Upgrading the VRS add-on appliance fails with the error
Operation Failed Failed to install update: Operation not permitted: The connection was closed by the remote end during handshake..
This may be due to the VRS deployment using the VRMS OVF files, instead of the addon OVF files, by mistake.
If the add-on VRS appliance were initially installed with the standard OVF files and not the addon OVF files the appliance can still work as a VRS appliance but may fail to upgrade.
From /opt/vmware/var/log/preupdatelog log we see
+ echo 'tdnf install --disablerepo='\''*'\'' -y '
+ rpm -e --nodeps hms-fileintegrity
error: package hms-fileintegrity is not installed
++ exec /opt/vmware/bin/ovfenv -k deployment_scenario
+ deployment_scenario=VRMS
+ '[' VRMS == VRMS ']'
+ checkDbAvailability
++ exec /opt/vmware/hms/bin/embedded-db.sh
++ grep offline
psql.bin: error: connection to server on socket "/tmp/.s.PGSQL.5432" failed: No such file or directory
Is the server running locally and accepting connections on that socket?
+ status=offline
+ [[ offline == \o\f\f\l\i\n\e ]]
+ echo 'Database is offline. Abort upgrade'
Database is offline. Abort upgrade
+ exit 1
from the ovfenv.xml file we see
<PropertySection>
<Property oe:key="deployment_scenario" oe:value="VRMS"/>
VRS appliance were initially installed with the standard OVF files and not the add-on OVF files
If there are VM's configured for replication using these add-on appliances then new appliances will need to be installed using the add-on OVF files as shown in https://docs.vmware.com/en/vSphere-Replication/9.0/administration-guide/GUID-DF5B80D3-5127-4BC0-B050-B9464F2D1120.html and the VM's migrated to the new add-on appliances before the original ones are unregistered form the VRMS and removed from inventory/deleted.