stackpack

• Get all snapshot

select * from STATS$snapshot;

• Check database instance

select * from STATS$DATABASE_INSTANCE; – in case database restarted, an error will be thrown as ORA-20200: The instance was shutdown between snapshots 981 and 982

select * from v$instance;

• Run stackpack

sqlplus perfstat/password@“(DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=server)(PORT=1521))(CONNECT_DATA=(SERVICE_NAME=xxrac)))”

@$ORACLE_HOME/rdbms/admin/spreport.sql

OR

@?/rdbms/admin/spreport.sql

• Purge stackpack

@$ORACLE_HOME/rdbms/admin/sppurge.sql

OR

@?/rdbms/admin/sppurge.sql

• Oracle jobs

select * from user_jobs;
SELECT job, next_date, next_sec, failures, broken, SUBSTR(what,1,40) DESCRIPTION FROM dba_jobs;
EXEC dbms_job.broken(24, false); EXEC dbms_job.run(24, false); – reset job when failure 16 attemps to run

• Oracle job for schedule run sppurge

---Check the snap present before 20 days----

select count(*) from stats$snapshot where snap_time < sysdate-20

----Create the procedure which will purge statspack snapshot after 20 days gap---------
create or replace procedure statspackpurge is
var_lo_snap number;
var_hi_snap number;
var_db_id number;
var_instance_no number;
noofsnapshot number;
n_count number ;
begin

n_count := 0;

select count(*) into n_count from stats$snapshot where snap_time < sysdate-20;

if n_count > 0 then

select min(s.snap_id) , max(s.snap_id),max(di.dbid),max(di.instance_number) into  var_lo_snap, var_hi_snap,var_db_id,var_instance_no
  from stats$snapshot s
     , stats$database_instance di
 where s.dbid              = di.dbid
   and s.instance_number   = di.instance_number
   and di.startup_time     = s.startup_time
   and s.snap_time < sysdate-20;

  noofsnapshot := statspack.purge( i_begin_snap      => var_lo_snap
                                      , i_end_snap        => var_hi_snap
                                      , i_snap_range      => true
                                      , i_extended_purge  => false
                                      , i_dbid            => var_db_id
                                      , i_instance_number => var_instance_no);

 dbms_output.Put_line('snapshot deleted'||to_char(noofsnapshot));

end if;
end;
/

----check the job already exits in the database------------------------
SELECT  job, next_date, next_sec, failures, broken, SUBSTR(what,1,40) DESCRIPTION FROM dba_jobs;


------scheduled the  job at mid night----------------------
declare
  my_job number;
begin
  dbms_job.submit(job => my_job,
    what => 'statspackpurge;',
    next_date => trunc(sysdate)+1,
    interval => 'trunc(sysdate)+1');
end;
/

--------------  Another way----------------------------------------------------------

variable v_jobno number;
begin
dbms_job.submit(:v_jobno,'statspack.purge(i_num_days=>15,i_extended_purge=>TRUE);',trunc(sysdate)+1+2/24,'SYSDATE',TRUE);
commit;
end;
/

• Reference

Stackpack

http://www.oracle-base.com/articles/8i/statspack-8i.php

http://docs.oracle.com/cd/B10500_01/server.920/a96533/statspac.htm

Oracle job

http://www.orafaq.com/wiki/DBMS_JOB

http://www.dba-oracle.com/tips_oracle_statspack_purge_utility.htm

http://myoracleworld.hobby-electronics.net/DB-statspack.html

http://dbaworks-sunny.blogspot.hk/2012/12/statspack-purge-script-for-scheduling.html

http://docs.oracle.com/cd/B19306_01/appdev.102/b14258/d_job.htm#i1000769