log_checkpoint_interval in oracle8i

[Vani]

Hi,
Here is one more head proning topic.
With respect Oracle 8i

If the redo log file size is 1MB,that is 2000 OS blocks; if the OS block size is 512 KB,then setting the parameter log_checkpoint_interval to 500 will initiate a checkpoint when the redo log file is 1500 blocks full.

Now my question:Is there any formula to caluculate it?.I tried most possible methods to get 1500 and Hmmmm,I could not succeed.

Help me ....

[marist89]

http://otn.oracle.com/docs/products/...nst.htm#434295

[marist89]

OH, and your OS block size is probably 512 bytes, not 512K.

[Vani]

I had gone through the site which you had forwarded already.

I want an explanation with example

Please explain in detail.In the Book I am studying it says 512 KB only

[tamilselvan]

If you set LOG_CHECKPOINT_INTERVAL = 500,
OS Block Size is 512 Bytes and
the redo log file size is 1MB, then for every 256,000 bytes written in the redo log file, checkpoint will occur.

Also note down, if Logical Volume Manager is configured, then LVM OS Block Size will override the OS Block Size.

There is no formula.
Just set LOG_CHECKPOINT_INTERVAL=1500 so that when even 3/4th of redo is filled, checkpoint will occur.



[Edited by tamilselvan on 03-14-2002 at 04:16 PM]

[Vani]

Ok After writing 256000 bytes to Redo logfile Check Point Occurs.
Hint : 500 * 512.
But in the Book he says setting parameter to 5oo will initiate a checkpoint when redolog file is 1500 blocks full.
How ?

[tamilselvan]

Which book?

Tamilselvan

[Vani]

Biju Thomas
I mean
Oracle 8i Architecture and administration and Backup and Recovry Study Guide
By
Doug Stuns
Biju Thomas

[tamilselvan]

It was a mistake if they wrote it. I think probably it could be a typo.

[Vani]

OK then I assume that after writing 256000 bytes to redologfile,checkpoint occurs

Thanx