Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Reorganizing a database that has raw partions
Hi,
I recently inhereted a database (Oracle Server 7.3.2) that was intially
set up with the data files on raw AIX 4.2 partitions. We are now at the
point where we must reorganize the tablespaces to distribute I/O load. I
am not familiar with raw partitions and would like some advice on how
remove the data from these raw devices and place them back in journaled
file systems. During this time we will also be restructering the
tablespaces for performence reasons. (We realize the error of our ways
now......)
I am currenlty in the middle of loading a "marketing" database. The
database is quite large when fully populated. Right now it is partially
populated.
Our plan is to export all populated tables drop all others. reomove the
current instance and abandon it. Recover the disks used by the raw
devices and with a new instance rebuild the database correctly with the
proper I/O load balance by creating Volume Groups for each datafile
contained in a particualr tablespace. Does this sound feasable or is
there a better way?
Issues we are dealing with are:
Identifying what pysical partitions are associated to what raw device on
what pysical volume (disk). Anybody have any shell scripts that help
with this...? I have been manually poking my through but it is going
slow...
There are three instances on the same IBM R40 using RAW partitions for ALL of the data files.
I need to get one of them redone in a week....
We are thinking of upgrading to Oracle 8 shortly after? or During? the restructuring. I would assume after would be the logical choice.
Any input would be greatly appreciated...
Brad Odland Received on Wed Nov 26 1997 - 00:00:00 CST