parallel recovery push change
A stuck nanosleep call impairs a parallel recovery process. The affected parallel recovery process spends 100% of its time in the undocumented parallel recovery push change wait event. Continue Reading
DTrace
DTrace
Redo Copy Latch Contention 2: MTTR Advisor
MTTR advisor data gathering can cause redo copy latch contention on servers with many CPUs. Checkpoint queue internals. An example of analyzing a compiled function in a proprietary code without disassembling it. Continue Reading
Tracing Latches: Redo Copy Contention
Tracing latches with OS tools, case study: redo copy contention caused by a logon storm Continue Reading
Tracing Library Cache Locks
Tracing Library Cache Locks with DTrace Continue Reading
Estimating DB Time With ASH: Logon Storm Pitfall
Logon storms may substantionally contribute to DB time which isn’t captured in active session history. Continue Reading
Troubleshooting java.lang.OutOfMemoryError in the Oracle Database
Internal processing of critical Java errors in the Oracle database Continue Reading
Warning: log write elapsed time
Oracle database internals – (redo) “log write elapsed time” instrumentation Continue Reading
IO Sort Cost Calculation
IO Sort Cost Calculation: formulae and internals Continue Reading
Enriching Trace with Call Stacks
A DTrace script for capturing the call stack when a line gets written into the trace file. It’s useful, for example, for finding out the functions participating in a particular query transformation. Continue Reading
Dead Transaction Rollback
A crashed process can leave a dead transaction behind. Generally, SMON is the one who’ll be rolling it back. Under certain circumstances, though, a server process might lend a helping hand. That’s exactly the scenario we’ll be focusing on. Continue Reading