jonlangemak / ansible_kubernetes Goto Github PK

Hi there,
I am very new to kubernetes and tried following your post about using sensible to deploy a local cluster. Ansible playbook run completes with no errors, but when I try to get cluster-info or get pods, I get the following error:
$kubectl get pods
The connection to the server localhost:8080 was refused - did you specify the right host or port

Even if I do specify the --server option, I get a similar error.

thanks for your help and excellent posts

Hi Jon,
Thank you very much for the tutorial (http://www.dasblinkenlichten.com/getting-started-kubernetes-using-ansible/) , it Rocks... :) .
Today I tried to reproduce it on my lab and everything was passing perfect, until i tried to find Pod networking allocation with:
kubectl get nodes --output=jsonpath='{range .items[*]}{.status.addresses[?(@.type=="InternalIP")].address} {.spec.podCIDR} {"\n"}{end}' which failed.

I'm not sure but maybe files/systemd_templatez/kube-controller-manager.service
should be corrected with:
diff files/systemd_templatez/kube-controller-manager.service files/systemd_templatez/kube-controller-manager.service.orig

"+" --allocate-node-cidrs=true
"+" --cluster-cidr={{ cluster_info.cluster_node_cidr }} \

"-" --allocate-node-cidrs=false \

Regards,
Radoslav Mitov

Hi Jon,
just found that if the networking should work as in the example from http://www.dasblinkenlichten.com/getting-started-kubernetes-using-ansible
The following changes should be applied in files/systemd_templatez/kubelet.service
"+" --network-plugin=kubenet
"+" --non-masquerade-cidr={{ cluster_info.cluster_node_cidr }} \

"-" --network-plugin=cni

Regards,
Radoslav Mitov

You mention it - but its not actually set in vars/main.yml

Hi Jon,

I got the following error messages when running ansible-playbook on my Ubuntu-1 master node.

ben.le@labs-ubuntu-k8-001:~$ ansible-playbook k8s_isntall.yml --ask-become-pass
SUDO password:
ERROR! Unexpected Exception: sequence item 0: expected string, AnsibleMapping found
to see the full traceback, use -vvv

ben.le@labs-ubuntu-k8-001:~$ ansible-playbook k8s_isntall.yml --ask-become-pass -vvv
Using /etc/ansible/ansible.cfg as config file
SUDO password:
ERROR! Unexpected Exception: sequence item 0: expected string, AnsibleMapping found
the full traceback was:
Traceback (most recent call last):
File "/usr/bin/ansible-playbook", line 109, in
exit_code = cli.run()
File "/usr/lib/python2.7/dist-packages/ansible/cli/playbook.py", line 154, in run
results = pbex.run()
File "/usr/lib/python2.7/dist-packages/ansible/executor/playbook_executor.py", line 81, in run
pb = Playbook.load(playbook_path, variable_manager=self._variable_manager, loader=self._loader)
File "/usr/lib/python2.7/dist-packages/ansible/playbook/init.py", line 54, in load
pb._load_playbook_data(file_name=file_name, variable_manager=variable_manager)
File "/usr/lib/python2.7/dist-packages/ansible/playbook/init.py", line 99, in _load_playbook_data
entry_obj = Play.load(entry, variable_manager=variable_manager, loader=self._loader)
File "/usr/lib/python2.7/dist-packages/ansible/playbook/play.py", line 115, in load
data['name'] = ','.join(data['hosts'])
TypeError: sequence item 0: expected string, AnsibleMapping found

Any suggestions?

Hello,I'm trying this on openstack machines (ip's like 10.97.177....).
The installation has finished with no error but when i try to check the etcd cluster it's down
kubectl get componentstatus
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
etcd-0 Unhealthy Get https://192.168.0.49:2379/health: x509: certificate is valid for 10.11.12.1, 127.0.0.1, 10.97.177.xx, 10.97.177.xx, 10.97.177.xx, 10.97.177.xx, 10.97.177.xx, not 192.168.0.49
scheduler Healthy ok

What i've changed is:

1. user from "user" to "ubuntu"; this is a sudoer
2. IP's and machine names
3. on /etc/hosts on each machine i've added all the machines with names and the associated IP's

I don't understand why the etcd cluster it's down.
Please advice.
Thank You

hello
can you giveme a hint why i get this bad cert error ?

kubectl get componentstatus
NAME STATUS MESSAGE ERROR
scheduler Healthy ok
controller-manager Healthy ok
etcd-0 Unhealthy Get https://host01:2379/health: remote error: tls: bad certificate

I tried everything, but IO can't get out of the NotReady state.
However, kubectl get componentstatus is fine

I noticed some etcd errors. I also wonder if the older aAnsible cluster building posting still works with the new Calico centric GitHub changes
Is it OK that calicoctl.cfg in the endpoints use the IP in etcdendpoint ?
/etc/cni was not created but it is referred to in /etc/calico/calicoctl.cfg