RE: Black Change tracking (nice typo. I'm quite certain we're tracking blocks)

Since I have dataguard, whence shall I recover if I have a problem?  

With a local (at least) and an off campus dataguard image of your database, I would think that the fastest local recovery is from the files on the local dataguard image and that the fastest off campus recovery is switching (or failing) over.  

If you don’t have dataguard or a home built standby recovery system, then optimization of your RMAN backup or one of the other methods Tim Gorman mentioned is probably required, but it is a serious question whether or not you need that belt if you are already wearing standby recovery suspenders (aka braces). All the of thread I’ve seen so far is dedicated toward optimization of the backup. I prefer to plan for recovery.  

Three questions I ask at the start specifying the architecture for a given recovery strategy:

1. Is it safe enough for the failure modes to be handled that were specified by management?
2. Is recovery fast enough for the failure modes I need to handle?
3. Is the price reasonable for the recovery specified by management?

Answering question #2 usually requires understanding the slope of increasing outage windows for the services that require the database to be open. Short outages (unless you’re doing air traffic control or nuclear power or grid management) are often nearly free. The cost per minute of outage may rise dramatically as hours pass by. The insurance value of extra small systems that can be recovered quickly should be considered in consolidation exercises. Often, for example, a standalone security system that only depends on itself and communications with it may be justified at the highest levels of redundancy and nearly instantaneous recovery even in a site disaster. Management should have this clearly in mind when they supply the answers to #1 and #3.  

Hint: Be very certain to NOT embarrass management if your questions are the first time they think about any of this. Possibly start very quietly with the line operations manager. When the manager’s eyes open widely with grateful recognition might be a good point in time to mention something like Axxana to make certain your off campus dataguard image can be completed to the last commit and that you’d really like to have two local dataguard images (space be damned) with one of them lagged sufficiently to recover from any software or hack blunder.  

mwf  

From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Ram Raman Sent: Tuesday, March 02, 2021 7:20 PM
To: ORACLE-L
Subject: Black Change tracking  

List,  

I am trying to come up with a backup strategy for a multi TB database, say 18Tb. We are going to be backing it up on disks (over NFS). I am thinking of enabling BCT in the database so backups will be quicker and use less resources. This DB is going to have nightly loads of several million rows and also have Data guard. v19c on Linux.  

I have had issues with BCT in older versions of oracle. Is anyone using BCT in a similar environment (big DB with loads and DG) and things are running good?  

We are not looking into backing up the DB from the standby site as things may change in that area in future.    

Thanks,

Ram.  

--
   

--

http://www.freelists.org/webpage/oracle-l Received on Wed Mar 03 2021 - 17:45:14 CET