Following the draft for migration guide from template based OCP3.x installation to OCP4.x operator based deployment outlined in THREESCALE-2929 THREESCALE-2935 the failure of restore of zync-database appears when executing:

{{
oc rsh $(oc get pods -l 'deploymentConfig=zync-database' -o json | jq -r '.items[0].metadata.name') bash -c 'psql -f ${HOME}/zync-database-backup'
}}

The failure is due to open connections to DB, here is message printed at the very beginning of restore process:

{{
psql:/var/lib/pgsql/zync-database-backup:14: ERROR:  database "zync_production" is being accessed by other users
DETAIL:  There are 8 other sessions using the database.
}}

Result: DB is not populated with new data (many more errors are printed)

Both upstream & customer documentation seems to have missing instructions to stop the services using this db

Dev notes

Shutdown zync-* pods before doing any database dump/restore