Description of problem: Attempt deployment of two CNS clusters associated with a single OCP cluster managed by a single heketi instance/pod -- based on the description of parameter glusterfs_cluster this should be supported: "...a single heketi service is expected to manage multiple distinct clusters. NOTE: For natively-hosted clusters, all pods will be in the same OpenShift namespace " [1] Official docs [2] also indicate management of multiple clusters is supported: "Heketi allows administrators to add and remove storage capacity by managing either a single or multiple Red Hat Gluster Storage clusters." The limitation with openshift-ansible deployment is that both clusters will reside in the same namespace/project, which is acceptable. Deployment of CNS is succcessful, but only a single cluster is configured, despite hostvars configured to deploy multiple CNS clusters out of box. Version-Release number of the following components: atomic-openshift-utils-3.7.42-1.git.2.9ee4e71.el7.noarch yum info atomic-openshift-utils Loaded plugins: product-id, search-disabled-repos, subscription-manager Installed Packages Name : atomic-openshift-utils Arch : noarch Version : 3.7.42 Release : 1.git.2.9ee4e71.el7 Size : 157 k Repo : installed From repo : rhel-7-server-ose-3.7-rpms Summary : Atomic OpenShift Utilities URL : https://github.com/openshift/openshift-ansible License : ASL 2.0 Description : Atomic OpenShift Utilities includes : - atomic-openshift-installer : - other utilities How reproducible: Always Steps to Reproduce: 1. Configure inventory with 2 sets of 3 gluster nodes, with each set containing hostvars for a unique glusterfs_cluster ID (e.g. glusterfs_cluster=1, glusterfs_cluster=2) 2. Execute openshift-ansible config.yml Actual results: No errors but all glusterfs nodes are associated with a single cluster -- observed via 'heketi-cli cluster list' Expected results: 'heketi-cli cluster list' shows two unique clusters provisioned. Additional info: Inventory attached (private)
Johnny, can you verify if this is still an issue in 3.10?
@Prasanth, could cns-qe team verify if this is still an issue in 3.10?
Closing this bug due to OpenShift versions < 3.11 exiting the full support phase. If this bug is still impacting releases >= 3.11 and your use cases, please reopen with those additional details and update the target release where you are impacted on a fully supported version of OpenShift.