Home
> Uncategorized > generate dumps – overview
generate dumps – overview
Some nice – important dump options 🙂
and still under constuction
find sga ID
in Unix environments with mutiple instances running within the same os-use (e.g. oracle) you cannot see which memory segment is allocated to which instance.
sqlplus / as sysdba oradebug setmypid oradebug ipc oradebug tracefile_name
the column Shmid shows the dependend memory structures
dump data block
alter system dump datafile <file#> block <block#> alter system dump datafile <file#> block min <block#> block max <block#>
dump file headers
oradebug setmypid oradebug unlimit oradebug dump file_hdrs 10
dump controlfile
oradebug setmypid oradebug unlimit oradebug dump controlf 10
dump library cache
oradebug setmypid oradebug unlimit oradebug dump library_cache <>
1 library cache stats
2 dumps hash table summary
4 library cache with basic info
8 objects with detailed info (incl. child reference, pin waiters…)
16 dump heap size
32 dump heap info
process state dump
alter session set events 'immediate trace name PROCESSSTATE level 10';
after ora-4020
alter session set events '4020 trace name PROCESSSTATE level 10';
shared server state dump
oradebug setmypid oradebug unlimit oradebug dump shared_server_state <level>
1 dump systemwide stats
2 dump queue
3 dump circuits on service queue
4-6 dump dispatcher
7-8 dump shared server
9 dump inactive dispatcher and server slots
10-13 dump info interesting circuits
14 dump info all circuits
systemstate dump
oradebug setmypid oradebug unlimit oradebug dump systemstate 10
Categories: Uncategorized
execution plan, optimizer, trace
Comments (0)
Trackbacks (0)
Leave a comment
Trackback