Description of problem: Unable to add RHS2.0+ nodes: 1. To a "Gluster Cluster" inside a Data Center having Compatibility version set to 3.2 and 2. To a "Gluster Cluster" with Compatibility version set to 3.2 inside a Data Center having Compatibility version set to 3.1 Version-Release number of selected component (if applicable): From RHEL6.4 engine: rhevm-3.2.0-10.10.beta1.el6 From RHS2.0+ node: ISO Used: RHS-2.0-20130219.3-RHS-x86_64-DVD1.iso # cat /etc/issue Red Hat Storage release 2.0 for On-Premise Kernel \r on an \m # rpm -qa |grep vdsm vdsm-cli-4.9.6-19.el6rhs.noarch vdsm-python-4.9.6-19.el6rhs.x86_64 vdsm-4.9.6-19.el6rhs.x86_64 vdsm-gluster-4.9.6-19.el6rhs.noarch vdsm-reg-4.9.6-19.el6rhs.noarch # rpm -qa |grep glusterfs glusterfs-3.3.0.6rhs-2.el6rhs.x86_64 glusterfs-devel-3.3.0.6rhs-2.el6rhs.x86_64 glusterfs-fuse-3.3.0.6rhs-2.el6rhs.x86_64 org.apache.hadoop.fs.glusterfs-glusterfs-0.20.2_0.2-1.noarch glusterfs-rdma-3.3.0.6rhs-2.el6rhs.x86_64 glusterfs-geo-replication-3.3.0.6rhs-2.el6rhs.x86_64 glusterfs-server-3.3.0.6rhs-2.el6rhs.x86_64 glusterfs-debuginfo-3.3.0.6rhs-2.el6rhs.x86_64 # rhn-channel -l rhel-x86_64-server-6-rhs-2.0 rhel-x86_64-server-6.2.z rhel-x86_64-server-sfs-6.2.z How reproducible: Always Steps to Reproduce: Case A: 1. Create a Data Center with "Compatibility Version" set to 3.2 2. Create a "Enable Gluster Service" Cluster with "Compatibility Version" set to 3.2 (Here only 3.2 can be selected) 3. Add the latest RHS2.0+ ISO (RHS-2.0-20130219.3-RHS-x86_64-DVD1.iso) installed Host to the gluster cluster created above. 4. Watch the Events tab and engine logs. Case B: 1. Create a Data Center with "Compatibility Version" set to 3.1 2. Create a "Enable Gluster Service" Cluster with "Compatibility Version" set to 3.2 (here, currently we are allowed to set the compatibility version to either 3.1 or 3.2, bug???) 3. Add the latest RHS2.0+ ISO (RHS-2.0-20130219.3-RHS-x86_64-DVD1.iso) installed Host to the gluster cluster created above. 4. Watch the Events tab and engine logs. Actual results: Add Host fails with the following error in the "Events" tab in both the cases: --- Host RHSHost is compatible with versions (3.0,3.1) and cannot join Cluster My32DC32GlusterCluster which is set to version 3.2. Host RHSNode is compatible with versions (3.0,3.1) and cannot join Cluster My31DC32GlusterCluster which is set to version 3.2. --- Expected results: Add Host should Pass, if we are allowing the user to add RHS2.0+ hosts to a 3.2 compatible cluster. If we don't allow RHS2.0+ hosts to a 3.2 cluster, please clarify on the same. Additional info: engine logs and vdsm logs are attached.
Created attachment 704125 [details] engine.log
Created attachment 704126 [details] vdsm.log
error is perfectly clear - vdsm in the RHS node isn't compatible with a 3.2 cluster. you need a more advanced vdsm version to be in a 3.2 cluster (4.10, etc.).
There is no feature upgrade of RHS-C planned in RHS2.0+, and hence it will continue to have a vdsm that supports 3.1 only.
This needs to be documented in our Quick Start Guide. Re-opening and re-assigning.
(In reply to comment #5) > There is no feature upgrade of RHS-C planned in RHS2.0+, and hence it will > continue to have a vdsm that supports 3.1 only. Note that when I say it will support 3.1 only, I mean it can be added to a cluster with compatibility level set to 3.1. So it should work with RHEV-M 3.2 also, as long as the correct compatibility level is set on the cluster.
As per comments above, closing NOT A BUG