30 ERROR invalid entity name or password | 07:45 |
| Yash | ERROR invalid entity name or password | 07:45 |
| admcleod | Yash: do you still have this 'upgrade in progress' message somewhere? | 07:47 |
| admcleod | Yash: also i forget, which juju version are you on? (juju --version) | 07:50 |
| admcleod | Yash: also can you pastebin the full results of 'juju status --format=yaml' | 07:51 |
| Yash | juju 2.0 7 beta | 07:56 |
| Yash | nova-compute/9 waiting executing 2.0-beta7 2 10.100.100.200 Incomplete relations: image, storage-backend neutron-openvswitch/36 error idle 2.0-beta7 10.100.100.200 hook failed: "install" ntp/73 error idle 2.0-beta7 10.100.100.200 hook failed: "install" | 07:56 |
| admcleod | Yash: hmm | 08:21 |
| admcleod | Yash: how did you resolve the issue you were having yesterday? with connecting to the ip / port? | 08:23 |
| admcleod | Yash: also can you upgrade juju to the latest version? | 08:25 |
| Yash | admcleod : I restarted machine after login using container exec command | 08:31 |
| Yash | 2.0-beta7-xenial-amd64 | 08:31 |
| Yash | How to upgrade? | 08:31 |
| Yash | using ppa? | 08:31 |
| Yash | devel one? | 08:31 |
| admcleod | yes | 08:31 |
| admcleod | beta8 is available | 08:32 |
| Yash | ppa:juju/devel | 08:34 |
| Yash | right? | 08:34 |
| Yash | And do I need to bootstarp again? | 08:34 |
| Yash | babbageclunk: welcome :) | 08:35 |
| Yash | *bootstrap | 08:35 |
| Yash | amd64 2.0-beta9-0ubuntu1~16.04 | 08:36 |
| Yash | 2.0-beta9 | 08:36 |
| Yash | its downloading 9 | 08:36 |
| Yash | is that ok? | 08:36 |
| Yash | admcleod: please confirm | 08:37 |
| admcleod_ | Yash: im sorry i seem to have disconnected and only saw you ask 'devel?' to which i responded 'yes', was there something else? | 08:38 |
| Yash | ppa:juju/devel | 08:38 |
| Yash | amd64 2.0-beta9-0ubuntu1~16.04 | 08:38 |
| Yash | is that ok? | 08:38 |
| Yash | np :) | 08:39 |
| Yash | Also do I need to bootstrap again? | 08:39 |
| admcleod_ | beta9 -even better :) | 08:40 |
| admcleod_ | you should bootstrap again, i think there is something wrong with the current deployment you have there | 08:40 |
| admcleod_ | Yash: ^ | 08:41 |
| Yash | ok | 08:41 |
| Yash | Thanks let me try with beta 9 | 08:41 |
| Yash | Thank you | 08:41 |
| Yash | Is there anytime when 2.0 stable will release? | 08:42 |
| === verterok` is now known as verterok | ||
| === admcleod_ is now known as admcleod | ||
| admcleod | Yash: i was told, but ive forgotten and dont want to guess so ive asked | 08:44 |
| admcleod | yash: (the date) | 08:44 |
| Yash | which date... I don't get it? | 08:45 |
| admcleod | Yash: the release date for 2.0 stable | 08:45 |
| Yash | ohh..my bad ok :) | 08:45 |
| Yash | admcleod: Thanks | 08:46 |
| admcleod | Yash: let us know if re-bootstrapping with the new version fixes things please? | 08:48 |
| Yash | admcleod: ok..I'm doing it now. I will keep you posted. | 08:49 |
| Yash | admcleod: is there any tester position? I can join. ;) | 08:49 |
| admcleod | Yash: haha im not sure, maybe :) | 08:50 |
| Yash | admcleod: please ask and let me join your team :) ..lol | 08:51 |
| admcleod | Yash: :D | 08:54 |
| === jacekn_ is now known as jacekn | ||
| Yash1 | https://api.jujucharms.com/identity/v1/idp/usso/callback?waitid=a593011ca1d5b53b399df26515435a53&openid.assoc_handle=%7BHMAC-SHA1%7D%7B5763c2e4%7D%7BCAq6Vg%3D%3D%7D&openid.claimed_id=https%3A%2F%2Flogin.ubuntu.com%2F%2Bid%2FhpYfrJt&openid.identity=https%3A%2F%2Flogin.ubuntu.com%2F%2Bid%2FhpYfrJt&openid.mode=id_res&openid.ns=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0&openid.ns.sreg=http%3A%2F%2Fopenid.net%2Fextensions%2Fsreg%2F1.1&openid.op_endpoint=https%3A%2F%2F | 09:29 |
| Yash1 | {"message":"cannot get user details for \"https://login.ubuntu.com/+id/hpYfrJt\": not found: not found","code":"forbidden"} | 09:29 |
| Yash1 | juju authentication problem | 09:29 |
| jamespage | cholcombe, hey - can you join #openstack-charms as well pls | 11:55 |
| jamespage | cholcombe, I tidied https://review.openstack.org/#/c/328374/ in the interest of getting that cleared for the landscape folk | 11:56 |
| codepython777 | anyone here? | 12:06 |
| === frankban is now known as frankban|afk | ||
| jamespage | bdx, hey - could you respond to my re-licensing request on openstack-dev pls (ideally want all contributors to agree before we switch to Apache 2.0) | 13:27 |
| === fginther is now known as fginther` | ||
| valeech | is there a way with a manual cloud to prevent juju from removing the machine after all services are removed from it? | 17:39 |
| lazyPower | valeech - there is a setting to tweak the machine reaping, but i dont recall what it is offhand | 17:40 |
| gennadiy | hi everybody, we have got issue with juju-gui-130. it doesn't display changes. i have found error in dev console - "Unknown delta type: actionInfo" | 17:49 |
| valeech | Thx lazyPower | 17:55 |
| lazyPower | valeech - i'm still looking | 17:55 |
| lazyPower | thats buried in one of our FAQ documents i think | 17:55 |
| catbus1 | Hi, JUJU 2.0 will create a linux bridge interface as "br-eth0" on the node with eth0 interface. Is there a way to stop juju from creating that bridge interface? | 17:55 |
| dooferlad | catbus1: As it stands the bridge is always created. | 18:00 |
| dooferlad | catbus1: is it actually causing problems, or is it just untidy? | 18:00 |
| catbus1 | dooferlad: I will find out if it's causing problems from this charm partner. | 18:02 |
| bdx | lazyPower, valeech: https://www.jujucharms.com/docs/devel/howto-harvesting | 18:03 |
| valeech | bdx: w00t! thanks. One more question, where is the environments.yaml in juju 2.0??? | 18:10 |
| magicalt1out | gone away | 18:10 |
| magicalt1out | never to be seen again | 18:11 |
| magicalt1out | banished, if you will | 18:11 |
| valeech | haha I just read that | 18:11 |
| valeech | ok so reading the 2.0 docs, I set the provisioner-harvest-mode to none with this command: juju set-model-config provisioner-harvest-mode=none | 18:40 |
| valeech | I then spun up some services on a machine, verified they worked and then destroyed them. After the services were removed juju stopped the machine and them removed it. Am I missing something? | 18:41 |
| lazyPower | valeech - it may be prudent to send that to the list - juju@lists.ubuntu.com, that way the core devs can get a chance ot see it and respond, as a lot of them are over in europe and way past EOD for them. | 18:59 |
| valeech | lazyPower got it! | 18:59 |
| === tvansteenburgh1 is now known as tvansteenburgh | ||
| lazyPower | bdx yo | 19:28 |
| bdx | yo | 19:47 |
| bdx | lazyPower: writing documentation like a madman :-) | 19:50 |
| lazyPower | bdx - http://146.148.77.23/app/kibana#/dashboard/Dockerbeat-Dashboard | 19:50 |
| bdx | lazyPower: no | 19:51 |
| bdx | redic | 19:51 |
| bdx | ulous | 19:51 |
| lazyPower | awww | 19:51 |
| lazyPower | he told me no ;_; | 19:51 |
| bdx | :-) | 19:51 |
| lazyPower | it just keeps on giving man. | 19:51 |
| lazyPower | now if only it supported triggers and notifications like prometheus, and we'd have a killer ops app on our hands. | 19:52 |
| bdx | no doubt ... did dockerbeat pre-exist, or did you build that? | 19:53 |
| lazyPower | its a community beat :) | 19:54 |
| lazyPower | https://github.com/Ingensi/dockerbeat | 19:54 |
| bdx | I have a feeling the beats stack is going to gain traction real fast .... its sooo useful and lightweight. Once people start to understand how simple and powerful it is .... game over | 20:00 |
| lazyPower | interesting prediction :) | 20:08 |
| lazyPower | I hope it does happen though, its a great stack to work with. Super simple and straight forward | 20:08 |
| bdx | thedac: nice mp for dnsha | 21:12 |
| thedac | bdx: thanks. It is coming along. Should be in our 16.07 release | 21:15 |
| thedac | bdx: follow along if you care to https://review.openstack.org/#/q/topic:dnsha | 21:16 |
| bdx | thedac: thats great. thanks, I will | 21:41 |
| bdx | I'm documenting like a monster as of late .... what do you guys think about these top level juju admin categories -> http://imghub.org/image/HHh4 | 22:02 |
| valeech | I have made a lot of progress on my juju understanding last night and today. Thanks for all of the help! | 22:14 |
| valeech | I am now stuck and can’t seem to find an answer. | 22:14 |
| valeech | I blew my whole setup away ealier and started with 3 fresh Trusty machines. | 22:15 |
| valeech | I have installed juju 2.0 beta 9 on the first, created a new model and add a manual cloud with 2 machines. I need to setup a postgres cluster. So I ran juju deploy cs:trusty/postgresql-101 —to 0. After postgres came all the way up I then ran juju add-unit postgres —to 1. Postgres installed on the other node and everything looked great until I get “Failed to clone postgresql/0” | 22:18 |
| valeech | That failure shows up in juju status | 22:19 |
| valeech | It changes between that and “(leader-settings-changed) Failed to clone postgresql/0” | 22:19 |
| valeech | How do I troubleshoot this one? | 22:20 |
| valeech | Fixed it. | 22:22 |
| valeech | I added my machines to maas using host names not IPs. The two machines weren’t able to resolve each other’s names so postgres could replicate. | 22:23 |
| valeech | I love this channel! you all are so helpful :) | 22:25 |
| arosales | valeech: good to see your making progress | 22:44 |
| valeech | arosales thx! | 22:53 |
| lazyPower | valeech tell your friends about us ^_^ | 22:55 |
| valeech | lazyPower That’s why I’m here. I need friends :) | 22:57 |
| lazyPower | :) We welcome you with open arms my friend | 23:00 |
| valeech | So I have juju 2.0 beta 9 working well on trusty. I would now like to delpoy maas 2.0. If I look at the apt-cache I see the maas versions are old. Is there a way I can specify what version of maas I want loaded at deploy time? Do I have to add the maas dev repositories to the 2 trusty machines so that when the charm installs it gets the latest version? | 23:16 |
| bdx | valeech: 'sudo add-apt-repository ppa:maas-maintainers/experimental3' will get you the daily | 23:20 |
| bdx | valeech: then you must 'sudo apt-get update && sudo apt-get install --upgrade maas' | 23:21 |
| valeech | bdx: I only do the upgrade if I have maas installed, right? I don’t have maas installed yet. I am using a manual cloud to deploy maas and then I will connect juju to that cloud once installed | 23:22 |
| arosales | have a good weekend folks | 23:26 |
| valeech | arosales you too | 23:27 |
| === cmars` is now known as cmars | ||
| valeech | bdx: apparently the maas-maintainers don’t have a trusty experimental release. | 23:41 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!