Experience: is what you get soon after you need it.

Experience: is what you get soon after you need it.

Rasul Allah (sal Allahu alaihi wa sallam) said: "Restore the trusts of those who trust you, and deal not falsely with him who deals falsely with you." [Abu Dawud, Tirmidhi]

Search This Blog

Monday, July 18, 2011

11gR2 : Restarting DUPLICATE After a Failure

A hasle in 10g & 11gR1 is we cannot "re-start" the failed duplicated database command from the point where it failed. We have to start all over again.

Where as in 11gR2 as per the docs: Now we can ...need to test this scenario.

http://download.oracle.com/docs/cd/E11882_01/backup.112/e10642/rcmdupdb.htm#i1008564

Restarting DUPLICATE After a Failure

RMAN automatically optimizes a DUPLICATE command that is a repeat of a previously failed DUPLICATE command. The repeat DUPLICATE command notices which datafiles were successfully copied earlier and does not copy them again. This applies to all forms of duplication, whether they are backup-based (with or without a target connection) or active database duplication. The automatic optimization of the DUPLICATE command can be especially useful when a failure occurs during the duplication of very large databases.

If a DUPLICATE operation fails, you need only run the DUPLICATE comand again, using the same parameters contained in the original DUPLICATE command. The second DUPLICATE operation:

  • Locates the datafiles that were successfully duplicated by the initial DUPLICATE command.

  • Displays a message similar to the following for each datafile that it does not need to duplicate again:

    RMAN-05560: Using previous duplicated file /oradata/new/data01.f for datafile 1 with checkpoint SCN of 1654665 
  • Restores only the missing or incomplete datafiles, thereby avoiding re-copying and restoring all the datafiles.

Before you attempt to resume a failed DUPLICATE operation, you must reset the auxiliary instance to NOMOUNT mode. One way to do this is to exit RMAN, use SQL*Plus to reset the auxiliary instance to NOMOUNT mode, start RMAN and then repeat the DUPLICATE command. If you do not want RMAN to automatically recover from a failed DUPLICATE operation, specify the keyword NORESUME to disable the functionality. Using the keyword NORESUME in the first invocation of DUPLICATE prevents a subsequent DUPLICATE command for the new database from using this automatic optimization.

8 comments:

Walter said...

Hi Sameer. I too could not find any reference to this in the 10g documentation. However, you'll be glad to know that 10gR2 (10.2.0.5.0) RMAN DOES support a restart of a failed DUPLICATE command. While creating a standby, I ran out of disk space for the auxilliary db. I received ORA-19502 and ORA-27063 errors initially when 4 channels were trying to restore 4 different datafiles to the same LUN. We added more space and I simply re-ran the DUPLICATE command. RMAN skipped all of the previously restored datafiles, then focused on the incompletes, all without error. Before adding disk space, I wasted much time trying to restore a single datafile, intending to move the other 3 to LUNs with enough room. I was blocked at every turn. Just re-run the DUPLICATE command and you'll sleep well. Walter Nirenberg, Sr. DBA, Veterinary Pet Insurance, wnirenberg@petinsurance.com

me said...

i just experienced this for the first time after a bad path on the aux server.

it would not resume after re-running.

then i realized that my controlfile has been modified by the duplicate.

i replaced the modified controlfile with the minimal one for the duplicate, created the right directory, restarted in nomount, and it resumed successfully.

RAGS said...

Hi Sameer,


"i replaced the modified controlfile with the minimal one for the duplicate".

What does the above statement mean.Can you please elaborate on it.

As per my understanding, duplicate command copies the control file and restores it automatically.

RAGS said...

Hi Sameer,


"i replaced the modified controlfile with the minimal one for the duplicate".

What does the above statement mean.Can you please elaborate on it.

As per my understanding, duplicate command copies the control file and restores it automatically.

Mubarak mustafa ginawi said...

Assalamualykum Sameer
(OUR Production database size is 11 TB)VLDB, keep it in your mind.

we have standby database up an running and every thing was good, but suddenly the storage that we store the data is failed with 7 disks and we lost all the data and stby configuration.
we decide to start the configuration and creation for the stby from beginner,, and we used the 11.2.0.4 version but When we executing the active duplicate command from target database in RMAN ,it hanging for looong time and gave us in the alert log the ( Destination database instance is 'Started' not 'mounted').
please if u can help us reply me as soon as possible.
big thanks for your time.

Sandeep Amrutia said...

Thanks for the nice blog, its very useful feature that we can resume failed duplicate db restoration.

I am having 19 TB DB and using restart feature i have saved almost 20hrs.

Thanks,
Sandeep.

mubarak masud said...

There are no minor phases in preparation for MBA. Attention should be paid to all aspects. When writing an essay, be sure to show you know how to assess your abilities and opportunities, set goals and achieve them. resume for mba

محمد محود said...

Restarting DUPLICATE
---------------

شركة تنظيف بالرياض