What prevents split brain from happening in 11.2 oracle dataguard with fast-start failover
edited Mar 28, 2013 2:46PM in High Availability Data Guard, Sharding and Global Data Services (MOSC) 4 commentsAnswered ✓
Can someone explain if split-brain can happen on a failover..
example the observer -- in 1 location; primary in another location; standby in another location;
the mpls to the primary is lost between observer & standy;
then this causes the observer to failover to the standby --- makes the standby -- the primary....
meanwhile the original primary is still working and users updating the original primary-- because they are local to the original primary ...
users in other parts (outside the local) connect & are now working on the standby (which became the primary after the failover)....
CAN this scenario happen? If not, what prevents it from happening?
0