Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: LOG_CHECKPOINT_INTERVAL
I think there you are confused between LOG_CHECKPOINT_INTERVAL and
LOG_CHECKPOINT_TIMEOUT. you should set timeout as zero to disable it.
Actually this parameter defines time in seconds after which an automatic
checkpoint is taken. Setting it to zero disables this feature, therfore
there would not be any time based checkpoints. The log_checkpoint_interval
parameter defines in number of redo log buffers written after which an
automatic checkpoint will be taken, if you set it to a very small value,
very frequent checkpoints will occur causing a lot of activity. You should
set the value to higher than the size of your redo log file size to disable
automatic checkpointing.
i thought the oracle server tuning guide was very clear on these points.
hope this helps
Vishal
Andy Carruthers wrote:
> We're having a discussion regarding this parameter and what happens if
> it is set to 0.
>
> Oracle Support say that this is a __BAD__ thing to do but haven't given
> a substansive reason why except to say that it forces lots of activity.
>
> On a recent Oracle Performance Tuning (v7) course, my boss was told to
> set this to 0 in order to disable checkpoints (except when the log file
> changes over).
>
> The Oracle Performance Tuning manual (O'Reilly & Associates) implies
> that the only valid parameters are 2 and above.
>
> Any ideas?
>
> Thanks, Andy
Received on Thu Jun 18 1998 - 04:54:52 CDT