How to Resume a Failed RMAN Duplicate Due to Missing Backups or ArchiveLogs

It is relatively easy to resume the rman restore and recovery process if it has failed due to missing backups or archivelogs.  You most often see the following error messages complaining that something is missing. RMAN-00571: ===========================================================
RMAN-03002: failure of Duplicate Db command at 05/24/2010 12:52:44
RMAN-03015: error occurred in stored script Memory Script
RMAN-06053: unable to perform media recovery because of missing log
RMAN-06025: no backup of log thread 1 seq 6253 lowscn 1663211060 found to restore

You may have forgotten to copy them over to the same local disk location and/or  failed to restore due to tape issues. After you are sure that you have everything you need in the location RMAN expects them,  shutdown the database that is being duplicated (this is assumed that the controlfile has been restored and the database mounted), start it backup in nomount mode and just rerun the same commands.
RMAN automatically (tested using Oracle RDBMS Enterprise Edition 10.2.0.4) runs through the commands and continues the restore and recovery process from where it left off.

For example: There was a gap in the archive logs in the flash recovery area. I have an alternate archive destination that contains the missing log.

After the first duplicate command failed, I used RMAN to catalog the existence of the  missing log:

CATALOG ARCHIVELOG '/u01/arc/ORACLE_SID/1_6253_DBID.arc';

see MOS Document: How To Catalog Backups / Archivelogs / Datafile Copies / Controlfile Copies [ID 470463.1]  Then shutdown, startup nomount the database being duplicated. Reran the exact same duplicate command:

archive log thread 1 sequence 6251 is already on disk as file /backup/flash_recovery_area/ORACLE_SID/archivelog/2010_05_22/o1_mf_1_6251_5zk173bs_.arc
archive log thread 1 sequence 6252 is already on disk as file /backup/flash_recovery_area/ORACLE_SID/archivelog/2010_05_22/o1_mf_1_6252_5zkhfw99_.arc
archive log thread 1 sequence 6253 is already on disk as file /u01/arc/ORACLE_SID/1_6253_DBID.arc
archive log thread 1 sequence 6254 is already on disk as file /backup/flash_recovery_area/ORACLE_SID/archivelog/2010_05_23/o1_mf_1_6254_5zl68htc_.arc

As you can see from the above output…that it applies the archive logs including the single cataloged piece in the alternate location.  This may save someone else time and trouble when they have the same issue.

Advertisements

About April C Sims

Oracle DBA for over a decade...enough said.
This entry was posted in Recovery, RMAN and tagged , , , , . Bookmark the permalink.

4 Responses to How to Resume a Failed RMAN Duplicate Due to Missing Backups or ArchiveLogs

  1. สาโรช says:

    good !!

  2. Rich Bernat says:

    Hi April,

    This has been out for quite some time (I’ve used this note since 2006 fo resume/restart a DUPLICATE from various failure points. It is especially useful when duplicating a multi TB system only to fail for something silly at the OS.. or a tape went offsite unexpectedly.

    Manual Completion of a Failed RMAN Duplicate [ID 360962.1]

    Regards,
    Rich

  3. Lynn Sattler says:

    I gave this a try and in the process shutdown my production database. Yes production came down issuing an rman shutdown command from the remote host attempting an rman duplicate command. (we run the rman duplicate commands on the remote host, “to clone machine”.

    I had to use this kind of a catalog command because I was trying to catalog an rman backup of archive logs, not regular archive log files.

    CATALOG backuppiece ‘/usr/local/oracle/……arch2012.09.13.xx.yy.zz.dbname_3961_1″;
    these worked fine, but then I shot myself in the foot doing the shutdown.

    However when I went back into rman to move on after the accidental shutdown, rman did attempt to finish the recovery quickly. (I believe it failed because I was mistakenly trying to clone databases at different versions “11.2.0.2 vs 11.2.0.3”)

    If I had it to do over again, I would have done the shutdown with sqlplus and restarted the rman session all over again with all the identical commands issued the first time. If I had the right matching versions, I believe the duplicate would have finished right and very quickly.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s