Re: V$SYSTEM_WAIT_CLASS

On Mon, 13 Aug 2007 21:16:17 -0700, dean <deanbrown3d_at_yahoo.com> wrote:

>I've read to the contrary several times. Consider what would happen if
>Oracle thought it saved a change to disk, and then there was a power
>failure before the cache was written to disk? It would then be in a
>corrupt state.

Consider there is redolog, which is always written *before* the data files are written.
Consider these redologs are multiplexed and on mirrored drives. So how likely is the fairy tale you describe above. Consider you are in dire need to revise your policy of not reading any manual.

-- 
Sybrand Bakker
Senior Oracle DBA