Hi,
We were discussing new RMAN features with our DBA team and touched the
topic of incrementally updated backups on an image copy (disk),
allowing for incremental-backups only, since the last incremental will
be merged back into the image copy (my own wording for this
mechanism.)
And this in combination with Block Change Tracking (BCT)
A theoretical drawback of BCT could be of that not all blocks are
checked for corruption during a level 0/1/2 backup, since in my
understanding, on every level, without BCT, the _whole_ database is
read by RMAN.
1 - is there indeed a risk of not being early warned for a bad block
when using BCT?
2 - what would be a wise mitigation of this risk?
Regards, Erik Ykema
Received on Wed Jul 18 2007 - 16:13:11 CDT
