This sequence of events can lead to the issue being encountered -
The following entries are observed in the /var/log/vmware/vcf/domainmanager/domainmanager.log file at the time of the failure -
YYYY-MM-DDThh:mm:ss.sss+0000 ERROR [vcf_dm,xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx,xxxx] [c.v.v.c.f.p.n.s.s.InventoryMgmtServiceImpl,dm-exec-17] Host with name xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx not found in inventoryYYYY-MM-DDThh:mm:ss.sss+0000 ERROR [vcf_dm,xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx,xxxx] [c.v.v.c.f.p.n.p.a.PrepareNsxtInfoFromInventoryAction,dm-exec-17] UNABLE_TO_GET_INPUT_FOR_NSXT_CONFIGURATIONjava.lang.NullPointerException: Cannot invoke "com.vmware.evo.sddc.common.core.error.ErrorCode.provideExtraContextAttributes()" because "errorCode" is null
VCF on VxRail 5.2.x
The host IDs in the SDDC Manager database are different from the host IDs present in the workflow execution context.
This is due to the previously failing task having been manually deleted and the initiate_stretch_cluster_vxrail python script having been incorrectly triggered a second time, rather than debugging and fixing the prior error and retrying the failing task from the SDDC Manager UI.
Please open a case with Broadcom Support for assistance.