=== frankban|afk is now known as frankban
[12:22] <hbogert_> Anyone experience with using maas/juju/kubernetes-core ? My node with kubernetes-worker went down. Now I'm trying to add a new kubernetes worker, simply by doing: `juju add-unit kubernetes-worker` but that never finishes. Looking on the node itself it fails to start kubelet and kube-proxy because it's missing files in /root/cdk/, e.g., kubeconfig and kubeproxyconfig
[13:10] <gunix> hey stokachu
[13:10] <gunix> are you around ?
[13:11] <gunix> we had a chat a month ago a month ago about juju, ubuntu, rancher, kubernetes and so on
[13:11] <gunix> you told me to try everything else on the market and after that come back to you when i am dissapointed :D
[14:13] <hbogert_> gunix: well I've been there and here as well, and I'm still disappointed :P
[14:44] <knobby> hbogert_: I use kubernetes-core and added a worker just as you described, but everything was cool and went well. Was the juju status stable?
[14:50] <catbus_> Juju bootstrapping on a maas cloud stops at 'Running machine configuration script...' for long. What does it do there? Running juju command with --debug doesn't give much info.
[14:55] <gunix> hbogert_: this should be done manually
[14:55] <gunix> i mean in theory it is possible, but there is not example online
[15:27] <knobby> catbus_: juju debug-log --replay might help
=== freyes__ is now known as freyes
[16:01] <hbogert_> gunix: what is manually in this case?
[16:08] <hbogert_> knobby: yeah it used to work for me as well. But can you confirm that you alsof went from zero to one worker?
[16:27] <Stormmore> morning juju world!
[16:36] <agprado> Morning Sotormore
[17:00] <jose-phillips> hi
[17:00] <jose-phillips> question
[17:00] <jose-phillips> is posible when i use manual deployment
[17:01] <jose-phillips> create a container on a remote machine with the network configuration
[17:01] <jose-phillips> ?
=== frankban is now known as frankban|afk
[17:13] <knobby> hbogert_: a good distinction. I replaced a worker, but I still had workers when I did it.
[19:25] <Budgie^Smore> so we use the Gerrit Trigger plugin, is there a way of configuring that in a Jenkinsfile?
[19:32] <gunix> hbogert_: manually is just manually installing every service and manually configuring every network interface
[19:32] <Budgie^Smore> for the record, I am trying to use a Multibranch Declarative Pipeline that should trigger off any changeset / commit
[19:53] <catbus_> knobby: juju debug-log needs to reach juju api server for info, at that point, the juju api server isn't up yet.
[19:55] <catbus_> I am running juju 2.3.1 on xenial. I have never seen this 'Running machine configuration script...' before.
[21:59] <jose-phillips> hi im installing openstack-dashboard
[21:59] <jose-phillips> with juju but the themes are not working
[21:59] <jose-phillips> because im getting 403
[21:59] <jose-phillips> im installling on xenial using openstack-origin cloud:xenial-pike
[22:17] <kwmonroe> jose-phillips: i haven't used openstack-dashboard, but could it be you're hitting the wrong url?  from the readme, looks like you'll need to use http(s)://<ip>/horizon.
[22:22] <kwmonroe> Budgie^Smore: generically, there seems to be some 'lightweight checkout' option that looks related with the gerrit trigger plugin: https://stackoverflow.com/questions/46204276/how-to-execute-jenkinsfile-with-gerrit-trigger-plugin-for-changeset-in-jenkins
[22:22] <kwmonroe> ^^ or are you asking for a way to configure plugins via the jenkins charm deployed with juju?
[22:25] <koaps> hey guys, does anyone know the channel for the openstack charms?
[22:25] <kwmonroe> koaps: @openstack-charms
[22:25] <koaps> thanks
[22:25] <kwmonroe> np
[22:25] <koaps> ceph is giving me issues
[22:25] <kwmonroe> not possible!
[22:25] <kwmonroe> ;)
[22:25] <koaps> seems like the install is broken
[22:26] <koaps> trys to use the python modules before they are installed
[22:26] <Budgie^Smore> kwmonroe, no actually posted that question in the wrong channel :-/ I want to write my triggers into the Jenkinsfile as I am using Multibranch Pipelining
[22:27] <Budgie^Smore> thanks for responding though
[22:27] <kwmonroe> heh, no worries.  good luck with your big words!
[22:30] <Budgie^Smore> Infra-as-Code FTW ;-)
[22:33] <kwmonroe> jose-phillips: you disconnected at the totally wrong time :)  i sent this shortly after you dropped re your dashboard issue:
[22:33] <kwmonroe> jose-phillips: i haven't used openstack-dashboard, but could it be you're hitting the wrong url?  from the readme, looks like you'll need to use http(s)://<ip>/horizon.
[22:34] <kwmonroe> also jose-phillips, for the container question, you can deploy stuff to containers on remote juju units like this: juju deploy ubuntu --to lxd:0
[22:34] <kwmonroe> ^^ that deploys the ubuntu charm to a container on machine 0.
=== JoseeAntonioR is now known as jose