issue with finding very low library cache gethitratio
have an oracle 10G database (i know we should have upgraded by now but this may apply to all versions......)
re nagios monitoring checks we are using for v$librarycache
find that after 3 weeks of restarting a database getting a very low ratio value of gethitratio (less than 10%) but healthy value for pinhitratio (98%)
total of pins is 200 times gets
the applications run packages opening many cursors within cursors etc.
added more memory recently and all of this extra seemed to be added to the buffer_cache not the shared_pool (we are using automatic SGA management)
presumably are are getting issues with locks rather than pins