I've encountered this issue when upgrading from 7.0.x to 7.2.x.
The solution that worked for me was to first determine captaincy within the SHC.
Cycle Splunk on the captain, which will force a new election.
One Splunk is again healthy on the previous captain, force a re-sync of the KV Store.
Validate via splunkd logs and/or DMC.
(always test in non-prod!)
I understood, but I have a standalone instance and not a search head cluster.
Probably this solution would not work for me but I'll try to resynchronize the kvstore.
Taking into account what you said, I did a clean up at the KV Store.
This command solved for me:
SPLUNK_BIN/splunk clean kvstore --local