I found these logs in maxscale logs:
2023-11-27 12:19:48 error : [mariadbmon] The backend cluster does not support failover/switchover due to the following reason(s):
Slave connection 'cdscp' from server7 to [10.13.134.38]:3306 is not using gtid-replication.
2023-11-27 12:19:48 notice : [mariadbmon] Disabling automatic cluster operations for 5 monitor ticks.
2023-11-27 12:19:48 notice : Server changed state: server8[
10.13.134.47:3306]: lost_slave. [Slave, Running] -> [Running]: Host: [10.13.134.46]:3306,IO Running: Yes, SQL Running: No, SQL Error: