After upgrading to VMware Cloud Foundation (VCF) Operations 9.1, users may experience failures during inventory synchronization for components managed by Fleet Lifecycle.
A primary symptom is a significant discrepancy between the license cores assigned and the actual usage reported in the Operations Manager (e.g., a discrepancy of thousands of cores). This issue occurs because the database node table becomes populated with duplicate entries during the upgrade process, which blocks the inventory sync required to finalize the upgrade.
The issue is caused by a software defect in the sddc-build-service pagination logic within the getAllComponentNodes() API. When the total component count exceeds 20, the service incorrectly resets the loop index to 0. This results in the repeated fetching and insertion of the first 20 nodes into the database, creating duplicates.
Resolution To resolve this issue, contact Broadcom support.
Workaround There is no viable manual workaround for this issue other than the application of the engineering hotpatch.
We are aware of this issue and expect a resolution to be released in 9.1.1ep1