Errors like this are seen in the postgres-0 logs on the primary node, /services-logs/prelude/postgres-0/file-logs/postgres.log
####-##-## ##:##:##.### UTC [#######] STATEMENT: update ebs_instance set leader=true where service_id=$1
####-##-## ##:##:##.### UTC [#######] ERROR: duplicate key value violates unique constraint "ebs_instance_leader_idx"
####-##-## ##:##:##.### UTC [#######] DETAIL: Key (leader)=(t) already exists.
VMware Aria Automation 8.x
This error is benign and can be ignored.
Aria Automation's Event Broker Service has one leader node that is allowed to update Event Trace Entries in a scheduled job.
Every node tries to become a leader every minute, but if there is a leader already, the attempt fails with this exception. This doesn't obstruct the functioning of EBS.