At the end of reconfigure process, vRA is synchronizing machine information with VM record in vRA database, the value becomes negative in the database if bigger than 2048GB.
As a side-effect of this, deleting a VM with a ‘negative disk size’ disk fails. The Dispose VM step keeps retrying and generating errors: "Value was either too large or too small for a UInt64" until inventory data collection fixes the issue.
Resolution
This is a known issue affecting VMware vRealize Automation versions 7.2 and 7.3. This issue is resolved in VMware vRealize Automation 7.4, available at VMware Downloads.
Workaround:
To work around this issue, manually run an inventory data collection against the compute resource affected.
Select Infrastructure > Endpoints > Endpoints.
Point to the endpoint for which you want to run data collection and click Data Collection.
Click Start.
(Optional) Click Refresh to receive an updated message about the status of the data collection you initiated.