Re: log_checkpoint_interval

Hi,

You should first know whether you need to have check point to occur at half of redo log or full.

The value depends of the OS block size.

Divide the redolog file size by the OS block size.

If the redlog size is 500K then if the os size is 512 bytes the 250K/521 ~ 500.

If you want the check poin to occur at half the redo log file then make the value

LOG_CECK_POINT_INTERVAl = 250 ( unit is in bytes).

then you will have check point twice per redolog file (if it is 250K size).

You should shutdown and start up the db after the parameter change.

take care

vimal

alt_at_interchange.co.uk wrote in article <7705mk$pkb$1_at_nnrp1.dejanews.com>...
> Hi
>
> Oracle 7.3.2.3.2
>
> Decunix
>
>
> Can someone tell me if the value for log_checkpoint_interval is set in
bytes
> or multiples of the blocksize (as with DB_BLOCK_BUFFERS) I can't find a
> reference to the value size in the manual other than it should be larger
than
> the size of the redo log files.
>
> Thanks
>
> Al
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
 

>
Received on Wed Jan 06 1999 - 11:56:43 CST