Db = 10.2.0.5
OL5
Hi ALL.
We have been experiecing ORA-01555 when purging data in a large history table.
And this weekend, we will run again a large purge program which can not be done by chunks.
Please help how to monitor.
Cause: request failed due to ORA-01555: snapshot too old: rollback segment number 16 with name "_SYSSMU16_1232623801$" too small
Our UNDO tablespace is set to autoextend on, so space is not a probem.
Can I say that the problem is our UNDO_RETENTION parameter? How long can I set its value to? Can I set it to 48 hrs?
SQL> show parameter undo
NAME TYPE VALUE
------------------------------------ ----------- ------------------------------
undo_management string AUTO
undo_retention integer 2000 << is this value in seconds?
undo_tablespace string APPS_UNDOTS1
Can you share me script how to monitor if ORA-01555 is about to happen?
Do I need to create a new UNDO tablespace so that I can assign the PURGE program to it?
Please help...
Kind regards,
jc