Hello all
We have inherited a couple of new databases which have been configured
with the default "Oracle suggested backup strategy"
The script is as below and this runs every day
RMAN redundancy is set to 1
BACKUP OPTIMIZATION is OFF
BLOCK CHANGE TRACKING has not been enabled
run {
allocate channel oem_disk_backup device type disk;
recover copy of database with tag 'ORA\$OEM_LEVEL_0';
backup incremental level 1 copies=1 for recover of copy with tag 'ORA\$OEM_LEVEL_0' database;
}
For all our existing databases, we are using below backup strategy
FULL LEVEL 0 on Sat night and Tuesday nights
INCR LEVEL 1 on Sunday,Monday,Wednesday,Thursday and Friday
We are looking for some help in understanding how this "Oracle suggested backup strategy" works and if this is an appropriate
strategy for any OLTP 24 x 7 production database.
Questions:
1) If we have a 400 GB production database, and the database has been up and running for 30 days, and this same script is running every day, what is happening in the background?
2) Day 1, there will be full backup of all the data files and then from day 2nd thru day 30th, changed blocks would be backed up? so the backup from Day 1 would be the biggest and rest of the 29th backups much smaller?
3) If there is the need for recovery on day 31st, would this mean the base (full backup from day 1) backup would be used first and then all
the changes from there and onwards would be applied using the backups from day 2nd thru 30th?
4) If there is any data corruption in one of the data files lets say on day 23rd, what would this mean in
We are trying to understand the pros and cons of this backup strategy .. since this is totally new to us and wondering if we should
replace this with our existing backup strategy
5) What if there is a data corruption in one of the data files on day 15th? What would happen then?
We also found this excerpt from one of the articles online
"The Oracle Suggested Backup (portrayed in DB Console) may be suitable for a starter database in a test environment, but it is a good example of what is not appropriate for any production environment. It makes one baseline backup of your database the first day, and then forever after makes incrementally updated backups. There are two problems with this: First, there is only one actual backup in one place. Second, incrementally updated backups simply take changed blocks from the database and substitute them for the original blocks in the very first backup, but the baseline backup is never repeated. Eventually, one can see the potential for bad blocks to creep in and never be detected (until, of course, the day you try to recover from your one and only backup!)."
We would really appreciate your guidance and help in trying to understand this if we should use this oracle suggested backup strategy or change this to our more tested and tried back up strategy using a combination of full plus incremental plus archive log backups.
Thanks in advance for your assistance