Description of problem:

When deploying logging and specifying gluster as the dynamic storage provider, generated PVC's are missing storage-class.


Version-Release number of the following components:

* openshift-ansible-3.11.123
* ansible-2.6


How reproducible: Consistently


Steps to Reproduce:
1. Deploy logging using glusterfs dynamic storage for persistence

# +++++++++++++++++++++++
# Cluster Logging
# ++++++++++++++++++++++++
openshift_logging_install_logging: true
openshift_logging_es_cluster_size: 3
openshift_logging_elasticsearch_storage_type: pvc
openshift_logging_es_pvc_storage_class_name: glusterfs-registry
openshift_logging_es_pvc_size: 20Gi
openshift_logging_es_pvc_dynamic: true
openshift_logging_es_number_of_shards: 3
openshift_logging_es_number_of_replicas: 1
openshift_logging_es_nodeselector: {node-role.kubernetes.io/infra: "true"}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


Actual results:

Playbook runs successfully. However, Elasticsearch PVC's are missing gluster storage-class reference, even though it was specified in the Ansible variables:

$ oc get pvc
NAME           STATUS    VOLUME    CAPACITY   ACCESS MODES   STORAGECLASS   AGE
logging-es-0   Pending                                                      1d
logging-es-1   Pending                                                      1d
logging-es-2   Pending                                                      1d


Expected results:

Elasticsearch PVC's would correctly reference gluster storage-class and bind automatically.