Thanks for the question, sh.
Asked: August 11, 2022 - 1:34 pm UTC
Last updated: August 12, 2022 - 4:48 am UTC
Version: 12.1
Viewed 1000+ times
You Asked
I tried to duplicate source database to create physical standby database using the below command
duplicate target database for standby FROM ACTIVE DATABASE USING COMPRESSED BACKUPSET
SPFILE
nofilenamecheck;
I have a very big database size with multiple big data files.
After finish 70% of database duplication, i discovered one of the big data files will take very long time to be finished.
So i want to use section size to speed up the duplication using the below command
duplicate target database for standby FROM ACTIVE DATABASE SECTION SIZE 500M USING COMPRESSED BACKUPSET
SPFILE
nofilenamecheck;
My question:
============
if i cancel the running duplication script and re_run it again, i will lose all the progress done before.
- It will start duplication from scratch?
- The big file which reach to around 50% of duplication, it will resume or starting from scratch.
regards,
and Connor said...
RMAN duplicate is designed to be re-runnable simply be re-running the same duplicate command.
I am not 100% sure if you now *vary* the command whether that will still be viewed as a resumption or a restart.
But if you are going to section sizes, then I'm pretty sure the unit of work for resumption is a datafile, so that big datafile which is partially done will resume from scratch.
If you don't get a nice resumption, you could always follow MOS notes 1602916.1 and/or 360962.1 which describe manually completing a partially completed duplicate.
Is this answer out of date? If it is, please let us know via a Comment