Configure iptables in one of the master nodes, disconnect SB DB/ NB DB from its RAFT cluster, check from Web console Observer-Metrics page, there is one redundant entry for the disconnected database with server_status "cluster_member"

Client Version: 4.10.3
Server Version: 4.12.0-0.ci.test-2022-08-26-074740-ci-ln-qxdfjmt-latest
Kubernetes Version: v1.24.0-2390+a097e267c1ffde-dirty
Pre-merge PR: https://github.com/openshift/cluster-network-operator/pull/1526See 
commit titled: OVN-K alerts: add ovn db cluster member error

100%

1.Configure iptables in one of the master nodes, disconnect SB DB/ NB DB from its RAFT cluster
for example, SB DB:
% oc debug node/ip-10-0-140-107.us-west-2.compute.internal
sh-4.4# chroot /host 
sh-4.4# iptables -A INPUT -p tcp --dport 9644 -j DROP
sh-4.4# iptables -A OUTPUT -p tcp --dport 9644 -j DROP

2.Check metric ovn_db_cluster_server_status from Web console, Observe-Metrics page, please see the picture attached

there is one redundant entry for the disconnected database with server_status "cluster_member"

there is no redundant entry for the disconnected database, only one entry with server_status "disconnected from the cluster (election timeout)"