one Table generating gc traffic upon single active instance cluster

Hello,

We are running an Oracle RAC database across two nodes with only one active instance i.e. second node is in secondary mode. We are using RAC for availability and resilience only and not scalability and every two weeks the primary role is passed from one node to the other. This has been running without any issues for the last 4 years. However over the last few days there have been some recorded instances of selects upon one table 'replication_queue_mes_b' giving poor response times. So far these incidents of poor response time have occurred when the primary instance was running upon