Bug 1420703

Summary: Document changes to take over exisiting cluster with rebased ceph ansible
Product: [Red Hat Storage] Red Hat Ceph Storage Reporter: Tejas <tchandra>
Component: DocumentationAssignee: Aron Gunn <agunn>
Status: CLOSED CURRENTRELEASE QA Contact: ceph-qe-bugs <ceph-qe-bugs>
Severity: high Docs Contact:
Priority: unspecified    
Version: 2.2CC: aschoen, asriram, hnallurv, kdreyer, tchandra
Target Milestone: rc   
Target Release: 2.2   
Hardware: Unspecified   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2017-03-21 23:49:56 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Tejas 2017-02-09 10:53:19 UTC
Description of problem:
       In this doument:
https://access.qa.redhat.com/documentation/en/red-hat-ceph-storage/2/single/installation-guide-for-red-hat-enterprise-linux/#taking_over_an_existing_cluster

The take-over-existing-cluster.yml script fails to execute since the group_vars is in a new location now.

We need to create a symlink from /etc/ansible/group_vars to /usr/share/ceph-ansible.


Also step3:
group_vars is located at /atc/ansible/group_vars.

Comment 3 Tejas 2017-02-10 05:52:44 UTC
I think it maybe a bit confusing for the users to run the playbook from the /etc/ansible directory.

Andrew,
  Would this be ok?

Thanks,
Tejas

Comment 4 Andrew Schoen 2017-02-10 12:19:38 UTC
Tejas,

The user should not run the playbook from /etc/ansible, this would mean that the playbook would not be able to find the roles defined in /usr/share/ceph-ansible.

Your group_vars need to be defined in /etc/ansible/group_vars and then /etc/ansible/group_vars should be symlinked to /usr/share/ceph-ansible. The playbook run must then happen in /usr/share/ceph-ansible, with the take-over-existing-cluster.yml playbook moved up a directory from infrastructure-playbooks.

Does that match the workaround you tested?

Thanks,
Andrew

Comment 8 Tejas 2017-02-13 03:43:34 UTC
The doc looks good. Thanks Andrew and Aron.