Re: Why pctincrease set to 1 for SYSTEM tablespace ?

You might like to have a look at my
web site for the article on how SMON
works - its getting a bit dated now as
it was written against 7.3 but in the area of coalescing etc. it is still accurate.

--

Jonathan Lewis
Yet another Oracle-related web site: http://www.jlcomp.demon.co.uk

Paul Bennett wrote in message <386BA706.A2430220_at_cc.gatech.edu>...
>If this just sets the default options for objects that are created in the
>tablespace, why would this effect SMON? Shouldn't SMON run on a tablespace
>if there was an object created with PCTINCREASE > 0 and that object asked
>for a next extent?
>
>Because, you could set DEFAULT PCTINCREASE 50, then create a bunch of
>objects, then set it back to 0.
>
>Or, an object could be asking for extents of different sizes for a month,
>then you could explicitly change the objects PCTINCREASE to 0. SMON still
>needs to run on this tablespace incase a bunch of rows are deleted and two
>of those extents are freed.
>
>What am I missing?
>
>-- Paul
>
>Sybrand Bakker wrote:
>
>> On Wed, 29 Dec 1999 19:28:46 +0100, "Matthieu Jobert"
>> <mjobert_at_freesurf.fr> wrote:
>>
>> >In the script generated by Oracle 8.1.5 NT database creation assistant,
>> >there is
>> >REM ********** ALTER SYSTEM TABLESPACE *********
>> >ALTER TABLESPACE SYSTEM
>> >DEFAULT STORAGE ( INITIAL 100K NEXT 100K MINEXTENTS 1 MAXEXTENTS 300
>> >PCTINCREASE 1);
>> >
>> >Why is PCTINCREASE set to 1 and not to 0 ???
>> >
>> >
>> >
>> Because this will allow SMON to coalesce the SYSTEM tablespace
>> periodically.
>>
>> Hth,
>>
>> Hth,
>>
>> Sybrand Bakker, Oracle DBA
>
Received on Thu Dec 30 1999 - 12:58:15 CST