re-org: Move openshift deployment to playbooks dir #84
Conversation
|
If you think that this change is ok, I will also send a PR for updating the docs. |
gbenhaim
force-pushed
the
re_org_part_1
branch
from
cec548d
to
a9ca557
Compare
| state: stopped | ||
| enabled: no | ||
| name: firewalld | ||
| failed_when: "result|failed and not 'Could not find the requested service' in result.msg" |
There was a problem hiding this comment.
will not result|failed be enough?
There was a problem hiding this comment.
No, in case that firewalld is not installed, the result|failed is true, but we don't care because firewalld is not installed.
| - name: Load openshift facts | ||
| openshift_facts: | ||
| role: common | ||
|
|
||
| - name: Set facts for docker-storage-setup | ||
| block: | ||
| - name: Detecting Operating System from ostree_booted |
|
|
||
| - name: install epel repository | ||
| yum: | ||
| name: https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm |
There was a problem hiding this comment.
and I'm not sure about the syntax here, see: http://docs.ansible.com/ansible/latest/yum_repository_module.html#examples
There was a problem hiding this comment.
The module that you have pasted should be used when you want to manually configure a repository. In our case, we use a release RPM.
|
|
||
| - name: install epel repository | ||
| yum: | ||
| name: https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm |
There was a problem hiding this comment.
and I'm not sure about the syntax here, see: http://docs.ansible.com/ansible/latest/yum_repository_module.html#examples
| - sos | ||
| - psacct | ||
| - docker | ||
|
|
There was a problem hiding this comment.
I deployed a cluster using the standard CentOS image from https://cloud.centos.org/centos/7/images/CentOS-7-x86_64-GenericCloud.qcow2
To get openshift-ansible to run, I had to install:
yum-utils ansible wget git net-tools bind-utils iptables-services bridge-utils bash-completion kexec-tools sos psacct docker sshpass NetworkManager centos-release-openshift-origin.noarch ceph-common docker centos-release-openshift iptables-utils NetworkManager atomic-openshift-utils
And ensure NetworkManager is started and enabled
There was a problem hiding this comment.
Prerequisites that don't appear here (such as NetworkManager, and centos-release-openshift) are being cared as part of the openshift-ansible modules.
There was a problem hiding this comment.
Didn't work for me until I manually installed them before running deploy_cluster.yml
| @@ -1,3 +1,3 @@ | |||
| - import_playbook: "{{ playbook_dir }}/deploy-with-{{ provider | default('lago') }}.yml" | |||
There was a problem hiding this comment.
I would move all playbooks (including control.yml) under playbooks dir
There was a problem hiding this comment.
I will be done in the following patches.
There was a problem hiding this comment.
I didn't want to do the entire re-org in one patch because it makes the code review overwhelming.
gbenhaim
force-pushed
the
re_org_part_1
branch
2 times, most recently
from
e143ee9
to
8d0fa57
Compare
| - "{{ playbook_dir }}/vars/{{ openshift_ver }}.yml" | ||
|
|
||
| - name: set SELinux to permissive mode | ||
| command: setenforce 1 |
There was a problem hiding this comment.
This is setting to enforcing mode. Did you mean setenforce 0? Or is the text wrong?
There was a problem hiding this comment.
Correct, I'll fix it.
| - name: set SELinux to permissive mode | ||
| command: setenforce 1 | ||
|
|
||
| - name: set SELinux to permissive mode under configuration file |
There was a problem hiding this comment.
Can you run this without become: true?
There was a problem hiding this comment.
You can see at the top of the playbook that we use remote_user: root.
If you think that we should use become, we can do it a different PR.
- Custom variables per Openshift version can be declared under playbooks/cluster/openshift/vars/ - Use a static version for Openshift 3.9 (v3.9.0-alpha.4) - Rename deploy-openshift.yml to config.yml. - Add the prerequisites of openshift to config.yml - there is no reason to keep them in a separate file. - Add a global variable file at vars/global_vars.yml Signed-off-by: gbenhaim <galbh2@gmail.com>
gbenhaim
force-pushed
the
re_org_part_1
branch
from
8d0fa57
to
f36b1f8
Compare
|
@rthallisey I've fixed the |
|
This change is not reflected in README, and made it outdated, please try to keep readme uptodate. |
Custom variables per Openshift version can be declared under
playbooks/cluster/openshift/vars/
Use a static version for Openshift 3.9 (v3.9.0-alpha.4)
Renamed deploy-openshift.yml to config.yml.
Added the prerequisites of openshift to config.yml - there is no
reason to keep them in a separate file.
Signed-off-by: gbenhaim galbh2@gmail.com