Failing with ERROR Orchestration failed : 'ansible_user'

[Dannyb48]

2021-11-03 20:59:35,547 INFO Executing playbook : .tox/teflo/interop_qe_openshift/ansible/playbooks/setup_test_driver.yml
2021-11-03 20:59:35,626 INFO 10.0.162.107
2021-11-03 20:59:35,627 ERROR Orchestration failed : 'ansible_user'
2021-11-03 20:59:35,698 ERROR Failed to run orchestration Configure OpenShift test driver 
2021-11-03 20:59:35,699 ERROR Orchestration failed : Failed to perform  Configure OpenShift test driver

We have a use case where we don't use ansible_params on provisioning our resources but we've specifed the ansible_user in a group_vars file and ansible_ssh_private_key_file has been specified in the ansible.cfg. So ansible is able to run but teflo fails with the above error.

Reproduce with the following scenario

• Provision an asset without using ansible_params
• Use the asset in an orchestrate/execute task in the hosts: key
• Ansible playbook that templates the hosts key hosts: {{ hosts }}

When provision completes your master inventory file should look like

[dummy-group:children]
dummy-asset

[dummy-asset]
10.0.114.99

[dummy-asset:vars]

What I believe is happening is that the ssh_retry helper function is dependent on that groups vars but since it's empty we get python keyerror. https://github.com/RedHatQE/teflo/blob/develop/teflo/helpers.py#L861

teflo should be flexible to this use case and have other methods for validating connectivity without relying on that specific group_var

[dno-github]

Jira ticket CCITCARBON-388 has been created for this issue. Url : https://issues.redhat.com/browse/CCITCARBON-388

[rujutashinde]

closed with #212