"no backend available" and bad user

[Dev Dewboy]

Hello,
I have a 4 node Group Replication cluster Single Primary. We have 2 load balanced MySQL Routers. All on v8.0.34.

We have an application that establishes connections to the databases using MySQL Connector J 8.0.33.

About every over day, the application stop connecting to the primary database through one of the MySQL Router. There are no errors on the DB server logs.The MySQL Router has this set of errors in the error log:

2023-10-11 12:02:22 routing INFO [7fab556f8640] Stop accepting connections for routing routing:group_rw listening on 3306
2023-10-11 12:02:22 routing ERROR [7fab54ef7640] no backend available to connect to

On the Primary Database
select * from performance_schema.host_cache\G
*************************** 1. row ***************************
IP: 10.11.111.101
HOST: mr-one.dns.local
HOST_VALIDATED: YES
SUM_CONNECT_ERRORS: 1
COUNT_HANDSHAKE_ERRORS: 20
FIRST_SEEN: 2023-10-10 15:53:16
LAST_SEEN: 2023-10-11 12:02:14
FIRST_ERROR_SEEN: 2023-10-10 16:09:54
LAST_ERROR_SEEN: 2023-10-11 12:02:14

Lastly and most importantly I query the connection_control_failted_login_attempts table and it shows this:
select * from information_schema.CONNECTION_CONTROL_FAILED_LOGIN_ATTEMPTS;
| USERHOST | FAILED_ATTEMPTS |
|''@'issp-mr1.ia.local' | 20 |

The number of handshake errors matches up with this user's failed attempt. And when the SUM_CONNECT_ERRORS is one, I can no longer connect to the Primary Database through this MySQL Router. I have to FLUSH HOSTS; on the DB and restart the MySQL Router and it start accepting connections again.

This ''@'issp-mr1.ia.local' is causing the Router from connecting to the DB Server. SUM_CONNECT_ERRORS increments to 1 on the DB server and prevent future connections from the router to it.
How can this be sent - ''@'issp-mr1.ia.local'.
Faulty code, bug in DB or Router or MySQL Connector J? Is there any way I can mitigate or resolve?

Thanks,
DD