"TNS:operation timed out" on primary replicated database
Hi
We have two 11.2.0.3 databases (Oracle patch 18) each on a sepearte WIndows 2008 R2 servers, coonected using Advanced MultiMaster Replciation.
Intermitantly we are finding that the Push job from Secondary to Primary is broken. If we unbreak, it breaks immediatly again. We can succesfully ping and tnsping in both directions. When we try "select count(*) from dual@primary_db_link" on secondary we get a connection lost error.
This is the case until the entire Primary server is rebooted when this issue may not occur again for serval days.
We have completely cleared the SQLNET parameters in sqlnet.ora but the issue still occurs.