wait event : cursor: pin S wait on X

You could see high such waits in your database if you are using automatic memory management of the SGA ( the new feature in Oracle 10G )
Solution : Set manually  the memory components of the SGA's database .

see also http://www.nocoug.org/Journal/NoCOUG_Journal_201008.pdf

how to collect statistics in exadata

In case you have an oracle exadata database machine, always use 100% percent for collecting optimizer statistics on your tables/schemas , in order to benefit from the smart scans .



asynch descriptor resize

High Numbers of 'asynch descriptor resize' Wait Events Seen [ID 1273748.1]

bad performance with ODI in AIX

Timeouts And Extremely Poor Performance On AIX 64 Bit Servers With JVM 1.6 [ID 968592.1]

error in datapump : -bash: syntax error near unexpected token `('

Usually this error appears when you try to export/import with datapump from command line in bash/ksh , and your command  has ( or ) inside .
example:  expdp directory=export__directory dumpfile=expdp_dumpfiles_%U.dmp schemas=(...,....,....) COMPRESSION=ALL parallel=4

Solution : use escape characters  \(  or \) 

example:  expdp directory=export__directory dumpfile=expdp_dumpfiles_%U.dmp schemas=\(...,....,....\) COMPRESSION=ALL parallel=4 



If your database does not start (stays forever during in mount state) , you should check for file $ORACLE_HOME/dbs/oranfstab .

This applies  in case you use DNFS ( direct nfs ) and you changed your volume names, mount points or ip servers .


how to align partitions in vmware


Our testing validates the assumption that VMware VMFS partitions that align to 64KB track
boundaries result in reduced latency and increased throughput. Partition alignment on both
physical machines and VMware VMFS partitions prevents performance I/O degradation due to
unaligned tracks

how to tune exadata

Oracle Sun Database Machine Performance Best Practices [ID 1067520.1]

smart scan in exadata

Smart scan : Find the statistics related to cell offload [ID 1438173.1]


cardinality feedback - a nasty feature in 11G

“I have a statement that runs relatively quickly the first time I run it (around 12 seconds). Subsequent executions always run much slower, usually around 20 or 30 minutes. If I flush the shared pool and run it again elapsed time is back to 12 seconds or so.”

 --> just do alter system set "_optimizer_use_feedback" = FALSE:

Sometimes, there are also other causes : histograms on the columns involved

See also Cardinality Feedback - Frequently Asked Questions [ID 1344937.1]


high "direct path read" waits in 11G


In 10g, serial full table scans for “large” tables used to always go through cache (by default). If table was small it was placed at most recently used (MRU) end of the buffer cache. If table is large it is placed at least recently used (LRU) end of the buffer cache.
In 11g, full table scan do not always go through the buffer cache. Decision to read via direct path or through cache is based on the size of the table, buffer cache and various other stats. Table is considered to be small or large based value of _small_table_threshold internal parameter.  Default value of this parameter is 2% of buffer cache size and is specified in blocks. This means any object (table) smaller than 2% of the buffer cache will be read via the buffer cache and not using direct path read. And tables larger than 2% of the buffer cache are read via direct path read and not via buffer cache. With AMM (Automatic Memory Management) or ASMM (Automatic Shared Memory Management), buffer cache could drop to a lower value if memory is falling short for shared pool. In such a case after restart of instance, _small_table_thresholdparameter would become even lower due to decreased buffer cache.


“Tirul te învață o lecție  fundamentală în viață: nu poți controla decît ceea ce depinde de tine. Dacă te consumi la infinit pentru ceea ce nu ține de puterile tale, atunci o iei razna”


Exadata X2-8 bug

ALERT - Exadata X2-8 systems affected by Linux bug 14258279: scheduling clock overflows in 208 days [ID 1473825.1]

datapump ORA-44002

Datapump Export Fails ORA-39006 ORA-39065 ORA-44002 [ID 435344.1]


high CPU and IO when using temporary tablespace group



high CPU usage in AIX 7.1




leap second in oracle linux

Leap Second Hang - CPU Can Be Seen at 100% [ID 1472421.1]


when to use oracle database smart flash cache

How To Size the Database Smart Flash Cache [ID 1317950.1]

oracle database smart flash cache




what can go wrong when installing oracle database RAC

TOP Note: Solutions for Typical 11gR2 Grid Infrastructure/RAC Database runInstaller Issues [ID 1056713.1]


oracle rac 29 june 2012 - problems with 1 second

                The kernel attempts to print a message when a leap second is inserted or removed. 
This can cause kernel versions prior to 2.6.29 to hang, due to a deadlock on xtime_lock.

See also the metalink note below .