The database is hanging randomly (with some leniency I would say almost each hour). Top session is:
select privilege#
from sysauth$
where (grantee#=:1 or grantee#=1) and privilege#>0
or
select /*+ connect_by_filtering */ privilege#, level
from sysauth$ connect by grantee#=prior privilege# and privilege#>0
start with grantee#=:1 and privilege#>0
In ash I saw these top wait events:high concurrency on event library cache: mutex X and latch: shared pool
This problem occurred after upgrading from 11.2.0.1 to 11.2.0.3.
These queries are almost always in top sessions but the wait event when the database is not hanging is CPU. In 11.2.0.1 they didn't appear. We are using DRCP.