During an SSP Installer upgrade, the Workload Cluster creation or upgrade phase fails with the following symptoms:
Machine Status: New Nodes/Machines are stuck in the Provisioning state and do not transition to Provisioned or Running.
Timeout: The upgrade process eventually times out (typically after 3600 seconds).
Controller Logs: The CAPV controller show the following reconciliation error:
E0311 ... controller.go:347] "Reconciler error" err="failed to reconcile VM: no compatible datastores found for storage policy: [STORAGE_POLICY_NAME]"
vDefend Security Services Platform 5.1
The issue is caused by a configuration mismatch between the SSP Installer's Management parameters and the actual vCenter environment.
Even if the Storage Policy exists and is healthy, the CAPV cannot find a valid placement because:
Named Datastore Mismatch: The platform configuration specifies a hardcoded Datastore name (or Content Library backing datastore) that has been renamed, deleted, or does not exist in the target vSphere Cluster.
Scoping Issues: The automation is looking for a specific datastore that is not "visible" or "mapped" to the ESXi hosts in the designated Compute Resource/Resource Pool.
Filtered Compatibility: While the Storage Policy is technically compatible with some datastores, the specific datastore defined in the "vCenter Parameters" of the platform is not among the compatible list.
Log into the vSphere Client.
Navigate to Policies and Profiles > VM Storage Policies.
Select the policy mentioned in the error logs and check the Storage Compatibility tab.
Note the exact names of the datastores listed as "Compatible."
In the SSP-I UI, under Instance Management, locate the vCenter Parameters.
Compare the configured "VM Datastore" or "Content Library" name against the names found in Step 1.
Confirm if the configured name is missing from the vCenter inventory or has been renamed.
Edit the Connection/Parameters: Update the platform configuration to point to a valid, existing, and compatible Datastore.
Retry Provisioning: * If the upgrade is paused, resume or retry the task.
The Controller should now successfully reconcile the VSphereVM object, find the compatible storage, and trigger the VM power-on sequence.