Unlike in a non-CDB, ADDM does not report the following issues in a PDB, because these issues apply to a CDB as a whole and do not apply to an individual PDB:
• I/O problems due to:
– undersized buffer cache
– undersized streams pool
– excessive temporary writes
– excessive checkpoint writes
– excessive undo writes
– excessive PQ checkpoint writes
– excessive truncate writes
– excessive tablespace DDL checkpoint
– I/O capacity limit
• SQL hard parsing issues due to:
– cursor aging
– out-of-memory failed parse
• SGA sizing issues
ADDM also does not report the following issues in a PDB, because these issues cannot be resolved at a PDB level:
• Cluster messaging related issues, such as network latency, congestion, contention, and
lost blocks
• Log file switch waits on archiving and on checkpoint incomplete
• Too many free-buffer waits
• Contention on log buffer waits
• Waits due to CPU bottleneck
• Operating system VM paging
• Session slot wait event
• CPU quantum wait event
• RMAN related wait events, such as PQ queued wait event, PGA limit wait event,
and I/O queue wait event
