Re: Help with SQL Trace

The database version is: 9.2.0.5. Per below, the query returned one row with no where constraint, therefore it can only have one row. I will try to get the total size of the table using extent information. It will take a little time since it is a remote database that I do not have direct access to.

Assuming the table is only one block, is this possible? Could some strange code such as locking a table for read consistency for the duration of processing or something goofy like that cause this output?

On 2/8/06, Larry Lutz <larry.j.lutz_at_gmail.com> wrote:
>
> The SQL below is simple since dual is a one row table.
> The plan is as exected. However, I would expect the
> logical and physical blocks read to be 2 at the most,
> and the elapsed time to be less than a second. Twenty
>
> minutes is obscure. Any one know how or if this can be
> possible? It came straight out of a production system:
>
> select to_char(sysdate,'YYYY/MM/DD HH24:MI:SS')
> Process_End_Time from dual
>
> call count cpu elapsed disk query current rows
>
> ------- ------ -------- ---------- ----------
> ---------- ---------- ----------
>
> Parse 1 0.00 0.00 0 0 0 0
> Execute 2 1271.80 1318.39 475439 21600279 2563509 1
> Fetch 2 0.00 0.00 0 3 0 1
> ------- ------ -------- ---------- ----------
>
> ---------- ---------- ----------
>
> total 5 1271.80 1318.39 475439 21600282 2563509 2
>
> Misses in library cache during parse: 0
> Optimizer goal: CHOOSE
> Parsing user id: 111 ()
>
> Rows Execution Plan
>
> -------
> ---------------------------------------------------
> 0 SELECT STATEMENT GOAL: CHOOSE
> 0 TABLE ACCESS (FULL) OF 'DUAL'
>
>

--
http://www.freelists.org/webpage/oracle-l